loading page

Regionalization in a global hydrologic deep learning model: from physical descriptors to random vectors
  • +9
  • Xiang Li,
  • Ankush Khandelwal,
  • Xiaowei Jia,
  • Kelly Cutler,
  • Rahul Ghosh,
  • Arvind Renganathan,
  • Shaoming Xu,
  • J L Nieber,
  • Christopher J Duffy,
  • Michael Steinbach,
  • Vipin Kumar,
  • Kshitij Tayal
Xiang Li
University of Minnesota, University of Minnesota, University of Minnesota

Corresponding Author:[email protected]

Author Profile
Ankush Khandelwal
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
Xiaowei Jia
University of Pittsburgh, University of Pittsburgh, University of Pittsburgh
Author Profile
Kelly Cutler
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
Rahul Ghosh
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
Arvind Renganathan
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
Shaoming Xu
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
J L Nieber
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
Christopher J Duffy
Pennsylvania State University, Pennsylvania State University, Pennsylvania State University
Author Profile
Michael Steinbach
University of Minnesota, University of Minnesota, University of Minnesota
Author Profile
Vipin Kumar
Department of Computer Science and Engineering, University of Minnesota, Department of Computer Science and Engineering, University of Minnesota, Department of Computer Science and Engineering, University of Minnesota
Author Profile
Kshitij Tayal
University of Minnesota Twin Cities, University of Minnesota Twin Cities
Author Profile

Abstract

Streamflow prediction is a long-standing hydrologic problem. Development of models for streamflow prediction often requires incorporation of catchment physical descriptors to characterize the associated complex hydrological processes. Across different scales of catchments, these physical descriptors also allow models to extrapolate hydrologic information from one catchment to others, a process referred to as “regionalization”. Recently, in gauged basin scenarios, deep learning models have been shown to achieve state of the art regionalization performance by building a global hydrologic model. These models predict streamflow given catchment physical descriptors and weather forcing data. However, these physical descriptors are by their nature uncertain, sometimes incomplete, or even unavailable in certain cases, which limits the applicability of this approach. In this paper, we show that by assigning a vector of random values as a surrogate for catchment physical descriptors, we can achieve robust regionalization performance under a gauged prediction scenario. Our results show that the deep learning model using our proposed random vector approach achieves a predictive performance comparable to that of the model using actual physical descriptors. The random vector approach yields robust performance under different data sparsity scenarios and deep learning model selections. Furthermore, based on the use of random vectors, high-dimensional characterization improves regionalization performance in gauged basin scenario when physical descriptors are uncertain, or insufficient.