A locally explained heterogeneity model for examining wetland disparity
ABSTRACTIdentifying the factors influencing wetland variations is crucial for understanding the relationship of climate change with wetland conservation and management. The wetland distribution is associated with multiple variables, and the interactions among these variables are complex. In this stu...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2023-12-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/17538947.2023.2271883 |
| _version_ | 1827775979316051968 |
|---|---|
| author | Yang Li Peng Luo Yongze Song Liqiang Zhang Ying Qu Zhengyang Hou |
| author_facet | Yang Li Peng Luo Yongze Song Liqiang Zhang Ying Qu Zhengyang Hou |
| author_sort | Yang Li |
| collection | DOAJ |
| description | ABSTRACTIdentifying the factors influencing wetland variations is crucial for understanding the relationship of climate change with wetland conservation and management. The wetland distribution is associated with multiple variables, and the interactions among these variables are complex. In this study, we aim to explore an interpretable and quantitative analysis of factors related to wetland spatiotemporal variations on the Tibetan Plateau (TP). By combining SHapley Additive exPlanations with a spatially stratified heterogeneity model, we propose a locally explained stratified heterogeneity (LESH) model that well reveals the effects of multiple variable interactions on the spatiotemporal variations of wetlands. The results show that topographic variables are the most important variables related to the spatial distribution of wetlands on the TP, and climatic variables are the most relevant factors for the increase in the wetland area on the TP from 2015 to 2019. In addition, the interactions among multiple variables strongly influence wetlands on the TP. Among them, when other geographic variables interact with the evaporation variable, its explanatory power on wetland distribution is significantly increased. Knowledge of wetland distribution determinants can help us understand the evolution of wetlands and the impacts of climate change on wetland variations. |
| first_indexed | 2024-03-11T13:57:40Z |
| format | Article |
| id | doaj.art-f15dbc2cffb74bfd9da896e29cc82e9f |
| institution | Directory Open Access Journal |
| issn | 1753-8947 1753-8955 |
| language | English |
| last_indexed | 2024-03-11T13:57:40Z |
| publishDate | 2023-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | International Journal of Digital Earth |
| spelling | doaj.art-f15dbc2cffb74bfd9da896e29cc82e9f2023-11-02T05:56:16ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552023-12-011624533455210.1080/17538947.2023.2271883A locally explained heterogeneity model for examining wetland disparityYang Li0Peng Luo1Yongze Song2Liqiang Zhang3Ying Qu4Zhengyang Hou5Key Laboratory of Remote Sensing Science, Faculty of Geographical Science, Beijing Normal University, Beijing, People’s Republic of ChinaChair of Cartography and Visual Analytics, Technical University of Munich, Munich, GermanySchool of Design and the Built Environment, Curtin University, Perth, AustraliaKey Laboratory of Remote Sensing Science, Faculty of Geographical Science, Beijing Normal University, Beijing, People’s Republic of ChinaKey Laboratory of Remote Sensing Science, Faculty of Geographical Science, Beijing Normal University, Beijing, People’s Republic of ChinaKey Laboratory of Remote Sensing Science, Faculty of Geographical Science, Beijing Normal University, Beijing, People’s Republic of ChinaABSTRACTIdentifying the factors influencing wetland variations is crucial for understanding the relationship of climate change with wetland conservation and management. The wetland distribution is associated with multiple variables, and the interactions among these variables are complex. In this study, we aim to explore an interpretable and quantitative analysis of factors related to wetland spatiotemporal variations on the Tibetan Plateau (TP). By combining SHapley Additive exPlanations with a spatially stratified heterogeneity model, we propose a locally explained stratified heterogeneity (LESH) model that well reveals the effects of multiple variable interactions on the spatiotemporal variations of wetlands. The results show that topographic variables are the most important variables related to the spatial distribution of wetlands on the TP, and climatic variables are the most relevant factors for the increase in the wetland area on the TP from 2015 to 2019. In addition, the interactions among multiple variables strongly influence wetlands on the TP. Among them, when other geographic variables interact with the evaporation variable, its explanatory power on wetland distribution is significantly increased. Knowledge of wetland distribution determinants can help us understand the evolution of wetlands and the impacts of climate change on wetland variations.https://www.tandfonline.com/doi/10.1080/17538947.2023.2271883Wetland distributionspatial heterogeneityspatial associationsSHAP |
| spellingShingle | Yang Li Peng Luo Yongze Song Liqiang Zhang Ying Qu Zhengyang Hou A locally explained heterogeneity model for examining wetland disparity International Journal of Digital Earth Wetland distribution spatial heterogeneity spatial associations SHAP |
| title | A locally explained heterogeneity model for examining wetland disparity |
| title_full | A locally explained heterogeneity model for examining wetland disparity |
| title_fullStr | A locally explained heterogeneity model for examining wetland disparity |
| title_full_unstemmed | A locally explained heterogeneity model for examining wetland disparity |
| title_short | A locally explained heterogeneity model for examining wetland disparity |
| title_sort | locally explained heterogeneity model for examining wetland disparity |
| topic | Wetland distribution spatial heterogeneity spatial associations SHAP |
| url | https://www.tandfonline.com/doi/10.1080/17538947.2023.2271883 |
| work_keys_str_mv | AT yangli alocallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT pengluo alocallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT yongzesong alocallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT liqiangzhang alocallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT yingqu alocallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT zhengyanghou alocallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT yangli locallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT pengluo locallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT yongzesong locallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT liqiangzhang locallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT yingqu locallyexplainedheterogeneitymodelforexaminingwetlanddisparity AT zhengyanghou locallyexplainedheterogeneitymodelforexaminingwetlanddisparity |