Home | english  | Impressum | Sitemap | KIT

High resolution temperature observations to identify different runoff processes

High resolution temperature observations to identify different runoff processes

Martijn Westhoff 


Delft Cluster, TU Delft 





Headwater catchments are important contributors to streamflow. They are small, but all combined they influence river flow significantly. To be able to make proper runoff predictions under different climate conditions and changing land use, it is important to have detailed understanding of the discharge processes in the headwater catchments.

In this research we explore the possibilities of fibre optic Distributed Temperature Sensing (DTS) to obtain more insight in temporal and spatial discharge dynamics during stormflow. DTS is a technique capable of measuring temperature along a fibre optic cable of up to 10 km length at a spatial resolution of 1 to 2 m and a temporal resolution as short as 10 s

In order to be able to quantify the hydrological fluxes within a first order stream in Luxembourg, we coupled an energy balance model with a routing and advection-dispersion model. Through this coupling temperature has become a high resolution tracer. The difference between observed and simulated temperature should then be caused by different exchange fluxes, which subsequently can be quantified.

During the research we found heat exchange with in-stream rock clasts appeared to be important. The many abundant rock clasts present on top of the streambed store heat, resulting in retarded heat transport.

We also developed a new method to quantify surface water-groundwater interactions (or hyporheic exchange). Stream water infiltrates into the subsurface (hyporheic zone) where it remains for a while before it returns to the stream. This process influences in-stream temperature, making it possible to quantify this flux at a high resolution with more flexibility in experimental design.

In the last step we extended the model to unravel spatial and temporal dynamics in discharge during a short but intensive summer rainstorm. Using the model as a learning tool, we showed that for such an event, gains of water remained constant over the event, stream losses increased with increasing discharge, and hyporheic exchange appeared to increase with discharge for part of the stream. From the modelling results, we also concluded that a side channel becomes active several hours after the start of the rainfall event.

This study has shwon that high resolution temperature observations can be instrumental in understanding detailed processes in hydrological systems.