Global S&T Development Trend Analysis Platform of Resources and Environment
DOI | 10.1029/2019WR025924 |
Challenges in Applying Machine Learning Models for Hydrological Inference: A Case Study for Flooding Events Across Germany | |
Schmidt, Lennart1,2; Hesse, Falk2,3; Attinger, Sabine2,3; Kumar, Rohini2 | |
2020-04-23 | |
发表期刊 | WATER RESOURCES RESEARCH |
ISSN | 0043-1397 |
EISSN | 1944-7973 |
出版年 | 2020 |
卷号 | 56期号:5 |
文章类型 | Article |
语种 | 英语 |
国家 | Germany |
英文摘要 | Machine learning (ML) algorithms are being increasingly used in Earth and Environmental modeling studies owing to the ever-increasing availability of diverse data sets and computational resources as well as advancement in ML algorithms. Despite advances in their predictive accuracy, the usefulness of ML algorithms for inference remains elusive. In this study, we employ two popular ML algorithms, artificial neural networks and random forest, to analyze a large data set of flood events across Germany with the goals to analyze their predictive accuracy and their usability to provide insights to hydrologic system functioning. The results of the ML algorithms are contrasted against a parametric approach based on multiple linear regression. For analysis, we employ a model-agnostic framework named Permuted Feature Importance to derive the influence of models' predictors. This allows us to compare the results of different algorithms for the first time in the context of hydrology. Our main findings are that (1) the ML models achieve higher prediction accuracy than linear regression, (2) the results reflect basic hydrological principles, but (3) further inference is hindered by the heterogeneity of results across algorithms. Thus, we conclude that the problem of equifinality as known from classical hydrological modeling also exists for ML and severely hampers its potential for inference. To account for the observed problems, we propose that when employing ML for inference, this should be made by using multiple algorithms and multiple methods, of which the latter should be embedded in a cross-validation routine. |
英文关键词 | machine learning inference floods |
领域 | 资源环境 |
收录类别 | SCI-E |
WOS记录号 | WOS:000537736400023 |
WOS关键词 | SOIL-MOISTURE ; WATER FLUXES ; UNCERTAINTY ; INTERPRETABILITY ; PRECIPITATION ; CAPABILITIES ; RISK |
WOS类目 | Environmental Sciences ; Limnology ; Water Resources |
WOS研究方向 | Environmental Sciences & Ecology ; Marine & Freshwater Biology ; Water Resources |
URL | 查看原文 |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | http://119.78.100.173/C666/handle/2XK7JSWQ/249191 |
专题 | 资源环境科学 |
作者单位 | 1.UFZ Helmholtz Ctr Environm Res, Dept Monitoring & Explorat Technol, Leipzig, Germany; 2.UFZ Helmholtz Ctr Environm Res, Dept Computat Hydrosyst, Leipzig, Germany; 3.Univ Potsdam, Inst Earth & Environm Sci, Potsdam, Germany |
推荐引用方式 GB/T 7714 | Schmidt, Lennart,Hesse, Falk,Attinger, Sabine,et al. Challenges in Applying Machine Learning Models for Hydrological Inference: A Case Study for Flooding Events Across Germany[J]. WATER RESOURCES RESEARCH,2020,56(5). |
APA | Schmidt, Lennart,Hesse, Falk,Attinger, Sabine,&Kumar, Rohini.(2020).Challenges in Applying Machine Learning Models for Hydrological Inference: A Case Study for Flooding Events Across Germany.WATER RESOURCES RESEARCH,56(5). |
MLA | Schmidt, Lennart,et al."Challenges in Applying Machine Learning Models for Hydrological Inference: A Case Study for Flooding Events Across Germany".WATER RESOURCES RESEARCH 56.5(2020). |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论