Identifying Governing Factors of Streamflow Recession Behavior
Abstract
Catchment storage-discharge relation has been widely studied using the parameters of streamflow recession analysis, including streamflow stability (log(a)) and streamflow nonlinearity (b). These parameters provide a process-based insight on how catchments store and release water as well as the extent to which catchments are sensitive to changes in input climatic signals. Here, we tailored a newly developed statistical approach, Marginal Contribution Feature Importance (MCI), to hydrologic context in order to investigate governing physio-climatic factors of the catchment recession behavior. We explored this along a sample of more than 1,000 snowmelt and rain-dominated catchments across the United States and Canada. Results show that streamflow stability in medium size catchments (between 50 and 1000 km2) is heavily influenced by climatic attributes in both rain-dominated and snow-dominated catchments. The relative importance of snowmelt related attributes (e.g., maximum snow depth and snow density) in snow-dominated catchments are greater than in rain-dominated ones. Other climatic attributes describing catchment overall water balance (e.g., aridity index), and soil and bedrock properties (e.g. hydraulic conductivity) are less important in the snow-dominated catchments than in rain-dominated ones. Streamflow nonlinearity is heavily influenced by catchment slope and geological properties in both rain-dominated and snow-dominated catchments, while the influences of geological attributes on streamflow nonlinearity are lesser in snow-dominated catchments compared to rain-dominated catchments. As scale increase, i.e. in large (>1000 km2) catchments, the relative importance of climatic attributes, describing the catchment overall water balance, increases for both streamflow stability and non-linearity, and becomes the sole dominant driver of recession parameters, while the relative importance of soil and bedrock hydraulics are all dropped. The findings of this study on understanding governing factors of recession parameters and the interpretation of the cause of spatial variability of these parameters, could inform developing a process-based generalizable understanding of streamflow sensitivity to climate change in different environmental settings.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.H22G..10L