Data Driven Discoveries in Streamflow, Vadose zone, and Baseflow

Abstract

Chapter 1: 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. Chapter 2: Machine learning has been broadly applied for vadose zone applications in recent years. This article provides a comprehensive review of such development. ML applications for variables corresponding to different complex vadose zone processes are summarized mostly in a prediction context. By analyzing and assessing these applications, we discovered extensive usages of classic machine learning models with relatively limited applications of Deep Learning (DL) approaches in general. We also recognized a lack of benchmark datasets for soil property research as well as limited integration of physics-based vadose zone principles into the machine learning approaches. To facilitate this interdisciplinary research of machine learning in vadose zone characterization and processes, a paradigm of knowledge guided machine learning is suggested along with other data driven and machine learning model-based research suggestions to advance future research. Chapter 3: Baseflow recession analysis is essential for catchment hydrology and modeling storage discharge relationships. Recently, it has been recognized an inconsistency between theoretical recession characteristics as indicated from classic Boussinesq solutions and empirical recession characteristics. We hypothesized that this inconsistency is because of the ambiguity of baseflow recessions and streamflow recessions. The empirical recession analysis does not completely exclude quick flow components in streamflow-based recessions and thus leads to a mixing effect of surface runoff and baseflow. Thus, in a continental scale across United States, we performed recession analysis over streamflow dataset and digital filter derived baseflow dataset over two recession criteria. The gage averaged recession parameter a is found statistically different between baseflow and streamflow. In particular, the baseflow derived a is more theoretically consistent to what Boussinesq solutions indicate in short and intermediate time domains and supports our hypothesis. This discovery suggests that baseflow recession analysis results are likely impacted by mis-included quick flow and evapotranspiration effects, which shed light on improving recession extraction criteria as well as advancing relevant groundwater study.