Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland
Predicting drought and flood disaster-prone alternating zones and enhancing cropland disaster resilience are critical for agricultural water management, mitigating meteorological disaster risks, and ensuring food security. However, the spatial prediction of climate disaster vulnerability at the loca...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Agricultural Water Management |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S0378377425000630 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1825206944411942912 |
|---|---|
| author | Yongheng Shen Qingxia Guo Zhenghao Liu Yanli Shen Yikun Jia Yuehan Wei |
| author_facet | Yongheng Shen Qingxia Guo Zhenghao Liu Yanli Shen Yikun Jia Yuehan Wei |
| author_sort | Yongheng Shen |
| collection | DOAJ |
| description | Predicting drought and flood disaster-prone alternating zones and enhancing cropland disaster resilience are critical for agricultural water management, mitigating meteorological disaster risks, and ensuring food security. However, the spatial prediction of climate disaster vulnerability at the local scale faces challenges such as data gaps and insufficient resolution, which results in a lack of relevant research. This study uses a coupled model of particle swarm optimization (PSO), long short-term memory (LSTM), and graph attention network (GAT) integrating historical data to predict drought- and flood-prone areas in 2035 in Heshun County, Shanxi Province, a typical small-scale inland mountainous region of China. Additionally, the study assesses cropland resilience using the TOPSIS method, and based on the spatial distribution of drought and flood disasters, proposes a Flood-Drought-Resilience Analysis (FDRA) framework, further formulating a site selection strategy for future High Standard Farmland (HSF) projects. The overall findings indicate that: (1) Precipitation (Pr) and the Standardized Precipitation-Evapotranspiration Index (SPEI) have increased in recent years, with Pr expected to continue rising until 2035. (2) The integration of historical data with the predictions from the PSO-LSTM-GAT model reveals significant spatial overlap between historical and future disaster-prone areas and intensive cropland, especially in the central region. (3) Compared to single models, the PSO-LSTM-GAT model demonstrates significantly improved performance and precision in predicting drought- and flood-prone areas. (4) Through the FDRA integrated adjustment mechanism, 6.6668 km² of unsuitable land was identified, and 6.7349 km² of high-quality land was selected as the proposed site for the next round of HSF projects. In the final part of the study, management zoning plans were designed for other areas vulnerable to drought and flood disasters, and specific recommendations for enhancing cropland resilience were provided. This study provides a theoretical basis for enhancing agricultural disaster resilience and sustainable development in localized areas, offering scientific decision-making support for policymakers to address future climate change and disaster risks. |
| format | Article |
| id | doaj-art-5803bc4689bd4c7f915faf1b17edcfcb |
| institution | Kabale University |
| issn | 1873-2283 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Agricultural Water Management |
| spelling | doaj-art-5803bc4689bd4c7f915faf1b17edcfcb2025-02-07T04:46:50ZengElsevierAgricultural Water Management1873-22832025-03-01309109349Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmlandYongheng Shen0Qingxia Guo1Zhenghao Liu2Yanli Shen3Yikun Jia4Yuehan Wei5College of Resources and Environment, Shanxi Agricultural University (Shanxi Academy of Agricultural Sciences), Jinzhong, Shanxi Province 030801, ChinaCorresponding author.; College of Resources and Environment, Shanxi Agricultural University (Shanxi Academy of Agricultural Sciences), Jinzhong, Shanxi Province 030801, ChinaCollege of Resources and Environment, Shanxi Agricultural University (Shanxi Academy of Agricultural Sciences), Jinzhong, Shanxi Province 030801, ChinaCollege of Resources and Environment, Shanxi Agricultural University (Shanxi Academy of Agricultural Sciences), Jinzhong, Shanxi Province 030801, ChinaCollege of Resources and Environment, Shanxi Agricultural University (Shanxi Academy of Agricultural Sciences), Jinzhong, Shanxi Province 030801, ChinaCollege of Resources and Environment, Shanxi Agricultural University (Shanxi Academy of Agricultural Sciences), Jinzhong, Shanxi Province 030801, ChinaPredicting drought and flood disaster-prone alternating zones and enhancing cropland disaster resilience are critical for agricultural water management, mitigating meteorological disaster risks, and ensuring food security. However, the spatial prediction of climate disaster vulnerability at the local scale faces challenges such as data gaps and insufficient resolution, which results in a lack of relevant research. This study uses a coupled model of particle swarm optimization (PSO), long short-term memory (LSTM), and graph attention network (GAT) integrating historical data to predict drought- and flood-prone areas in 2035 in Heshun County, Shanxi Province, a typical small-scale inland mountainous region of China. Additionally, the study assesses cropland resilience using the TOPSIS method, and based on the spatial distribution of drought and flood disasters, proposes a Flood-Drought-Resilience Analysis (FDRA) framework, further formulating a site selection strategy for future High Standard Farmland (HSF) projects. The overall findings indicate that: (1) Precipitation (Pr) and the Standardized Precipitation-Evapotranspiration Index (SPEI) have increased in recent years, with Pr expected to continue rising until 2035. (2) The integration of historical data with the predictions from the PSO-LSTM-GAT model reveals significant spatial overlap between historical and future disaster-prone areas and intensive cropland, especially in the central region. (3) Compared to single models, the PSO-LSTM-GAT model demonstrates significantly improved performance and precision in predicting drought- and flood-prone areas. (4) Through the FDRA integrated adjustment mechanism, 6.6668 km² of unsuitable land was identified, and 6.7349 km² of high-quality land was selected as the proposed site for the next round of HSF projects. In the final part of the study, management zoning plans were designed for other areas vulnerable to drought and flood disasters, and specific recommendations for enhancing cropland resilience were provided. This study provides a theoretical basis for enhancing agricultural disaster resilience and sustainable development in localized areas, offering scientific decision-making support for policymakers to address future climate change and disaster risks.http://www.sciencedirect.com/science/article/pii/S0378377425000630Climate changeDrought-flood prone areasFarmland resilience enhancementParticle swarm optimization（PSO）Composite deep learning models |
| spellingShingle | Yongheng Shen Qingxia Guo Zhenghao Liu Yanli Shen Yikun Jia Yuehan Wei Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland Agricultural Water Management Climate change Drought-flood prone areas Farmland resilience enhancement Particle swarm optimization（PSO） Composite deep learning models |
| title | Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland |
| title_full | Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland |
| title_fullStr | Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland |
| title_full_unstemmed | Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland |
| title_short | Prediction of drought-flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high-standard farmland |
| title_sort | prediction of drought flood prone zones in inland mountainous regions under climate change with assessment and enhancement strategies for disaster resilience in high standard farmland |
| topic | Climate change Drought-flood prone areas Farmland resilience enhancement Particle swarm optimization（PSO） Composite deep learning models |
| url | http://www.sciencedirect.com/science/article/pii/S0378377425000630 |
| work_keys_str_mv | AT yonghengshen predictionofdroughtfloodpronezonesininlandmountainousregionsunderclimatechangewithassessmentandenhancementstrategiesfordisasterresilienceinhighstandardfarmland AT qingxiaguo predictionofdroughtfloodpronezonesininlandmountainousregionsunderclimatechangewithassessmentandenhancementstrategiesfordisasterresilienceinhighstandardfarmland AT zhenghaoliu predictionofdroughtfloodpronezonesininlandmountainousregionsunderclimatechangewithassessmentandenhancementstrategiesfordisasterresilienceinhighstandardfarmland AT yanlishen predictionofdroughtfloodpronezonesininlandmountainousregionsunderclimatechangewithassessmentandenhancementstrategiesfordisasterresilienceinhighstandardfarmland AT yikunjia predictionofdroughtfloodpronezonesininlandmountainousregionsunderclimatechangewithassessmentandenhancementstrategiesfordisasterresilienceinhighstandardfarmland AT yuehanwei predictionofdroughtfloodpronezonesininlandmountainousregionsunderclimatechangewithassessmentandenhancementstrategiesfordisasterresilienceinhighstandardfarmland |