A distant source of runoff can only contribute to stream flow during the times at which it is efficiently connected to the stream network.
Photo: M. Rinderer
ICaRuS - Investigating Catchment Runoff Response by Assessing Spatial Patterns of Groundwater Dynamics
The project builds upon previous work on topographic controls on catchment-scale groundwater dynamics (Rinderer et al., 2014, Rinderer et al., 2016) in deriving spatial patterns of groundwater response under consideration of the catchment physiographic characteristics, rainfall characteristics and antecedent soil saturation using time series clustering (Rinderer et al., 2017). Distance-based similarity measures (the integral of two time series) captured similarity in the shape and amplitude of the groundwater time series and were correlated with topographic indices. Correlation-based similarity measures (based on the cross-correlation of two time series) were more sensitive to seasonal differences in groundwater dynamics.
Based on these results it was possible to transfer groundwater dynamics from more than 50 monitoring sites to all non-monitored locations of a Swiss pre-alpine catchment. The groundwater dynamics of midslope locations were found to be most difficult to predict but had already been highlighted in earlier studies to be of key-importance in establishing hydrological connectivity between uphill sites and the lower hillslopes and streams, respectively (Rinderer et al., 2017).
An analysis of 19 months-long time series showed that the area with groundwater response that also was connected to the stream and therefore likely contributed to runoff, expanded and contracted predominantly aligned with the channel network. Isolated areas with groundwater response got connected to the stream network during short periods of time that were associated with highest streamflow (Rinderer et al., in preparation). Clear exponential relations between specific discharge and connected area or total catchment storage were identified that were very similar for the rising and falling limb of event hydrographs and for events with multiple peaks (Rinderer et al., in preparation).
The results of this project show the importance of hydrological connectivity to better understand the often non-liner runoff response in headwater streams.
B. McGlynn (Earth and Ocean Sciences, Duke University, NC, USA)
I. van Meerveld (University of Zurich, CH)
Funded by: Swiss Science Foundation, SNF
Rinderer et al., 2017