Early in the project, we identified five rivers across the UK with contrasting climatological, hydrological and geological (and hence, river temperature) characteristics. A representative section of each of these rivers was subsequently instrumented with a range of sensors in order to characterise river temperature patterns and heat transfers to and from the river. Following installation of these instruments, we used a drone equipped with a thermal infrared (TIR) camera to map surface water temperature patterns in each of the rivers. Comparison of the river temperature data obtained from the drone against ‘real’ temperature data recorded by temperature sensors installed in the rivers indicated that the drone was not sufficiently accurate to allow for large-scale characterisation of water temperature heterogeneity. However, the drone-based data was nonetheless useful for mapping small ‘point’ cool water inputs that engendered small-scale changes in river temperature.
The next phase of the project consisted of mapping the environmental context of each river location. We combined GIS mapping with field data from the river corridor and drone-based topographic surveys to produce detailed databases of each river (necessary for the subsequent implementation of river temperature models). The drone-based topographic surveys (carried out in collaboration with the project partner, Marine Scotland Science) resulted in the development of a new methodology for assessing the impact of tree shading on river temperature patterns (detailed below); this new technique has the potential to substantially aid the management of river temperature extremes through the improved identification and modelling of tree shading effects.
These data collection phases of the project were followed by the development of river temperature models for each site. We conducted an exhaustive meta-analysis regarding the advantages and disadvantages of the various river temperature models available to identify the most appropriate one to use, subsequently publishing our findings in a scientific journal (Earth-Science Reviews). The results of this analysis will be particularly beneficial to river managers who are unsure as to the most appropriate approaches for modelling river temperature patterns. Our chosen river temperature model was subsequently supplied with data from the data collection phase of the project, and used to simulate temperatures in the study rivers; results of these analyses clearly show the role of a variety of different processes (eg. tree shading, evaporative heat loss, groundwater inflow) in generating patterns of stream temperature heterogeneity. We have not yet completed model implementation at all sites, so the final objective of the project (an assessment of the impacts of climate/land-use change on future river temperature heterogeneity) is still ongoing. However, work to achieve this aim is currently underway.
Results of these objectives were presented at a range of international conferences, including the EnviroDrones workshop (Darmouth College, USA; May 2017), HydroEco 2017 (Birmingham, UK; June 2017), Americal Geosciences Union fall meeting (New Orleans, USA; December 2017) and European Geosciences Union annual meeting (Vienna, Austria; April 2018).
