Regional Patterns and Physical Controls of Streamflow Generation Across the Conterminous United States

Abstract

A long-standing, foundational research topic in hydrological science is understanding streamflow generation mechanisms during storm events, including how the occurrence and relative dominance of different streamflow generation mechanisms change with their climatic and catchment physiographic controls. A particular goal has been to construct a universal conceptual framework that permits the a priori determination of dominant streamflow generation mechanism(s), that is, the mechanism(s) that have the greatest contribution to streamflow in any given catchment, based on available climatic and catchment physiographic characteristics. Such a conceptual framework could assist in the selection of appropriate model structures for hydrological simulations and predictions (

Strength and Memory of Precipitation's Control Over Streamflow Across the Conterminous United States

“…Based on a large-sample analysis of six different hydrological signatures across catchments in the United States, Wu et al (2021) Wu et al's six signatures (2021) distinguished among saturation overland flow and two subsurface quickflow mechanisms, which they referred to as SSF1 (lateral quickflow through shallow soils that arises when the water table rises into transmissive near-surface layers) and SSF2 (lateral quickflow in macropores or along a poorly transmissive interface in the subsurface). We compared the results of test Case 2 to studies of streamflow generation and catchment classification by Wu et al (2021) and other investigators to extend our evaluation of the utility of TE-based tools for model-free inference of streamflow generation and catchment behavior to real datasets (Figure 8). For ease of comparison of continuous TE metrics with discrete categories, we discretized 𝐴𝐴 TE𝑚𝑚𝑚𝑚𝑚𝑚𝑚𝑚→𝑄𝑄 and TE max timing into "high" or "low" values, based on whether they were above or below the data set median.…”

A Hydrologic Functional Approach for Improving Large‐Sample Hydrology Performance in Poorly Gauged Regions

“…As recently emphasized through a statistical analysis conducted by Wu et al. (2021), the magnitudes of water input intensity, subsurface storage and water input volume are not sufficient to distinguish among four categories of streamflow generation mechanisms. Further distinguishing among saturation excess overlandflow, subsurface stormflow, transmissivity feedback, and groundwater flow requires the knowledge of catchment slope, soil‐bedrock conductivity contrast and depth to bedrock (see table 3 of Wu et al., 2021).…”

“…Smaller catchments were discarded because of the downscaling uncertainty in climatic attributes, while larger catchments were discarded owing to a potentially large river routing time delay impact on streamflow response that was not considered in our indices/attributes. These cut‐offs for catchment size were adopted based on the experience of similar LSH studies (e.g., Sawicz et al., 2011; Wu et al., 2021). Catchments with a coefficient of variation (i.e., the relative variation around the average catchment‐scale value) of soil and geological attributes larger than 50%. Catchments with relative surface water (e.g., wetland, pothole) area larger than 10% of the total catchment area.…”

“…Specifically, this study responds to calls made by Oudin et al (2010), Troch, Berne, et al (2013), and more recently by Knoben et al (2018), Gnann et al (2021) and Wu et al (2021), to create new process-based descriptors of catchment functional behaviors for applications in large-sample or comparative hydrology, in order to improve the estimation of (dis)similarity in hydrologic behaviors among catchments and to improve the predictions of streamflow signatures in poorly gauged regions. In doing so, we introduce three interactive functional indices to assess whether the addition of the functional indices to climatic indices, or the use of the interactive functional indices instead of their individual climatic and physical components, as catchment descriptors, could:…”

Identification of the controlling factors for hydrological responses by artificial neural networks

“…The coupled effects of meteorological, topographic, pedologic, and vegetation factors on hydrological responses, lead to a complicated and nonlinear relationship between rainfall and runoff (Ares et al, 2020; Mathias et al, 2016). A foundational research topic in hydrological processes requires further exploration to facilitate the understanding of hydrological responses at the basin scale (Bartlett et al, 2016; Jin, Zhang, Gu, Tian, & He, 2015; Van Nieuwenhuyse et al, 2011; Wu et al, 2021).…”