Estimating the Routing Parameter of the Xin’anjiang Hydrological Model Based on Remote Sensing Data and Machine Learning

Abstract

The parameters of hydrological models should be determined before applying those models to estimate or predict hydrological processes. The Xin’anjiang (XAJ) hydrological model is widely used throughout China. Since the prediction in ungauged basins (PUB) era, the regionalization of the XAJ model parameters has been a subject of intense focus; nevertheless, while many efforts have targeted parameters related to runoff yield using in-site data sets, classic regression has predominantly been applied. In this paper, we employed remotely sensed underlying surface data and a machine learning approach to establish models for estimating the runoff routing parameter, namely, CS, of the XAJ model. The study was conducted on 114 catchments from the Catchment Attributes and MEteorology for Large-sample Studies (CAMELS) data set, and the relationships between CS and various underlying surface characteristics were explored by a gradient-boosted regression tree (GBRT). The results showed that the drainage density, stream source density and area of the catchment were the three major factors with the most significant impact on CS. The best correlation coefficient (r), root mean square error (RMSE) and mean absolute error (MAE) between the GBRT-estimated and calibrated CS were 0.96, 0.06 and 0.04, respectively, verifying the good performance of GBRT in estimating CS. Although bias was noted between the GBRT-estimated and calibrated CS, runoff simulations using the GBRT-estimated CS could still achieve results comparable to those using the calibrated CS. Further validations based on two catchments in China confirmed the overall robustness and accuracy of simulating runoff processes using the GBRT-estimated CS. Our results confirm the following hypotheses: (1) with the help of large sample of catchments and associated remote sensing data, the ML-based approach can capture the nonstationary and nonlinear relationships between CS and the underlying surface characteristics and (2) CS estimated by ML from large samples has a robustness that can guarantee the overall performance of the XAJ mode. This study advances the methodology for quantitatively estimating the XAJ model parameters and can be extended to other parameters or other models.

Forest Tree Species Diversity Mapping Using ICESat-2/ATLAS with GF-1/PMS Imagery

“…As the base learner of GBRT, each new regression tree fit is taught by the residuals taught by the prior regression tree. It continuously reduces residuals in a gradient-boosting manner and adjusts the learning rate to enhance the learning effect [67,68]. The gradient boosting regression tree GBRT algorithm can be broken down into several fundamental stages, as shown in Table 4 [67], based on a summary of related research.…”

Sensor Networks, Data Processing, and Inference: The Hydrology Challenge

“…[67], [69] [86], [131], [148], [149] [47] [38] [39], [40] [83], [84], [85], [101], [105], [134], [136], [137], [147], [150] [42], [43], [44] [75] [132] using theoretical models (namely, Green-Ampt model, see later) during storm events generating runoff. Comparing the first and second plots, we can observe that the hyetographs for storm events that result in runoff are typically skewed to the right, denoting a tendency for the storm to increase in intensity after some time from the start of rainfall.…”

“…Previous attempts in using ML algorithm for parametersć alibration simply seek to provide efficient tools for finding the values of the model´s parameters in their actual configuration. With this aim, researchers have leveraged the ML algorithms as optimization tools using mainly deep learning [147]- [149] and other algorithms as regression trees [150]. However, ML algorithms should pave the way for streamlining models, and shaping supervised data-driven models in view of physical ones can help deduce a better parameters' structure identifying self-correlations, redundant information, or patterns between dependent and independent variables that were previously unknown.…”

A Synthesis of Global Streamflow characteristics, Hydrometeorology, and catchment Attributes (GSHA) for Large Sample River-Centric Studies

“…A recent study filled in the discontinuity and latency of gauge records, and provided streamflow for over 45,000 gauges with improved data quality (Riggs et al, 2023). These global-scale datasets have been widely used in data driven machine learning models (Kratzert et al, 2019a(Kratzert et al, , 2019bRen et al, 2020), physical hydrological models (Aerts et al, 2022;Clark et al, 2021), and parameter estimation and regionalization studies (Addor et al, 2018;Fang et al, 2022).…”