Greenhouse Gas Release from Reservoirs in Scotland and North Wales

Abstract

Inland waters play an important role in the transport, transformation, loss and storage of carbon (C) and nitrogen (N) in the pathway between terrestrial and marine systems. Inland waters can act as both sources to the atmosphere through evasion of GHGs such as CO2, CH4 and N2O, but also as sinks through C burial in sediments. To date, there are no major studies on GHG release from reservoirs in the UK, despite drinking water provision being heavily reliant on peatland-fed reservoirs. The overall aim of this thesis was therefore to quantify, compare and understand C and N dynamics, both spatially and temporally, within 30 temperate reservoirs in North Wales (n = 15) and Scotland (n = 15). Chapter 3 presents a broad overview of these 30 reservoirs, where the role of catchment characteristics in determining spatial variation of reservoir biogeochemistry is investigated. Three of the Scottish reservoirs included in Chapter 3 were also selected for greater in-depth sampling at weekly to fortnightly intervals to provide greater temporal understanding of C and N, and become the two remaining data chapters of this thesis. Chapter 4 quantifies GHG export from the catchments of Baddinsgill (moorland) and Black Esk (forested) reservoirs, whilst Chapter 5 determines the importance of water level drawdown on GHG emissions from Waltersmuir reservoir. In Chapter 3, results show that the 30 sampled reservoirs were overall, oversaturated in CO2, CH4 but to a lesser extent in N2O, largely reflecting differences in catchment land cover and soil types. Large temporal and spatial differences in C and N concentrations were also observed, with seasonal differences occurring between reservoirs in Scotland and North Wales. Influx of CO2 was occasionally seen, with one reservoir showing overall negative evasion which was linked to the presence of cyanobacteria. Chapter 4 revealed that GHG evasion from inlets was higher than the reservoirs at both Baddinsgill and Black Esk, emphasising the need for integrated, catchment wide monitoring across stream and reservoir systems. Annual and areal Fluvial C and N fluxes were calculated for both catchments, and the input-output balance also calculated to determine whether the reservoirs were net C and N sources or sinks. Results revealed both reservoirs as overall net sinks for C and N, apart from at Baddinsgill, which was a net CH4 source. Dissolved inorganic carbon (DIC) was the largest C input to both reservoirs, whilst nitrate (NO3-) was the largest N input. All aquatic systems experience natural fluctuations in water level, this is often more extreme in reservoir environments due to seasonal demands or operational maintenance. Chapter 5 determines the impact of a controlled water level drawdown on GHG emissions from both sediments and the surface water area at Waltersmuir reservoir. As the water level drops and sediments become exposed to the air and begin to dry out, biogeochemical processes are temporarily altered. Results showed pulses of GHG fluxes from the reservoir during the drawdown period. Such episodic events were indicative of physical displacement of accumulated sediment gases, which can contribute significantly to total reservoir emissions. The three month drawdown period at Waltersmuir reservoir contributed disproportionately to total annual emissions (66% of CO2 equivalent-weighted emissions). This study adds to a growing body of evidence that suggests reservoir drawdown zones are active areas of biogeochemical cycling and can stimulate CO2, CH4 and N2O release. Conclusively, this thesis shows that GHG concentrations and fluxes from UK reservoirs are highly temporally and spatially variable, with magnitudes comparable to other studies in temperate and boreal regions. Fluctuating water levels can regulate emissions and should be considered in budget studies, with the relative importance expected to increase under climate scenarios. Future studies which aim to quantify GHG dynamics from UK reservoirs should focus on integrated catchment wide monitoring (i.e. including any inflows, the reservoir itself, and outflows) to gain fuller understanding of the reservoir impact on C and N, and also potential impacts of catchment management on fluxes. Such monitoring is resource intensive so it is recommended that future efforts focus on reservoirs across different tropic states (i.e. oligotrophic to eutrophic) covering both lowland and upland catchments.