NB27 Summer low flow events in Scottish rivers (Normalised Flow Index)

Despite generally being considered a wet country, Scotland can be vulnerable to prolonged periods of dry weather, which can result in pressure upon the environment and water users in some areas (SEPA, 2014).  In some regions of the country, particularly in the North and West Highlands, water supply sources have relatively little storage capacity in comparison to the larger supply sources of the Central Belt.  This makes the task of maintaining the public water supply without resorting to special measures more vulnerable to the types of dry summer episodes highlighted by this indicator.

Whilst the ecology of rivers can generally adapt to a natural range of conditions and water levels, prolonged dry periods can lead to detrimental impacts on the environment. Aquatic ecology can be vulnerable to low flows in summer when water temperature is at its highest and wetted habitat space and dissolved oxygen are at their lowest. During low flow events, there is reduced dilution capacity in rivers and, as a consequence, the concentrations of pollutants can increase markedly.  As river levels drop, changes in flow may fragment the river and reduce suitable habitat space. A low flow in a river may also prevent migratory fish from moving upstream and disrupt food supplies by impacting on aquatic invertebrates (SEPA, 2014).

Climate change projections indicate hotter, drier summers for much of Scotland by the 2050s, potentially leading to an increase in periods of water scarcity.

Analysis of historical drought impacts in Scotland indicates that it is departure from normal rather than changes in absolute water supply that determines whether water scarcity impacts occur (Gosling, 2014).  Water scarcity indices provide a measure of how exceptional a period is in comparison to a long term average. In order to improve their ability to plan for and respond to water scarcity events, SEPA utilise a number of measures to monitor conditions across Scotland (SEPA, 2014). This indicator draws upon analysis of low flow events in Scotland’s rivers as defined by the Normalised Flow Index (NFI).  This measures how much the flow over a 30 or 90 day period deviates from the long term average. 

The two time periods are presented in order to assess the different drought types that can impact water resources in Scotland (both acute and chronic). In regions with little storage capacity, risk of water supply and environmental impacts may best be indicated by one month duration, whereas regions with significant storage are likely to be more vulnerable to anomalies of a longer duration (Gosling, 2014).

Related indicators:

NA13 Abstraction of water for irrigation

NB33 Progress towards the environmental objectives of the River Basin Management Plans

BW7 Customers and zones vulnerable to supply deficit

Figures 1 and 2, and Table 1, show the low flow situation during summer 2014.  The summer 30-day NFI showed that there were mild low flow conditions across much of the country with the largest deficit occurring in the Clyde region.  The 90-day index illustrates the longer term situation across June, July and August.  Again, although not particularly severe, the Clyde experienced the lowest flows relative to the long-term average.

Table 1 Condition classes for 30-day (1^st August 2014) and 90-day (1^st September 2014) low flow indices (NFI) for SEPA Area  Advisory Group regions

Over the period 1981 to 2014, there has been very little evidence of any trends in the severity of summer low flow events (Tables 2 and 3).  Only one gauging station (in Argyll) has shown a significant increase in the severity of 30-day summer low flow events.  As for the 90-day index, nine stations have shown a trend of decreasing severity of these longer duration summer low flows but the majority indicate no trend.

Table 2 Changes in severity of summer 30 day low flow events by SEPA Area Advisory Group between 1981 and 2014. Numbers refer to number of gauging stations in each category.

Table 3 Changes in severity of summer 90 day low flow events by SEPA Area Advisory Group between 1981 and 2014. Numbers refer to number of gauging stations in each category.

Using the UKCP09 climate model outputs as inputs to rainfall-runoff models it has been shown that reductions in summer flows are projected across many of Scotland’s rivers by 2050 (Haxton et al., 2012).  These flows have been used to project changes in the indices of low flow events presented here i.e. the 30 and 90-day normalised flow indices (Gosling 2014).  These results indicate a projected increase in severity and frequency of summer and autumn low flow events.  Exceptional summer low flow events with a return period of 1 in 40 years for the current baseline period (1961-90) are projected to have median return periods as low as 1 in 9 years by the 2050s.

The lack of observable trend in historical records of summer low flows may not contradict the projected changes in flows implied by UKCP09. Several studies have recently shown that, given the rate at which climate is expected to change and given the high natural variability of the maritime climate of the UK, it may not be possible to detect climate change in river flows for some time to come (Fowler et al., 2010; Radziejewski & Kundzewicz, 2004).

The indices are only calculable at gauging stations measuring near natural flows and which have a long period of record.  There are 147 such stations covering the period 1981 to present day.  In some areas such as the North West Highlands there are few gauging stations which fit these criteria and there are none in Orkney and Shetland.  For assessing trends over a longer period, fewer records are available.  Shorter record lengths inhibit the ability to detect statistically significant trends in the flow data which have high year on year variability.  Having said this, 147 gauging stations with records longer than 30 years provides a good base for detecting trends in the future.

Fowler, H. J., Wilby, R. L., Cooley, D., Sain, S. R. & Thurston, M. (2010) Detecting changes in UK precipitation extremes using climate model projections: Implications for managing

fluvial flood risk. Proceedings of the Third BHS International Symposium, Newcastle University, UK, July 2010.

Gosling, R. (2014) Assessing the impact of projected climate change on drought vulnerability in Scotland. Hydrology Research 45 (6), 806-815

Haxton, T., Crooks, S., Jackson, C. R., Barkwith, A. K. A. P., Kelvin, J., Williamson, J., Mackay, J. D., Wang, L., Davies, H., Young, A. & Prudhomme, C. (2012)  Future Flows Hydrology. Available at http://dx.doi.org/10.5285/f3723162-4fed-4d9d-92c6-dd17412fa37b.

Radziejewski, M. & Kundzewicz, Z. W. (2004) Detectability of changes in hydrological records. Hydrological Sciences Journal 49 (1), 39–51.

SEPA (2014) Scotland’s National Water Scarcity Plan (Consultation document Dec 2014). Available online at: https://consultation.sepa.org.uk/water-unit/water-scarcity-plan/supporting_documents/Consultation%20Scotlands%20National%20Water%20Scarcity%20Plan.pdf

Scotland’s National Water Scarcity Plan (consultation document)

https://consultation.sepa.org.uk/water-unit/water-scarcity-plan/supporting_documents/Consultation%20Scotlands%20National%20Water%20Scarcity%20Plan.pdf

Development of this indicator and primary author of this document:  Richard Gosling, Scottish Environment Protection Agency (SEPA)

Roisin Murray-Williams of SEPA for analysing the trends in the flow indices