The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning
IntroductionSoil erosion is a critical issue faced by many regions around the world, especially in the purple soil hilly areas. Rainfall and slope, as major driving factors of soil erosion, pose a significant challenge in quantifying their impact on hillslope runoff and sediment yield. While existin...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2025-04-01
|
| Series: | Frontiers in Environmental Science |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fenvs.2025.1580149/full |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850199404297846784 |
|---|---|
| author | Naichang Zhang Zhaohui Xia Peng Li Qitao Chen Ganggang Ke Fan Yue Fan Yue Yaotao Xu Tian Wang |
| author_facet | Naichang Zhang Zhaohui Xia Peng Li Qitao Chen Ganggang Ke Fan Yue Fan Yue Yaotao Xu Tian Wang |
| author_sort | Naichang Zhang |
| collection | DOAJ |
| description | IntroductionSoil erosion is a critical issue faced by many regions around the world, especially in the purple soil hilly areas. Rainfall and slope, as major driving factors of soil erosion, pose a significant challenge in quantifying their impact on hillslope runoff and sediment yield. While existing studies have revealed the effects of rainfall intensity and slope on soil erosion, a comprehensive analysis of the interactions between different rainfall types and slope is still lacking. To address this gap, this study, based on machine learning methods, explores the effects of rainfall type, rainfall amount, maximum 30-min rainfall intensity (I30), and slope on hillslope runoff depth (H) and erosion-induced sediment yield (S), and unveils the interactions among these factors.MethodsThe K-means clustering algorithm was used to classify 43 rainfall events into three types: A-type, B-type, and C-type. A-type is characterized by long duration, large rainfall amounts, and moderate intensity; B-type by short duration, small rainfall amounts, and high intensity; and C-type is intermediate between A-type and B-type. The Random Forest (RF) algorithm was employed to assess the impacts of these factors on runoff and sediment yield, along with a feature importance analysis.ResultsThe results show that rainfall amount has the most significant impact on runoff and sediment yield. Under different rainfall types, the ranking of the effects of rainfall amount and I30 on H and S is as follows: rainfall amount (C>A>B), I30 (A>B>C). The impact of slope follows a trend of first increasing and then decreasing, with varying degrees of influence on H and S depending on the rainfall type.DiscussionThe novelty of this study lies in combining machine learning techniques to systematically evaluate, for the first time, the interactions between rainfall type and slope and their impact on hillslope runoff and sediment yield in purple soil hilly areas. This research not only provides a theoretical basis for soil erosion control but also offers scientific support for the precise prediction and management of soil conservation measures in purple soil regions. |
| format | Article |
| id | doaj-art-348fd02501f946eab92bc533bf48a6ab |
| institution | OA Journals |
| issn | 2296-665X |
| language | English |
| publishDate | 2025-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Environmental Science |
| spelling | doaj-art-348fd02501f946eab92bc533bf48a6ab2025-08-20T02:12:38ZengFrontiers Media S.A.Frontiers in Environmental Science2296-665X2025-04-011310.3389/fenvs.2025.15801491580149The impact of rainfall and slope on hillslope runoff and erosion depending on machine learningNaichang Zhang0Zhaohui Xia1Peng Li2Qitao Chen3Ganggang Ke4Fan Yue5Fan Yue6Yaotao Xu7Tian Wang8Power China Northwest Engineering Corporation Limited, Xi’an, ChinaPower China Northwest Engineering Corporation Limited, Xi’an, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an, ChinaPower China Northwest Engineering Corporation Limited, Xi’an, ChinaShaanxi Union Research Center of University and Enterprise for River and Lake Ecosystems Protection and Restoration, Xi’an, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an, ChinaState Key Laboratory of Eco-Hydraulics in Northwest Arid Region of China, Xi’an University of Technology, Xi’an, ChinaIntroductionSoil erosion is a critical issue faced by many regions around the world, especially in the purple soil hilly areas. Rainfall and slope, as major driving factors of soil erosion, pose a significant challenge in quantifying their impact on hillslope runoff and sediment yield. While existing studies have revealed the effects of rainfall intensity and slope on soil erosion, a comprehensive analysis of the interactions between different rainfall types and slope is still lacking. To address this gap, this study, based on machine learning methods, explores the effects of rainfall type, rainfall amount, maximum 30-min rainfall intensity (I30), and slope on hillslope runoff depth (H) and erosion-induced sediment yield (S), and unveils the interactions among these factors.MethodsThe K-means clustering algorithm was used to classify 43 rainfall events into three types: A-type, B-type, and C-type. A-type is characterized by long duration, large rainfall amounts, and moderate intensity; B-type by short duration, small rainfall amounts, and high intensity; and C-type is intermediate between A-type and B-type. The Random Forest (RF) algorithm was employed to assess the impacts of these factors on runoff and sediment yield, along with a feature importance analysis.ResultsThe results show that rainfall amount has the most significant impact on runoff and sediment yield. Under different rainfall types, the ranking of the effects of rainfall amount and I30 on H and S is as follows: rainfall amount (C>A>B), I30 (A>B>C). The impact of slope follows a trend of first increasing and then decreasing, with varying degrees of influence on H and S depending on the rainfall type.DiscussionThe novelty of this study lies in combining machine learning techniques to systematically evaluate, for the first time, the interactions between rainfall type and slope and their impact on hillslope runoff and sediment yield in purple soil hilly areas. This research not only provides a theoretical basis for soil erosion control but also offers scientific support for the precise prediction and management of soil conservation measures in purple soil regions.https://www.frontiersin.org/articles/10.3389/fenvs.2025.1580149/fullpurple soilrandom forestclustering algorithmslope scalesoil erosion |
| spellingShingle | Naichang Zhang Zhaohui Xia Peng Li Qitao Chen Ganggang Ke Fan Yue Fan Yue Yaotao Xu Tian Wang The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning Frontiers in Environmental Science purple soil random forest clustering algorithm slope scale soil erosion |
| title | The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning |
| title_full | The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning |
| title_fullStr | The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning |
| title_full_unstemmed | The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning |
| title_short | The impact of rainfall and slope on hillslope runoff and erosion depending on machine learning |
| title_sort | impact of rainfall and slope on hillslope runoff and erosion depending on machine learning |
| topic | purple soil random forest clustering algorithm slope scale soil erosion |
| url | https://www.frontiersin.org/articles/10.3389/fenvs.2025.1580149/full |
| work_keys_str_mv | AT naichangzhang theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT zhaohuixia theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT pengli theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT qitaochen theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT ganggangke theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT fanyue theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT fanyue theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT yaotaoxu theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT tianwang theimpactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT naichangzhang impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT zhaohuixia impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT pengli impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT qitaochen impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT ganggangke impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT fanyue impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT fanyue impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT yaotaoxu impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning AT tianwang impactofrainfallandslopeonhillsloperunoffanderosiondependingonmachinelearning |