Research completed: July 2024

Non-technical summary

This is a non-technical summary to a report published separately as GB wholesale electricity market reform: impacts and opportunities for Scotland. The reader is invited to refer to the full report for detail.

Context

This study assesses the impact that the introduction of locational marginal pricing (LMP) to the Great Britain (GB) wholesale electricity market would have for Scotland, as well as the impact of potential alternatives. LMP has been proposed as a potential reform in the UK Government’s Review of Electricity Market Arrangements (REMA) consultation, which aims to reform electricity markets to enable a net zero energy system. LMP would be a significant reform and is of particular interest to Scotland, as the country is likely to be affected differently to other parts of GB.

We conducted a literature review and assessment of LMP and its alternatives between September 2023 and January 2024. It is an independent review and is not the view of the Scottish Government. This included a detailed assessment of quantitative and qualitative literature, as well as input from an expert advisory panel. The panel was invited to attend two 2-hour discussions, commented on, and reviewed interim findings. It consisted of stakeholders across government, energy research centres, renewables developers, flexibility providers, industry and business representatives, energy suppliers, large consumers of electricity in Scotland, a community energy group, and a consumer protection and advocacy body. Its views have been considered and included in the development of this review. This is the non-technical summary, with a detailed report published separately.

Locational marginal pricing

The wholesale electricity market is where electricity is bought and sold before it is delivered to consumers. Its main participants are electricity generators and suppliers. The current wholesale market is national and marginal. This means that electricity can be bought and sold anywhere in GB at a single national price^[1], regardless of the physical constraints, or bottlenecks, on the transmission network. An example of this can be found at the B6 boundary that separates the transmission network between Scotland and England. Constraints arising here limit power flow (typically southward). Generators and consumers are not directly incentivised by the wholesale market to place and operate physical assets that generate or consume electricity in a way that is efficient for their specific location on the electricity network.

Electricity is traded in advance based on a predicted amount of electricity demand. The amount of electricity generated in real-time is adjusted by the electricity system operator (National Grid ESO) to meet the actual, rather than the predicted demand. The cost incurred by National Grid ESO is then passed on to consumers through their electricity bills. When traded generation is expected to exceed the maximum power flow of the network (creating a constraint), additional trades need to be made by National Grid ESO in affected areas to change the expected operating schedules of generators or consumers. With network build not keeping up with the growth in renewables, this inefficiency is accelerating and contributing to higher electricity bills for consumers (National Grid ESO, 2022a).

LMP could help reduce this inefficiency by splitting the national market into smaller geographic areas called zones or nodes (see Figure 1). This creates smaller markets that reflect the supply and demand in an area, and the constraints of the network. Areas where the supply is higher than demand will see prices fall, and areas with higher demand will see prices rise. This could incentivise generation and demand to locate where they do not exacerbate constraints. However, it is necessary to consider the wider, non-price factors that also influence decisions by generators and consumers on where to locate. These include the availability and quality of renewable resources (e.g. wind speed or seabed space), supply chains, skills, planning and consenting.

Additionally, the daily variation in price within these locational markets would reflect the instantaneous state of the local network. The result of this would be to create better signals that indicate how to operate flexible assets such as battery storage, international interconnectors, demand side response and dispatchable low-carbon generation (such as hydrogen or biomass) more efficiently. This helps to balance generation and demand and reduce constraints on the network. This further reduces operating costs for National Grid ESO, which are passed directly to consumers.

LMP could, however, make investment in renewable electricity generation less attractive in certain areas of the UK. Without appropriate investment support, it would place additional risks on market participants and create market uncertainty due to the radical nature of the reform. This could have positive impacts on investment in new sources of flexibility (such as storage), but negative impacts on renewables ambitions, particularly in Scotland. Policies could be put in place to mitigate these risks. The impact of LMP on renewable energy development in Scotland will be highly sensitive to whether such policies are implemented effectively.

Objectives of the Scottish Government

The Scottish Government outlined key ambitions in the Draft Energy Strategy and Just Transition Plan (ESJTP 2023), amongst other strategy papers. This review was completed before the publication of the final Energy Strategy and Just Transition Plan in 2024.

The review aims to discuss how LMP could impact the Scottish Government in achieving these ambitions. They have been summarised using four broad categories most relevant to wholesale market reform:

• Support ambitions to scale up low-cost renewable energy.
• 8-11GW of offshore wind by 2030 (ambitions from draft ESJTP).
• 20GW of onshore wind by 2030 (ambitions from draft ESJTP).
• Adhere to the principles of a fair and just transition.
• Deliver affordable energy that isn’t subject to global fossil fuel price volatility.
• Enable community participation.
• Incentivise wider economic benefit including jobs, skills, supply chains and investment.
• Support accelerated decarbonisation of heat, transport and industry, including through carbon capture and hydrogen.
• Decarbonise heat and transport using renewable electricity/hydrogen.
• Scale hydrogen generation and develop carbon capture in Scotland.
• Enable a secure and flexible net zero energy system, which is not dependent on fossil fuels.
• Enable energy security through the development of own resources and energy storage.
• Invest in grid infrastructure at pace to allow for a net zero transition.

Key outcomes for wholesale market reform

Wholesale market reform will have widespread impacts on Scotland’s energy strategy, as well as wider social and economic implications. By reviewing Scottish Government strategy papers and assessing where wholesale market reform has significant impact, the authors have developed key outcomes that need to be prioritised for electricity market reform to align with Scotland’s ambitions:

• Strategic coordination of renewable development and network investment is required to ensure that renewables stay in Scotland and net zero is achieved.
• Local price signals are necessary to encourage investment in and optimise the use of flexible assets, such as batteries, and enable an efficient use of renewables.
• Mechanisms that allow electricity users to benefit from low-cost renewable generation are required.
• Benefits and costs of a green transition need to be shared fairly to consumers, communities, and businesses.

Findings

In this section we present the key findings on how LMP and its alternatives could impact Scotland’s energy transition ambitions. This is split into four broad categories:

• The scale up of low-cost renewable energy.
• The fair and just transition.
• The decarbonisation of heat, transport, and industry.
• Enabling a secure and flexible net zero energy system.

Scale up of low-cost renewable energy

LMP would create regional differences in wholesale prices across GB, which depend on local levels of generation and demand. Areas such as the south of England, where demand is higher than supply, would likely see wholesale prices rise. Areas with an oversupply of renewable generation, such as Scotland, would see wholesale prices fall. The primary purpose of LMP is to create a market that is more reflective of the cost of delivering electricity to specific locations on the grid. In doing so, this encourages the placement of generation and demand where it is most suitable and cost-effective for the energy system. The wholesale price signal seen by renewables developers in Scotland could disincentivise investment, as market revenues would decline. Modelling by Aurora (2023) and FTI Consulting (2023) suggest a general southern shift in solar generation, away from Scotland. Changes in the buildout of on- and off-shore wind are more contested due other non-price factors such as the effective on-shore wind ban in England, as well as limited off-shore site availability due to leasing rounds from the Crown Estate. Certain market arrangements could be developed to help shield generators from excessively low local wholesale prices, however this would somewhat diminish the benefit of LMP.

Additionally, LMP introduces a change to the rights of access participants have to the market. Currently, electricity generators can sell electricity on the wholesale market regardless of transmission network constraints. They have firm access rights to the market. Under LMP, generators lose their firm access to the network. As a result, they can only sell their electricity within their zone/node or when it can be transmitted to consumers. This introduces a significant risk for generators in Scotland, as there are times when more electricity is produced from wind in Scotland than can be transmitted to domestic and commercial consumers within Scotland and to the rest of the UK. National Grid has proposed to significantly upgrade the network to 2035, however some excess flows from Scotland are likely to persist even after the new transmission is built.

The new risks created by LMP, combined with additional implementation uncertainty (as a result of reforming wholesale market arrangements), could lead to increases in the cost of capital. The cost of capital reflects the cost of money (e.g. interest on debt) required to finance projects. It represents the return required for an investment to be worthwhile and increases with project risk. As renewables require major upfront investment, the cost of capital has a significant impact on investment levels and the final cost of electricity for consumers. Overall, modelling completed by Aurora (2023), FTI Consulting (2023) and AFRY (2023) shows that small increases in the cost of capital caused by introducing LMP could wipe out any benefits linked to cost savings resulting from LMP.

UK decarbonisation relies on significant renewables capacity in Scotland. As such, the introduction of LMP alone would risk Scottish renewables deployment and therefore GB decarbonisation ambitions. To mitigate this, a possible solution is to reform the renewables support scheme, referred to as Contracts for Difference, to reduce risk in low carbon electricity generation development. This solution must be explored further for possible options and feasibility. Alternatively, improved Transmission Use of System Charges (TNUoS) could provide similar locational investment signals to LMP. These charges are paid by generators and suppliers to recover the cost of installing and maintaining the transmission network. However, reformed TNUoS would lack the operational incentives for flexible assets that LMP could provide.

Fair and just transition

If LMP benefits are realised, the total cost of running the electricity system should decrease moderately as a more efficient electricity system is developed. If these benefits are not offset by increases in the cost of capital for renewables, the modelled annual net economic benefit to the cost of the electricity system lies between £0.2bn-1.6bn (AFRY, 2023; Aurora, 2023).

Due to significant existing renewables capacity, LMP could see Scottish consumers benefitting from wholesale electricity prices lower than current prices as well as prices in other regions of GB. This benefit would reduce over time, though according to one study, Scottish prices would remain as some of the lowest in Europe (FTI Consulting, 2023). As transmission network is reinforced to 2035 and more electricity generation facilities are built closer to where they are needed, prices across GB will converge.

However, initially prices would rise in some areas in GB, although not as much as they would decrease in Scotland (FTI Consulting, 2023). It is possible that some consumer groups, e.g. domestic customers, would be shielded from wholesale prices through arrangements with their electricity retail companies, or UK Government policy design. Additionally, energy suppliers may not pass savings directly to customers, as their costs may rise in other regions. As wholesale electricity prices only constitute a proportion of the domestic electricity bill, with other components including network charges and green levies, the impact of LMP on overall domestic electricity bills will depend on the proportion of the bill that wholesale prices make up at any given time.

Overall, the benefits are more likely to be seen by commercial and industrial consumers in Scotland, who are less likely to be shielded from wholesale prices. The extent to which these benefits are realised depends on when LMP is implemented. The modelling shows the earlier it is implemented, the greater the benefit, as networks are reinforced and become less constrained to 2035 and beyond. However, National Grid ESO suggests LMP will take at least four to eight years for implementation (National Grid ESO, 2022a), limiting the benefits that can be attained.

The development of employment opportunities and other wider economic benefits due to accelerated renewables development is a significant benefit for Scotland. To ensure this, continued development of renewables is necessary through supporting policy. LMP also provides a significant economic opportunity through investment in new demand and industrial sectors. Lower electricity prices could attract investment in sectors such as green hydrogen, data centres or green steel – though none of the reviewed studies directly model this. The Fraser of Allander Institute study (FAI, 2023) shows that the renewable energy sector already supported more than 42,000 jobs across the Scottish economy and generated over £10.1 billion of output in 2021. With decarbonisation seeing the decline of the Scottish oil & gas industry, renewable energy and new demand sectors could provide significant employment opportunities and economic growth.

Decarbonisation of heat, transport and industry

Overall, the modelling in reports published by Aurora (2023) and FTI Consulting (2023) suggests that even if LMP is implemented successfully, it would not significantly affect the pace of decarbonisation of the electricity system. In fact, implementation of LMP without appropriate accompanying mitigations could risk UK decarbonisation efforts through a hiatus in renewable generation investment. The main benefit of LMP is that it could reduce the cost of decarbonisation, especially in Scotland, where the price of electricity could decrease the most.

The electrification of heat and transport is a significant aspect of decarbonisation. Lower wholesale costs under LMP in Scotland can contribute to heat pump and electric vehicle (EV) uptake. This is more likely for heat pumps, as electricity cost is a larger proportion of the total lifetime cost compared to EVs. Analysis by the authors indicates that a 35% reduction in wholesale cost in Scotland would reduce the total cost of ownership of an EV (in years 1-4) by 2%, and 10% for heat pumps. Both still have significant upfront costs that would need to be addressed.

LMP could make the development of green hydrogen more attractive in Scotland. Aurora’s modelling (2023) suggests hydrogen produced in Northern Scotland could have some of the lowest costs in Europe. This is because electricity is one of the main cost components of hydrogen electrolysis. This could generate a hydrogen export economy that could also benefit the decarbonisation of other industrial processes.

Carbon capture on the other hand is not likely to benefit from LMP. The implementation of carbon capture is linked to identifying industrial sites with good transport and carbon storage opportunities.

Enable a secure and flexible net zero energy system

LMP incentivises the optimal location and operation of flexible assets. Flexible assets can shift the consumption or generation of electricity in time or location. The significant capacity of renewable generation in Scotland means that prices in the wholesale market would show significant variation. This would attract investment of flexible assets in Scotland, as operators can access higher revenues. A system with a large proportion of renewable generation requires greater capacity of flexible assets. Such assets relieve network constraints and reduce the overall requirement for generation capacity and network build. Both Aurora (2023) and FTI (2023) show a significant increase in the capacity of battery storage in Scotland due to the implementation of LMP.

Under LMP, the operation of flexible assets is more efficient. A national wholesale market sends the same price signal to all flexible assets, anywhere in the country, regardless of local constraints. This would be improved under LMP, as flexible assets would respond to wholesale price variation, which would reflect local grid requirements. A particular benefit seen is the improved use of interconnectors to other countries. Overall, this enables a cheaper, more secure power system.

Local constraint markets (LCMs) could provide alternative locational signals for flexibility in this respect. LCMs are new electricity markets designed around network constraints. They provide incentives for operators to change their generation/consumption schedules, so that limits on the network are not exceeded. LCMs could, to an extent, replicate LMP market signals for flexibility. However, they would likely create additional barriers and be more complex by creating multiple markets and signals for flexibility to respond to.

Conclusions

The conclusions are based on the authors full independent assessment of the opportunities and threats that LMP and wider electricity market reform could have on the Scottish Government’s ambitions. Based on the findings of this study, the Scottish Government should support the development of a GB-wide strategic plan for renewables and network investment. The Scottish Government should fully explore the implementation of LMP with accompanying reformed support for renewable generation, specifically Contracts for Difference, to ensure continued investment in Scotland.

On the basis of this assessment, the following conclusions are presented in order of importance.

• Scotland must prioritise and coordinate a strategic plan for renewable generation and network reinforcement with the UK Government.

Without support mechanisms for renewables that shield energy generators from LMP, there would be additional risks that disincentivise renewables development in Scotland. Delays to transmission network reinforcement would exacerbate this. Long-term locational signals to strategically locate investment is essential to achieve a low-cost net zero power system. LMP, alongside support mechanisms for renewables, could provide these signals and continue to enable renewables development in Scotland. It is essential that mechanisms such as reformed Contracts for Difference are tested for feasibility before implementation. Alternatively, improved Transmission Network Use of System charges could provide the market with similar signals that indicate the best locations to invest, although this will not improve dispatch signals in the way LMP would.

• LMP would provide the clearest dispatch signal for flexibility, delivering efficient investment in and operation of flexibility.

Maximising the use of renewables can only be done with significant electricity system flexibility. LMP can provide effective investment signals for its development in Scotland and improve operational signals to optimise its use. This would reduce whole system investment requirements in generation capacity and network, reducing bills for consumers. LCMs could be an alternative in this regard, however, could also result in more complex markets and are unlikely to fully replicate the benefits created by LMP.

• The potential benefits of LMP for consumers are greater the earlier it is introduced.

LMP would create the most significant benefit for Scottish consumers before the transmission network is reinforced to 2035, and therefore, would need to be implemented quickly to maximise benefits. The extent to which this can be achieved is limited, as National Grid assumes implementation may take 4-8 years. A well-developed plan to implement LMP is required that accounts for the creation of support mechanisms which protect renewable generation, ensuring benefits are realised.

• Careful implementation of LMP is required to address regional differences in price.

Scottish consumers benefitting from lower wholesale prices would be a clear winner of LMP. However, this is not evenly spread across the rest of GB and must be considered.

Scotland has a clear opportunity to benefit from a net zero power system by making the most of low-cost renewable energy and distributing those benefits to consumers. Proposed changes to wholesale electricity markets could improve system-wide efficiency and offer cheaper electricity in Scotland. However, it could increase risk associated with investment in Scottish renewables, increasing costs. The Scottish Government needs to engage carefully with the electricity market reform process to ensure that prospective benefits are realised, and that potential disbenefits are avoided or mitigated.

© The University of Edinburgh, 2024
Prepared by Environmental Resources Management Ltd. on behalf of ClimateXChange, The University of Edinburgh. All rights reserved.

While every effort is made to ensure the information in this report is accurate, no legal responsibility is accepted for any errors, omissions or misleading statements. The views expressed represent those of the author(s), and do not necessarily represent those of the host institutions or funders.

If you require the report in an alternative format such as a Word document, please contact info@climatexchange.org.uk or 0131 651 4783.

1. The national price for all generators is set by the most expensive generation selling power on the wholesale market in the period (marginal). ↑

   
   

Research completed in April 2024

DOI: http://dx.doi.org/10.7488/era/4747

Executive summary

To deliver climate change mitigation and adaptation, nature restoration and high quality food production, the Scottish Government produced their vision for agriculture, along with the next steps, to encourage sustainable and regenerative farming in Scotland. A programme of work is underway to reform agricultural payments with a greater emphasis placed on delivering environmental outcomes with a proposed structure of four payment tiers tied to a suite of potential measures that will deliver tangible outcomes.

This study identified the most suitable metrics that could be used to monitor the success of the proposed measures in the agricultural reform programme against environmental outcomes. This includes consideration of cost-effectiveness, practicalities and the skills and capabilities of those tasked with monitoring.

Findings

We found potential metrics for assessing the success of the measures for all outcomes. Most metrics can already be applied as the methods are available, whilst a small number are under development and could be applied in the near to medium term. These metrics fell into several categories:

• Emissions cannot be measured directly, so we suggest using current farm-level tools to assess GHG emissions, known as carbon audits. A field level, real time GHG emission model is in development as well as a tool for doing this for ammonia.
• Many metrics depend on direct sampling of soil or biodiversity and can’t be realistically replaced by proxies or existing data. However, well designed sampling programmes can maximise the efficiency of sampling, e.g. sampling for soil carbon, nutrients, pH and eDNA can be done at the same time.
• The outcomes associated with animal health, nutrition and breeding must be largely monitored through proxy metrics. These are relatively easy to measure and provide useful information directly to the land manager.
• A few metrics, such as pesticide usage data or area under permanent habitat, collected as part of the agricultural census, can be derived from existing data.
• Some of the metrics in development could take advantage of samples/data collected at the start of any monitoring programme (e.g. soil eDNA, acoustic monitoring) and others would come online later (e.g. LIDAR-derived hedge data).
• The measure ‘retain traditional cattle’ could not be related to the outcomes.
• Deciding on a suitable suite of metrics to assess the benefits of the Agriculture Reform Programme is only one step as there are issues related to design, sample size and data to be considered.

Recommendations

A full list of suitable metrics for each measure from the Agricultural Reform list of measures is supplied in an accompanying spreadsheet “MeasuresXMetrics.xlsx”. The spreadsheet can be filtered to look at what metrics are suitable for each measure, which outcome they relate to, whether the metric is suitable for direct assessment, if it provides additional useful information or if the metric is still in development, whether the metric is suitable against multiple outcomes and who can carry out the monitoring.

Table 1 summarises the spreadsheet by showing which metrics relate to each outcome. The final choice of which metrics to collect will depend on two main factors:

• The availability of resources to carry out any monitoring programme
• The sampling philosophy adopted; whether widespread collection of a few metrics, where data collection could be partly done by land managers, versus a programme designed to give accurate data at the national level by sampling intensely from a representative sample of locations with mainly expert-led sampling.

Combining information on who can do the monitoring and potential likely costs of expert-led monitoring, we suggest the following monitoring philosophy is appropriate:

• All enterprises to assess soil erosion and buffer strip effectiveness.
• All livestock enterprises to record growth rate, milk yields, mortality, conception rates, replacement rates, age at slaughter for sheep and cattle.
• ScotEID to require information on sires.
• All enterprises to use farm tool calculators (carbon audits) to model GHG emissions. Livestock enterprises to model ammonia emissions when a suitable tool is available. The requirement to model might be limited to enterprises above a certain size to reduce costs.
• The remaining outcomes would be best assessed using expert-led monitoring in a sample-based programme similar in philosophy to the Welsh approach. The resources available for monitoring and statistical power analysis would be key inputs into developing a sampling approach with decisions about the trade-off between number of metrics recorded versus sample size needing to be made.

Table 1. Metrics identified as worthy of adoption in future monitoring, listed by outcome – a full list of which metrics are suitable to assess each measure are shown in the spreadsheet. Metrics are divided into three categories: Suggested metric – a suitable metric for monitoring the relevant outcome(s) that can be applied now; Additional metric – a useful set of additional information or approaches; and Metric in development – analytical methods are still in development, but samples/data can be collected and archived for future analysis. Metrics suitable for use for multiple outcomes are shown in bold.

Glossary / Abbreviations table

Introduction

This report examines the potential metrics for assessing the environmental outcomes of measures identified in the Scottish Agricultural Reform Programme.

Policy environment

Agriculture is a major contributor to Scottish greenhouse gas (GHG) emissions; currently, it is responsible for c. 20 % of countrywide emissions (Brodie 2023). Agricultural management has also been a major driver of the declines in above- and belowground biodiversity (Walton et al. 2023) and puts significant pressure on Scottish water bodies, preventing them from reaching Good Ecological Status (Environmental Standards Scotland 2022).

Following on from the Scottish Government’s Vision for Agriculture, a new Agriculture and Rural Communities (Scotland) Bill has been passed, which will allow for a new framework for future support payments for farmers (“farmer” is used in this report to cover both farmers and crofters), including for environmental goods. This will encourage sustainable and regenerative farming practices that will help Scotland transition towards net zero, reverse the decline in biodiversity, and improve soil health and water quality.

It is anticipated that there will be a new framework for agricultural payments focused on key outcomes of high-quality food production; climate mitigation and adaptation; nature restoration; and wider rural development alongside a just transition. Greater conditionality will be key, with a transition towards shifting 50% of direct payments to climate action and funding for on-farm nature restoration and enhancement by 2025.

At present a draft list of measures (Appendix A) is being appraised by Scottish Government that covers both land-based and animal-based actions that should lead to improvements in biodiversity, climate, flooding, soil health, water quality and animal health and welfare. However, a system of monitoring and verification is needed to ensure compliance and that the measures are delivering the desired outcomes.

Aims

The aims of this project were:

• To identify potential metrics that could be used to monitor the success of the proposed measures in delivering the desired environmental outcomes (Appendix B). Those metrics that could be used in practice will have to be cost-effective, practical and within the skills and capabilities of those tasked with the monitoring.
• To take an overview across all the metrics and outcomes to refine the list of metrics to avoid duplication and maximise the usefulness of information collected.

Considerations for selecting appropriate metrics

Introduction

To determine whether any changes over time are the result of direct action through applied measures, it is important to be able to compare areas where measures have been applied with other similar areas that are not in the scheme (control sites). Without this, it is not possible to determine whether any change detected is due to the measures or to other drivers.

It is also possible that even if an improvement is not detected on sites where measures have been applied, the measures might mean that a negative change, that would otherwise have occurred, has been avoided.

A Before-After-Control-Impact (BACI) design is commonly used for monitoring the effect of environmental interventions. However, a difficulty is that areas which are originally selected as controls may join the scheme later. Also, as pointed out by Emmett et al. (2014), it can be difficult to select appropriate controls given the numerous other factors, including field contents, size, and boundary characteristics that would need to be held constant across matched pairs. Even if the areas selected as controls are not part of the current scheme, they may not be true controls as they may have benefitted from similar environmental measures under legacy schemes.

As a result of these issues, it can be difficult and costly to assess outcomes at the level of individual farms, though overall performance of measures can be assessed through an appropriate monitoring scheme.

Requirements

Effective monitoring requires an appropriate baseline for measuring outcomes against (Pakeman et al. 2020). A proper baseline gives power to any analysis, as it is detecting change against known values for indicators. For example, agricultural soil monitoring as part of scheme monitoring will need to align with the national soil monitoring programme that is in development.

Similarly, identifying an appropriate sampling design is critical. It needs to cover enterprises in different situations and localities and have the appropriate statistical power to give good evidence on the performance of each measure in at least the medium-term (i.e., to inform revisions to agricultural support schemes). Some outcomes may be detectable quickly, but others, like soil carbon, may take longer to be detectable within realistic sampling regimes (Saby et al. 2008). For other measures it may be difficult to separate the effects of the scheme from market-driven effects, such as the breeding of livestock for reduced methane production, which could be driven by the price of carbon rather than the support from any scheme (Cottle & Conington 2012, 2013).

Selecting metrics

The selection of metrics depends on several factors, including the design of any monitoring scheme, what is being monitored, for whom and for what purpose, and needs to take account of the trade-offs associated with the approach taken. These can be seen as different aspects of taking either a “broad and shallow” or a “narrow and deep” approach to data gathering for the same amount of effort. Data gathered from a “narrow and shallow” approach will be less detailed and likely less robust, whereas a “broad and deep” approach may be too costly to deploy widely.

Sample or population

Taking a sample of the population and focussing monitoring has the benefit of concentrating resources if it is understood that any sampling design has some measure of uncertainty built in. This type of approach has been adopted in monitoring programmes such as Countryside Survey (e.g., Carey et al. 2008) and the monitoring of the Welsh agri-environment scheme Environment and Rural Affairs Monitoring and Modelling Programme (ERAMMP), which focusses monitoring on 300 1 km x 1 km grid squares and assesses the impact of the scheme using information on how much land in each square is under Glastir funded management (see Section 8.1.1). The approach allows for efficient linkage between changes in different outcomes, but with the proviso that there is uncertainty and that it can only give a national-level picture.

Citizen scientist or specialist

For agriculture, options will include asking the farmer or land manager to gather information, drawing data from wider datasets, or drawing in specialists to sample and process data. There are advantages and disadvantages to asking land managers, as opposed to specialists, to carry out the monitoring. Land managers differ from citizen scientists in other monitoring, e.g., the British Trust for Ornithology’s Breeding Bird Survey, which is undertaken by volunteers with a high degree of skill at bird recognition. Expectations would have to be tempered in terms of what can be provided.

Consequently, the advantage of monitoring by the land manager is that it is effectively free, it can be repeated frequently and provides information direct to the land manager. This must be viewed against the benefits of sampling with more accuracy and precision by specialists.

It may be possible to develop hybrid monitoring strategies using the advantages of the different groups, either using land managers to take samples (e.g., soils), which are then sent away for analysis, or deploying monitoring equipment, with the specialists undertaking data analysis. Specialist data analysis is preferable from the point of view of scientific robustness, although monitoring equipment does need expert maintenance, calibration and quality control and is more costly. Alternatively, a tiered approach to monitoring could be followed, with land managers collecting some data whilst more specialised data collection is undertaken on a sample of farms.

Meaningful scales of monitoring

The appropriate scale of monitoring is inherent in what is being monitored. For plants, relatively small areas (a few square metres) tend to be monitored, whilst for butterflies and bees, the area might be a transect 100 m long and 5 m wide, and for birds, the British Trust for Ornithology uses 1 km x 1 km grid squares as the basis of their Breeding Bird Survey.

In consequence, the scale of monitoring for different aspects of the environment and biodiversity will not be the same for all outcomes. There is, therefore, some constraint on the overall approach as it is dependent on finding the most appropriate scale for each outcome.

Who is the monitoring for?

The vision for agriculture includes provision for payments that deliver to defined outcomes. If the aim is to inform management at the farm-scale or smaller, in effect using the results of monitoring in adaptive management, then there may be a benefit to a broad and shallow approach. There is also value in aligning monitoring with appropriate advice and resources for decision making. However, if the monitoring is just aimed at showing which measures are value for money, then a national level focus is more appropriate.

Understanding what is driving change

If measured changes can be linked directly to the impact of targeted funding, or with conditions for an agri-environment scheme, then this is a direct demonstration of the efficacy of the scheme.

However, a narrower set of more detailed monitoring may be better placed to understand more precisely what is driving change as a greater range of measured parameters can be used to examine the processes that lead to change. This improved knowledge might be more useful in developing future schemes and inform adaptive management. A tiered approach to monitoring may deliver the best information.

Can you monitor outcomes, or just activity?

It is possible that suitable methods to measure outcomes at the desired scale are not available or practical. Consequently, it may be that measuring actions or activity remain the only option to assess whether management is driving change in the desired direction. However, there would need to be some form of outcome monitoring at a wider scale to assess overall performance of the scheme.

Does land manager-led monitoring need supervision?

This is a contentious issue, but in other spheres such as sampling for water industry and fish farm compliance there are quality assurance assessments of ‘operator collected data’. Some are targeted based on evidence of some kind, but there is a random element to create pressure to conform.

There is a need to consider whether an inspection system is required to ensure there is pressure to maintain high standards of monitoring. Northern Ireland has decided that the best way to obtain robust data for monitoring is to employ people to do the measurement and use techniques such as GPS monitoring to check sample collection protocols are being followed (https://www.afbini.gov.uk/articles/soil-nutrient-health-scheme).

Methodology

We used an expert led rapid evidence assessment to look for different ways of assessing the success of each measure against environmental outcomes. This involved a multistep approach to developing appropriate metric recommendations to monitor the environmental outcomes of the new agricultural support system.

Step 1

For the land-based proposed measures only, we assessed each proposed measure (Appendix B) to identify which of the outcomes it was relevant to. For example, there are nine outcomes listed for In Field – Cultivated Soils, but not all outcomes are relevant for each measure. For example, the outcomes Reducing Soil Greenhouse Gas (GHG) emissions and Increasing soil carbon/organic matter content are unlikely to be affected by Efficient/Reduced use of synthetic pesticides so it would not be useful to monitor those if this was the sole measure in place.

This step was undertaken by individuals with expertise in each outcome.

Step 2

For each combination of relevant outcomes and measures, we used expert knowledge and a search of relevant literature to identify potential metrics that could be employed to assess compliance and/or the success of the measure in reaching the desired outcome (Appendix C). These were categorised in the following ways:

• Compliance or outcome-based
• Already collected under the current payment scheme, by agencies or third parties, or if novel data metrics will be required
• Practical for field-level monitoring, holding-based monitoring or for national-scale monitoring only, or unsuitable for routine monitoring.

This step was undertaken with the expertise of the research team backed up by literature searches. However, for the land-based measures, one individual was tasked with identifying appropriate metrics across all measures relevant to a particular outcome to ensure a consistency of approach. In contrast, the livestock-based measures are more holistic and required an expert to consider the actions around these in the round to identify appropriate metrics.

Step 3

The assessment in Step 2 generated a large list of metrics with associated methods that could be employed to assess the success of the scheme. A series of three workshops was used to consolidate these to ensure that where possible the same method can be employed across as many measures as possible for simplicity and to help in scaling up from individual measures to the success of the whole scheme. This stage delivered a shortlist of metrics that could be used to assess the success of the measures in delivering the desired outcomes, i.e., cost-effective, practical and within the skills and capabilities of those tasked with implementing the metric(s).

Step 4

This step focussed on identifying data collection approaches for consideration, as well as considering requirements for establishing an initial baseline and for future data collection to assess both compliance/activity and outcomes. Data collected could be integrated into existing data sets, such as the National Soil Inventory of Scotland, to give a longer perspective of change.

Potential metrics for each outcome

The outputs from Steps 1 and 2 are presented in Appendices B and C but are summarised below. Step 3 identified a set of metrics that could be employed in monitoring outcomes. This section identifies those metrics that would provide practical and cost-effective information. Potential metrics are categorised into three levels:

• Suitable metric – a suitable and available metric for monitoring the relevant outcome(s).
• Additional metric – a useful set of additional information or approach.
• Metric in development – analytical methods are still in development, but samples/data can be collected for future analysis.

Reducing soil greenhouse gas (GHG) emissions

The outcome

Greenhouse gas emissions from agriculture are a significant part of the national total. Reducing these emissions is a key goal of the Agricultural Reform Programme and the Climate Change Plan.

Considerations with a metric

Current methods required for direct measurement of GHG fluxes are not suitable for wide-scale use as they are dependent on relatively expensive equipment and a high degree of specialist knowledge to run the equipment.

We suggest that instead of this a modelling approach, based on existing or in development farm/field GHG calculators, is used that would estimate CO₂, N₂O and CH₄ emissions. These are also known as Carbon Audits and are currently funded as part of the Preparing for Sustainable Farming initiative. However, several issues would need considering:

• There are several modelling tools on the market (see section “Reducing Soil Greenhouse Gas (GHG) emissions” in the Appendix), so an updated review (see Leinonen et al. 2019) of their capabilities would be needed to ensure that only suitable products were used, and to ensure consistency of outputs.
• Assistance may be needed, and hence need paying for, in setting up the calculators in the first instance, as in the Carbon Audits in the Preparing for Sustainable Farming initiative.
• Outputs from the calculator depend on the quality of the primary data gathered, which means data quality checks may be a requirement.
• Feed and forage quality might be useful information to feed into the calculators – see section below on Animal health and nutrition.

Land managers will benefit from these whole farm or field-level calculators with the potential to identify cost reductions or increases in productivity through improved forage and manure management. This could be supported by the soil organic matter and nutrient data collected.

Suggested metrics

Suitable metric: Modelled farm emissions of CH₄, CO₂, N₂O

Metric in development: Modelled gas fluxes in real time at the field scale.

Increasing soil carbon/organic matter content

The outcome

Increasing the levels of soil carbon through regenerative agriculture can make agricultural land a sink for carbon and facilitate the journey to net zero.

Issues with a metric

Soil organic carbon can be routinely measured. There are different laboratory methods available, all of which work well, but a standardised approach would need to be selected for any scheme. Dry combustion (Dumas method) is widespread in its application and thought of as the best chemical method for soil carbon determination (Chatterjee et al. 2009). In addition, some consideration needs to be given to dealing with soil samples from calcareous soils where inorganic carbon levels are high (mainly carbonates), which though rare do include soils like machair soils. Additionally, by linking soil carbon to clay content (measured when characterising soil texture) a measure of the land parcel’s status regarding storing carbon is produced. Thresholds of 13:1, 10:1, and 8:1 clay to soil organic carbon could potentially be applied to arable, arable ley, and woodland systems (Prout et al., 2022).

Laboratory measurement is straightforward, but to calculate stocks, there also needs to be a measurement of soil bulk density (total dry mass per unit volume). Consideration of sampling depth(s) is important as some changes, such as a switch to deeper rooting crops may increase subsoil carbon, while changes in soil tillage might affect the vertical distribution of soil carbon. A standardised sampling protocol needs to take this into account. The approach being taken in Northern Ireland is informative. Every farm and every field are being sampled for carbon and nutrients and soil testing is a precondition of eligibility for environmental payments. Soil carbon stocks are large and are heterogeneously distributed, meaning that quantifying changes over short time periods is seldom possible. For instance, the proposal for a directive on Soil Monitoring and Resilience (Soil Monitoring Law) will require samples to be taken every five years. However, to ensure agronomic management changes will deliver and to identify which ones deliver, actions such as the employment of minimum tillage, use of winter cover crops, inputs of organic wastes and increases in permanent vegetation cover (woodland, hedges, grassland) need to be recorded at the field level alongside actions that will reduce soil carbon such as the removal of permanent vegetation cover and ploughing of grasslands.

Further considerations in developing this sampling include:

• Sampling to be carried out by land manager or by experts. There is a trade-off between cost and reliability but given the range of other soil metrics that need to be sampled to assess other outcomes, we suggest that soil sampling is expert led.

• Should samples from the same field be bulked to reduce costs or should they be analysed separately (expensive) to provide measures of error/heterogeneity and the possibility to statistically assess change at the field level rather than at the farm or national level? For instance, the Soil Nutrient Health Scheme in Northern Ireland analyses a bulked sample of 25 cores but this can miss coldspots and hotspots of nutrients (Hayes et al. 2023). The Welsh Soil Project splits each field into three before the W-shaped sampling is done. There is a direct trade-off between the number of fields that can be sampled and the number of samples per field. We suggest that the most useful information comes from sampling as many fields as possible, so a bulked sample per field would be an appropriate sample to measure. Some within field stratification could be done if there was a clear internal boundary, e.g., between dry slope and wetter flat ground.

• Collecting additional information such as the current and past management and cropping at field level would enhance interpretation.

• Several companies already operate soil testing services. In a competitive market, there is a question regarding how consistency is guaranteed and whether a consistency check should be carried out by a third party. United Kingdom Accreditation Service (UKAS) accreditation would be a minimum standard for participating laboratories.

• Sampling of enclosed land with a single habitat per field is straightforward. However, consideration needs to be given on how to sample from unenclosed land which may contain multiple habitats and a wide range of soil types.

Suggested metrics

• Suitable metric: Soil carbon stock, Area under permanent vegetation or other carbon positive management
• Additional metrics: Soil clay content
• Metric in development: Indicators based on soil FTIR spectroscopy.

Increasing resilience to weather events

The outcome

Soils are vulnerable to runoff and erosion after heavy rain and to drought. Improving the resilience of soils will safeguard their continuing productivity, reduce their susceptibility to the runoff of water and nutrients, and subsequent downstream impacts on flooding and water quality.

Issues with a metric

Resilience is a synthetic metric and can be best seen as a multi-dimensional concept. In addition, the thresholds for resilience will depend on soil type. Regarding improving soil resilience, mineral soils that have greater soil carbon concentrations tend to retain water and have better soil structure, allowing water flow through them rather than across them. Soils that show water percolating (high permeability) rather than flow across the surface are at lesser risk of runoff and erosion, whereas compacted soils with lesser porosity and greater bulk densities are much more vulnerable to weather events. Compacted soils also restrict water availability and nutrient dynamics impacting crop growth. The presence of water stable aggregates also helps prevent water and wind breaking down the soil and hence lower the risk of erosion. These indicators are covered elsewhere in this report (see sections Increasing soil carbon/organic matter content, Improving soil nutrient content and Reducing diffuse pollution) and hence not covered here in detail.

Suggested metrics

• Suitable metric: Soil carbon stock, Water stable aggregates, Soil bulk density and porosity, Erosion monitoring
• Additional metric: Visual Evaluation of Soil Structure (VESS)
• Metric in development: Indicators based on soil FTIR spectroscopy

Improving soil nutrient content

The outcome

Maintaining soil nutrient supply to ensure high levels of productivity is important for efficient farming. However, an oversupply of nutrients can lead to losses as emissions of ammonia and nitrous oxide, or as increased nutrient loadings of freshwaters. While Scotland has no widespread and high impact nutrient issues such as Lough Neagh in Northern Ireland, there are localised issues that have been identified through designations such as Nitrate Vulnerable Zones that might be more cost effective/appropriate to measure.

Issues with a metric

The total concentrations of the various soil nutrients are relatively straightforward to sample and analyse and could be combined with sampling for soil carbon. Analysis methods depend on whether a restricted set of macro-nutrients is the focus, or whether micronutrients and heavy metals are also of interest.

Total nutrient levels work well for some nutrients, but there may be an interest in looking at available nutrients where there is an extraction/exchange step to assess what is available to plants and leaching processes. There are standard laboratory methods for this, particularly for nutrients such as potassium and calcium, but phosphate extraction methods have been developed to be specific for different soil acidity levels (pH).

Unfortunately, neither total nutrient levels nor extractable/exchangeable levels work well for nitrogen, as nitrate is very quickly absorbed by roots, leached, or transformed (e.g., to nitrous oxide). Here, an incubation step is needed, meaning that getting a good understanding of available nitrogen requires sampling, dividing the sample, extracting immediately from one half of the sample, incubating the other half for a set time under standard conditions, and then calculating the release of nitrogen by the soil.

There is an immediate trade-off with adding fertiliser to raise nutrient levels, as excess nutrients can be leached and end up in the aquatic environment, or excess nitrogen can be lost as N₂O. Hence, a balance must be reached where inputs meet plant requirements, while also fostering accumulation of soil organic matter to maximise intrinsic soil nutrient cycling. Current agronomic practice is to apply inorganic fertiliser at rates based on an understanding of plant uptake, but application rates often exceed those which are required as soil-specific variability in supply of nutrients from soil organic matter is usually not accounted for. Tools such as PLANET (Planning Land Applications of Nutrients for Efficiency and the environmenT), a nutrient management decision support tool for farmers and advisers to carry out field level nutrient planning and for demonstrating compliance with the Nitrate Vulnerable Zone (NVZ) rules, could be useful in this regard.

Maintaining optimal pH for crop growth also appears to reduce soil greenhouse gas emissions (Wang et al., 2021; Zhang et al, 2022), but there is a degree of context specificity, and this may not be appropriate for soils of high organic matter content.

Suggested metrics

• Suitable metric: Mineralisable N and available P, Soil pH
• Metric in development: Indicators based on soil FTIR spectroscopy

Reducing diffuse pollution

The outcome

Diffuse pollution has severe impacts on freshwater biodiversity and water quality with risks that climate change (low and high flow extreme increases, warmer temperatures) exacerbates effects such that moderate nutrient loading improvements may not lead to improved water quality.

Issues with a metric

Monitoring of diffuse pollution operates across scales, from the field scale, to highlight local improvements, to the catchment scale to understand cumulative effects and impacts (Bieroza et al. 2021). Field-scale predictions and observations of runoff prevalence and pathways, monitoring of soil compaction (measured by soil porosity) and soil chemistry (particularly nitrogen and phosphorus levels) provides an idea of risk, as does monitoring of in-field erosion (Hayes et al. 2023). Management at the edge of fields, e.g., buffer strips are designed to reduce diffuse pollution, but for best effectiveness, their location and design need to be targeted to ensure that they effectively treat converging runoff pathways and critical delivery points to the channel network (Stutter et al. 2021). Similarly, nutrient losses from field drains also need to be monitored as these can only be mitigated by specially designed and strategically located buffer strips.

Water sampling provides integrative evidence of the effectiveness of measures as it reflects management upstream in the catchment. Whilst monitoring of chemistry, biodiversity (invertebrates) and sediment will provide an understanding of upstream issues, it may be difficult to attribute impacts to diffuse or point source pollution (Glendell et al. 2019).

Water quality is closely linked with soil nutrient status, particularly nitrogen and phosphorus status of the soils, so relevant information can be acquired by soil sampling. However, there is also the need to monitor runoff generation and pathways, soil erosion, sediment flows and drainage waters. Monitoring is especially useful during extreme events, including high and low flows. An understanding of pollutant concentration changes over differing flow stages (e.g., inter-storm sampling) brings a wealth of information beneficial to management about source and transport behaviours at field to catchment scales.

We suggest that land managers are given responsibility for assessing erosion and water flow pathways and the subsequent monitoring of erosion and sediment flows, and potentially taking water samples of drainage waters for analysis by specialist laboratories. This would mean farmers assessing whether individual buffer strips were effective at preventing water flows, or whether their design allowed for flow around their edges by visiting them during periods of heavy rain. Future erosion pathways could be identified using fine-scale elevation data from LIDAR to model the flow of water across the surface of land (e.g. Reaney et al. 2019. Aquatic biodiversity requires specialist surveyors and could be done at the same time as the above-ground biodiversity assessment (Section 6.9).

SEPA currently collect a wide range of data from multiple sites. We suggest that it would be of benefit to use the current SEPA monitoring of agricultural catchments as the basis for studies linking agricultural management and water quality, by ensuring studies are joined up. This may mean enhancing the range and/or frequency of measures taken. A nested design could be followed, whereby field- and farm-scale sampling are nested within these catchments representing different land use typologies in Scotland, with water quality being monitored at the catchment outlet. The detailed knowledge from these catchments could be linked to farm-level data to make national estimates of benefits.

Farm-level models for looking at nutrient inputs and losses have been developed for England and Wales, e.g., FARMSCOPER. However, the extent to which it can be applied to the soils, climate and farming systems in Scotland has not been tested and this would need carrying out before it could be recommended as a metric for use in assessing the efficacy of measures.

Suggested metrics

• Suitable metric: Mineralisable nitrogen, available phosphorus and pH, Soil bulk density and porosity, Erosion monitoring and effectiveness of buffer strips (including other enhancements e.g., wetlands, wet woodland, sediment traps)
• Additional metric: Visual Evaluation of Soil Structure (VESS), Detailed monitoring in representative SEPA and other research catchments for process-based understanding on management impacts
• Metric in development: Runoff evaluation using LIDAR derived fine resolution topographic data

Improving water and air quality

The outcome

Water quality is tightly linked to freshwater biodiversity. However, it also has implications for the cost of water treatment downstream. Air pollution, particularly of ammonia, can also severely impact local biodiversity.

Issues with a metric

There can be a disconnect between actions at the field scale to reduce nutrient loss and water quality as actions can be poorly sited, poorly implemented and miss important routes of pollutant movement. However, there is clear evidence that reduction in soil nutrient status is the most likely route to deliver improvements in water quality, so monitoring for water quality is intrinsically linked to monitoring of soil nutrient status (Hayes et al. 2023).

High-resolution water quality monitoring that would represent the temporal and spatial variability is expensive and the movement of water in catchments may make linking it to the actions of individual farms problematic. Consequently, we suggest a combination of field/farm-level monitoring of soil nutrient status (i.e., soil organic matter, plant available (mineralisable) N, biologically available P and pH) and detailed monitoring of several representative catchment outlets to improve the understanding of processes. These could be based around SEPA’s existing catchment observation platforms, with additional investment to maximise the robustness of collected evidence.

Further action to reduce point source pollution, such as slurry pit overflow, farmyards and septic tanks, should not be overlooked (Harrison et al. 2019). Monitoring of this would be in the form of capital spend. Best practice should be followed for digestate and slurry application to land.

Currently available sensors for monitoring ammonia emissions tend to be expensive, require technical expertise and are sensitive to meteorological conditions and other atmospheric gases. Lower cost passive samplers, which could be deployed by non-specialists are less accurate, have lower temporal resolution, and require laboratory analysis (Insausti et al., 2020). A similar approach to that proposed for water quality could be implemented, with intensive monitoring of key areas with intensive livestock production systems, coupled with national scale monitoring utilising the National Ammonia Monitoring Network which monitors atmospheric ammonia concentrations monthly. A farm-level calculator for ammonia emissions is in development as part of the Scottish Government’s Strategic Research Programme. This would be the most cost-effective way forward for wide deployment of monitoring.

Suggested metrics

• Suitable metric: Mineralisable nitrogen, available P and pH, Soil bulk density and porosity, Erosion monitoring and effectiveness of buffer strips
• Additional metric: Intensive farm-scale monitoring of ammonia emissions in livestock intensive areas, Visual Evaluation of Soil Structure (VESS), Detailed monitoring in SEPA catchments for process-based understanding on management impacts
• Metric in development: Modelled farm emissions of ammonia

Improving soil water retention and flow

The outcome

Soil water retention is important in reducing soil erosion and diffuse pollution. If water flows through the soil it is slowed, reducing flood peaks, and there is greater interaction between the soil and water reducing the risk of nutrient loss. In contrast, water flowing across the surface of soils leads to erosion and nutrient runoff.

Issues with a metric

There are several detailed methods available to understand water retention and flow through soils, but they are not appropriate for wide-scale monitoring, apart from their potential use in the detailed monitoring of test catchments. These include detailed measures of soil texture, as well as laboratory measures of hydraulic conductivity. Direct measures of soil compaction with penetrometers suffer from variability due to soil water content, stoniness of the soil and differences between manufacturers. They are not suitable for wide-scale monitoring.

However, a set of straightforward measures are available to assess how soil water behaves. As part of the sampling of soil for soil carbon measurements, bulk density is measured to calculate carbon stocks from carbon concentrations. However, topsoil bulk density can vary seasonally and with respect to management. Subsoil bulk density is an indicator in the draft EU soil monitoring and resilience law and provides a more consistent measure of how the soil is behaving. This is a key parameter for understanding the effect of management on this outcome. However, the additional effort of also recording specific gravity of the soil will allow the calculation of soil porosity, another key parameter that is important for assessing soil water retention.

The Visual Evaluation of Soil Structure (VESS) is a qualitative metric that could also be used to supplement other measures and provide land managers with direct information at the field level on the degree of soil compaction, especially if this included both topsoil and subsoil. For quantitative measures of soil structure, the measurement of Water stable aggregates (WSA) should be considered and removes the potential for subjectivity.

Suggested metric

• Suitable metric: Sub-soil bulk density and porosity, Water stable aggregates, Erosion monitoring
• Additional metric: Visual Evaluation of Soil Structure (VESS)

Improving soil biodiversity

The outcome

Maintaining a healthy soil ecosystem is critical to the regulation of key processes, as soil organisms are critical to the cycling of nutrients and to plant growth. For instance, soil animals like earthworms are highly important to water movement in soils.

Issues with a metric

Soil biodiversity, whilst a key soil health indicator (Neilson et al. 2021), is unlikely to be practically assessed by the land manager. Identification of surface-dwelling invertebrates, such as beetles and earthworms, requires specialist taxonomic skills; even for earthworms a total count does not work as all functional groups need to be present for good soil health. Existing data is not available for surface dwelling invertebrates, but data collection methods with pitfall traps are standardised, for example by the Environmental Change Network. However, these methods require at least two visits, so may not be cost-effective. Previous earthworm surveys have been carried out (Boag et al. 1997, Carpenter et al. 2012), we suggest that methodologies should be kept consistent.

Molecular methods have been employed for bacteria, fungi and nematodes. However, methods to characterise complete soil biodiversity using eDNA (environmental DNA) are now emerging. As is typical with emerging technologies, there are issues surrounding data interpretation, thresholds and developing and/or defining baseline comparators. It is, perhaps, too early to suggest using this as a monitoring method, as the science relating molecular data to improvements in soil health is in its infancy. However, as soil sampling is likely to be used to monitor other outcomes, samples could be taken and archived for future use as a baseline to assess change.

Pesticide usage could be a proxy for the pressure on biodiversity, and hence pesticide usage data would be a useful addition to direct monitoring. It is already collected in Scotland, but refining the data to consider impacts on soil organisms and the different application rates would be necessary.

Further consideration needs to be given to:

• Collecting contextual information such as the current and previous crops.
• Whether the optimum times for sampling in spring and autumn coincide with the optimum times for sampling soil carbon and nutrients.

Suggested metrics

• Suitable metric: Surface dwelling invertebrates and earthworm functional group abundance
• Additional metric: Pesticide usage data
• Metric in development: eDNA samples archived as interpretation needs to improve

Removing drivers for biodiversity loss

The outcome

As much of Scotland is affected by agriculture, sensitive agricultural management is important to delivering the goals of the Scottish Biodiversity Strategy.

Issues with a metric

Biodiversity is intrinsically multi-dimensional, but typical agri-environmental monitoring targets habitat diversity, birds, pollinators and plants, as they give information at different scales.

In most schemes, biodiversity monitoring is done by specialists, as it is the status of priority species that has been the driver for the development of the scheme. However, that is not practical in terms of cost at the farm level, so a choice must be made between:

• Land manager-led monitoring aided by tools such as report cards and identification guides. Bird surveys could allow different levels of precision from individual species to groups (e.g., finches). Similarly, pollinator surveys could record at the level of group (bumblebee, honeybee, butterfly, hoverfly) or plant surveys, by numbers of different types of flower (e.g., daisy, pea types) in a set area. Alternatively, there is the possibility of sub-contracting to specialists if grant payments included money for monitoring. Land managers setting out acoustic recording devices also fits into this space. The resulting files could be uploaded to a central organisation responsible for analysis. The methodologies for data analysis are still in development, but sound files could be archived for later analysis when the methodologies have matured to deal with high levels of false positive identifications. The biodiversity audit as part of the whole farm plan also falls into this category.
• Specialist surveys on samples of farms with the sampling design considering the implementation of measures (Pakeman et al. 2020) or being large enough to assess change for most measures, however, they are distributed across the landscape (e.g., the Welsh approach to monitoring Glastir).

There is a clear trade-off here between broad and shallow versus narrow and deep approaches. To enable adaptive management at the farm level, then land manager-led monitoring is important, but there is the risk that the measures deliver higher numbers of generalist species, do not benefit species that are a conservation priority, but the data is incapable of showing this. It may be that a hybrid approach is necessary, so that field/farm-level data is complemented by detailed measures on a sample of land holdings. However, sample sizes need to be sufficient to confidently assess change. Previous monitoring studies, e.g., Perry et al. (2003), could be used to identify appropriate levels of sampling needed.

Currently collected biodiversity data is not appropriate for agri-environment monitoring for a range of reasons, mainly due to mismatches in scale between land holdings and the specific sampling method used. In the case of breeding bird data, it has been used as a measure of general farmland diversity against which the performance of in-scheme farms has been judged.

Proxies for habitat diversity currently collected by RPID would be useful data, but it only characterises area and has no measure of quality associated. Alternatives include using remote sensing data (e.g., habitat maps or LIDAR derived information on hedgerow extent and conditions) that provides information on land cover and structure, but these are only proxies for biodiversity.

Finally, pesticide usage is a clear driver of biodiversity loss. Usage statistics are already collected using a sampling approach to assess a Scotland-level picture. However, the diversity of chemicals applied, and their different application rates would require methodological developments to combine their usage into meaningful statistics.

Suggested metrics

• Suitable metric: Bird, pollinator and plant composition and diversity
• Additional metric: Farmland habitat diversity, pesticide usage survey data
• Metric in development: Acoustic diversity, LIDAR derived hedge data

Improving animal nutrition

The outcome

Improving animal nutrition will reduce the time taken to deliver animals to market. This reduces lifetime emissions especially of methane.

Issues with a metric

Improving livestock nutrition leads to increased animal performance and reduced methane, nitrous oxide and ammonia emissions. Monitoring of nutrition can be undertaken through laboratory analysis of feedstuffs. The key analyses are forage digestibility – which can easily be undertaken by many feed companies – and dietary crude protein. There is also an important trade-off already mentioned between optimising nutrition and the increased fertiliser use, leading to greenhouse gas emissions and/or pollution of water courses. However, these are very much business-related metrics, and their collection may not be informative as a means of national monitoring, particularly as silage quality varies between fields, time of year and across years. The need for its collection as part of a national monitoring scheme is, therefore, debateable.

Instead, we suggest that simple measures of animal performance are collected and form part of routine monitoring of flock/herd status. These reflect actual performance rather than inputs into the system and are easier to record.

Suggested metrics

• Suitable metrics: Growth rates, Milk yields (Dairy cattle only), Mortality, Conception rates, Replacement rates, Age at slaughter
• Additional metric: Feed analysis for digestibility/protein

Improving animal breeding

The outcome

Focusing on animal breeding can improve the productivity of farming systems and, also, increase the quality of products like meat and milk. In terms of reducing methane production, breeding can directly reduce emissions, but also quicker growing animals will release less over their lifetimes.

Issues with a metric

Selective breeding for improved productivity, improved efficiency or reduced methane emissions could drive permanent and cumulative improvement in performance and/or reductions in methane emissions. Monitoring of selective cattle breeding for specific traits could be undertaken through applications for calf passports to ScotEID, but this would rely upon sire details being recorded on passports (which is currently not mandatory) and on the sire’s genetic potential for selected traits being known.

Proxy measures such as growth rates, milk production, conception rates and days to slaughter could also be used to monitor improvements over longer time periods but could be confounded with improvements in nutrition and health.

Suggested metrics

• Suitable metric: Sire details included on applications to ScotEID for calf/lamb passports
• Additional metric: Growth rates, Milk production, Conception rates, Age at slaughter

Improving animal health

The outcome

Improved animal health has a direct benefit to animal welfare. However, it also reduces losses during the production process, improving productivity and reducing methane emissions on a lifetime basis.

Issues with a metric

Several endemic (and exotic) diseases and syndromes can impact on the production efficiency and associated GHG emissions of farmed livestock. Some diseases have a direct impact on individual animals and metrics such as growth rates, reproductive success, and replacement rates. Others have a more indirect impact at herd/flock and national level, through how diseased animals are managed following diagnosis. Data and metrics on the prevalence of key priority diseases and health conditions at a national level are currently not collected, but would be invaluable, if logistically challenging. Eradication may be feasible for some diseases, e.g., Bovine Viral Diarrhoea (BVD), but requires the relevant tools, e.g., vaccines, and diagnostics to be available, in addition to coordination and buy-in across the industry.

The most straightforward was to assess animal health would be to collect a common set of proxy measures, e.g., growth rates, age at slaughter, conception rates, replacement rates and mortality rates will be the most feasible approach to measuring progress on animal health. This approach could also be applied to animal breeding and nutrition.

Recording all these metrics would be useful to both national-level monitoring of performance and for the land manager’s care for their livestock. This could also include records of veterinary medicines used to gauge movement towards sustainable prescribing, though this complex topic (Humphry et al. 2021) is outwith the scope of this report.

Suggested metrics

• Suitable metrics: Growth rate, Milk yields (Dairy cattle only), Mortality, Conception rates, Replacement rates, age at slaughter

Methane suppression

The outcome

Methane is a greenhouse gas with much higher global warming potential than carbon dioxide (methane from non-fossil fuel sources has a global warming potential of 27 times that of C0₂ with a 100-year time horizon, IPCC 2021). Enteric methane is released by ruminants such as cattle and sheep as part of the natural digestion of plant material by their associated microbiota. Methane is a significant part of agricultural emissions and so reducing it is key to reaching net zero emissions.

Issues with a metric

Selective breeding for reduced enteric methane emissions/increased animal efficiency (section 6.11) is a long-term strategy. In the short term, feed supplements designed to suppress enteric methane production could be used to drive down emissions. Sexed semen could be used to optimise herd dynamics by reducing numbers of male dairy calves and increasing male beef and dairy-beef calves. Direct measurement of methane emissions depends upon specialised equipment and is therefore not practical at the herd or flock level.

Two potential options are available. Firstly, to monitor the usage of methane-reducing feed supplements and calculate emission reductions based on their reduction factors. However, appropriate reduction factors for all feed products may not be available for all systems. The other option is to use current carbon footprinting tools (e.g., Agrecalc, Cool Farm Tool), but these need a subscription, may need the help of a consultant to set up and would benefit from information on forage quality and the impacts of feed supplements (so in effect replacing the need for providing information separately on feed supplements). The use of a standard tool across herds/flocks would allow for comparison.

Suggested metrics

• Suitable metric: Modelled farm emissions of CH₄, CO₂, N₂O

Nutrient management

The outcome

Poor nutrient management can lead to the emissions of nitrous oxide, methane and ammonia. It also runs the risk of point source and diffuse pollution into watercourses.

Issues with a metric

Organic manures help recycle nutrients and build soil organic matter. However, there is the potential for them to be a source of ammonia, methane and nitrous oxide, as well as nutrient runoff in water courses. Much can be done to alleviate this, with well-designed and covered manure stores as well as appropriate application techniques. Gaseous emissions are difficult to monitor directly, so these would have to be modelled using a farm calculator. Impacts on soil nutrients and water quality are dealt with in previous sections so a separate metric for nutrient management is not necessary.

Suggested metrics

• Suitable metric: Modelled farm emissions of CH₄, CO₂, N₂O, Mineralisable nitrogen and available phosphorus, Effectiveness of buffer strips
• Metric in development: Modelled farm emissions of ammonia

Coordinated metric collection

This section examines opportunities to synthesise across the required outcomes to minimise the number of metrics to be collected.

Why is this important?

Any monitoring must be as cost-effective as possible. Consequently, during the design phase decisions should be focussed on making the recording of metrics as straightforward as possible and to build efficiency into any monitoring programme, for example, by sampling multiple metrics on the same visit.

Greenhouse gas emissions

Gas emissions cannot be realistically measured directly. Using current farm-level tools to assess GHG emissions will deliver against multiple outcomes [Reducing Soil Greenhouse Gas (GHG) emissions, Livestock emissions, Nutrient management]. In addition, a tool for estimating ammonia emissions [Improving water and air quality] is in development, as is a field-level, real time emission model. These will further enhance capability in this area.

Coordinated sampling strategies

Many metrics depend on the direct sampling of soil or biodiversity and can’t be realistically replaced by proxies or existing data. However, well designed sampling programmes can maximise the efficiency of sampling, e.g., sampling for soil carbon, nutrients, pH and eDNA can be done at the same time. Even if this were not possible, sampling of soil nutrients, particularly mineralisable nitrogen and available phosphorus, would deliver against multiple outcomes [Improving soil nutrient content, Reducing diffuse pollution, Improving water and air quality, Nutrient management]. Similarly, monitoring soil bulk density is important for multiple outcomes [Increasing soil carbon/organic matter content, Improving soil water retention and flow], as is Water stable aggregates.

Soil monitoring

A range of soil monitoring is already being carried out for different purposes. There is a need to consider how future monitoring could supplement or replace existing work in this area, including:

• Scottish Government’s National Test Programme’s includes soil carbon measurement in Preparing for Sustainable Farming (PSF).
• The Agriculture and Horticulture Development Board Scorecard for soil health which includes soil structure (using VESS Visual Evaluation of Soil Structure), pH, extractable nutrients (phosphorus, potassium and magnesium), earthworms, soil organic matter.
• Linking Environment And Farming (LEAF).
• Red Tractor.
• Soil Association.

Livestock

The outcomes associated with animal health, nutrition and breeding must be largely monitored through proxy metrics, but these are relatively easy to measure and provide useful information direct to the land manager. However, it would be difficult to disentangle the differing contributions of nutrition, health and breeding on the overall performance of the flock/herd. At present, the separate contributions of improving animal nutrition, improving animal health and improving the genetics of the flock/herd are not easily separated but offer three routes for livestock managers to improve performance and consequently reduce emissions, one or more of which can be followed.

National level data

A few metrics can be based on existing data such as data collected as part of the agricultural census or can be derived from existing data such as satellite habitat maps. These are useful additional data, but do not provide the best metrics to assess the success of outcomes. They include: Area under permanent vegetation or other carbon positive management, Detailed monitoring in SEPA catchments to include water quality (nitrate, phosphate etc.), SEPA regulatory monitoring, Pesticide Usage Survey data, Farmland habitat diversity. They can be identified by filtering column I in the terrestrial sheet of MeasuresXMetrics.xlsx file.

Metrics currently in development

There are a range of metrics that are in development, some of which could take advantage of samples/data collected at the start of any monitoring programme (e.g., soil eDNA, acoustic monitoring) but others would come online later (e.g., LIDAR derived hedge data).

Cost-effective data acquisition strategies

Where this report recommends farmer-led metric recording, then this would provide a whole population value that can be followed through time. However, where only a proportion of the population of farms/fields can be sampled there has to be a statistically sound design adopted. This would include a comparison between areas where measures have been applied with other similar areas that are not in the scheme (control sites). A Before-After-Control-Impact (BACI) design is commonly used for monitoring the effect of environmental interventions. One issue to be addressed is that areas which are originally selected as controls may join the scheme later, so starting with a larger control population may guard against this.

An example – Wales

In Wales, the Glastir Monitoring and Evaluation Programme (GMEP) sample consisted of a stratified random sample 150 “Wider Wales” 1 km squares and 150 targeted at priority areas for the agri-environment scheme. It should be noted, however, that the “Wider Wales” squares do include land which is in the scheme, and that even the targeted 1 km squares contain differing amounts of land where specific management options have been applied. As it was found that the “Wider Wales” squares had considerable coverage of the scheme, in the more recent ERAMMP National Field Survey the aim has been to capture as much in-scheme and counterfactual land as possible within the full set of 300 squares.

To allow sampling effort to be spread across years and provide both temporal and spatial coverage, a rolling monitoring programme was followed by GMEP, in which sites are revisited, for example, every four or five years but different sites are sampled in years two and three. This allows better spatial coverage than if each site was revisited every year, while at the same time providing a more powerful estimate of change over say a five-year period, than not revisiting sites at all. The GMEP scheme uses a four-year rolling monitoring programme. Countryside Survey is also now following a rolling programme. Power analysis for the GMEP scheme (Emmett et al., 2014) suggested that across a variety of metrics, around 45 squares per year was the minimum number that need to be monitored before losing significant power to detect change over a period of 8 years (two cycles of the rolling programme).

Other considerations

Soil nutrients and soil carbon would be most appropriately measured at the scale of fields within farms, as this is the level at which relevant measures are applied. On the other hand, surveys of birds and pollinators, which are mobile over a larger area, might be more appropriately recorded for parcels of land, such as 1 km squares, although it is unlikely that the same measures will have been applied consistently across a whole 1 km square. To provide a common spatial unit across different metrics, the GMEP survey used 1 km squares, but, as it is not possible to sample vegetation and soils over the entire square, five randomly placed plots in each square were used for vegetation monitoring and soil samples were taken from the same plots. Vegetation was also recorded in other plot types, for example, along boundaries and field margins.

If fields rather than squares are to be used for soil monitoring, a representative sample for a particular field or part of a larger field can be obtained by bulking individual cores. For example, in the Soil Nutrient Health Scheme in Northern Ireland samplers follow a ‘W’ shaped track and take 25 cores. This should give a good estimate of the mean for an individual field but unless replicate cores are analysed individually it does not provide an estimate of the variability within the field. As a result, it is not possible to determine whether a change in a particular field between two sampling occasions is statistically significant. Under the Scottish Government’s National Test Programme 20% of arable and improved grassland can claim funding for soil testing each year. If this scheme is continued, it could mean complete coverage of all arable and improved grassland fields after five years. The recommendation of Scotland’s Farm Advisory Service (FAS) is that larger fields should be divided into 4 ha units, potentially with the help of the 1:25000 soil map. This approach might provide sufficient replicate samples across a farm as a whole to allow a change to be detected on a specific farm, although it should be noted that unless a suitable control is available it may not be possible to attribute any change to particular measures and that soil carbon changes in response to measures might take longer than five years to be detectable.

A note on data

Monitoring across a range of outcomes and metrics will generate a considerable volume of data. This will require a significant investment in design and development of the databases and in the staff required for data curation.

Alongside the technical aspects of database curation and management, consideration should be given to who owns the data – whether the land manager as it concerns their land holding or the taxpayer as they paid for it, who has access to it? – a narrow access regime provides increased security, especially around GDPR, but wider access allows for a broader range of analyses to be carried out. Furthermore, an overall data controller/owner would likely need to be appointed to comply with GDPR. It should be possible to develop data frameworks, where analysis without direct access to locations is possible (similar to medical data where analysis is separated from any data identifying subjects) and comply with Freedom of Information requests. Arguably, data should follow FAIR data principles and be open access as it has been funded from the public purse, as in the European Soil Observatory.

Conclusions

Some metrics will clearly be valuable in identifying the benefits of future agri-environmental management. For example, the collection of data on soil carbon and methane emissions clearly supports the Scottish Government’s climate ambitions. Others will support policies regarding sustainability (soil erosion) and the health of Scotland’s freshwater resources (reducing diffuse pollution). There is a mix of field data collection, farmer-collected data and modelled information with some usage of existing data.

It should be noted that several metrics have been identified that may only be proxies of the outcome they relate to, such as area of non-farmed habitats or pesticide usage, but they have the advantage of being based on already collected data with the cost savings this brings. Other metrics are still in development but should either be available by the start of the scheme or where samples can be collected for future analysis.

There are several outcomes that are closely related and need consideration together. Improving animal nutrition requires maintaining soil nutrients at a level where protein is not limiting growth. This may require the application of organic and/or inorganic fertilisers. However, excess nutrients can end up as N₂O and ammonia emissions from slurry and the leaching of nitrates into freshwater. Careful management to optimise nutrient use is, therefore, required to reach all the desired outcomes: improving soil nutrient content, reducing diffuse pollution, improving water and air quality, livestock nutrition and nutrient management.

A second set of outcomes are also closely related, those dealing with livestock genetics, health and nutrition, alongside reducing methane emissions. Improved efficiency across the livestock sector should increase margins but at the same time reduce the methane footprint of meat.

Some metrics are not useful in isolation and need to be collected as a set to be useful. This is particularly true for animal health, nutrition and genetics where a range of data on growth rates, milk yields, mortality, conception rates and replacement rates are needed to get a full picture.

The final choice of which metrics to collect will depend on the availability of resources to carry out the monitoring and the type of sampling philosophy adopted. Assembly, curation and analysis of the data will all add costs to metric collection but it is important to get the most out of the data. Data ownership is also a key consideration.

Given the division between farmer-led and expert-led monitoring highlighted in the spreadsheet and in Section 6, we suggest the following:

• All enterprises to assess soil erosion and buffer strip effectiveness as this is highly site specific.
• All livestock enterprises to record growth rate, milk yields, mortality, conception rates, replacement rates, age at slaughter for sheep and cattle.
• ScotEID to require information on sires.
• All enterprises to use farm tool calculators to model GHG emissions. Livestock enterprises to model ammonia emissions when a suitable tool is available. The requirement to model might be limited to enterprises above a certain size to reduce costs.
• The remaining outcomes are best assessed using expert-led monitoring in a sample-based programme similar in philosophy to the Welsh approach. Resources available for monitoring and statistical power analysis would be a key part of how to structure this monitoring. They would also determine whether to focus on a small number of metrics and outcomes and cover a larger sample size, or to cover all outcomes on a smaller sample size. The outcomes monitored in this way include those focused on soils, waters and biodiversity.

References

Bieroza, M.Z., Bol, R. & Glendell, M. (2021) What is the deal with the Green Deal: Will the new strategy help to improve European freshwater quality beyond the Water Framework Directive? Science of The Total Environment, 791, 148080. https://doi.org/10.1016/j.scitotenv.2021.148080

Boag, B., Palmer, L.F., Neilson, R., Legg, R. & Chambers, S.J. (1997). Distribution, prevalence and intensity of earthworm populations in arable land and grassland in Scotland. Annals of Applied Biology, 130, 153-165. https://doi.org/10.1111/j.1744-7348.1997.tb05791.x

Brodie, E. (2023) Farm for Scotland’s Future: the case for change. A report for Scottish Environment LINK. https://farmforscotlandsfuture.scot/wp-content/uploads/2023/04/Farm-for-Scotlands-Future-full-report.pdf

Carey, P.D., Wallis, S., Chamberlain, P.M., Cooper, A., Emmett, B.A., Maskell, L.C., McCann, T., Murphy, J., Norton, L.R., Reynolds, B., Scott, W.A., Simpson, I.C., Smart, S.M. & Ullyett, J.M. (2008) Countryside Survey: UK Results from 2007. NERC/Centre for Ecology & Hydrology, 105pp. (CEH Project Number: C03259) https://nora.nerc.ac.uk/id/eprint/5191/

Carpenter, D., Sherlock, E., Jones, D.T., Chiminoides, J., Writer, T., Neilson, R., Boag, B., Keith, A.M. & Eggleton, P. (2012). Mapping of earthworm distribution for the British Isles and Eire highlights the under-recording of an ecologically important group. Biodiversity and Conservation, 21, 475-485. https://doi.org/10.1007/s10531-011-0194-x

Chatterjee, A., Lal, R., Wielopolski, L., Martin, M.Z. & Ebinger, M.H. (2009) Evaluation of different soil carbon determination methods. Critical Reviews in Plant Science, 28, 164-178. https://doi.org/10.1080/07352680902776556

Cottle, D.J. & Conington, J. (2012) Breeding for reduced methane emissions in extensive UK sheep systems. Journal of Agricultural Science, 150, 570-583. doi:10.1017/S0021859611000967

Cottle, D.J. & Conington, J. (2013) Reducing methane emissions by including methane production or feed intake in genetic selection programmes for Suffolk sheep. Journal of Agricultural Science, 151, 6, 872-888. doi:10.1017/S0021859612001037

Emmett, B.E., Abdalla, M., Anthony, S., Astbury, S., August, T., Barrett, G., Biggs, J., Botham, M., Bradley, D., Brown, M., Carter, H., Chadwick, D., Cigna, F., Cooper, D., Cooper, J., Cosby, BJ., Creer, S Cross, P., Edwards, F., Edwards, M., Evans, C., Ewald, N., Fitton, A., Garbutt, A., Grebby, S., Greene, S., Halfpenney, I., Hall, J., Harrison, S., Harrower, C., Henrys, P., Hobson, R., Hughes, S., Isaac, N., Jackson, B., Jarvis, S., Jones, D., Keith, A., Kelly, M., Korenko, J., Lallias, D., Leaver, D., Lebron, I., Malcolm, H., Maskell, L., McDonald, J., Moxley, J., Norton, L., O’Hare, M., Owen, A., Pereira, M., Peyton, J., Powney, G., Pywell, R., Rawlins, B., Robinson, D.A., Rorke, S., Rowland, C., Roy, D., Scarlett, P., Scholefield, P., Scott, A., Scott, L., Scott, R., Siriwardena, G., Smart, S., Smith, P., Swetnam, R., Taylor, R., Tordoff, G., Van Breda, J., Vincent, H., Wagner, M., Waters, E., Watkins, J., White, J., Williams, B., Wood, C. & Wright, S. (2014) Glastir Monitoring & Evaluation Programme. First Year Annual Report to Welsh Government (Contract reference: C147/2010/11). NERC/Centre for Ecology & Hydrology (CEH Project: NEC04780), pp.442. https://gmep.wales/sites/default/files/GMEP-First-Year-Annual-Report-2014-05-17.pdf

Environmental Standards Scotland (2022) Baseline Evidence Review – Water (Strategy and Analysis). https://www.environmentalstandards.scot/wp-content/uploads/2022/09/Baseline-Evidence-Review-Water-2022-09.pdf

Estrada-Carmona, N., Sánchez, A.C., Remans, R. & Jones, S.K. (2022) Complex agricultural landscapes host more biodiversity than simple ones: A global meta-analysis. Proceedings of the National Academy of Sciences, 119, e2203385119. https://doi.org/10.1073/pnas.2203385119

Giles, M.E., Caul, S., King, D., Mitchell, S., Sim, A., Neilson, R. & Paterson, E. (2023). Grass variety selection of microbial community composition is associated with differences in soil CO₂ emissions. Applied Soil Ecology, 190, 104968.

Glendell, M., Palarea‐Albaladejo, J., Pohle, I., Marrero, S., McCreadie, B., Cameron, G. & Stutter, M. (2019) Modeling the ecological impact of phosphorus in catchments with multiple environmental stressors. Journal of Environmental Quality, 48, 1336-1346. https://doi.org/10.2134/jeq2019.05.0195

Harrison, S., McAree, C., Mulville, W. & Sullivan, T. (2019) The problem of agricultural ‘diffuse’ pollution: Getting to the point. Science of the Total Environment, 677, 700-717. https://doi.org/10.1016/j.scitotenv.2019.04.169

Hayes, E., Higgins, S., Geris, J., Nicholl, G. & Mullan, D. (2023) Weighted risk assessment of critical source areas for soil phosphorus losses through surface runoff mechanisms. Catena, 225, 107027. https://doi.org/10.1016/j.catena.2023.107027

Humphry, R., Skuce, P., Hayward, A., Henry, M., Hutchinson, I., Gunn, G. & Costa, M. (2021) Measurement of antimicrobial usage: what can we learn across Livestock Sectors? https://sefari.scot/research/measurement-of-antimicrobial-usage-what-can-we-learn-across-livestock-sectors

Insausti, M., Timmis, R., Kinnersley, R. & Rufino, M.C. (2020) Advances in sensing ammonia from agricultural sources. Science of The Total Environment, 706, 135124. https://doi.org/10.1016/j.scitotenv.2019.135124

IPCC (2021) Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Eds Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Péan, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R., Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2391 pp. doi:10.1017/9781009157896.

Leinonen, I., Eory, V., MacLeod, M., Sykes, AS., Glenk, K., & Rees, RM. (2019). Comparative analysis of farm-based carbon audits. ClimateXChange. https://www.climatexchange.org.uk/media/3584/farm-based-carbon-audits-final.pdf

Neilson, R., Aitkenhead, M., Lilly, A. & Loades, K. (2021) Monitoring soil health in Scotland by land use category – a scoping study. James Hutton Institute. https://doi.org/10.7488/era/1293

Pakeman, R.J., Beaton, J., Fielding, D., Hewison, R., McKeen, M., Potts, J., Quinzo-Ortega, L., Stephen, T. & Stockan, J. (2020) Evaluation of the Biodiversity outcomes of the 2014-20 SRDP Agri-Environment Climate Scheme through a selection of case studies. Report to NatureScot. https://www.nature.scot/doc/naturescot-research-report-1254-evaluation-biodiversity-outcomes-2014-20-srdp-agri-environment.

Perry, J.N., Rothery, P., Clark, S.J., Heard, M.S. & Hawes, C. (2003) Design, analysis and statistical power of the Farm‐Scale Evaluations of genetically modified herbicide‐tolerant crops. Journal of Applied Ecology, 40, 17-31. https://doi.org/10.1046/j.1365-2664.2003.00786.x

Prout, J.M., Shepherd, K.D., McGrath, S.P., Kirk, G.J.D., Hassall, K.L. & Haefele, S.M. (2022) Changes in organic carbon to clay ratios in different soils and land uses in England and Wales over time. Scientific Reports, 12, 5162. https://doi.org/10.1038/s41598-022-09101-3.

Reaney, S.M., Mackay, E.B., Haygarth, P.M., Fisher, M., Molineux, A., Potts, M. & Benskin, C.M.H. (2019) Identifying critical source areas using multiple methods for effective diffuse pollution mitigation. Journal of Environmental Management, 250, 109366. https://doi.org/10.1016/j.jenvman.2019.109366

Saby, N.P., Bellamy, P.H., Morvan, X., Arrouays, D., Jones, R.J., Verheijen, F.G., Kibblewhite, M.G., Verdoodt, A.N.N., Üveges, J.B., Freudenschuss, A. & Simota, C. (2008) Will European soil‐monitoring networks be able to detect changes in topsoil organic carbon content? Global Change Biology, 14, 2432-2442. https://doi.org/10.1111/j.1365-2486.2008.01658.x

Skuce, P., Holland, J., Kenyon, F., Wells, B., McCracken, D., Neilson, R. & Bartley, D. (2022) Anthelmintics and the Environment – opening a whole can of worms? https://sefari.scot/research/anthelmintics-and-the-environment-%E2%80%93-opening-a-whole-can-of-worms

Staggenborg, J. & Anthes, N. (2022) Long‐term fallows rate best among agri‐environment scheme effects on farmland birds—A meta‐analysis. Conservation Letters, 15, e12904. https://doi.org/10.1111/conl.12904

Stutter, M., Costa, F.B. & Ó hUallacháin, D. (2021) The interactions of site-specific factors on riparian buffer effectiveness across multiple pollutants: A review. Science of the Total Environment, 798, 149238. https://doi.org/10.1016/j.scitotenv.2021.149238

Walton, P., O’Brien, D., Smart, J., Burns, F., Basset, D., Bradfer-Lawrence, T., Foster, S., James, B.D., Mancini, F., Mordue, S., Pakeman, R.J., Pescott, O.L., Simkin, J., Stanbury, A.J. & Towers, M. (2023) State of Nature Scotland 2023. The State of Nature Partnership. https://stateofnature.org.uk/wp-content/uploads/2023/09/TP26056-SoN-Scotland-summary-report-v5-1.pdf.pagespeed.ce.EIp-TYaoGQ.pdf

Wang, Y., Yao, Z., Zhan, Y., Zheng, X., Zhou, M., Yan, G., Wang, L., Werner, C. & Butterbach‐Bahl, K. (2021) Potential benefits of liming to acid soils on climate change mitigation and food security. Global Change Biology, 27, 2807-2821. https://doi.org/10.1111/gcb.15607

Zhang, H.M., Liang, Z., Li, Y., Chen, Z.X., Zhang, J.B., Cai, Z.C., Elsgaard, L., Cheng, Y., van Groenigen, K.J. & Abalos, D. (2022) Liming modifies greenhouse gas fluxes from soils: A meta-analysis of biological drivers. Agriculture, Ecosystems & Environment, 340, 108182. https://doi.org/10.1016/j.agee.2022.108182

List of Appendices

Appendix A. List of measures under consideration

Appendix B. Correspondence between land-based measures and relevant outcomes

Appendix C. Potential monitoring metrics and methods

Measures and metrics spreadsheet

Appendices

© Published by The James Hutton Institute, 2024 on behalf of ClimateXChange. All rights reserved.

While every effort is made to ensure the information in this report is accurate, no legal responsibility is accepted for any errors, omissions or misleading statements. The views expressed represent those of the author(s), and do not necessarily represent those of the host institutions or funders.

DOI: http://dx.doi.org/10.7488/era/4346

Executive summary

Aims

This research is a rapid review, presenting and examining evidence relating to climate change and digital connectivity such as:

• whether investment in digital connectivity can support reductions of greenhouse gas (GHG) emissions and, if so, how
• examples of relevant policies and impacts
• the best options for assessing emissions from digital connectivity and services in Scotland
• key evidence gaps in these areas.

The review was undertaken between October and the middle of December 2023 and was focused on and bounded by specific criteria set out by the Scottish Government and ClimateXChange steering group. The study team were asked to only include information and projects that were current and operating, not theoretical. Search terms were selected and agreed with the steering group.

We used a methodology known as “claim, argument, evidence” to assess whether claims made within certain arguments were true or false. We classified evidence as having low, moderate or good confidence levels, based on its volume and quality, and the level of agreement in the literature reviewed.

Findings

We have found mixed evidence of the decarbonisation impact of digital connectivity and whether it contributes to adaptation and a just transition. Our main findings on basis of the literature reviewed are:

• The Information Communications Technology (ICT) sector is a source of GHG emissions. The sector’s energy consumption and generation of electronic waste (e-waste) generates GHG emissions directly. This is despite the possibility that it can reduce emissions indirectly by increasing efficiency and through behaviour changes like reduced travel due to remote meetings. Studies point out a need for a holistic approach in calculating GHG emissions of the ICT sector, to fully account for indirect emissions and emissions from end-of-life.
• ICT technology and digitalisation reduce GHG emissions in other industries. Heavy industry and the energy sector would benefit the most from digitalisation. To achieve this, there would need to be widespread high-speed internet coverage, which would likely generate further GHG emissions. On their own, digital connectivity infrastructures do not support emissions reduction. They provide a mechanism to support decarbonisation of other sectors.
• The GHG emissions associated with e-waste are of growing concern internationally. Even though ICT use could help reduce GHG emissions in other sectors, it is uncertain whether this can outweigh the direct emissions of the ITC sector. It gives us only moderate confidence that the ICT sector can help reduce more emissions than are inherent in the manufacture, use and disposal of the equipment used.
• The indirect impact of ICT technologies can either lead to a net reduction in carbon emissions or to a net increase. The overall effects depend on context. Rebound effects can lead to increases in emissions. Policy and measurement do not usually account for these effects. Human behaviour plays a part in whether the indirect impacts on emissions are positive or negative. This means that it is not solely down to technology and therefore we are only moderately confident that the challenge of emissions reduction can solely be met by utilising digital technology.
• We are unable to say whether digital connectivity supports climate adaptation because of the small number of ex-post studies in this area. With regard to a just transition, digital connectivity and ICT can have either a positive or a negative effect, either addressing or exacerbating existing inequalities such as access to digital connectivity and skills. Studies repeat the need for strong policy in this area.

Within the literature reviewed as part of the study, we identified gaps in knowledge, including:

• Lack of evidence on whether investment in digital connectivity directly reduces GHG emissions, or contributes to a just transition and how.
• There are varying approaches to quantifying direct and indirect emissions of ICT and to comparing the GHG emissions of digital and non-digital practices and solutions.
• Climate adaptation in relation to ICT is either an afterthought or future looking, with few real-world examples.
• Case studies of digital technologies saving money, power or water in municipalities focused on the GHG emissions reduced or averted, with no acknowledgment of rebound effects, which literature states is important.
• The GHG emissions of data collection and use necessary to digitalisation are opaque and limited to specific studies on data centres.
• Lack of evidence of policy to address GHG emissions of e-waste or the embedded emissions from extraction of raw materials and production of ICT equipment.
• Lack of best practice for measuring, monitoring and assessing the GHG footprint of electronic communications services.

Glossary

Introduction

Need for this research

The extent to which the development and deployment of digital and data solutions supports the reduction of a country’s greenhouse gas footprint, assists in adaptation, and contributes to a just transition is unclear. Digital technologies have become an integral part of our lives, but they also have an environmental impact, including the production of greenhouse gas emissions (GHG) during their manufacturing, use and disposal.

In recent years, over £1 billion has been invested in programmes to enhance digital connectivity in Scotland, for a variety of anticipated outcomes relating to regional equity and opportunity. These include Digital Scotland Superfast Broadband (DSSB), Reaching 100% (R100), Scottish 4G Infill (S4GI), and the Scotland 5G Centre, with a regional network of 5G Innovation Hubs to facilitate widespread deployment of 5G.

Digital connectivity, and increasing access to it, is the focus of many Scottish Government policies. The Digital Strategy, ‘A changing nation: how Scotland will thrive in a digital world,’ is the policy backbone, setting out actions on: broadband and connectivity; data and statistics; digital inclusion and ethics; digital, data and technology profession, skills and capability; Transforming public services; and the Technology Assurance Framework. Enhancing Scotland’s digital infrastructure, both nationally and internationally, has also been a stated priority in successive Programmes for Government and the 10-year National Strategy for Economic Transformation (NSET) published in 2022. There is a lack of current evidence on the extent of the potential contribution of digital connectivity to Scotland’s climate change goals, not least of achieving net zero by 2045.

Project aim

The aim of the project is to examine recent research on climate change and digital connectivity to answer the following questions:

• To what extent is there evidence that investment in digital connectivity can support emissions reduction, climate adaptation and a just transition?
• If so, what are the key mechanisms by which this could occur (for example, reduction in travel, investment in green data centres or other mechanisms suggested in the evidence)?
• What are key examples of existing policies (in Scotland, such as in Local Authorities, the UK and/or international examples from comparable countries) designed to support emission reductions, adaptation, and/ or just transition through digital connectivity? Is there any evidence for the impact of such policies?
• What are the different options suggested within the literature for Scotland to provide a baseline assessment of, and monitor carbon emissions from digital infrastructure, technologies, and associated activities?
• What are the other key gaps in existing knowledge where further research is required to support digital connectivity and Scotland’s climate change goals?

These questions are answered in Sections 5, 6, 7, 8 and 9. By better understanding the mechanisms in which digital connectivity supports Scotland reaching net zero, policy makers will know how to influence what they want to occur.

Key terms used throughout the report are explained in the Glossary in Section 2.

Components of the digital landscape covered by this research

The focus of the research is digital connectivity. This can encompass a wide range of products and services. Figure 1 sets out the boundaries of the research undertaken to inform this paper, including:

• infrastructure such as fixed broadband, mobile connectivity and data centres
• application, use and behaviours such as artificial intelligence and the Internet of Things, data driven products and services, and practices such as home working
• the list of countries with applicable learning for Scotland.

Figure 1 – The landscape of digital connectivity defined as within scope of this research.

Approach to the research

This section provides an overview of the research approach. Our full methodology is outlined in Appendix 1.

Methodology for collecting evidence

Frazer-Nash Consultancy (Frazer-Nash) was tasked with completing this research for ClimateXChange (CxC) on behalf of the Scottish Government Digital Connectivity Division. A steering group was set up consisting of representatives from Scottish Government, CxC and Frazer-Nash.

We followed an approach based on the Double Diamond approach of Discovery and Define^[1], including literature gathering, revising and providing initial conclusions, and further developing conclusions before developing the report. We socialised the initial and final conclusions with the steering group. Keywords for the literature search were also agreed with the steering group. The literature reviewed was identified through google and google scholar searches. The review was focused on and limited by specific criteria, such as the non-inclusion of theoretical studies around “what is possible”, with an emphasis on current and recent experience.

Methodology for policy review

One of our research questions requests a review of policies which were designed to support emission reductions, adaptation or just transition through digital connectivity. To determine the geographic scope of the research, we chose countries analogous to Scotland facing similar digital connectivity challenges, that is, large landmasses with areas of low population density, and a number of isolated and rural remote communities. This list was agreed with the steering group and consists of: Finland, Wales, Portugal, Norway, Sweden, Estonia, Canada (Ontario), New Zealand, Denmark, and Iceland.

We use a star key (Table 3) to rate the extent that digital connectivity and emissions reduction are linked within a country’s policy.

Section 7 sets out the policies we found and reviewed.

How we have presented our findings

By following this methodology, we came up with a series of statements based on the findings from our research. These are presented in Section 5 and 6 with a structure as follows:

• Claim: a conclusion formatted in bold and accompanied by a statement of confidence in our conclusion.
• Argument: concise statements explaining how we arrived at the conclusion.
• Evidence: synthesis of literature in support of our argument.

We have provided a confidence level based on the extent of agreement in the literature and the robustness of evidence. We follow a methodology similar to the one developed by the Intergovernmental Panel on Climate Change for the fifth assessment report and used for the sixth for the consistent treatment of uncertainties.

Figure 2 sets out what constitutes low, moderate, and good confidence in our claims.

Low agreement is where sources do not agree.

Medium agreement is where sources make broadly similar conclusions but the data or evidence they use to support their conclusions are very different.

High agreement is where sources independently make similar conclusions and underlying data are similar despite being independent.

Limited evidence is some evidence available but largely anecdotal and not from recognised peer reviewed sources. Availability of data was low.

Medium evidence is information from peer reviewed sources or official sources.

Robust evidence is a greater volume of information from peer reviewed sources and official sources.

The combination of low agreement and limited evidence provides the lowest level of confidence and the combination of high agreement and robust evidence provides a good level of confidence, with combinations in-between generating moderate confidence.

Our definitions for “limited”, “medium” and “robust” evidence are described in Appendix 1: Detailed Methodology & Approach to the Research. This means that when we say we have “good confidence” in a finding, we are content that there is medium to high agreement in the literature and medium to robust evidence provided for that claim.

Figure 2 – How extent of literature agreement and evidence robustness combines into our stated confidence level.

Investment in digital connectivity and emissions reduction

Digital connectivity, technologies and GHG emissions

We have good confidence in the evidence that, taken on its own, digital connectivity and digital technologies are sources of GHG emissions.

Digital connectivity enables a range of ICT applications. The underlying infrastructure that makes it all work often gives rise to GHG emissions. It depends on the structure of the primary energy and electricity generation sectors of the countries where ICT goods are produced and used, as well as the materials used, such as plastic. These emissions arise across communication equipment such as fixed and mobile broadband, datacentres, cables, and the computers or devices themselves.

The ICT sector is responsible for around 3% to 4% of global greenhouse gas emissions (UNEP, 2021). In Scotland, using domestic output and supply and the environmental input-output model greenhouse gas effects data, the sector contributes around 2% of direct and indirect emissions of carbon dioxide equivalent^[2]. It is also true that regions and countries with higher levels of digital economy development have higher GHG emissions (Wang, et al., 2023). Between 1995 and 2015, GHG emissions of ICT manufacturing have doubled and demand for materials to develop more ICT equipment has quadrupled in the same time period (Itten, et al., 2020).

Besides the high-energy consumption of ICT and electronic equipment, many energy-intensive infrastructures such as backhaul and data centres need to be built to achieve digital connectivity (Lin & Huang, 2023). This means that GHG emissions will increase as a country or region digitalises, up to a certain point (explored further in section 5.1.2). On their own, digital connectivity infrastructures do not support emissions reduction. They provide a mechanism to support decarbonisation of other sectors.

Digital connectivity, emissions reduction and other economic sectors

We have good confidence in the evidence that digital connectivity can only support emissions reduction when paired with other economic sectors.

Digital connectivity is hailed as an enabler for decarbonisation. Despite being a source of GHG emissions themselves, they enable other sectors to digitise in ways that improve productivity and efficiency. The mechanisms by which this is achieved are explored more in Section 6. Essentially, ICT products and services allow traditional industry to change their methodologies to curb GHG emissions (Wang, et al., 2023). Many policymakers hope that the reduction in GHG emissions achieved by these sectors will outweigh the ICT sector’s emissions, as suggested by the European Commission, which states: “If properly governed, digital technologies can help create a climate neutral, resource-efficient economy and society, cutting the use of energy and resources in key economic sectors and becoming more resource-efficient themselves. When implemented under the right conditions, digital solutions have demonstrated significant reduction in greenhouse gas emissions, increased resource efficiency and improved environmental monitoring.” (European Commission, 2022)

Many policies are reliant upon a viewpoint that, on balance, digital innovation to reduce GHG emissions will outweigh the emissions cost of producing and maintaining the necessary ICT networks and components. It is less common for papers to acknowledge there is an initial increase in GHG emissions (particularly from energy use) at the onset of digitalisation. Nor are there many papers discussing the point at which digitalisation starts to reduce emissions.

Lin &Huag is another paper that does address this issue (Lin & Huang, 2023). They state that with increased digitalisation, the resulting increased digital connectivity meant that an energy saving effect could be scaled up across the economy. This marginal energy saving effect exceeds energy consumption of the system – this could be seen as the point at which the GHG savings which result from efficiencies outweigh the GHG emissions from the energy use, production of devices and so on involved in digitalisation. Lin and Huang refer to this point as ‘digitalisation level 0.43’ (Lin & Huang, 2023). The digitalisation level indicator used in the paper is based on data on digital infrastructure, such as internet access and bandwidth, digital application, e.g. fixed and mobile subscription and digital skills and on aggregate ranges from 0 to 100%, using a weighted average for the component elements of the indicator. The paper stipulates that most developed countries have passed the 0.43 point, and it is reasonable to assume that this is the case for Scotland. The assumption of the inverted U-shaped relationship is well tested in the paper, see image in Figure 4. However, the slope of the downward curve is not specified and therefore the applicability of the analysis to Scotland is uncertain, however it is likely to depend on other factors such as the structure of the economy. (Lin & Huang, 2023) make no comment on obsolescence or upgrades to physical equipment.

Figure 4 – Country-level energy intensity against digitalisation; adapted from (Lin & Huang, 2023).

Paired with industrial sectors, there is therefore good evidence that digital connectivity supports emission reductions.

Indirect impacts from ICT use on GHG emissions

We have good confidence in the evidence that indirect impacts from ICT use can be both positive and negative for GHG emissions.

ICT can have both increasing and decreasing effects on GHG emissions. These can be direct or indirect. Direct impacts include energy consumption while the device is in use. Indirect impacts include secondary benefits such as more people being able to work from home and associated reduction in commuting emissions. While digital connectivity can reduce transport through, for example, hosting virtual meetings, some studies postulate that it could also increase emissions from transport by creating the desire to travel to places seen on the internet (Bieser and Hilty, 2018; Hilty and Bieser, 2017; Wang, et al, 2023). Many studies which look at quantifying both, the direct and indirect effects of ICT use, often conclude that the indirect effects are favourable (i.e., reducing GHG emissions) and far outweigh direct effects of energy use. However, these studies often neglect factors such as stimulating transport demand, rebound effects, behaviour changes of humans using these systems, or the embedded carbon of the product or service (Itten, et al., 2020).

The CxC study on emissions impact of home working in Scotland found a small reduction in commuting and office emissions and an increase in home emissions. However, how these changes in emissions balance for each individual defines the net emissions impact from homeworking (Riley, et al., 2021).

Therefore, we conclude from the literature reviewed, that the evidence remains divided in which is more significant: increasing or decreasing effects on GHG emissions.

Emissions reduction and digital technologies that rely on connectivity

We have good confidence in the evidence that the challenge of emissions reduction cannot be met without digital technologies that rely on connectivity.

A great number of the studies and policies we read stated strongly that the challenge of emissions reduction and climate adaptation will not be met without the intervention or use of digital technology and tools (including Royal Society, 2020; Exponential Roadmap Initiative, 2023). The three technologies most often hailed as transformative to all sectors of the economy are 5G, the Internet of Things (IoT, connected devices pooling data often in real time for decision-making) and artificial intelligence (AI, computer-based machine learning).

Many papers assert that digital technology has the potential to assist the transition to a low carbon world, enabling global emission reductions while limiting the emissions created by ICT use (Royal Society, 2020). Some claim that if the currently available digital solutions were used at scale, there would be the potential to reduce GHG emissions in the three highest-emitting global sectors (energy, materials, and mobility) by 20% by 2050 (World Economic Forum, 2022). There was no concrete evidence in these papers that connectivity would enable these goals to be met, only claims.

Sectors that will benefit the most from digital connectivity

We have good confidence in the evidence that suggests that the sectors that will most benefit from digital connectivity are industrial in nature and will vary from country to country.

The sources above state the energy sector would benefit the most from digital solutions. In Scotland, the energy sector is the fourth highest emitter at 4.9 million tonnes of CO₂e in 2021 (Scottish Government, 2021). With regards to electricity in particular, the complexity and scale of integrating more renewable energy generation and increasing the distribution capacity of the electricity grid will not be possible without digital technologies (Energy Systems Catapult, 2023), especially with increasing requirements for data sharing and more effective system planning and operation. Renewable generation is intermittent and requires active grid management. Digital technologies can help balance the supply side (electricity producers) and the demand side (consumers) management for a more agile, stable and reliable electricity grid for industrial, commercial and household users.

The industry sector is globally responsible for 37% of total final energy consumption and about 20% of GHG emissions. In Scotland, industrial processes and business account for 20% of CO₂e in 2021.

As described in 5.1.4, digital technologies will be important to manage the supply and demand of large industrial energy users in a system with diverse sources and feedstock (European Commission, 2022).

The effective use of these digital technologies relies on connectivity. Without it, none of the claims explored in literature can come to fruition.

The emissions intensity of digital connectivity

We have moderate confidence in the evidence that the lowest emissions form of digital connectivity is currently fibre.

One study has found that fibre is the most energy efficient technology for broadband access networks, compared with the family of Direct Subscriber Line (DSL) technologies delivering network access through voice lines and Data Over Cable Service Interface Specification (DOCSIS) which delivers network access through cable television (European Commission, 2020). Studies brought together by Europacable also demonstrated that fibre is the most energy efficient technology for internet access compared to microwave, millimetre wave, copper, satellite, and laser (Europacable, 2022). This is because there are fewer intermediate devices and amplifiers, and glass fibre is largely passive and requires little energy to function.

Although 5G networks are touted to be more energy efficient than 4G networks, the overall energy and emissions impacts are still uncertain. 5G antennas use three times as much energy as a 4G antennae, and a higher network density will be required (International Energy Agency, 2023). Literature on the energy use of 5G is found to be dominated by small-scale, single technology assessments. Embedded energy use and indirect energy use effects are largely overlooked (Williams, et al., 2022).

Satellite broadband is a less disruptive approach to connect rural areas to the internet, requiring less work on land to lay cables, however the GHG emissions of Low Earth Orbit (LEO) satellites are only recently being explored. An October 2023 study estimates worst-case emissions to be 31-91 times higher than equivalent terrestrial mobile broadband (Osoro, et al., 2023). It is unclear whether the terrestrial mobile broadband used in this comparison is sufficiently representative of a rural broadband connection or fixed broadband.

The World Bank identified a strong statistical connection between the capacity of the network (the number of users and the amount of data they require) and the level of GHG emissions (World Bank, 2022). Fibre has a high data capacity, but is only one component of a network. There are other critical parts of the network infrastructure such as data centres which drive this trend.

The most efficient data centres and emissions reduction

We have moderate confidence in the evidence that hyperscale and co-located data centres are the most efficient and offer a high potential for reducing emissions.

Data centres and data transmission networks account for approximately 1-1.5% of global electricity use, making them responsible for 1% of energy-related GHG emissions (IEA, 2023). Rapid growth in demand at large data centres has resulted in a substantial increase in energy use in this sector, growing 20-40% annually over the past several years (IEA, 2023). As a result of this, the International Energy Agency (IEA) has given data centres the “More Efforts Needed” rating, which means that data centres need to do more to align to the IEA’s Net zero by 2050 Scenario. Progress is assessed at the global level against the IEA’s net zero by 2050 Scenario Trajectory for 2030 (IEA, n.d.), and recommendations are provided on how they can get “on track” with this pathway. Recent trends on reducing the environmental impacts of data centres have generally been in the right direction to match this trajectory; however, without acceleration it will fall short (IEA, n.d.).

The carbon footprint of a data centre is affected by three factors:

• electricity consumption
• water consumption
• lifetime of the equipment.

When analysing these factors, it can be seen in Table 1 that hyperscale and co-located data centres are far more efficient (including accounting for water consumption) than internal data centres. This is driven by better energy utilisation, more efficient cooling systems and increased workloads per server (Lavi, 2022). As a result, they are less carbon intensive per tonne of GHG emissions per workload than internal data centres.

Table 1 – Impacts of hyperscale, colocation and internal data centres. Adapted from Lavi, 2022.

With Scotland’s electricity maintaining a grid intensity of below 50 grams of CO2e per kilowatt hour delivered across 2017-2020 (Scottish Government, 2023), as opposed to the UK average of 149 grams of CO2e per kilowatt hour delivered in 2023 (National Grid, 2023), the emissions intensity of datacentres in Scotland is likely to be significantly lower.

Summary

Investment in digital connectivity can support emission reductions for those primarily industrial sectors which benefit from efficiency. ICT reliant on digital connectivity is supposed to help meet challenges of emission reduction although there is a lack of evidence for these claims.

Digital technology is a source of emissions in and of itself which tends to be overlooked.

As a result of these, we cannot say for certain whether the indirect effects of digitalisation (e.g., saved emissions from home working, see Section 5.1.3) will reduce overall emissions.

Climate adaptation, just transition and investment in digital connectivity

This section sets out the evidence we have been able to find that meets our criteria. Although just transition and adaptation are important policy areas, the steering group wished to focus primarily on Net zero targets and emissions reduction with this research. The Steering Group also emphasised the need to only include information and projects that were current and operating, not theoretical.

The resulting research has emphasised how these concepts are new and emerging. As novel as the concepts such as just transition and adaptation are, the evidence base is being created. As the situation progresses, more and more evidence will be developed to revise the assertions below.

Digital connectivity and adaptation strategies

With the evidence we have been able to find that meets our criteria, we have low confidence in the evidence that digital technologies, which rely on connectivity, support climate adaptation strategies.

ICT technology is an integral component of many proposed mitigation measures (Dwivedi, et al., 2022), but less so for adaptation. Mitigation is reducing and stabilising levels of GHG emissions; adaptation is adapting to life in a changing climate. It is considered by many that digital connectivity and the ability to communicate and share data will be important for adaptation, especially in rural communities.

The example we have been able to find include the European Commission Farmers Measure Water project, where one farmer described how decisions need to be made quickly: “We need fast internet in rural areas because a lot of farmers and water authorities have to make decisions on an hourly basis. If we take a measurement and only see the results in a week’s time, it is too late: the problem has already occurred. If you have fast internet, you have direct access to your data and can decide on the spot what to do” (European Commission, 2022).

Digital technologies can also support climate-resilient agriculture by helping farmers assess weather forecasts and mitigate impacts on crop yields and productivity (United Nations Development Programme, 2023).

In terms of what the ICT sector itself is doing to adapt to climate change, in 2018 TechUK submitted a report to the Department for Environment, Food and Rural Affairs (Defra) on behalf of the ICT sector outlining how the sector intends to adapt to climate change. Within it, they state that ICT infrastructure including connectivity has unique characteristics that make it more resilient (TechUK, 2018). These include:

• Asset life is relatively short. So more resilient assets can be deployed as part of the normal replacement cycle.
• There is built in redundancy in ICT infrastructures so that if same proportion of ICT assets is damaged or affected by climatic events, there are backups.
• Technology development is fast particularly around threats.

The first two of these are in direct conflict with reducing the direct GHG emissions of ICT and digital connectivity delivery. Programmes that mandate less redundancy or longer asset life may affect the ICT industry’s ability to adapt to climate change. The final point reinforces the ICT sector’s claim that it will innovate out of problems, without evidence to support it.

Digital connectivity and a just transition

With the evidence we have been able to find that meets our criteria, there is moderate evidence that digital connectivity supports a just transition.

There is debate in the literature over whether digital connectivity supports a just transition. Views are largely that it may help when accompanied by strong policy. One study shows that the Just Transition Score may increase as digitalisation increases (Wang, et al., 2022), but this could be a correlation rather than indicating causation. The mechanisms are also little explored: for example, one paper sets out that the digital economy indirectly improves just transition by increasing the level of human capital and financial development (Wang, et al., 2022). There is no further investigation into how this takes place.

There are a few points of information related to how digital connectivity relates to just transition:

• People with low and medium income are more vulnerable to the impacts and costs of economic transitions. Transitions may include job automation, increasing need for access to digital solutions and digital public services, higher energy and food prices, or transport poverty (European Commission, 2022).
• Some articles link which digital solutions can be justice and equity enablers. Examples include
• smart energy management and decentralised and distributed energy production and sale (United Nations Development Programme, 2023)
• an easy-to-use and reliable public transport system that improves mobility for all (United Nations Development Programme, 2023).

This indicates that digitalisation may enhance a just transition.

• Collecting data and use of data is highlighted as important for justice and social good (Friends of Europe, 2021). Many smart solutions require a level of monitoring to maintain the efficiency of the service. Regulation, oversight and controls on appropriate data collection and use will be key. This indicates that policy implemented through digital solutions may become increasingly important in relation to a just transition.
• Across many policies, a just transition is also linked to skills development, with the Climate Change Committee (2023) stating digital skills as a fundamental enabler of net zero. The Welsh Government state the need to “prevent existing labour market inequalities being carried through into the new net zero and digital economies” (Welsh Government, 2022), recognising that employers are actively seeking employees with digital skills.

Summary

The evidence base related to digital connectivity and adaptation in relation to concrete real-world examples is very limited among the literature we have reviewed.

There is no direct evidence to date that investment in digital connectivity supports a just transition, but there are many suggestions for mechanisms by which it might influence a just transition. One of these mechanisms is skills development.

Key mechanisms by which digital connectivity influences emissions reduction

Digital connectivity, primary needs for travel and GHG emissions

We have moderate confidence in the evidence that digital connectivity can reduce primary needs for travel, although we have low confidence to whether this reduces GHG emissions in total.

The assumption that digital can replace physical goods or services completely and therefore avert emissions underpins a great deal of policies supporting digitalisation. It is true digitalisation can substitute certain products or GHG generating activities, such as an e-reader capable of displaying hundreds of books or videoconferencing and telework replacing physical travel. Methodologies to measure the true GHG emissions savings of these substitutes are not rigorous or consistent (Hook, et al., 2020). At the same time, demand for travel is still growing (Itten, et al., 2020; Statista, 2023).

Differences in methodology, scope and assumptions make it difficult to estimate average energy savings of working from home versus working in the office. Rebound effects and home energy use is often overlooked, and where they are included, studies find smaller savings (Harvard Business Review, 2022). Rebound effects include increased non-work travel and more short trips. For example, Harvard Business Review found that a decrease in vehicle miles driven is accompanied by a 26% increase in the number of trips taken (Harvard Business Review, 2022). Trips which would have been taken anyway, such as taking children to school, are also not included.

In the report “Emissions impact of home working in Scotland” concludes that working from home leads to a reduction in commuting and office emissions and an increase in home emissions. How these changes in emission balance out for each individual defines whether their net impact from home working will be positive or negative. The authors state that across their scenarios, the overall impact on emissions will be small (Riley, et al., 2021).

Due to the ambiguities in methodologies, the actual or potential GHG emission reductions of teleworking remain uncertain. Economy-wide savings are likely to be modest (Riley, et al., 2021), and in many circumstances could be negative or non-existent (Hook, et al., 2020).

Public sector digital technology use

We have good confidence in the evidence that the public sector is using technology to solve problems linked to sustainability – but the evidence is not accompanied by reports on the effects of technology use on GHG emissions.

The mechanism of reducing GHG emissions by public sector authorities using digital technology is mainly around energy efficiency. Many documents include a wealth of examples of cities using technology to save energy (European Commission, 2022) – but the GHG emissions associated with implementation or life cycle of this equipment have not been considered.

Main source of emissions from digital connectivity and associated ICT

We have good confidence in the evidence that the largest proportion of emissions from digital connectivity and associated ICT equipment comes from waste management after use.

The ICT sector tends to focus on energy use of their products as the largest influence on the carbon footprint. Therefore, there are calls for energy sources to be decarbonised (Ericsson, n.d.). Independent academic studies are more likely to conclude that the carbon footprint or life cycle emissions of a digital product is dominated by electronic waste or e-waste (Itten, et al., 2020 and Dwivedi, et al., 2022). Figure 5 shows the result of a study into video streaming from device purchase, which identifies that 78% of the GHG emissions are from e-waste (Itten, et al., 2020). This illustrates our claim that the largest proportion of emissions from the use of devices comes from waste management after use (please note, extraction of materials and production was not included in this study, which focused on impacts from consumer behaviour).

Figure 5 – Proportion of GHG emissions from the use case of streaming videos (Itten, et al., 2020).

In its 2020 report on e-waste, the International Telecommunication Union (ITU) estimates that 15 million tonnes of CO₂e were averted by the recovery of iron, aluminium, and copper from processed e-waste (International Telecommunications Union, 2020). The ITU report also disclosed that less than 18% of all e-waste can be accounted for, meaning that almost 83% of e-waste is likely not properly disposed of. The sector’s emission reductions may be limited because of the uncertain fate of e-waste.

Human behaviour and digital connectivity

We have good confidence in the evidence that human behaviour plays a role in digital trends, rebound effects, and responsible use of digital connectivity.

Academic papers point out that whilst digital technologies are becoming more efficient individually, the higher demand for computing power, storage capacities, transmitted data and devices per person is systematically compensating for this progress (Aebischer & Hilty, 2015) (Hischier & Wager, 2014). This trend can be partially explained by rebound effects regarding time, volume, weight, and price (Itten, et al., 2020), but also human behaviour. Technology can act as a fashion or wealth statement, with the average person owning more and more connected devices such as smartphones and smart watches. These are often replaced with the latest model far sooner than is required on a technology replacement cycle (Itten, et al., 2020).

Future ICT sector energy consumption reduction

We have moderate confidence in assertions that the ICT sector will continue to innovate to reduce energy consumption.

Deployment of next generation low-power chips and more efficient connectivity technologies (5G and 6G, networks powered by artificial intelligence) is repeatably hailed as the way to reduce the overall footprint of ICT (European Commission, 2022).

Each switch to new standards or technologies requires a massive replacement of equipment. For example, 5G and 6G will require users to replace equipment, due to lack of backwards compatibility of existing smartphones, tablets, and computers. Also, as a growing fraction of products become smart or part of the Internet of Things (IoT), overall resource demand could decrease in theory. In practice, the opposite happens because software-controlled objects are also prone to software-induced obsolescence (Kern et al., 2018; NGI, 2020). While each new model is likely to be more energy efficient than the last, and while smaller smart IoT devices may not consume large amounts of energy in use, 85-95% of their lifecycle energy footprint is created in production. The sheer number and variety make them particularly susceptible to obsolescence once software or hardware support runs out (NGI, 2020).

The fast-evolving nature of digital technologies and the possible sharp increase in digitally enabled services is likely to reinforce the ICT sector’s growing emissions (European Commission, 2023). The European Commission has set out that unless digital technologies are made more energy-efficient, their widespread use will increase energy consumption.

Summary

The key mechanisms that ICT and digitalisation can reduce GHG emissions described by literature include replacing the need to travel, although there is evidence that these savings may not be as high as first thought. The largest source of emissions from ICT equipment is after use, as e-waste, something that changing standards and upgrading systems can increase. Human behaviour plays a role in the resulting emissions from ICT and digitalisation.

Key examples of digital connectivity policies

We studied international policies associated with digital connectivity and decarbonisation, adaptation and just transition in 10 countries, selected based on the methodology in Section 4, to gather important contextual information for Scotland. The degree to which each country links their digital goals and strategy has been given a score, with five representing explicit mention of the GHG or carbon impacts of increased digitalisation, and one representing no mention or linking of decarbonisation within the policy, see Appendix 1 for further detail on the scoring.

Appendix 2: Summary of digital policies across 10 countries provides further detail on individual policies.

Decarbonisation impact of these policies

Policy measures to support emission reductions, adaptation or just transition

Few of the countries we studied for this research have set policy measures designed to support emission reductions, adaptation, or just transition in direct association with digital technologies.

No evidence of impact has been identified during this review. This does not prove a lack of progress or attention. There are other jurisdictions outside the scope of this research which may have evidence of policy impact. An example is the European Union Declaration on Digital Rights and Principles. This promotes digital products and services with a minimum negative impact on the environment and on society, as well as digital technologies that help fight climate change (European Commission, n.d.).

Sustainability considerations of using ICT and digital infrastructure

We have good confidence that European countries are starting to look at the sustainability considerations of using ICT and digital infrastructure.

The European Commission is leading the way in setting net zero or climate neutrality targets for certain elements of ICT infrastructure. In the “Fit for the Digital Age Strategy”, the Commission sets ambitious goals such as the climate neutrality of data centres in the EU by 2030 (European Commission, 2023). Measures to improve the circularity of digital devices and to reduce electronic waste include the Right to Repair Directive (European Commission, 2023) and the recently issued eco-design criteria for mobile phones and tablets (European Commission, 2023). These should have a corresponding positive impact on lifecycle emissions from digital technologies. Efforts are also ongoing to develop low-energy chips under the European Processor Initiative (European Processor Initiative, 2023).

The European Commission is starting to look at policy and governance around ICT direct and indirect emissions: “Until recently, the digital transition progressed with only limited sustainability considerations. To diminish adverse side effects and deliver its full potential for enabling environmental, social, and economic sustainability, the digital transition requires appropriate policy framing and governance” (European Commission, 2022).

“Digitalisation is an excellent lever to accelerate the transition towards a climate-neutral, circular, and more resilient economy. At the same time, we must put the appropriate policy framework in place to avoid adverse effects of digitalisation on the environment.” Svenja Schulze, Federal Minister for the Environment, Nature Conservation and Nuclear Safety of Germany (European Council, 2020).

Policy development programmes for datacentre best practice

We have good confidence that countries are starting to drive policy for data centre best practice.

In Estonia, the government has moved to the use of the Estonian Government Cloud (Riigipilv) for ‘Infrastructure, Platform and Software as a Service.’ Analysis of this pointed out that eliminating in-house servers and server rooms, instead relying on cloud services via data centres, offers the biggest potential for reducing emissions (Vihma, 2022). Data centres of the Estonian Government also use the ISO50001 energy management certification.

In Germany^[3], the Government launched the Green IT initiative in 2008 to reduce the energy consumption and GHG emissions of its ICT operations. One objective set for the 2022 to 2027 phase of Green IT initiative includes that ‘main’ data centres (>100kW ICT load) owned by the government should meet the German Federal Government Blue Angel criteria for energy efficient data centres (Blume & Keith, 2023). From the start of the initiative, energy consumption has fallen by 49% from 649.65 GWh in 2008 to 334.54 GWh in 2021. This reduced consumption resulted in budgetary savings of €546 million (Blume & Keith, 2023).

In Denmark, the Agency for Digital Government examined which environmental requirements the public sector can include in tenders for data centres and concluded that the EU’s Green Public Procurement criteria is the most appropriate to use (Agency for Digital Government, n.d.).

In China, the Government has called for an average Power Usage Effectiveness of 1.25 in the east and 1.2 in the west of the country as part of its Eastern Data and Western Computing Project. Major cities now have maximum Power Usage Effectiveness requirements for new data centres, including Beijing (1.4), Shanghai (1.3) and Shenzhen (1.4) (IEA, 2023). Power Usage Effectiveness is the metric used to determine the energy efficiency of a data centre.

The private sector is also taking action to reduce the environmental impacts. In January 2021, date centre operators and industry association in Europe launched the Climate Neutral Data Centre Pact, pledging to make data centres climate-neutral by 2030 with intermediate (2025) targets for PUE and carbon-free energy (IEA, 2023).

Baseline assessments and monitoring

We have good confidence that there is no current framework for baseline assessment or monitoring of the environmental impact of increased digitalisation which also considers the indirect benefits and potential rebound effects.

There is a need to develop consistent metrics to measure the impact of technology on the environment (United Nations Environment Programme, 2021).

The European Commission identifies a need for a science-based assessment methodology on the ‘net environmental impact’ of increased digitalisation that consider both the benefits and the possible rebound effects (European Commission, 2022). The Commission has therefore launched dedicated research and innovation initiatives, saying that it will launch a project under Horizon Europe, to develop a methodology and common indicators for measuring the footprint of ICT (European Commission, 2023). In the UK, Building Digital UK also recognises this as a gap and will be reporting on environmental benefits of their interventions (Building Digital UK, 2023). Similarly, EU Member States are collaborating on the Toulouse call for a Green and Digital Transition in the EU. This looks to monitor the impact of digitalisation on the environment and contribute to the development of measurement tools (Presidence Francaise, 2022).

While the framework does not exist to quantify the full scope of direct and indirect effects, a number of standards exist for some elements.

Global standards to support carbon accounting in the ICT sector

There are global standards that can support carbon accounting (the method used to calculate a carbon footprint) in the ICT sector.

The main ones recognised and accepted by ICT bodies are:

• Greenhouse Gas Protocol ICT Sector Guidance. This builds on the internationally accepted GHG Protocol Product Life Cycle Accounting and Reporting Standard (GeSI and Carbon Trust, 2017).
• Recommendation ITU-T L.1470 (01/2020) (International Telecommunications Union, 2020).
• “Guidance for ICT companies setting science-based targets.” (Science Based Targets Initiative, 2022).

In summary

Our literature review has found no good examples of international experience of applying standard carbon accounting in the ICT sector, and this gap is recognised at the European Union level. Standards exist at the corporate or product level which could be adapted.

Conclusions

This work is the start of a process. As a rapid review, we were able to quickly identify information which fit our criteria, but there may be areas we have missed. Digital connectivity infrastructure and ICT are highly interconnected and overlapping with our behaviours and geography, and so this exercise has also highlighted we are having to pull together disparate pieces of information, research and case studies to try to come to conclusions. A key challenge is in understanding the “net” picture – there are disparate sources citing the means by which digital connectivity can impact on emissions, but it is not possible to combine this evidence to form a complete picture.

We were asked to research five key questions and found the following answers:

• To what extent is there evidence that investment in digital connectivity can support emissions reduction, climate adaptation and a just transition?

We have found mixed evidence of the decarbonisation impact, adaptation and just transition of digital connectivity. The sector produces direct emissions from energy consumption and generation of e-waste. This is despite the possibility that it can reduce indirect emissions through increasing efficiency and behaviour changes such as reduced travel linked to working from home. Studies point out a need for a holistic approach in calculating GHG emissions of the ICT sector, including rebound effects and emissions from the end-of-life. This would ensure indirect emissions and emissions from end-of-life are fully accounted for.

Investment in digital connectivity can support emissions reduction for those primarily industrial sectors that benefit most from efficiency. ICT technology and digitalisation can and does reduce GHG emissions in other industries. Heavy industry and the energy sector would benefit the most from digitalisation. ICT reliant on connectivity is supposed to help meet challenges of emissions reduction although there is a lack of evidence for these claims.

The ICT sector is a source of GHG emissions, which tends to be overlooked. It is our view that the reduction in indirect GHG emissions (largely driven by digitalising and making other sectors more efficient) does not negate the need to reduce the ICT sector’s direct impacts from energy consumption and generation of e-waste.

While the ICT sector focuses on the emissions associated with energy use, which is not insignificant, we have good confidence that the GHG emissions associated with e-waste are of growing concern internationally in terms of reaching climate goals. It is uncertain whether ICT’s GHG emissions reduction potential in other sectors can actually outweigh its direct emissions. It gives us only moderate confidence that the ICT sector can help reduce more emissions than are inherent in the manufacture, use and disposal of the equipment used to achieve those savings.

There is a great deal of speculation that digital technologies have the potential to aid adaptation to climate challenges, especially in rural areas, though with few concrete examples. While there is no direct evidence that investment in digital connectivity supports a just transition, there are many suggestions for mechanisms by which it might influence a just transition. One of these mechanisms is skills development, which is also recognised as a key enabler for net zero. Digital connectivity and ICT are capable of doing both good and bad, either addressing or exacerbating existing inequalities, as well as questions around access to connectivity and skills. Studies repeat the need for strong policy in this area.

• If so, what are the key mechanisms by which this could occur (for example, reduction in travel, investment in green data centres or other mechanisms suggested in the evidence)?

The key mechanisms by which ICT and digitalisation can reduce GHG emissions, as described by literature, include replacing the need to travel, although there is evidence that these savings may not be as high as first thought. The largest source of emissions from ICT equipment is e-waste, which changing standards and upgrading systems can increase. Human behaviour plays a role, either positive or negative, in the emissions from ICT and digitalisation.

• What are key examples of existing policies (in Scotland, such as in local authorities, the UK and/or international examples from comparable countries) designed to support emission reductions, adaptation and/ or just transition through digital connectivity? Is there any evidence for the impact of such policies?

No evidence of impact has been identified during this review. Many of the countries analogous to Scotland have policy that mentioned digitalisation as an enabler or essential piece of their decarbonisation, climate change or net zero agenda. None of them were accompanied by evidence of impact of their policies. This does not prove a lack of progress or attention. There are other countries outside the scope of this research that may have evidence of policy impact.

• What are the different options suggested within the literature for Scotland to provide a baseline assessment of, and monitor carbon emissions from digital infrastructure, technologies, and associated activities?

There are no good examples of what other countries are doing, and this gap is recognised at the European Union level. Standards exist at the corporate or product level, which could be adapted.

• What are the other key gaps in existing knowledge where further research is required to support digital connectivity and Scotland’s climate change goals?

Gaps include a need for a baseline assessment methodology, direct studies exploring the questions asked in this research and a consistent methodology for calculating direct and indirect emissions from ICT and digitalisation.

Gaps identified by this research

We have used specific search criteria and search words and applied them in google and google scholar. On basis of this search, we have found the following evidence gaps:

• There is no active study that has been found within this review that investigates whether investment in digital connectivity directly results in GHG emissions reduction.

There are varying approaches to quantifying direct and indirect emissions of ICT, with no academic or sector wide consensus.

There are different approaches and methodologies for calculating and comparing the GHG emissions of digital and non-digital practices and solutions, for example online versus in-person events. As an example, (Hook, et al., 2020) outlines that working from home evaluations should encompass the following:

• energy footprint
• transportation footprint
• technology footprint
• waste footprint.

The evidence we have found to investigate whether digital connectivity contributes to a just transition and the key mechanisms by which this occurs is not conclusive or good quality.

The ICT sector and literature focus on emissions reduction, with climate adaptation either an afterthought or future looking, with few real-world examples.

Case studies of digital technologies saving money, power or water in municipalities focus on the GHG emissions reduced or averted, with no acknowledgment of rebound effects, which literature states is important.

The GHG emissions associated with the collection and use of data, which is deemed to be necessary to digitalisation, are opaque and limited to specific studies on data centres. For example, the full lifecycle of the Internet of Things is not explored in the literature.

Lack of evidence of policy to address GHG emissions of e-waste e.g. from refrigerants leaking GHG.

Lack of evidence of policy to address the embedded GHG emissions from extraction of raw materials and production of the ICT equipment.

Lack of best practice for measuring, monitoring and assessing the GHG footprint of electronic communications services. The European Commission is also looking to develop this in a Horizon Europe project.

Bibliography

Aebischer & Hilty, 2015. The Energy Demand of ICT: A Historical Perspective and Current Methodological Challenges. In: ICT Innovations for Sustainability. s.l.:s.n., pp. 71-103.

Agency for Digital Government, n.d. Digital Green Transition. [Online]
Available at: https://en.digst.dk/digital-transformation/digital-green-transition/
[Accessed 30 11 2023].

Agency for Digital Government, n.d. Digital Green Transition. [Online]
Available at: https://en.digst.dk/digital-transformation/digital-green-transition/

Blume, A. & Keith, M., 2023. Germany’s green IT initiative is cutting energy consumption and GHG emissions in its public sector. [Online]
Available at: https://blogs.worldbank.org/governance/germanys-green-it-initiative-cutting-energy-consumption-and-ghg-emissions-its-public
[Accessed 30 11 2023].

Building Digital UK, 2023. Building Digital UK – environmental resource guide August 2023. [Online]
Available at: https://www.gov.uk/guidance/building-digital-uk-environmental-resource-guide-august-2023

Centre for Digital Public Services, 2022. Decarbonising Wales with digital technology. [Online]
Available at: https://digitalpublicservices.gov.wales/our-work/decarbonising-wales-digital-technology

Digital.Govt.NZ, 2022. Action Plan for the Digital Strategy for Aotearoa. [Online]
Available at: https://www.digital.govt.nz/digital-government/strategy/digital-strategy-for-aotearoa-and-action-plan/action-plan-for-the-digital-strategy-for-aotearoa/

Digital.Govt.NZ, n.d. The Digital Strategy for Aotearoa. [Online]
Available at: https://www.digital.govt.nz/digital-government/strategy/digital-strategy-for-aotearoa-and-action-plan/the-digital-strategy-for-aotearoa/

Digitalserings Partnerskabet, 2021. Visions and recommendations for Denmark as a digital pioneer. [Online]
Available at: https://en.digst.dk/media/24796/visions-and-recommendations-for-denmark-as-a-digital-pioneer-danish-government-digitisation-partnership.pdf

Dreyfuss, Gadson, Riding & Wang, n.d. The IT Productivity Paradox. [Online]
Available at: https://cs.stanford.edu/people/eroberts/cs201/projects/productivity-paradox/background.html
[Accessed 2023].

Dwivedi, et al., 2022. Climate change and COP26: Are digital technologies and information management part of the problem or the solution? An editorial reflection and call to action. International Journal of Information Management, Volume 63.

Kern et al., 2018. Sustainable software products—towards assessment criteria for resource and energy efficiency. Future Generation Computer Systems, Volume 86, pp. 199-210.

Energy Systems Catapult, 2023. Innovating to Net Zero: Digitalisation for Net Zero. [Online]
Available at: https://es.catapult.org.uk/report/innovating-to-net-zero-digitalisation-for-net-zero-emissions/

Ericsson, n.d. ICT and the climate. [Online]
Available at: https://www.ericsson.com/en/reports-and-papers/industrylab/reports/a-quick-guide-to-your-digital-carbon-footprint

Europacable, 2022. Fibre: the most energy-efficient solution to Europe’s bandwidth needs, White Paper. [Online]
Available at: https://europacable.eu/wp-content/uploads/2022/07/Europacable-Whitepaper-on-Energy-Efficiency-of-Fiber-networks-05-July-2022.pdf

European Commission, 2020. Fibre is the most efficient broadband technology. [Online]
Available at: https://digital-strategy.ec.europa.eu/en/library/fibre-most-energy-efficient-broadband-technology

European Commission, 2022. Strategic Foresight Report 2022 Twinning the green and digital transitions in the new geopolitical context. [Online]
Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:52022DC0289

European Commission, 2022. Study on Greening Cloud Computing and Electronic Communications Services and Networks: Towards Climate Neutrality by 2050. [Online]
Available at: https://digital-strategy.ec.europa.eu/en/library/study-greening-cloud-computing-and-electronic-communications-services-and-networks-towards-climate

European Commission, 2023. Circular economy: New rules to make phones and tablets more durable, energy efficient and easier to repair, enabling sustainable choices by consumers. [Online]
Available at: https://ec.europa.eu/commission/presscorner/detail/en/ip_23_3315

European Commission, 2023. Report on the state of the Digital Decade 2023. [Online]
Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52023DC0570

European Commission, 2023. Right to repair: Commission introduces new consumer rights for easy and attractive repairs. [Online]
Available at: https://ec.europa.eu/commission/presscorner/detail/en/ip_23_1794

European Commission, n.d. European Digital Rights and Principles. [Online]
Available at: https://digital-strategy.ec.europa.eu/en/policies/digital-principles

European Council, 2020. Digitalisation for the benefit of the environment: Council approves conclusions. [Online]
Available at: https://www.consilium.europa.eu/en/press/press-releases/2020/12/17/digitalisation-for-the-benefit-of-the-environment-council-approves-conclusions/

European Processor Initiative, 2023. Framework partnership agreement in European low-power microprocessor technologies. [Online]
Available at: https://www.european-processor-initiative.eu/

European Union Digital Skills and Jobs Platform, 2023. Estoanian Digital Agenda 2030. [Online]
Available at: https://digital-skills-jobs.europa.eu/en/actions/national-initiatives/national-strategies/estonia-estonian-digital-agenda-2030

European Union Digital Skills and Jobs Platform, 2023. Finland: a snapshot of digital skills. [Online]
Available at: https://digital-skills-jobs.europa.eu/en/latest/briefs/finland-snapshot-digital-skills

Eurostat, n.d. Glossary:Global-warming potential (GWP). [Online]
Available at: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Glossary:Global-warming_potential_(GWP)

Exponential Roadmap Initiative, 2023. Exponential Climate Roadmap. [Online]
Available at: http://www.exponentialroadmap.org/

Finnish Government, 2022. Government report on the digital compass sets the course for Finland’s digital transformation. [Online]
Available at: https://valtioneuvosto.fi/en/-/10623/government-report-on-the-digital-compass-sets-the-course-for-finland-s-digital-transformation

Freitag, et al., 2021. The real climate and transformative impact of ICT: A critique of estimates, trends, and regulations. Patterns, Volume 3, Issue 8.

Friends of Europe, 2021. Connected Europe: A digital brand for a just transition. [Online]
Available at: https://www.friendsofeurope.org/insights/connected-europe-a-digital-brand-for-a-just-transition/

GeSI and Carbon Trust, 2017. ICT Sector Guidance built on the GHG Protocol Product Life Cycle Accounting and Reporting Standard. [Online]
Available at: https://ghgprotocol.org/sites/default/files/2023-03/GHGP-ICTSG%20-%20ALL%20Chapters.pdf

Global Enabling Sustainability Initiative, 2020. Portugal joins Digital With Purpose to unite sustainability with technology. [Online]
Available at: https://gesi.org/posts/portugal-joins-digital-with-purpose-to-unite-sustainability-with-technology

Government of Canada, 2022. Canada’s Digital Ambition 2022. [Online]
Available at: https://www.canada.ca/en/government/system/digital-government/government-canada-digital-operations-strategic-plans/canada-digital-ambition.html

Government of Iceland, 2021. Policies and Strategies. [Online]
Available at: https://www.government.is/topics/information-technology/policies/

Harvard Business Review, 2022. Is Remote Work Actually Better for the Environment?. [Online]
Available at: https://hbr.org/2022/03/is-remote-work-actually-better-for-the-environment

HCL Tech, 2021. Artificial Intelligence, defined in simple terms. [Online]
Available at: https://www.hcltech.com/blogs/artificial-intelligence-defined-simple-terms

Hischier, R. & Wager, P., 2014. The Transition from Desktop Computers to Tablets: A Model for Increasing Resource Efficiency?. In: ICT Innovations for Sustainability. Advances in Intelligent Systems and Computing. s.l.:s.n.

Hook, A., Court, V., Sovacool, B. K. & Sorrell, S., 2020. A systematic review of the energy and climate impacts of teleworking. Environmental Research Letters, 15(9).

HP Enterprise, n.d. Workload. [Online]
Available at: https://www.hpe.com/us/en/what-is/workload.html

IBM, 2023. What is artificial intelligence (AI)?. [Online]
Available at: https://www.ibm.com/topics/artificial-intelligence

IEA, 2023. Data Centres and Data Transmission Networks. [Online]
Available at: https://www.iea.org/energy-system/buildings/data-centres-and-data-transmission-networks

IEA, n.d. Net Zero Emissions by 2050 Scenario (NZE). [Online]
Available at: https://www.iea.org/reports/global-energy-and-climate-model/net-zero-emissions-by-2050-scenario-nze

IEA, n.d. Tracking Clean Energy Progress 2023. [Online]
Available at: https://www.iea.org/reports/tracking-clean-energy-progress-2023#methodology
[Accessed 08 12 2023].

International Energy Agency, 2023. Data Centres and Data Transmission Networks. [Online]
Available at: https://www.iea.org/energy-system/buildings/data-centres-and-data-transmission-networks

International Telecommunications Union, 2020. ITU-T Recommendations. [Online]
Available at: https://www.itu.int/ITU-T/recommendations/rec.aspx?rec=14084

International Telecommunications Union, 2020. The Global E-waste Monitor 2020. [Online]
Available at: https://www.itu.int/en/ITU-D/Environment/Documents/Toolbox/GEM_2020_def.pdf

Itten, et al., 2020. Digital transformation—life cycle assessment of digital services, multifunctional devices and cloud computing. The International Journal of Life Cycle Assessment, Volume 25, p. 2093–2098.

Lavi, H., 2022. Measuring greenhouse gas emissions in data centres: the environmental impact of cloud computing. [Online]
Available at: https://www.climatiq.io/blog/measure-greenhouse-gas-emissions-carbon-data-centres-cloud-computing
[Accessed 30 11 2023].

Lin, B. & Huang, C., 2023. Nonlinear relationship between digitization and energy efficiency: Evidence from transnational panel data. Energy, Volume 276.

Ministry of Finance Finland, n.d. Sustainable Growth Programme for Finland – boosting reforms and investments. [Online]
Available at: https://vm.fi/en/sustainable-growth-programme-for-finland

National Grid, 2023. How much of the UK’s energy is renewable?. [Online]
Available at: https://www.nationalgrid.com/stories/energy-explained/how-much-uks-energy-renewable#:~:text=20%20April%202023%20saw%20the,achieved%20on%2018%20September%202023.

NGI, 2020. Internet of Waste: The case for a green digital economy. [Online]
Available at: https://media.nesta.org.uk/documents/Internet_of_Waste_-_The_case_for_a_green_digital_economy_1.pdf

Nordic Council of Ministers, n.d. Report: A study of the digital green transition in the Nordic Baltic countries. [Online]
Available at: https://en.digst.dk/media/27140/bilag-3a-study-of-the-digital-green-transition-in-the-nordic-baltic-countries.pdf

Norway and European Union, n.d. Norway EU Green Alliance. [Online]
Available at: https://www.regjeringen.no/contentassets/debc1b0c0a2f47d1b77beb96d896cf45/20230420eu-norway-green-alliance-final.pdf

Norway Grants, n.d. Estonia – Green ICT. [Online]
Available at: https://eea.innovationnorway.com/article/estonia-green-ict

Norwegian Minstry of Foreign Affairs, n.d. Digitalisation for Development: Digital strategy for Norwegian development policy. [Online]
Available at: https://www.regjeringen.no/globalassets/departementene/ud/dokumenter/utvpolitikk/digital_strategynew.pdf

OECD Library, 2023. Information and communication technology (ICT). [Online]
Available at: https://www.oecd-ilibrary.org/science-and-technology/information-and-communication-technology-ict/indicator-group/english_04df17c2-en

OECD, 2018. OECD Reviews of Digital Transformation. [Online]
Available at: https://www.oecd.org/sweden/going-digital-in-sweden.pdf

Ontario, 2020. Ontario Onwards. [Online]
Available at: https://files.ontario.ca/tbs-ontario-onwards-action-plan-en-2020-10-18.pdf

Ontario, n.d. Building a Digital Ontario. [Online]
Available at: https://www.ontario.ca/page/building-digital-ontario#section-0

Osoro, O., Oughton, E., Wilson, A. & Rao, A., 2023. Sustainability assessment of Low Earth Orbit (LEO) satellite broadband mega-constellations. [Online]
Available at: https://arxiv.org/abs/2309.02338

Portugal Digital, 2020. Action Plan for the Digital Transition: Strategy for the digital transition in Portugal. [Online]
Available at: https://portugaldigital.gov.pt/plano-de-acao-para-a-transicao-digital/

Presidence Francaise, 2022. Toulouse Call for a Green and Digital Transition in the EU. [Online]
Available at: https://www.economie.gouv.fr/files/files/2022/Call_for_Green_Digital_Transition_EU.PDF

Regeringskansliet, 2010. ICT for a greener administration – ICT agenda for the environment 2010–2015. [Online]
Available at: https://www.government.se/contentassets/4a75c8132fb84313a6a0d2f08a04ab76/ict-for-a-greener-administration—ict-agenda-for-the-environment-2010-2015-n2010.31

Riley, R. et al., 2021. Emissions impact of home working in Scotland, s.l.: Climate X Change.

Royal Society, 2020. Computing for net zero: how digital technology can create a ‘control loop for the protection of the planet’. [Online]
Available at: https://royalsociety.org/-/media/policy/projects/climate-change-science-solutions/climate-science-solutions-computing.pdf

Science Based Targets Initiative, 2022. Information and Communication Technology (ICT). [Online]
Available at: https://sciencebasedtargets.org/sectors/ict

Scottish Environmental Protection Agency, 2023. Scotland’s industrial greenhouse gas emissions revealed for 2022. [Online]
Available at: https://media.sepa.org.uk/media-releases/2023/scotlands-industrial-greenhouse-gas-emissions-revealed-for-2022/

Scottish Government, 2021. A changing nation: how Scotland will thrive in a digital world. [Online]
Available at: https://www.gov.scot/publications/a-changing-nation-how-scotland-will-thrive-in-a-digital-world/

Scottish Government, 2021. Green datacentres and digital connectivity: vision and action plan for Scotland. [Online]
Available at: https://www.gov.scot/publications/green-datacentres-and-digital-connectivity-vision-and-action-plan-for-scotland/

Scottish Government, 2021. Scottish Greenhouse Gas Statistics 2021, s.l.: s.n.

Scottish Government, 2023. Climate Change Plan Monitoring Report 2023: Electricity. [Online]
Available at: https://www.gov.scot/publications/climate-change-monitoring-report-2023/pages/3/#:~:text=Scotland%20has%20made%20significant%20progress,for%20the%20years%202017%2D%202020.

Scottish Government, 2023. Publication – Statistics Supply, Use and Input-Output Tables: 1998-2020. [Online]
Available at: https://www.gov.scot/publications/input-output-latest/

Siemens, n.d. Messages from 2030 Sweden: A picture of transformation. [Online]
Available at: https://assets.new.siemens.com/siemens/assets/api/uuid:3731ed91-6429-4cfb-bd02-b51cb9791ed7/Siemens-Report-Picture-of-Transformation.pdf

Social Progress Imperative, 2022. Just Transition Score. [Online]
Available at: https://www.socialprogress.org/static/e1977d5b833d24ddcfd4a0ad381262f9/Just%20Transition%20Score%20-%20Social%20Progress%20Imperative-%202022.pdf

Statista, 2023. Annual growth in global air traffic passenger demand from 2006 to 2022, with forecasts until 2023. [Online]
Available at: https://www.statista.com/statistics/193533/growth-of-global-air-traffic-passenger-demand/

Stjornarrad islands, 2023. Opinber innkaup / Public procurement. [Online]
Available at: https://www.stjornarradid.is/verkefni/rekstur-og-eignir-rikisins/opinber-innkaup/

TechUK, 2018. The UK’s core digital infrastructure: Data Centres Climate Change Adaptation and Resilience. [Online]
Available at: https://assets.publishing.service.gov.uk/media/5a81ce26e5274a2e8ab55eb7/climate-adrep-tech-uk.pdf

The Danish Government, 2022. National Strategy for Digitalisation. [Online]
Available at: https://en.digst.dk/media/27861/national-strategy-for-digitalisation-together-in-the-digital-development.pdf

UNEP, 2021. With new pact, tech companies take on climate change. [Online]
Available at: https://www.unep.org/news-and-stories/story/new-pact-tech-companies-take-climate-change

United Nations Development Programme, 2023. Four ways digital can power a just energy transition. [Online]
Available at: https://www.undp.org/blog/four-ways-digital-can-power-just-energy-transition

United Nations Environment Programme, 2021. With new pact, tech companies take on climate change. [Online]
Available at: https://www.unep.org/news-and-stories/story/new-pact-tech-companies-take-climate-change

Vihma, P., 2022. The carbon footprint of Estonian digital puublic services. [Online]
Available at: https://e-estonia.com/carbon-footprint-of-estonian-digital-public-services/
[Accessed 30 11 2023].

Wang, H., Yang, G. & Yue, Z., 2023. Breaking through ingrained beliefs: revisiting the impact of the digital economy on carbon emissions. Humanities and Social Sciences Communications, Volume 10.

Wang, J., Wang, K., Dong, K. & Shahbaz, M., 2022. How does the digital economy accelerate global energy justice? Mechanism discussion and empirical test. Energy Economics, Volume 114.

Welsh Government, 2022. Stronger, Fairer, Greener Wales. [Online]
Available at: https://www.gov.wales/sites/default/files/publications/2022-05/stronger-fairer-greener-wales-plan-employability-and-skills0.pdf

Williams, L., Sovacool, B. & Foxon, T., 2022. The energy use implications of 5G: Reviewing whole network operational energy, embodied energy, and indirect effects. Renewable and Sustainable Energy Reviews, 157(112033).

World Bank, 2022. How broadband infrastructure impacts greenhouse gas emissions: A Chile case study. [Online]
Available at: https://blogs.worldbank.org/digital-development/how-broadband-infrastructure-impacts-greenhouse-gas-emissions-chile-case-study

World Economic Forum, 2021. This year’s e-waste to outweigh Great Wall of China. [Online]
Available at: https://www.weforum.org/agenda/2021/10/2021-years-e-waste-outweigh-great-wall-of-china/

World Economic Forum, 2022. Digital Tech Can Reduce Emissions by up to 20% in High-Emitting Industries. [Online]
Available at: https://www.weforum.org/press/2022/05/digital-tech-can-reduce-emissions-by-up-to-20-in-high-emitting-industries/#:~:text=World%20Economic%20Forum-,Digital%20Tech%20Can%20Reduce%20Emissions%20by,20%25%20in%20High%2DEmitting%20Industries&text=Davos%2DKloste

Appendix 1: Detailed methodology and approach to the research

This Annex details the methodologies developed and used to complete this research project.

Definitions

The scope of this project defines ‘climate change’ as including Scotland’s interim and 2045 Net zero and emissions reduction targets. Although just transition and adaptation are important policy areas, the steering group wished to focus primarily on Net zero targets and emissions reduction with this research. The Steering Group also emphasised the need to only include information and projects that were current and operating, not theoretical.

By ‘digital connectivity’, we include the following:

Infrastructure

• fixed broadband (including subsea fibre, trunk or backhaul fibre, and access fibre)
• mobile connectivity (including 4G and 5G macro and 5G small cells)
• Datacentres Application, use and behaviours.
• existing applications such as artificial intelligence (AI) and Internet of Things (IoT)
• data-driven products and services
• practices such as working from home that are enabled by digital connectivity.

Geographical scope

• in Scotland and/or serves Scotland
• where there is applicable learning for Scotland.

Key dependencies

• digital skills
• renewable energy

These are all encompassed in our Figure 1 – The landscape of digital connectivity defined as within scope of this research. on page 14 of this report.

Literature review

Search terms

The search terms we used were agreed by the Steering Group on 24^th October 2023, these are set out below under the subheadings of Infrastructure, Application, use and behaviour, Geographical scope, and Key dependencies. We used Google and Google Scholar search engines.

Infrastructure:

• Digital connectivity infrastructure strategy
• Digital connectivity infrastructure policy
• Digital connectivity and climate change
• Digital transformation
• Broadband strategy
• Broadband policy
• Mobile network strategy
• Mobile network policy
• 5G strategy
• 5G policy
• Remote digital connectivity
• Rural digital connectivity
• Digital connectivity national plan
• Digital connectivity policy
• Economy strategy + datacentres (looking for links to environmental topics)
• Datacentres as opportunities for economic growth (looking for links to environmental topics)
• National Data Strategy (looking for links to environmental topics)
• Security of data infrastructure (looking for links to environmental topics)
• Resilience of data infrastructure (looking for links to environmental topics)
• Net zero digital connectivity infrastructure
• Net zero broadband infrastructure
• Net zero mobile network infrastructure
• Net zero datacentres
• Green digital connectivity infrastructure
• Green broadband infrastructure
• Green mobile network infrastructure
• Green datacentres
• ‘distribution’, ‘fairness’, ‘equality’ or ‘just transition’ of digital connectivity infrastructure
• ‘distribution’, ‘fairness’, ‘equality’ or ‘just transition’ of broadband infrastructure
• ‘distribution’, ‘fairness’, ‘equality’ or ‘just transition’ of mobile network infrastructure
• ‘distribution’, ‘fairness’, ‘equality’ or ‘just transition’ of datacentres
• Environmental impacts of digital connectivity infrastructure
• Environmental impacts of broadband infrastructure
• Environmental impacts of mobile network infrastructure
• Environmental impacts of datacentres
• Environmental footprint of digital connectivity infrastructure
• Environmental footprint of digital networks
• Sustainable digital infrastructure
• Sustainable broadband
• Digital carbon footprint
• Carbon emissions of datacentres
• Carbon emissions of digital connectivity infrastructure
• Carbon emissions of home working
• Environmental payback of digital connectivity infrastructure
• Environmental payback of broadband infrastructure
• Environmental payback of mobile network infrastructure
• Environmental payback of datacentres
• Life cycle assessment of working from home.
• Defra carbon factors for working from home.
• Life cycle assessment of broadband
• Life cycle assessment of mobile phones / mobile network infrastructure
• Life cycle assessment of datacentres
• Energy intensity of digital connectivity infrastructure
• Energy intensity of broadband infrastructure
• Energy intensity of mobile network infrastructure
• Energy intensity of datacentres
• Environmental sustainability of digital connectivity infrastructure
• Environmental sustainability of broadband infrastructure
• Environmental sustainability of mobile phones / mobile network infrastructure
• Environmental sustainability of datacentres.

Application, use and behaviours

• AI strategy
• Internet of Things (usually in the digital strategy and linked to Environmental departments of councils, waste etc).
• Working from home strategies across business groups and government.

Geographical Scope:

• The analysis of policies was focused on:
• Scotland
• Finland,
• Wales,
• Portugal,
• Norway,
• Sweden,
• Estonia,
• Canada (Ontario),
• New Zealand,
• Denmark and
• Iceland

These jurisdictions have topological and population density scale comparisons with Scotland and are likely to face similar digital connectivity issues (land mass, areas of low population density, rural communities).

Some further findings on China were identified as part of the research and included in the report.

Key dependencies

Key dependencies were discussed as part of the scoping analysis as follows:

• Digital skills strategies – from school age to beyond in target countries
• Government skills strategies
• Government “digital transformation” strategies – usually local government
• Digital inclusion strategies
• Renewable energy strategy (to see if there’s a link to digital)
• National Grid and Distribution Network Operators / Distribution System Operator and their requirements for digital connectivity (fixed and mobile)

Methodology for policy review

We undertook a desk review of existing policies from countries within our scope. We looked for whether their policies were designed to support emission reductions, adaptation or just transition through digital connectivity. Section 7 presents our results and uses a score to rate the extent that digital connectivity and emissions reduction is linked within a country’s policy, with five being explicit mention of the GHG or carbon impacts of increased digitalisation and one being no mention or linking of decarbonisation within the policy.

Table 3 – Key of star ratings used to assess country policies and their link between digital connectivity and emissions reduction.

Assessment of confidence

Following a methodology developed by the Intergovernmental Panel on Climate Change for the fifth assessment report and used for the sixth for the consistent treatment of uncertainties, we developed a confidence level based on the extent of agreement in the literature and the robustness of evidence. Figure 3 on page 15 sets out what constitutes low, moderate, and good confidence in our claims.

• Low agreement is where sources conflict.
• Medium agreement is where sources make broadly similar conclusions but the data or evidence they use to support their conclusions is very different.
• High agreement is where sources independently make similar conclusions and underlying data are similar despite being independent.
• Limited evidence is some evidence available but largely anecdotal and not from recognised peer reviewed sources. Availability of data was low.
• Medium evidence is information from peer reviewed sources or official sources.
• Robust evidence is a greater volume of information from peer reviewed sources and official sources.

Appendix 2: Summary of digital policies across 10 countries.

A summary of digital policies and their links with decarbonisation, across 10 countries selected based on the methodology in Section 4. The degree to which each country links their digital goals and strategy has been given a score, with five representing explicit mention of the GHG or carbon impacts of increased digitalisation, and one representing no mention or linking of decarbonisation within the policy.

The policies from the individual countries are presented in Appendix 2: Summary of digital policies across 10 countries.

Some policies explicitly mentioned a just transition, and reference to adaptation was not found in any of the policies we were able to identify. See Section 7.2.3 for specific comment on data centre related policy.

Finland

Score: five – digitalisation and decarbonisation linked heavily, and mentions of wider collaboration with other decarbonisation initiatives as well as of the GHG or carbon impacts of increased digitalisation.

Finnish policy does connect increased digitalisation with helping the green transition, but there is no explicit mention of the carbon impact of increased digitalisation on the environment.

The Finnish Government: Digital Compass was drawn up for the purpose of managing the development of the digital transformation in Finland. Based on European values and the Digital Decade 2030 programme. Promotes an economically, socially and ecologically sustainable digital green transition (Finnish Government, 2022).

Objective 9 of the Digital Compass states that Finland develops and applies digital technologies that respond to global climate and environmental challenges (European Union Digital Skills and Jobs Platform, 2023).

Ministry of Finance Finland: Sustainable Growth Programme for Finland aims to support growth that is ecologically, socially and economically sustainable in line with the aims of the Govt Programme. Funding will come mainly from EU Recovery Plan ‘Next Generation EU’ – one of four key elements ‘Digitalisation and a digital economy will strengthen productivity and make services available to all’ (Ministry of Finance Finland, n.d.)

Portugal

Score: two – minor mention of decarbonisation in the digital connectivity policy.

Portugal says digitalisation will contribute to decarbonisation.

Portugal’s Action Plan for Digital Transition (Measure 9) speaks to increased digitalisation of public services, which it reports will contribute to decarbonisation and environmental benefits. (Portugal Digital, 2020)

Portuguese Secretary of State for the Digital Transition has partnered with Digital With Purpose (2020 onwards) to acknowledge and deliver digital sustainability. (Global Enabling Sustainability Initiative, 2020)

Norway

Score: two – minor mention of decarbonisation in the digital connectivity policy.

Norwegian policy connects increased digitalisation with aiding the green transition.

The Norwegian Ministry for Foreign Affairs, Digitalisation for Development, Digital Strategy for Norwegian Digital Policy acknowledges climate change as an important priority but doesn’t directly acknowledge the climate impacts of ICT (Norwegian Minstry of Foreign Affairs, n.d.).

The Norway-EU Green Alliance acknowledges that digital transition is important for and contributes to the green transition (Norway and European Union, n.d.).

Sweden

Score: four – digitalisation and decarbonisation linked heavily, and mentions of wider collaboration with other decarbonisation initiatives.

Swedish policy links the use of ICT to decarbonisation effects, as well as acknowledging decarbonisation, circularity, conscious choices, and the energy transition as drivers for a sustainable world.

The Swedish Government’s ICT for a Greener Administration report outlined the importance of acquisition and public procurement, use of ICT in government agencies and digital tools to reduce business travel (OECD, 2018).

The focus of the Swedish Information Society policy is, among other things, to use ICT to promote sustainable growth (Regeringskansliet, 2010).

Estonia

Score: five – digitalisation and decarbonisation linked heavily, and mentions of wider collaboration with other decarbonisation initiatives as well as of the GHG or carbon impacts of increased digitalisation.

Estonian policy contains a clear and explicit mention of the carbon effects of increased digital footprints, and provides a commitment to reduce the effects.

The Estonian Digital Agenda 2030 stresses the activities of the Estonian development plan contribute through the use of innovative technologies and environmentally friendly solutions to reduce the impact of climate change. They are also meant to reduce the time required for covering distances and ensure a good living environment all across Estonia.

The Estonian government also has Green Digital Government Commitments, stating “we analyse the environmental impact of the Estonian digital government and ways to reduce it” (European Union Digital Skills and Jobs Platform, 2023).

Canada (Ontario)

Score: one – no mention of decarbonisation within digital connectivity policy.

Canadian policy contains no mention of the carbon or environmental impact of increased digitalisation.

Canada’s Digital Ambition 2022 mentions at a high level that their Digital Ambition aligns with the Greening Government Strategy – but delivery on specific plans is unclear from published policy and strategy (Government of Canada, 2022).

The Building a Digital Ontario – Ontario’s Digital Strategy does not mention environmental protection or any digital sector emissions (Ontario, n.d.).

The Ontario Onwards Action plan mentions the importance of environmental protection, but does not specifically link environmental protection with digital and sustainability.

“The Government of Canada’s Digital Ambition goes hand in hand with the Greening Government Strategy, which seeks to make Government of Canada’s operations low carbon through green procurement and clean technologies. Through the increased promotion of environmental sustainability, and by integrating environmental considerations in its procurement process, the federal government is in a position to influence the demand for environmentally preferable goods and services” (Ontario, 2020).

New Zealand

Score: one – no mention of decarbonisation within digital connectivity policy.

New Zealand policy does not explicitly mention the carbon or environmental impacts of increased digitalisation.

The Digital Strategy for Aotearoa proclaims: “we use data and digital technology to address big issues of our time like climate change. We also want the tech sector to play a key role in creating a more equitable, low-carbon future.” (Digital.Govt.NZ, n.d.)

However sustainable delivery or green ICT is not noted in any of the flagship initiatives of the Action Plan for the Digital Strategy for Aotearoa (Digital.Govt.NZ, 2022).

Denmark

Score: four – digitalisation and decarbonisation linked heavily, and mentions of wider collaboration with other decarbonisation initiatives.

Danish policy takes a holistic approach to digitalisation and digital section emissions, with direct considerations for green ICT acquisition and support for the EU’s Green Public Procurement criteria.

The Danish Ministry of Finance’s National Strategy for Digitalization focuses on digital as an enabler and doesn’t consider the impact of digital emissions (The Danish Government, 2022).

The “Visions and Recommendations for Denmark as a Digital Pioneer” document focusses on digitising energy and utility data as a prerequisite to understand the impact of increased digital connectivity. Heavier focus on using digital to achieve green transition (Digitalserings Partnerskabet, 2021).

The Agency for Digital Government – Digital Green Transition lays out plans for the EU’s Green Public Procurement criteria to have been tested throughout 2022 and 2023 (Agency for Digital Government, n.d.).

The Study on the Digital Green Transition in the Nordic-Baltic Countries does not explicitly mention the quantification of spend vs emissions (Agency for Digital Government, n.d.).

Iceland

Score: three – digitalisation recognised or reported as a contributor to green transition, but no mention of the GHG impacts of digitalisation.

Icelandic policy nods to sustainable procurement as a lever for green digitalisation, but provides no quantification.

The Digital Green Transition – Government of Iceland sets out the Icelandic ambition to leverage digital effects to achieve and accelerate the green transition (Nordic Council of Ministers, n.d.) (Government of Iceland, 2021) (Stjornarrad islands, 2023)

Wales

Score: four – digitalisation and decarbonisation linked heavily, and mentions of wider collaboration with other decarbonisation initiatives.

It is recognised that digitalisation will play a role in the transition to net zero in the Decarbonising Wales with digital technology website.

The policy for Wales also mentions a just transition and how skills are central to that (Centre for Digital Public Services, 2022).

Tech Net Zero discovery investigated greener government and third sector tech report came up with 6 recommendations of how public services could use digital technologies to reach net zero, one of which was to measure the carbon footprint of a digital service (Centre for Digital Public Services, 2022).

© Published by Frazer-Nash Consultancy, 2024 on behalf of ClimateXChange. All rights reserved.

While every effort is made to ensure the information in this report is accurate, no legal responsibility is accepted for any errors, omissions or misleading statements. The views expressed represent those of the author(s), and do not necessarily represent those of the host institutions or funders.

1. 
   

   Double Diamond Model: what is it? – Justinmind ↑

   
   
2. 
   

   Using the supply, use and input-output tables and 2019 data coupled with the Greenhouse Gas Effects 2024-2025 (Scottish Government, 2023) and includes the direct and indirect carbon dioxide equivalent emissions of the following sectors: Computers, electronics and opticals; Telecommunications; Computer services; and Information services. ↑

   
   
3. 
   

   Whilst undertaking the research to support this statement we have identified additional information, beyond the scope of our initial research, for Germany and China. ↑

   
   

December 2023

DOI: http://dx.doi.org/10.7488/era/4512

Executive summary

Background

Scotland recognises the significance of a place-based transition to net zero greenhouse gas emissions (GHG). As part of setting a target of net zero by 2045, the Climate Change (Scotland) Act 2009 places importance on the role of local authorities in achieving this target. Therefore, it is a priority for the Scottish Government to facilitate area-wide and locally-led efforts as part of a just transition to net zero.

Across the 32 local authorities in Scotland, 17 have set net zero targets specific to tackling territorial GHG emissions generated in their geographical area (from agriculture, buildings, industry, land use and land use change and forestry, transport and waste). This is in direct comparison to 26 local authorities that have set net zero targets to reduce their organisational GHG emissions.

This research examines local authority climate-relevant strategies and policies within them; the potential of these policies to reduce emissions if they were scaled to the national level; and the barriers that local authorities face in implementing these policies.

Main findings

We developed a register of 69 climate change strategies across all 32 local authorities. We found that local authorities are modelling exemplary action on climate change across many fronts through the benefit of deep-rooted relationships with local stakeholders and unparalleled knowledge of their area.

However, the level of detail and methodological evidence presented in climate change strategies are often sparse, with many strategies failing to model the scale of impact on GHG emissions.

From the 69 climate-related strategies, we selected six leading strategies for quantification and identified 13 policies within these that could be appropriate for scaling up. We undertook an initial estimate of the potential territorial emission reduction if they were replicated across all Scottish local authorities. We also assessed the likelihood for change at this scale, considering local authorities’ sphere of control, capacity and timescales, alongside the magnitude of potential change. Through this process we identified two policy areas with the potential for major impact on territorial greenhouse gas emissions:

1. Nature-based solutions: a combination of individual policies to green derelict land, restore damaged peatland and afforestation.
2. Net zero transport: several climate policy initiatives such as active transport, decarbonisation of public transport and low-emission vehicle licences for taxis.

The impact on Scotland’s national territorial emissions, should all local authorities adopt the leading policies, from nature-based solutions (5,497 ktCO₂e) and net zero transport (1,527 ktCO₂e) amounts to an estimated total potential reduction of 7,024 ktCO₂e by 2045. This is an indicative figure, illustrating the scale of change that could be possible.

We found that the Scottish Government have set a compelling ambition to closely support local authorities to develop locally owned and led climate action strategies to tackle territorial emissions.

However, we also found that local authorities are limited by a lack of clarity on their roles and responsibilities, and by a lack of best practice guidance or frameworks across all the territorial emission categories. They face barriers including lack of data maturity, capacity, specialist skills, accountability and funding.

Recommendations

Local authorities could be further supported to develop their climate policies. We recommend the establishment of best practice guidance on the development of climate policies. This would help improve clarity and consistency across local authorities.

Further research could expand on the capacity and capability requirements to deliver local authority climate policies between now and 2045, including methods by which the resourcing needs could be met. Further investigation could help quantify the funding available for tackling each GHG inventory, where further funding might best be directed and methods for administrating funding to ensure that national ambitions can be met.

Glossary and abbreviations

Introduction

Context

The recent parliamentary inquiry into the role of local government in delivering net zero stressed that it will be impossible for Scotland to reach net zero without local leadership spearheading area-wide decarbonisation efforts (Net Zero, Energy and Transport Committee, 2023). The inquiry recognised that achieving net zero cannot be dictated. It requires a collective effort between local government, which holds the local knowledge and fruitful partnerships across the public and private sectors, and national government which have the strategic capabilities and resources to support and coordinate local efforts.

The Scottish Government is continuing the drive toward empowering, building capacity, and providing the necessary foundations for local government to build their net zero programmes. The parliamentary inquiry also established that, while councils have at times been a model for net zero leadership, this needs to be rapidly scaled across all local authorities and all emission sectors in each local authority. The inquiry report noted that the Scottish Government must facilitate this scaling by providing local authorities with a comprehensive roadmap for net zero and “far more certainty than they have at present about the roles they are to play” (Net Zero, Energy and Transport Committee, 2023).

The Duties of Public Bodies: Reporting Requirements Order placed responsibilities on all public bodies, including local authorities, to report on scope 1 and 2 (and some scope 3) organisational emissions (Climate Change Order, 2015). As a result, all 32 local authorities have developed organisational emission inventories and in 2022 the Accounts Commission reported that 26 local authorities had developed organisational net zero targets (Audit Scotland, 2022). However, local authorities have some influence on certain emissions reduction beyond their organisational boundaries. These emissions produced within a local authority’s geographical area of responsibility are referred to as ‘territorial emissions’. Only 17 local authorities have developed territorial net zero targets and even fewer have developed policies for reducing territorial emissions. If this situation persists, it will present a major barrier to the success of Scotland’s national Climate Change Plan, which is heavily reliant on place-based and locally-led action (Scottish Government, 2020).

In their recent progress update to parliament, the Climate Change Committee noted that “momentum on a local level is increasing, but local action is uncoordinated” (Climate Change Committee, 2022, p. 53). There are pockets of exemplary action but also a lack of knowledge sharing across local authorities. This has led to policies being rolled out with different timescales, best practice not being disseminated and opportunities being missed to drive coordinated action across all local authorities. In November 2023 the Scottish Government launched a new Scottish Climate Intelligence Service to support local authorities to build capacity and capability for the development of area-wide programmes of emissions reduction for the benefit of their communities. This service will enable local authorities to deliver their own area-wide territorial net zero targets and to contribute to Scotland’s national commitment to net zero by 2045 (Improvement Service, 2023).

This research addresses some of the identified challenges by analysing the climate policies local authorities have developed to directly tackle territorial GHG emissions, and mapping their potential impact on territorial GHG emissions.

Project aims and research questions

The first aim of this project was to identify key GHG emission reduction policies developed by Scottish local authorities. We developed a comprehensive register of local authority climate-related strategies and associated policies and described the current action being taken by each local authority across all emission categories.

The second aim was to compile and undertake an initial estimate of the policies’ GHG emission reduction potential at both the local authority and national level. This aim was broken down into three sub-questions. Firstly, to identify what the key policies are that have significant GHG emission reduction potential. Secondly, to estimate their emissions reduction potential within their respective local authorities. Thirdly, to estimate what the emission reduction potential would be, should they be applied across all Scottish local authorities. This type of analysis has previously been conducted by the Edinburgh Climate Commission and Place-based Climate Action Network, although this was only in relation to policy scenarios at the local level (Williamson, et al., 2020).

The third aim was to engage with local authorities through a series of semi-structured interviews to understand how the most significant policies could be implemented across Scotland, including the role of Scottish Government and other public bodies in enabling this.

Overall, this project highlights area-based policy options for Scottish Government to consider for national deployment, whether as a statutory instrument, as in the case of Local Heat and Energy Efficiency Strategies (LHEES), or via other delivery approaches such as frameworks or guidance.

Defining the greenhouse gas emission inventory

The UK greenhouse gas inventory (GHGI) is published annually by the Department for Energy Security and Net Zero (DESNZ) and sets out the latest estimates in territorial GHG emissions for all 374 local authorities across the United Kingdom, including the 32 local authorities across Scotland. We have charted the latest DESNZ territorial GHGI publication data for Scotland (DESNZ, 2023) in Figure 1 below. This shows the total territorial GHG emissions split into the inventory categories (agriculture, buildings, industry, LULUCF, transport and waste) between 2005 and 2021. The dataset employs several different methodologies to calculate the spatially disaggregated emissions for each inventory category.

Table 1 provides a description of each of the GHGI categories. These are important for drawing boundaries around polices, determining which inventory a specific policy will impact.

It is possible for policies to transcend multiple emission inventories. For example, a policy that seeks to develop a green network to increase the level of active transport^[1] by improving tree canopy coverage and hedgerows would impact a transport and LULUCF inventory. There are activities and emission changes that would impact both inventories in this instance.

Methodology

This section provides a summarised version of the research methodology. A more detailed methodology is available in Appendix 13.1.

A steering group was established to support the delivery of the project, and consisted of representatives from the Scottish Government, ClimateXChange, Sustainable Scotland Network (SSN), and the Turner & Townsend research team. Findings and outcomes were reported to the steering group for comments and to confirm the research direction throughout the project. The project was divided into three tasks.

Evidence review

Task 1 was to compile a comprehensive policy register to understand the current climate action being taken by each local authority. This register provides a useful tool to view and analyse individual climate policies across Scotland. We applied the following process:

• Search: our search began with reviewing information available through the “Wider Influence” tab of local authority climate change submissions to SSN (SSN, 2023b). Where gaps existed, we supplemented these by conducting an online search of local authority websites and other public body sources for the relevant policy documentation.
• Classify: we utilised a rapid evidence assessment (HM Treasury, 2020) to classify each policy based on its high-level data, including years of coverage, policy owner, whether the policy is monitored, and any associated targets.
• Select: we developed screening criteria based on Scottish Government priorities for the current project and used this to recommend six strategies of significance to progress to Task 2.

We presented the key findings to the steering group and our assessment of the selected strategies. We asked the steering group for advice on the selection of the six strategies. This resulted in the addition of geographical criteria to our selection assessment, to ensure the research considered local authorities from rural and island communities.

Quantitative research

For Task 2 we developed a GHG profile for each of the six strategies selected from Task 1. This involved identifying the emission boundary of each policy within the strategies and the quantification of the potential impact on territorial emissions of the respective local authority. We then proceeded to calculate an aggregated figure to estimate the policies’ potential impact if rolled out at the national level. We approached this by:

• Assessment boundary: GHG boundaries were established using GHG Protocol Action Standard (Greenhouse Gas Protocol, 2014) to apportion the relevant sinks and sources to each policy and estimate potential emission impacts. This was used to determine the likelihood and magnitude of change.
• Policy scenario emissions: in the first instance, we used existing activity and emission factor information from the local authority policies to develop policy scenario emissions estimates. In the absence of information, we applied Intergovernmental Panel on Climate Change guidance. We then used the HM Treasury Green Book to approximate changes and associated emissions values to provide national-level policy scenario figures.

The more comprehensive methodology in Appendix 13.1 explains in detail the range of approaches and methodologies applied in the assessment of GHG boundaries, development of the policy scenario emissions estimations and the limitations of this approach. The findings from Task 2 were presented to the steering group with the objective of selecting two of the most likely and impactful areas of policy to be considered for national deployment by local authorities. These were developed into policy briefings for Scottish Government.

Qualitative research

For Task 3 we conducted interviews with representatives from two local authorities to gain their views on wider implementation of the selected policy areas, including the roles of local authorities, Scottish Government and other public bodies. We planned a third interview with one further local authority however, we were not able to agree a time and date for the interview to take place in the timescales of this research.

A topic guide was developed and formed the basis of 45-minute semi-structured interviews on Microsoft Teams. These aimed to collect the comprehensive views on the likelihood of wider adoption of the policies, including practicability, the capacity and capability required to deliver a new policy. We also included other open-ended questions, encouraging participants to expand further on topics they deemed relevant. The data from interviews was collated in a thematic analysis grid and key themes were extracted using an analytical approach guided by participant views.

We combined the data from all sources (the evidence review, quantitative potential emissions modelling, discussions with the steering group, and the qualitative research) to discuss the key challenges and the possible approaches to adopting the climate policies at a national scale. The conclusion is presented in Section 11.

Review of existing evidence

Overview

The aim of this review was to understand the climate strategy and policy landscape across Scottish local authorities. We created a Climate Strategy Register that involved the collation of climate action plans from all 32 local authorities including individual sector strategies such as transport plans, waste plans and local development plans (Appendix 13.3).

This report makes a distinction between a climate strategy and a climate policy in the context of the documents reviewed. Policies feed into sector strategies, which feed into a climate change strategy.

Most local authorities reviewed already have a top-level document we define as a climate change strategy. A climate change strategy refers to several planned actions and policies designed to outline an organisation’s approach to tackling climate-related challenges in their local region. Climate change strategies encompass other nomenclature such as a ‘climate action plan’. A climate change strategy will typically cover ambitions for all GHG emission inventories and may link to separate sector strategies that set out in further detailed policies specific to a singular emission inventory. For example, a climate change strategy might reference a separate transport emission sector strategy.

A climate policy encompasses an individual action or set of actions that deliver ambitions set out by a climate strategy. Policies will typically include setting of targets and key performance indicators to measure and verify the success of the policy’s intended impact. For example, a transport sector strategy might include a policy to increase electric vehicle charging infrastructure, and a policy to implement a low emission zone (LEZ) in a city centre.

We used several sources of information to inform our review of existing evidence. We started with reviewing the “Wider influences” local authority climate change report submissions to SSN (SSN, 2023b). The wider influences section of SSN reports was completed with varying degrees of information but overall, the level of detail was sparse. We supplemented this gap by searching each of the local authority websites for their climate action strategies. We found various types of initiatives at different levels of hierarchy.

We identified 69 strategies relevant to climate change across the 32 local authorities. We developed short summaries of each strategy document, which are set out in Appendix 13.2.

We developed a screening matrix to rank each of the strategies against five criteria outlined in Table 2 and determined the level of maturity by assessing the level of evidence provided in a climate change strategy as yes / no / partial. Each of the strategies was then assigned a relevance score to identify those that closely aligned with the research objectives.

Although some strategies where much more detailed than others in terms of the detail provided against individual policies, all the strategies provided sufficient information for us to understand how they would lead to an impact of the GHG emissions in their area. However, quantified information about the level of impact a strategy had was often high-level, not valued as an impact on territorial GHG emissions, or left as an open ambition.^[2]

Selected local authority strategies

From the existing evidence review, we identified five climate change strategies that scored well across all of the screening categories. These climate change strategies were discussed with the steering group and we identified that all of the selected climate change strategies were across the central belt of the country. We therefore added a sixth strategy from a more rural local authority to ensure that we had a more diverse geographical spread. The six climate strategies matching the criteria were taken forward to the next task of valuating climate policies. The local authorities selected are shown in Figure 3 below.

In the following paragraphs, we present two example climate change strategies as representative of the strategies we reviewed.

Stirling’s Climate and Nature Emergency Plan was the highest-ranking strategy (table 3) we reviewed. This was due to the large array of topics covered, efficient writing style, the explanation of policies and how those could be translated into other local authorities and regions. It provided several emission impact figures for policies and actions to show the effect on the environment and highlighted how these would be resourced in the region. Stirling’s Climate and Nature Emergency Plan was also one of the few climate change strategies to mention their current territorial emissions, which is the key focus of this project. Mention of territorial emissions is usually a strong indicator that a climate change strategy would give thorough information around carbon impacts and implementation. Stirling’s Climate and Nature Emergency Plan estimated a territorial emission reduction of 1/3 between 2005 and 2018 and mapped out their future to show where the local authority wanted to be by 2030. This was one of many examples from Stirling’s Climate and Nature Emergency Plan that set it apart from other climate change strategies and provided a clear understanding of how the local authority wanted to meet their targets for territorial GHG emissions.

The Glasgow Climate Plan (Glasgow City Council, 2022) and Stirling Climate and Nature Emergency Plan (Stirling Council, 2022) were key examples of detailed climate change strategies that could be deployed to support a national transition to net zero. Both strategies gave detailed explanations of the current regional context which was pivotal in explaining why certain policies or actions had a greater impact than others. The strategies also highlighted the importance of developing and investing in climate policymaking to ensure polices they set are appropriate for the regions as well as the communities they serve, whilst aiming to minimise the (negative) impact on residents as much as possible. Another key area both strategies explore is the financial implications of initiatives, indicating whether projects are either already funded, part funded or if they are being financed. This is something the Glasgow Climate Plan provided details on more than any other climate strategy reviewed. Importantly, the strategies outlined the capacity requirements to adequately resource their polices and provided timebound milestones to monitor progress against.

Additional findings

Territorial emissions impact

Of the 69 climate change strategies, 56 either partially valued their emissions impact or failed to value the scale of their impact on GHG emissions at all. A common theme in the absence of territorial GHG emission impact was to apply a bespoke indicator as a measure of success, such as increasing the number of staff working remotely. The majority of climate change plans did not outline the methodologies applied in gathering and quantifying performance measures and targets, so it was often unclear how impacts would be measured.

The key aim of this research was to identify policies that could impact territorial GHG emissions in a major way. The top performing policies against the criteria were scored well because they quantified the anticipated impacts. Emissions were typically quantified as either a tonnage reduction in GHG emissions (tCO₂e) or a percentage reduction against a baseline figure.

Resourcing, financing and timelines

56 of the 69 climate change strategies had fully or partially evidenced timescales for implementation and completion. adopt a unified approach.

The most mature climate change strategies also included considerations around cost, whether funding had been secured, who would be financing it and who would be delivering these policies. For example, Argyle and Bute’s Decarbonisation Plan (Argyll and Bute Council, 2021) outlines sources of funding against each individual policy, whether funding has been secured or still requires budget.

However, policies aimed at achieving the same outcome might do so on different timescales. There was no clear pattern across the climate change strategies on how timescales were decided upon. The exception to this rule was waste targets as they are set nationally, which is a good example of how other policy areas could do the same to territorial emissions and targets

Only 13 of the 69 climate change strategies cited their territorial emissions. Of those, only some set territorial emissions targets. It is not clear why this was the case. It could be due to local authorities not having updated information about their territorial emissions or because they were not confident in how they could enact change in their regions. Climate change strategies that specifically mentioned territorial emissions and set emissions targets for their area had more detailed action lists that went beyond council owned assets. This difference is important as it highlights some local authorities are being proactive in tackling territorial GHG emissions in the local authority area beyond just those of their own organisations.

Summary

The level of detail and consistency of targets and performance metrics showed that there was no clear and consistent approach to developing climate change strategies. This makes comparison and valuation of the climate strategies complex due to the non-uniform nature of presenting impact and the lack of detail around the methodological approaches applied.

The strategies we ranked high on our measures including scalability, replicability, and quantification of impacts, could form the basis of best-practice knowledge sharing, and setting of a national approach (see Appendix 13.1.1 for further detail). Our findings reflect those of recent research carried out by Environmental Standards Scotland (Environmental Standards Scotland, 2023) that recommended Scottish Government introduce a standardised Climate Plan template with mandatory reporting for local authorities. This recommendation would go some way to solving some of the challenges uncovered by this research.

Results of quantitative research

Overview

From the six climate strategies reviewed in detail (Figure 3), 61 distinct climate policies were extracted. The distribution of the policies across the GHG inventory categories is summarised in Figure 4.

Of the 61 policies extracted, most policies (26) targeted building emissions and are outside the scope of this research as they are covered by the exemplar LHEES approach that has already been rolled out nationally across all local authorities. This research intended to identify policies in other GHG inventory categories that have the same potential for rollout across local authorities. With building emissions excluded, the remaining 35 policies have the greatest numbers in transport (13), LULUCF (8) and industry (7) as shown in Figure 4.

Figure 4: Number of policies extracted, by GHG inventory.

Of the 35 policies, we could only collect sufficient information from 13 policies to be able to estimate potential GHG emission impact. These are described alongside example targets and KPIs in Appendix 13.5.

Policy scenario emissions

We analysed the 13 policies to estimate the potential GHG emission impact if they were to be scaled-up to the national level and enacted across all 32 local authorities. The potential GHG emission impacts are high-level indicative estimates using a basic methodological approach and incorporating multiple assumptions, as set out in Appendix 13.1.2 and 13.2. As such, the quantitative findings are indicative, illustrating the scale of potential impact that local authorities may have in tackling climate change. Further analysis would provide more accurate potential GHG impacts of policies.

The findings of this analysis are detailed Table 4. Each row in Table 4 contains a climate policy that originates from either a single local authority or multiple local authorities where policies were similar. Table 4 details that across the 13 policies assessed for their GHG emission impact, there is potential for an estimated 9 MtCO₂e overall change to territorial emissions, or 22% of the current inventory emissions for Scotland.

The full breakdown of the indicative estimated potential impact on each individual local authority’s GHG inventory is presented in Appendix 13.6 and sources for the assumptions and conversion factors are included at Appendix 13.2.

For each of these 13 policies valued, we also show in Table 4 our assessment of the likelihood of each policy to cause a change in emissions if rolled out nationally to all local authorities, taking account sphere of control, capacity and capability, and the timescale over which a policy would be enacted. We also assessed the magnitude of the potential change. Both of these methodologies are outlined in IPCC guidelines (IPCC, 2006) and set out in Appendix 13.4. There will be widely ranging factors and contexts at an individual local authority level which have not been accounted for and that would significantly impact implementation of the policies assessed. In addition, there are critical wider factors such as future national policy development and available budget that were not incorporated into this quantitative analysis.

Findings

Comparing the policies evaluated in Table 4 with the Climate Change Plan sector envelopes (Scottish Government, 2020, p. 253) indicates that both LULUCF and transport policies have the greatest potential to impact territorial GHG emissions, with a high likelihood of the local authority being able to influence their outcome. Table 3 below shows estimated potential GHG emission reductions in these policy areas if implemented in each local authority.

The other policy areas evaluated may also compare favourably with the Climate Change Plan sector envelopes but local authorities have a more limited control on the outcomes. This is the case with policies relating to changes in agricultural practices. In addition, while seven industrial emission-related policies were present amongst the six climate strategies finalised, none sought to value their impact on territorial GHG emissions and provided limited definitive action. Instead, the industrial-emission-related policies opted for a model of getting organisations to sign up to climate change pledges. Policies that were either outside the local authorities’ sphere of influence, or policies that impacted centralised issues, such as waste management, were also not carried forward to interviews with local authorities.

Results of qualitative research

Overview

The results of the quantitative research found that policies in LULUCF and transport showed potential in having significant impacts on local authority territorial GHG emissions. To find out more about how these policies were developed, and the potential pathways to implementing similar policies at the national level, we interviewed local authorities who had leading policies in nature-based solutions and net zero transport.

Findings

The findings below combine evidence from our review of existing data and assessment of the key themes identified through thematic analysis of interviews.

Capacity and capability

It was clear from the interviews that lack of capacity to develop and deliver policies would likely hamper efforts in expanding policies across all local authorities in Scotland. We found that some local authorities had the resource and ability to hire specialist skills into the organisation. Through this they could actively engage with teams across the organisation to ensure policy ambitions were carried out. An example of this given by one respondent:

“It’s imperative to ensure that any planting of new trees considered multiple planning and climate aspects, impacting the species of tree selected, factoring in considerations about the future microclimate and requirements for future flood prevention.”

However, local authorities do not always know what skills they need to deliver on a policy ambition. One respondent explained that many policies require both multi-disciplinary expertise, such as project management, as well as specialist skills, such as ArcGIS^[3], to properly manage the rollout of a policy.

One respondent explained that budget cuts mean that retaining enough resource within the organisation, with access to the right skills and expertise would be a defining factor in the success of climate policies’ targets. Respondents did signal that it was possible to access skills external to the local authority (e.g. through consultancy) but this was often ad hoc. Developing and implementing policies will require multi-year and decadal management to realise their full benefits. Not being able to retain the skills and resource within the local authority places their success at risk.

Data maturity

One of the respondents explained that having good quality data that is continually updated and shared across the organisation is critical to enabling policy development and delivery. The example provided was the data landscape for nature-based solutions policies, which is complex, onerous to compile and requires near-constant updating. For example, in the greening of derelict land, the classification of land as ‘derelict’ ebbs and flows as multiple stakeholders retain interest in the space. The local authority itself (potentially across multiple departments), private individuals, residents and developers may all have a stake in the use of the derelict land. Added to this is the difficulty of collecting accurate data about derelict land, such as carbon evaluation, existence of contaminants, appraisal of natural ecosystems and animal species, and importance to flood prevention. This information is needed to show causal links between greening derelict land and benefits such as heat reduction and carbon sequestration.

Data also enables a local authority to develop robust climate policies by identifying measurable KPIs and to set realistic timescales. Several climate change strategies we reviewed were at early stages of development and specifically referenced the need for additional research to complete the valuation of a policy’s impact. For example, several transport policies referenced other transport strategy documents in-development that sought to improve data maturity for the local area, and enable valuation of impacts and target setting. Timescales for the development of these strategies were not clear.

Collecting adequate data is key to the development, measurement, and success of a climate policy. However, the landscape is complex and demanding and interrelated to capacity and capability in the local authority as discussed above.

Geographical diversity

We found that the overarching aims of climate change strategies across Scotland are the same. However, sometimes these goals are were coupled with specific local issues. Therefore, motivations, KPIs, and targets by which the local authorities measure the performance of climate related policies often differ. This has a knock-on effect on the data and capacity needed to implement policy across diverse communities.

One clear example of this is in homeworking policies. In large island communities that have a widely dispersed rural communities, home working and flexible working has benefited commuters who do not need to travel great distances to reach their work location. One interviewee explained that the policy has helped island communities to overcome other issues such as the lack of public transport provision. Similar homeworking policies also exist in cities with a specific focus on reducing the amount of traffic congestion within the city centre at peak times. Both sets of policies have differing motivations for enacting homeworking polices but the end benefit of reduced air pollution is the same.

Accountability and ownership

We found that climate policies often span multiple departments within an organisation. In some circumstances this led to ambiguity around accountability for the successful delivery of a policy. One respondent explained that for nature-based climate policies, using afforestation as a specific case in point, the responsibility and budget for tree planting might fall with a local authority’s parks department. However, responsibility to actively manage LULUCF from a climate perspective might reside with the sustainability or planning departments. This leads to complexities around who in a local authority needs to be consulted for LULUCF projects and who has ultimate ownership of a policy being successfully enacted. Respondents referenced that it is not uncommon for there to be “a lot of silo working” across departments, so projects that might impact on a climate policy are not always communicated, or vice versa. Respondents also noted that there tends to be an aversion to taking on or sharing climate policy responsibilities because it is a change from how departments have functioned in the past,

“[we] have always done it this way so why would we do it another way”.

Funding

Funding, or the lack thereof, was a common theme across respondents. One respondent noted that there is a lack of funding available to commission external expertise, for example the delivery of a feasibility study. This hampered efforts to collect the information needed to develop robust policies and set realistic targets. It was clear from the strategies reviewed that only a few local authorities sought to quantify the funding requirement to deliver policies.

A strong theme was the lack of funding to attract and retain talent within the local authorities. One example given was that of senior planners, who are required within in a local authority to appropriately manage LULUCF. We were told:

“[Local authorities] advertised at between £39,000 and £48,000 per annum while the private sector advertises similar roles for between £48,000 and £68,00 per annum”.

This leads to expertise being stripped out of the public sector by the private sector after employees have gained a few years’ experience.

There are several avenues of funding available to Scottish local authorities. However, it was the view of respondents that funding was piecemeal, short-term where local authorities needed a longer-term financial commitment and finite, which leads to competition across local authorities. There was a view shared across respondents that funders such as Scottish Government and NatureScot should look to review how funding is administered. A model was suggested in which funders work directly with each individual local authority to identify areas where funding could have the greatest impact at the local level. There was appreciation though that both Scottish Government and NatureScot are themselves suffering from budget and resourcing pressures to many of the local authorities, which hampers efforts to change existing models.

Summary

While very limited, the qualitative evidence indicates that many of the barriers highlighted by the interviews are aligned to those presented in the climate strategy documentary review. Further, the interviews also indicate that these barriers are interlinked and require a holistic approach to be overcome. For example, the lack of funding directly impacts capacity and capability within local authorities to deliver climate policy. This in turn directly impacts the maturity of data across the sector and, again, the local authority’s ability to deliver robust climate policies.

Considering the identified barriers to enacting climate policies, local authorities have nevertheless made significant inroads to developing some best-in-class policies that go above and beyond national ambitions. This is evident in the detail and narrative presented in multiple climate strategies. This shows there is a major interest and commitment by local authorities to tackle their territorial emissions. While policymaking in this area is limited in its scope, scale and consistency, local authorities interviewed demonstrated keenness to increase action.

Combined results

Table 4 combines the quantitative and qualitative research’s estimated potential impacts for the policies should they be implemented nationally. Appendix 13.1.2 describes the methodology used to arrive at the figures included and Appendix 13.2 lists the sources used.

Policy briefing: Nature-based solutions

Background

Biodiversity loss and the destruction of natural habitats is directly linked to climate change. Scottish forests, peatlands and bogs contribute to healthy eco systems. These systems work to remove CO₂ from our atmosphere and in some areas become large carbon sinks. According to the Biodiversity Intactness Indicator, Scotland has seen a 15% decline in its natural capital since 1950 with only 64% of our protected woodlands being in a favourable or recovering condition (Scottish Government, 2022).

Figure 5 shows the total estimated impact on LULUCF territorial GHG emissions by each individual policy, moving the inventory from 2.1 MtCO₂e emission per annum to (negative) -3.4 MtCO₂e through a combination of three polices.

Figure 5: Potential impact on LULUCF territorial GHG emissions across Scotland for a nature-based solutions policy

Greening Derelict Land

The rewilding policy outlined in Glasgow’s Climate Plan (Glasgow City Council, 2022) was one of the most developed we found during the quantitative review. It was used as the foundation to value the potential impact of nation-wide greening of derelict land. NatureScot estimated the total area of urban vacant and derelict land in Scotland in 2017 to be 11,649 hectares (Nature Scot, 2022). Across Scotland, 35% (4,077 ha) of urban vacant and derelict land can be thought of as being uneconomic to develop and/or is viewed as suitable to reclaim for a ‘soft’ end use (i.e. non-built use). The most common new use for sites that were previously urban vacant and derelict land was for residential development, with 50% of sites reclaimed for this purpose (Nature Scot, 2022). Changing land use for derelict land comes with many challenges for local authorities to consider including potential decontamination, private ownership, stakeholder relations, and internal ownership of the policy (see findings from the qualitative research in Section 7).

We have given an interim target of 2025 for greening to reach an estimated net gain in carbon sequestration potential of 2.2 MtCO₂e across Scotland by 2040. This figure is an upper bound estimate and was calculated on the basis of the following significant assumptions:

• 50% of the uneconomic land could be ’greened’ as described above.
• Derelict land is assumed to be neutral grassland that can be converted to coniferous woodland, applying carbon stock estimates (tC / ha) by habitat type and converting to MtCO₂e (Carbon Rewild, 2020).
• Afforested trees would reach their peak potential sequestration between 16 and 25 years of age (Carbon Store, n.d.).

Peatland Restoration

Scotland’s Nature Agency estimates that Scotland has some 1.8 Mha of blanket bog, representing 23% of the total land area (NatureScot, 2023). It is estimated that up to 80% of the total peatland area (1.44 Mha) is damaged. We have drawn on several policies across three local authorities that had detailed peatland restoration ambitions. The policies we reviewed sought to meet the pace of restoration set by Scottish Government of 20,000 ha restored per annum, with a target of 250,000 ha restored by 2030 (Scottish Government, 2020). Maintaining this pace of change to 2045 would mean a potential restoration of 0.55 Mha of peatland by 2045. The International Union for Conservation of Nature (IUCN) estimates that up to 4.6 tCO₂e per hectare could be reduced by restored peatland (IUCN, 2010). This produces an estimated carbon reduction potential of 2.5 MtCO₂e.

A strong caveat to the total potential restoration area is that much of the peatland across Scotland is under private ownership. Local authorities have limited powers outwith their own land ownership and may face significant challenges in convincing some private landowners to restore the peat on their land. In the absence of clear data on the area of peatland under private ownership, or other ownership covenants, for the purposes of estimating a potential GHG emission reduction we have made the broad assumption that these challenges could be overcome. However, if these challenges cannot be overcome it would severely reduce achievable emissions reductions.

Afforestation

We have used Stirling’s Climate & Nature Emergency Plan (Stirling Council, 2022) reforestation policy to plant 360,000 new trees by 2030, and 1 million new trees by 2045 as the basis for the modelled figures. The average kilogram of carbon dioxide sequestered by a mature tree is between 10kg CO₂ and 40kg CO₂ depending on age, species, and growing environment (EcoTree, 2023). For the purposes of estimation, 25kgCO₂ / tree / per annum has been used. Scaling this ambition to the national level, the total estimated removal of 0.8 MtCO₂ per annum across Scotland.

There are significant assumptions that sit behind the above estimation. These include:

• Stirling’s policy does not specify the type of land that will be converted, the detailed timescales for planting (impacting when the new tree stock will be at maturity), nor the preferred species of tree to be reforested.
• The policy does not value the GHG emission impact of planting new trees.
• We have assumed that the afforested trees will sequester emissions at their peak potential (i.e. a mature forest). This means the estimated emissions removals are limited by the fact we have not modelled a progressive change in sequestration over time, accounting for the growth of new woodland, such as that outlined by the Woodland Carbon Code (UK Woodland Carbon Code, 2021).

Summary

During our research we found that local authorities were eager to develop and create policies for land use that could make a quantifiable impact. One common theme across all local authorities was the consideration of peatland as one of the most impactful policies to reduce their carbon emissions. There are abundant resources provided by the IUCN peatland code (IUCN, 2023) that local authorities could access to begin developing strong peatland restoration policies.

Policy briefing: Net zero transport

Background

Scotland has ambitious targets to reduce transport emissions to net-zero by 2045 (Transport Scotland, 2019a). Transport emissions are one of the largest GHG inventory categories, accounting for 24% of overall territorial emissions (DESNZ, 2023). This is reflected in the number of transport policies identified across local authority climate change strategies. The policies in the section below demonstrate how local authorities are driving forward transport solutions.

Figure 6: Potential impact on transport territorial GHG emissions across Scotland for a net zero transport policy

Figure 6 shows the total estimated impact on transport GHG emissions by each individual policy, moving the inventory from 9,878 MtCO₂e emission per annum to 8,351 MtCO₂e through a combination of seven polices.

The Scottish National Transport Strategy states that 40% of transport emissions come from fossil fuelled cars. Recognising the impact that internal combustion engine cars have, local authorities have started to introduce policies targeted specifically at reducing these emissions. (Transport Scotland, 2019a).

High private use car use does not just affect GHG emissions, it also has a significant impact on air quality, health and pedestrian safety. Private car use contributes to high pollutions levels and with transport contributing to 1/6 of Scotland’s particulate matter (PM10) it is clear this is an area for policy focus (Transport Scotland, 2019a).

Local authorities understand the need for potent policies to be in line with national targets such as the goal to reduce car kilometres driven by 20% by 2030. These range from encouraging more active travel through the creation of active travel corridors and implementing low emissions zones in congested zones.

Active transport

The figures for this policy were modelled using Argyll and Bute Council’s Decarbonisation Plan 2022-2025 (Argyll and Bute Council, 2021). £2.3 million has been invested in delivering a wide range of active travel initiatives such as improved pathways, community cycle repair stands, cycle parking and new cycling routes. Through a combination of similar initiatives, a viable aim would be to convert 47% of remaining road journeys of up to 3km to active travel, which was the average proportion of active travel journeys up to 3km in 2019 (Transport Scotland, 2019b). The Council has committed to develop an Active Travel Strategy that would drive the policy forward at a future stage, but up to this point, resource to deliver the policy is dependent on external funding awards and is not covered by council budgets.

Homeworking

This policy has been valued as a proportion of the 262,000 Scottish FTE public sector total workforce (Scottish Government, 2022) working from home for 50% of their contracted hours. Reducing the average commute of 20 km round trip to office locations made in 73% of circumstances by personal car (Scottish Government, 2022b). Further potential emission reductions could be achieved through reduced operation of offices, such as heating, lighting, equipment and other operational emissions, although these have not been factored into our current study. However, it should be noted that emissions from reduced transport are minimal due to increased emissions associated with staff working from home (Riley et al., 2021).

Low-emission zones

Currently, there are four low emission zones (LEZ) in Scotland with enforcement for Dundee, Aberdeen and Edinburgh being introduced in 2024. Glasgow’s LEZ is integrated with the City Development Plan 2, Glasgow Transport Strategy and their Climate Plan to implement the change. The LEZ has been operating since 2018 with the aim of encouraging more active travel and public transport use in the city centre. The policy was implemented in phases to ensure low levels of disruption for residents, which should be a key consideration if scaling this across Scotland. Using findings from the London LEZ (Mayor of London, 2023), we have assumed a 4% CO₂ saving on emissions from transport on minor roads, to account for the fact LEZs will likely be operational in urban areas.

Decarbonisation of public transport

Climate targets published in the Stirling Climate & Nature Emergency Plan (Stirling Council, 2022) aim to reduce GHG emissions from public transport by an interim target of 25% in 2030, with an overall target of 75% by 2045. This has been extrapolated using population as a function to estimate the number of people served by public transport. However, the provision of public transport across Scotland is dependent on several factors, including sparseness of the population and socioeconomic circumstance, which are not accounted for in the potential emissions impact estimation. Further work should be undertaken to quantify the benefits.

Decarbonisation of fleet vehicles

This policy’s emissions were modelled using the estimated number of 28,800 fleet vehicles in the Scottish public sector (Scottish Futures Trust, 2022). We applied a conversion factor for assumed petrol cars, diesel LGVs and HGVs (BEIS, 2023). The average number of kilometres travelled annually is 12,000 km (Scottish Futures Trust, 2022). Post-conversion to EV emissions are zero, as per emission factor guidance. It is worth noting that EV technology for HGVs is under development and may not play a major role until post-2030 (Transport & Environment, 2023).

Council business travel

These emissions were estimated based on climate targets published in the Stirling Climate & Nature Emergency Plan (Stirling Council, 2022). The plan sets out the ambition of reducing baseline transport emissions (4,450 tCO₂) by the interim target of 45% by 2030, and the overall target emission reduction of 90% by 2045. This has been applied across the other local authorities, using population as a proxy. Further research to quantify emissions for each local authority would need to be carried out to refine these estimates.

LEV taxi licences

Stirling Climate & Nature Emergency Plan (Stirling Council, 2022) sets out the authority’s commitment to 100% of all taxis operating in the region being EVs by 2045. Using this as a foundation, we have valued the policy ambition in potential national GHG territorial emission impact.

There are 20,396 taxi licences registered across the 32 local authorities in Scotland of which 9,928 were registered as of 2021 (Transport Scotland, 2021). 1.9% are thought to be ULEVS (DfT, 2023). The policy will seek to increase the share of ULEV licences to 100% by 2045 effectively curtailing the emissions from private car hire.

To calculate the GHG emission impact, we anticipate that the average number of kilometres travelled per annum per capita is 80.85 km taken from the average number of trips made in the UK, by mode of transport (DESNZ, 2023) across the population of Scotland (5,563,000). Assumed that most private hire taxis are diesel cars, we applied the emission factor for a diesel car from BEIS company reporting datasets (BEIS, 2023) to calculate a saving on emission of 76.36 ktCO₂e.

Summary

It is clear from our research that transport is a key focus for all local authorities across Scotland due to the interlinked impacts spanning multiple socio-economic factors. Transport policies are very publicly visual in their delivery, making it easy for local authorities to point toward action being taken. In this section we have outlined some of the transport-related policies that could potentially be rolled out across Scotland’s local authorities. There is great potential to support local authorities to drive ambitious change in transport emissions, many of whom are already showing innovative solutions to enacting change in their local area. We have also given high-level estimates of potential emissions reductions if some of the most mature existing travel policies were scaled up.

Conclusions

Through pursuit of Local Heat and Energy Efficiency Strategies (LHEES), the Scottish Government has set the foundations for local authorities to drive their own locally led net zero agendas, directly tackling territorial greenhouse emissions from buildings. This research sought to investigate the role of local authorities in addressing emissions across other inventory categories, to replicate the success and best practice generated by LHEES.

From the evidence reviewed and from the interviews with local authorities, it is clear that there is local authority ambition to deliver climate policies that tackle local climate challenges, at the same time as delivering emissions reductions that go above and beyond national targets. Our climate strategy register details 69 current local authority climate-relevant strategies and describes the action being taken across all emission categories. We uncovered several climate change strategies that clearly detail intent, value their potential impacts and address resourcing and funding needs. Further research could be carried out to establish best-practice guidance on the development of climate policies, using existing local authority approaches as the foundation. This would help improve consistency across local authorities in how they value policy impacts and Scottish Government’s understanding of the resourcing, skills and funding needed to deliver.

This research assessed local authority strategies and policies to find where the most mature and impactful local authority climate policies have been developed. We scaled-up the emission reduction potential of the strongest of these local policies to give high-level-indicative estimates of what the impact could be in other local authorities and at a national level. Combining all of the analysis, we identified the greatest potential for impactful local authority controlled policies on territorial emissions to be within the LULUCF and transport categories.

For these to be implemented across Scotland, we found that the Scottish Government has a key role to play. They can provide effective leadership through facilitating best-practice knowledge sharing, improved access to skilled resource and targeted funding initiatives.

Territorial GHG policies are complex and data-driven, requiring specialist resource to develop and deliver, which we found does not always exist within individual local authorities. The Scottish Climate Intelligence Service has recently been launched in response to this barrier for many local authorities. Further research could expand on the capacity and capability requirements to deliver local authority climate policies between now and 2045, including methods by which the resourcing needs could be met.

Finally, funding is key to driving forward all the strategies and policies we have reviewed in this research. There are many pockets of funding available to local authorities to deliver climate policies. However, the interviews show that the funding is often piecemeal and short-term. Further investigation could help quantify the funding available for tackling each GHG inventory, where further funding might best be directed and methods for administrating funding to ensure that national ambitions can be met.

References

Argyll and Bute Council, 2021. Decarbonisation Plan 2022-2025, s.l.: s.n.

Audit Scotland, 2022. Scotland’s councils’ approach to addressing climate change. s.l.:s.n.

BEIS, 2022. Digest of UK Energy Statistics – Annual data for UK, 2021, London: Crown.

BEIS, 2023. Government conversion factors for company reporting of greenhouse gas emissions. [Online]
Available at: https://www.gov.uk/government/collections/government-conversion-factors-for-company-reporting

BEIS, 2023. Green Book supplementary guidance: valuation of energy use and greenhouse gas emissions for appraisal, London: Crown Commerical.

BEIS, 2023. UK Greenhouse Gas Emission Statistics: Frequently Asked Questions, London: Crown Copyright.

Carbon Rewild, 2020. Exploring the Carbon Capture Potential of Different Land Types. [Online]
Available at: https://carbonrewild.com/exploring-the-carbon-capture-potential-of-different-land-types/#:~:text=Wetlands%20are%20globally%20considered%20a,grow%20in%20or%20on%20water

Carbon Store, n.d. Carbon Sequestration. [Online]
Available at: https://carbonstoreuk.com/publications/carbon-sequestration-the-basics/#:~:text=As%20trees%20grow%2C%20the%20process,CO%E2%82%82%20as%20they%20are%20sequestering.

Climate Change Committee, 2022. Progress in reducing emissions in Scotland – 2022 Report to Parliament, s.l.: s.n.

Climate Change Order, 2015. Climate Change (Duties of Public Bodies: Reporting Requirements) (Scotland) Order 2015. s.l.:s.n.

DESNZ, 2023. Average number of trips by purpose and main mode. [Online]
Available at: Average number of trips by purpose and main mode

DESNZ, 2023. UK local authority and regional greenhouse gas emissions national statistics, 2005 to 2021. [Online]
Available at: https://www.gov.uk/government/statistics/uk-local-authority-and-regional-greenhouse-gas-emissions-national-statistics-2005-to-2021
[Accessed 01 November 2023].

DfT, 2023. NTS0409: Average number of trips and distance travelled by purpose and main mode: England, 2002 onwards, s.l.: Department for Transport.

EcoTree, 2023. How much CO2 does a tree absorb?. [Online]
Available at: https://ecotree.green/en/how-much-co2-does-a-tree-absorb

Environmental Standards Scotland, 2023. An investigation into the effectiveness of the systems in place to support local authorities in their duty to contribute to the delivery of climate change targets, s.l.: s.n.

Glasgow City Council, 2022. Glasgow’s Climate Plan, Glasgow: Glasgow City Council.

Greenhouse Gas Protocol, 2014. Policy and Action Standard: An accounting and reporting standard for estimating the greenhouse gas effects of policies and actions, s.l.: World Resources Institute.

HM Treasury, 2020. Magenta Book: Central Government guidance on evaluation, London: UK Government.

Improvement Service, 2023. Climate Intelligence Service. [Online]
Available at: https://www.improvementservice.org.uk/products-and-services/consultancy-and-support/climate-change/climate-intelligence-service

IPCC, 2006. Guidelines for National Greenhouse Gas Inventories, s.l.: s.n.

IUCN, 2010. Peatlands and Greenhouse Gas Emissions Reduction in Scotland, s.l.: IUCN.

IUCN, 2023. Peatland Code. [Online]
Available at: https://www.iucn-uk-peatlandprogramme.org/peatland-code-0

Mayor of London, 2023. Inner London Ultra Low Emission Zone Expansion One Year Report. [Online]
Available at: https://www.london.gov.uk/programmes-strategies/environment-and-climate-change/environment-and-climate-change-publications/inner-london-ultra-low-emission-zone-expansion-one-year-report?auHash=IxeIM3L6iJh-CwYvb2wek2UKMCSJvpOqMgtpRAMt5B8

Nature Scot, 2022. LLC3 Urban Vacant and Derelict Land, s.l.: s.n.

NatureScot, 2023. Blanket bog. [Online]
Available at: https://www.nature.scot/landscapes-and-habitats/habitat-types/mountains-heaths-and-bogs/blanket-bog#:~:text=Blanket%20bog%20is%20one%20of,23%25%20of%20our%20land%20area.

Net Zero, Energy and Transport Committee, 2023. The role of local government and its cross-sectoral partners in financing and delivering a net-zero Scotland, Edinburgh: The Scottish Parliament.

Scottish Futures Trust, 2022. Phasing out petrol and diesel cars & vans from public sector fleet, Edinburgh: Scottish Futures Trust.

Scottish Government, 2020. Update to the Climate Change Plan 2018 – 2032: Securing a Green Recovery on a Path to Net Zero, s.l.: s.n.

Scottish Government, 2022a. About public sector employment statistics. [Online]
Available at: https://www.gov.scot/publications/about-public-sector-employment-statistics/pages/local-government-employment/

Scottish Government, 2022b. Census data – Transport. [Online]
Available at: https://www.scotlandscensus.gov.uk/census-results/at-a-glance/transport/

Scottish Government, 2022. Biodiversity strategy to 2045: tackling the nature emergency, Edinburgh: ScotGov.

Scottish Government, 2022. Public Sector Employment in Scotland Statistics for 2nd Quarter 2022. [Online]
Available at: https://www.gov.scot/publications/public-sector-employment-scotland-statistics-2nd-quarter-2022/pages/6/

SSN, 2023b. Reporting: Public Bodies Climate Change Duties. [Online]
Available at: https://sustainablescotlandnetwork.org/reports
[Accessed 07 07 2023].

Stirling Council, 2022. Climate and Nature Emergency Plan 2021 – 2045, Stirling: Stirling Council.

Transport & Environment, 2023. Electric HGVs will soon be cheaper overall than diesels in the UK. [Online]
Available at: https://www.transportenvironment.org/discover/electric-hgvs-will-soon-be-cheaper-overall-than-diesels-in-the-uk/

Transport Scotland, 2019a. National Transport Strategy, Edinburgh: s.n.

Transport Scotland, 2019b. Transport and Travel in Scotland 2019: Results from the Scottish Household Survey, Edinburgh: s.n.

Transport Scotland, 2021. Scottish Transport Statistics, Glasgow: ScotGov.

Transport Scotland, 2022. Scottish Transport Statistics (2021) Personal and cross-modal travel, s.l.: Transport Scotland.

UK Woodland Carbon Code, 2021. 3.3 Project carbon sequestration. [Online]
Available at: https://www.woodlandcarboncode.org.uk/standard-and-guidance/3-carbon-sequestration/3-3-project-carbon-sequestration

Williamson, R., Sudmant, A., Gouldson, A. & Brogan, J., 2020. A Net-Zero Carbon Roadmap for Edinburgh, s.l.: PCAN & Edinburgh Climate Commission.

Appendices

Detailed methodology

Selection of climate strategies

The research identified 69 separate climate-related strategies across the 32 local authorities. To determine which were the key strategies to take forward to develop greenhouse gas emission boundaries, we designed five selection criteria to score each of the strategies against the metrics in 3.

We developed a screening matrix that ranked the strategies against five criteria outlined in Table 5 and determined the level of maturity on a scale of 1-3, assessing the level of evidence provided in a climate change strategy as yes / no / partial. We further embellished the five section criteria to ensure the strategies selected covered, as a collective, each of the six greenhouse gas emission inventories

Following presentation of the final policies selected with the steering group, a further consideration was made to ensure that at least one climate strategy from a local authority located outside of Scotland’s central belt was included, to ensure a better geographical spread. This resulted in the addition of Dumfries and Galloway Council to the climate boundary task.

Greenhouse gas emission boundaries and scenario emissions calculations and limitations

It is impractical to measure greenhouse gas emissions impact in real time from every chimney, exhaust, or acre of land use. GHG emission estimates are based on a series of models that estimate emissions from different sources (BEIS, 2023). The calculations performed for each of the scenario emissions is in line with international guidance (IPCC, 2006). We used government conversion factors for company reporting of greenhouse gas emissions (BEIS, 2023), Green Book supplementary guidance on the valuation of energy use and greenhouse gas emissions for appraisal (BEIS, 2023) and from IPCC guidance (IPCC, 2006). Other sources were researched from literature in the absence of standardised sets of emission factors.

The basic equation used to quantify scenario emissions is:

Equation 1: GHG scenario emissions

• Activity data is a variable that is changed by a policy. For example, a policy may look to reduce the number of kilometres travelled by private car.
• Emission factor is a constant that is used to convert the activity data to an impact. In most cases, this will be a GHG emission conversion factor.
• The impact estimate can either form a policy target or metric by which to measure success. Typically, this will be a GHG emission saving but it could also include other benefits (e.g. societal).

An example of this methodology in practice would be estimating GHG emissions from vehicles. The activity data might be the total number of kilometres travelled by that type of vehicle and the emission factor would be the amount of CO₂ emitted per kilometre.

Emission factors for energy sources are either dependent on the fuel characteristics (for emissions of CO₂) or how the fuel is burned, for example the size and efficiency of equipment used. For other sources, the emission factor can be dependent on a range of parameters, such as feed characteristics for livestock or the chemical reactions taking place for industrial process emissions. Emission factors are typically derived from measurements on several representative sources and the resulting factor applied to all similar sources in the UK.

This approach follows the ‘Tier 1’ approach as set out in IPCC guidance for national greenhouse gas inventories (IPCC, 2006):

Table 5: Quantification of GHG emission impact

An example of how an emission factor was applied to an activity is converting 1 tonne of municipal waste to 1 tonne of recycled waste as part of a landfill reduction strategy. Using emission conversion factors from government conversion factors for company reporting (BEIS, 2023), 1 tonne of waste sent to landfill has a greenhouse gas intensity of 497 kgCO₂e/tonne. A tonne of waste recycled has a greenhouse gas intensity of 21 kgCO₂e/tonne. Comparisons made between the two indicate a net greenhouse gas benefit of avoiding waste going to landfill.

As noted in Table 5, this is a basic methodological approach, using emissions and conversion factors from representative sources not specific to Scottish local authorities. In some instances, population data has been used as a proxy where local authority specific data was not available. The activity data was also derived from a variety of sources encompassing a range of levels of confidence (see Appendix 13.2). As such there is a high level of uncertainty in the estimated projected emissions reductions.

Sources for emissions equations

As described in the methodology section above, the figures presented in Tables 3, 4, 12, 13, 14, 15 and 16 and Figures 5 and 6 used the basic equation activity data x emission factor. The emissions factors were primarily drawn from Green Book supplementary guidance: valuation of energy use and greenhouse gas emissions for appraisal (BEIS, 2023) and Guidelines for National Greenhouse Gas Inventories (IPCC, 2006). However, in some cases additional sources were drawn on. The activity data was calculated using a range of sources. The sources are presented in Table 7 below, by GHG inventory category.

Table 7: Sources for emissions calculations by inventory category

Climate change strategy register

Quantifying impact

In the development of the emission boundaries, we applied two measures of assessing impact: Likelihood and Magnitude.

Likelihood

Likelihood is defined as the probability or chance that a given policy will achieve its intended impact or target. We have applied IPCC Guidance (IPCC, 2006) to determine likelihood as outlined in Table 8.

There are several considerations made when assessing the likelihood, a policy has in achieving its intended outcomes.

• Sphere of control: a measure of how much control a local authority has over whether action is taken against a policy. This ranges on a scale from absolute where a policy is enacted through legislation, through to voluntary where a policy results in stakeholders making a pledge.
• Capacity and capability: whether the local authority have the resources it needs to actively measure and enforce the provisions within a policy once it is active.
• Timescale: the impacts of policies may require consistent action taken over several years, or even decades. This can prove difficult as socioeconomic needs shift over time meaning that policies may also need to adapt over time, changing impacts and targets.

An example of a policy that is ‘very likely’ to meet its intended targets is a Low Emission Zone whereby a local authority has absolute ability to determine the classification of vehicles that enter its zone. Compare this to a policy improving active travel provision whereby the intended benefits are somewhat dependent on stakeholders enacting the policy out of their own free-will.

Magnitude

Magnitude is a simple measure of a policy’s potential impact on an inventory’s emissions. Following IPCC guidance (IPCC, 2006), we have set the following impact boundaries to rank the valued policies:

Policy descriptions

Table 10: Descriptions of 13 climate policies collated from six chosen local authorities for valuation, including example targets and KPIs set by the local authorities

Valuing greenhouse gas emissions

Table 21: Total territorial greenhouse gas emissions (ktCO₂e), by inventory (BEIS, 2022)

Table 32: Estimated potential impact on greenhouse gas emissions (ktCO2e) from Agriculture and LULUCF policies

Table 13: Estimated impact on greenhouse gas emissions (ktCO2e) from Transport policies

Table 44: Estimated potential impact on greenhouse gas emissions (ktCO2e) from Waste policies

Table 15: Estimated potential impact on total territorial greenhouse gas emissions (ktCO2e), by inventory