Understanding aquatic carbon loss from upland catchments in south west Scotland during land use change from commercial forest to wind farm

Abstract

High concentrations and fluxes of dissolved organic carbon (DOC) in fluvial systems are associated with the dark brown water colour familiar in many upland, peat-dominated areas and may indicate a depletion of the terrestrial carbon store. The removal of this colour can also be problematic and expensive for water companies as well as affecting the ecological functioning of the water body through factors such as reduced light penetration through the water column. Disturbance resulting from activities such as land use change can also enhance the loss of carbon and this may manifest itself in elevated concentrations and fluxes of DOC from aquatic systems. This thesis describes and explains patterns of change in DOC quantity and quality from the Crosswater, Crosswater of Luce and Tig catchments draining Arecleoch Forest, a peatland in south Ayrshire, Scotland, from 2008 to 2010. This time period incorporates the installation of a 60-turbine wind farm built and operated by Scottish Power Renewables (SPR). Water samples were collected from Arecleoch at different spatial scales ranging from catchments to soil pore water and temporal scales ranging from daily to seasonally. Concentrations of DOC were measured and fluxes estimated at the catchment scale. DOC concentrations from all three catchments exhibited the well-established seasonal pattern with maxima in late August/early September and minima seen in February/March. The Tig catchment experienced the greatest burden of disturbance from the wind farm development and returned the highest DOC concentrations and fluxes. The Crosswater catchment, used as a control site due to its isolation from wind farm activities, had higher DOC concentrations than the Crosswater of Luce throughout the monitoring period possibly due to a greater proportion of forest cover. ii DOC flux ranged from 35.0 g C m-2 yr-1 from the Crosswater of Luce catchment in 2008 to 55.0 g C m-2 yr-1 from the Crosswater in 2009. The Tig catchment was not monitored for the whole period but returned the highest DOC fluxes of the three catchments between January and June 2010 (15.7 g C m-2). These values are considered high for UK peatlands. It is possible to make a tentative estimate of an extra 12 g C m-2 being exported from the Crosswater of Luce in 2009 that may have been a result of wind farm and/or forestry activities in the catchment. At the sub-catchment scale, “hot spots” of high DOC concentrations (up to 113.4 mg L-1) were found during the final survey of headwater streams inside the development area of the wind farm site during construction in August 2010. Further surveys are recommended to assess whether DOC concentrations have decreased since completion of the wind farm. Daily water samples were collected upstream and downstream of turbine 33 during the excavation of the turbine base. DOC concentrations were higher downstream before work began on the turbine base and although the gap between upstream and downstream DOC concentrations increased over the monitoring period, statistical comparisons of these differences before and after the start of excavation work were not significant at the 95 % confidence level. Challenges arose from the practicability of conducting robust research on a construction site and novel approaches to monitoring DOC were developed. Activity scores were used to quantify the effect of peatland disturbance on DOC concentrations at the catchment scale. The results suggest that this approach may have merit but requires comprehensive site records from the developer. The non-linear nature of the individual wind farm development and forestry activities made it impractical to disentangle the impact of each, particularly for forest harvesting. iii Activity scores could, together with other information gathered from site records, be useful to developers as an indicator of the most likely periods for peat disturbance. Knowledge of the differing disturbance potential of the various activities could also provide useful information to feed into the carbon payback calculator. DOC quality was explored using ultraviolet (UV) absorbance, specific UV absorbance (SUVA) and E4/E6 ratios. The latter metric identified changes in the composition of DOC related to disturbance with water samples from areas draining land subject to disturbance having lower E4/E6 ratios indicating a greater degree of humification of the DOC. This research provides one of only three studies to investigate concentrations and fluxes of DOC in water courses draining land subject to disturbance relating to wind farm construction. It is the only study that incorporates a period of time prior to work beginning and takes in the whole of the development phase. In this respect it provides a valuable addition to our understanding of the way in which peatlands respond to land use change and may provide useful tools to assist developers in minimising the impact of their activities on these valuable carbon stores.