Hydrology and Earth System Sciences (2020-11-01)

Predicting probabilities of streamflow intermittency across a temperate mesoscale catchment

  • N. H. Kaplan,
  • T. Blume,
  • M. Weiler

Journal volume & issue
Vol. 24
pp. 5453 – 5472


Read online

The fields of eco-hydrological modelling and extreme flow prediction and management demand detailed information of streamflow intermittency and its corresponding landscape controls. Innovative sensing technology for monitoring of streamflow intermittency in perennial rivers and intermittent reaches improves data availability, but reliable maps of streamflow intermittency are still rare. We used a large dataset of streamflow intermittency observations and a set of spatial predictors to create logistic regression models to predict the probability of streamflow intermittency for a full year as well as wet and dry periods for the entire 247 km2 Attert catchment in Luxembourg. Similar climatic conditions across the catchment permit a direct comparison of the streamflow intermittency among different geological and pedological regions. We used 15 spatial predictors describing land cover, track (road) density, terrain metrics, soil and geological properties. Predictors were included as local-scale information, represented by the local value at the catchment outlet and as integral catchment information calculated as the mean catchment value over all pixels upslope of the catchment outlet. The terrain metrics catchment area and profile curvature were identified in all models as the most important predictors, and the model for the wet period was based solely on these two predictors. However, the model for the dry period additionally comprises soil hydraulic conductivity and bedrock permeability. The annual model with the most complex predictor set contains the predictors of the dry-period model plus the presence of tracks. Classifying the spatially distributed streamflow intermittency probabilities into ephemeral, intermittent and perennial reaches allows the estimation of stream network extent under various conditions. This approach, based on extensive monitoring and statistical modelling, is a first step to provide detailed spatial information for hydrological modelling as well as management practice.