HoTRiverS project bearing fruit!

During 2017, we collected a vast amount of data from 5 rivers across England, Scotland and Wales, including river temperature measurements, meteorological observations and airborne thermal and optical imagery from drones.  Although we’re still fairly early in the analysis phase, we’re starting to see some really interesting results in our data:

Drone-based thermal imaging of river temperature
During May 2017, we used a drone equipped with a thermal infrared (TIR) camera to map surface water temperature patterns in Baddoch Burn, a tributary of the Aberdeenshire Dee (see image below).  Although previous research has shown how airborne thermal imaging surveys can be used to identify important cool-water habitats used by salmon and trout during summer high temperature, survey flights using conventional aircraft are very costly, and the location of these critical habitats often remains unknown. The lower cost of drone-based aerial surveys presents a potential solution to this problem. We were therefore interested to test how well miniaturised drone-based thermal imaging cameras performed when used to map water temperature.

Optical (left) and thermal (right) image mosaics of Baddoch Burn, Scotland

Although analysis of our TIR images is still ongoing, initial indications from our test flights at Baddoch Burn suggest that this thermal imaging technique will prove useful for mapping small-scale variations in surface water temperature of the kind that are useful to fish such as salmon and trout.

Following these good initial results from Baddoch Burn, we have subsequently acquired TIR imagery of the other 5 HoTRiverS study sites. We’re currently developing techniques to extract larger-scale river temperature patterns from the resulting thermal infrared image datasets. We hope to use these data to better understand how the different environmental processes across the HoTRiverS sites lead to variability in river temperature patterns; such data is highly important for understanding how climate change will impact different river environments in the future.

3D mapping of tree shade
River managers across the UK are currently planting trees on river banks to shade rivers during the hottest parts of the day. However, there is still considerable uncertainty regarding how and where trees should be planted in order to produce ‘optimum’ stream temperature reductions, due in part to the fact that detailed information regarding tree cover in the UK can be difficult to obtain, especially in remote locations. There is therefore a need to develop new techniques for understanding the impacts of tree cover on stream temperature.

On order to address this research need, we have developed a novel technique combining drones and computer models to simulate the impacts of shading on river temperature. During July 2017, we used a drone to obtain optical imagery of a 2 km section of Girnock Burn, another tributary of the Aberdeenshire Dee, Scotland.  Using a technique called ‘structure-from-motion photogrammetry’, which enables the extraction of 3D data from 2D photos (in a similar way to which stereo vision allows humans to perceive depth), we were able to map bankside tree heights with a very high degree of accuracy. This tree height data is subsequently input into a computer model to simulates the impacts that tree shading has on stream temperature, allowing us to better understand the role that shading has on stream temperature in a particular location.

When used to simulate stream temperatures in Girnock Burn, the new methodology highlights the extent to which tree shading reduces stream temperature. Indeed, our results show that on warm summer days, Girnock Burn could be >3 °C warmer than it currently is, were it not for the presence of tree shading.  We aim to roll out this technique to the other HoTRiverS study sites. By doing so, we hope that we will be able to better understand where and how bankside tree planting will have a greatest impact on reducing stream temperature, with a view to helping river managers protect streams against climate change.


Simulated impact of bankside tree shading on daily average stream temperature in lower Girnock Burn. The x-axis shows the distance from the confluence with the River Dee. Green line is the modelled temperature under current conditions. Red line is the modelled temperature under an imagined scenario with no trees.

New papers from the HoTRiverS team

The HoTRiverS team has recently published two new articles covering river temperature research.  The first of these two articles takes a look at how temperatures and heat exchange processes in streams are impacted by different types of riparian vegetation:

Dugdale, S.J., Malcolm, I.A., Kantola, K., & Hannah, D.M. (2018). Stream temperature under contrasting riparian forest cover: Understanding thermal dynamics and heat exchange processes. Science of The Total Environment, 610-611, 1375-1389

The second of these articles is a review of the current ‘state-of-the-art’ of process-based river temperature modelling, and will hopefully be useful for anyone looking to understand more about methods for stream temperature simulation:

Dugdale, S.J., Hannah, D.M., & Malcolm, I.A. (2017). River temperature modelling: A review of process-based approaches and future directions. Earth-Science Reviews, DOI: 10.1016/j.earscirev.2017.10.009

If you’d like a copy of either of the articles, please contact a member of the HoTRiverS team.

HoTRiverS project update

The intensive field season from spring – autumn 2017 is now almost over, and the time has come to start analysing our data.  Please watch this space for new updates from the HoTRiverS team.

ASCF/FCSA webinar on the remote sensing of river habitats

I was recently invited by the Atlantic Salmon Conservation Foundation / Fondation pour la conservation du saumon atlantique (Canada) to give a webinar on the use of remote sensing to map river habitats.  You can watch the video (in French only) here:

For more information about the ASCF / FCSA and the work they do, please visit their website.