Low flow controls on stream thermal dynamics

Water level fluctuations in surface water bodies, and in particular low flow drought conditions, are expected to become more frequent and more severe in the future due to the impacts of global environmental change. Variations in water level, and therefore in-channel water volume, not only have the potential to directly impact stream temperature, but also aquatic vegetation coverage which, in turn, may affect stream temperature patterns and dynamics. Manipulation experiments provide a systematic approach to investigate the multiple environmental controls on stream temperature patterns. This study aims to use temperature data loggers and fibre optic distributed temperature sensing (FO-DTS) to investigate potential drought impacts on patterns in surface water and streambed temperature as a function of change in water column depth. To quantify the joint impacts of water level and associated vegetation coverage on stream temperatures, investigations were conducted in outdoor flumes using identical pool-riffle-pool features, but with spatially variable water levels representative of different drought severity conditions. Naturally evolved vegetation growth in the flumes ranged from sparse vegetation coverage in the shallow flumes to dense colonization in the deepest. Observed surface water and streambed temperature patterns differed significantly within the range of water levels and degrees of vegetation coverage studied. Streambed temperature patterns were more pronounced in the shallowest flume, with minimum and maximum temperature values and diurnal temperature variation being more intensively affected by variation in meteorological conditions than daily average temperatures. Spatial patterns in streambed temperature correlated strongly with morphologic features in all flumes, with riffles coinciding with the highest temperatures, and pools representing areas with the lowest temperatures. In particular, the shallowest flume (comprising multiple exposed features) exhibited a maximum upstream-downstream temperature warming of 3.3 °C (T in = 10.3 °C, T out = 13.5 °C), exceeding the warming observed in the deeper flumes by ∼2 °C. Our study reveals significant streambed and water temperature variation caused by the combined impacts of water level and related vegetation coverage. These results highlight the importance of maintaining minimum water levels in lowland rivers during droughts for buffering the impacts of atmospheric forcing on both river and streambed water temperatures