{"title":"流域尺度下山区梯田废弃过程及驱动机制","authors":"Huili Guo , Liquan Sun , Shufang Wu , Kadambot H.M. Siddique","doi":"10.1016/j.ecoleng.2025.107582","DOIUrl":null,"url":null,"abstract":"<div><div>Terraces, an important cultivation resource in hilly and gully areas, are increasingly being abandoned due to various factors affecting the regional ecological environment. This abandonment poses a significant threat to national food security. Consequently, it is crucial to investigate the extent, spatial distribution, and driving factors behind terrace abandonment in these regions. The driving factors of abandoned terraces in mountainous areas are different from those in plain areas. However, our understanding of the driving factors of abandoned terraces in mountainous areas remains limited. To address this knowledge gap, we used deep learning and change detection methods to identify abandoned terraces in hilly and gully regions. We analyzed the temporal and spatial characteristics of multi-year terrace distribution, and used a geographic detector to determine the driving factors and degree of terrace abandonment. We found that: (1) between 2009 and 2012, the terrace abandonment rate ranged from 10.55 to 31.66 %, with an average annual growth rate of 7.03 %, and between 2013 and 2022, the abandonment rate of terrace ranged from 31.79 to 36.68 %; (2) terraces with prolonged abandoned durations were located primarily in high-elevation areas with limited accessibility. More than 50 % of the abandoned terraces have been unattended for more than 8 years, with the largest proportion (22.28 % of the total abandoned terrace area) abandoned for 11 years; (3) in a relatively consistent socioeconomic context, the distance to unclassified roads emerged as the main driving factor influencing terrace abandonment, explaining 80.5 % of the variation. Unlike unterraced land, the terrace platform size also played a significant role; (4) the global Moran's index of 0.562 indicates that adjacent terrace abandonment significantly affects terrace abandonment. Our study introduces novel methods for identifying abandoned terraces and highlights important driving factors behind their abandonment. These findings offer valuable insights and serve as a reference for future research on abandoned terraces.</div></div>","PeriodicalId":11490,"journal":{"name":"Ecological Engineering","volume":"215 ","pages":"Article 107582"},"PeriodicalIF":3.9000,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"The process and driving mechanism of abandoned terraces in mountain region at the watershed scale\",\"authors\":\"Huili Guo , Liquan Sun , Shufang Wu , Kadambot H.M. Siddique\",\"doi\":\"10.1016/j.ecoleng.2025.107582\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Terraces, an important cultivation resource in hilly and gully areas, are increasingly being abandoned due to various factors affecting the regional ecological environment. This abandonment poses a significant threat to national food security. Consequently, it is crucial to investigate the extent, spatial distribution, and driving factors behind terrace abandonment in these regions. The driving factors of abandoned terraces in mountainous areas are different from those in plain areas. However, our understanding of the driving factors of abandoned terraces in mountainous areas remains limited. To address this knowledge gap, we used deep learning and change detection methods to identify abandoned terraces in hilly and gully regions. We analyzed the temporal and spatial characteristics of multi-year terrace distribution, and used a geographic detector to determine the driving factors and degree of terrace abandonment. We found that: (1) between 2009 and 2012, the terrace abandonment rate ranged from 10.55 to 31.66 %, with an average annual growth rate of 7.03 %, and between 2013 and 2022, the abandonment rate of terrace ranged from 31.79 to 36.68 %; (2) terraces with prolonged abandoned durations were located primarily in high-elevation areas with limited accessibility. More than 50 % of the abandoned terraces have been unattended for more than 8 years, with the largest proportion (22.28 % of the total abandoned terrace area) abandoned for 11 years; (3) in a relatively consistent socioeconomic context, the distance to unclassified roads emerged as the main driving factor influencing terrace abandonment, explaining 80.5 % of the variation. Unlike unterraced land, the terrace platform size also played a significant role; (4) the global Moran's index of 0.562 indicates that adjacent terrace abandonment significantly affects terrace abandonment. Our study introduces novel methods for identifying abandoned terraces and highlights important driving factors behind their abandonment. These findings offer valuable insights and serve as a reference for future research on abandoned terraces.</div></div>\",\"PeriodicalId\":11490,\"journal\":{\"name\":\"Ecological Engineering\",\"volume\":\"215 \",\"pages\":\"Article 107582\"},\"PeriodicalIF\":3.9000,\"publicationDate\":\"2025-03-07\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Ecological Engineering\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0925857425000709\",\"RegionNum\":2,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ECOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ecological Engineering","FirstCategoryId":"93","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0925857425000709","RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ECOLOGY","Score":null,"Total":0}
The process and driving mechanism of abandoned terraces in mountain region at the watershed scale
Terraces, an important cultivation resource in hilly and gully areas, are increasingly being abandoned due to various factors affecting the regional ecological environment. This abandonment poses a significant threat to national food security. Consequently, it is crucial to investigate the extent, spatial distribution, and driving factors behind terrace abandonment in these regions. The driving factors of abandoned terraces in mountainous areas are different from those in plain areas. However, our understanding of the driving factors of abandoned terraces in mountainous areas remains limited. To address this knowledge gap, we used deep learning and change detection methods to identify abandoned terraces in hilly and gully regions. We analyzed the temporal and spatial characteristics of multi-year terrace distribution, and used a geographic detector to determine the driving factors and degree of terrace abandonment. We found that: (1) between 2009 and 2012, the terrace abandonment rate ranged from 10.55 to 31.66 %, with an average annual growth rate of 7.03 %, and between 2013 and 2022, the abandonment rate of terrace ranged from 31.79 to 36.68 %; (2) terraces with prolonged abandoned durations were located primarily in high-elevation areas with limited accessibility. More than 50 % of the abandoned terraces have been unattended for more than 8 years, with the largest proportion (22.28 % of the total abandoned terrace area) abandoned for 11 years; (3) in a relatively consistent socioeconomic context, the distance to unclassified roads emerged as the main driving factor influencing terrace abandonment, explaining 80.5 % of the variation. Unlike unterraced land, the terrace platform size also played a significant role; (4) the global Moran's index of 0.562 indicates that adjacent terrace abandonment significantly affects terrace abandonment. Our study introduces novel methods for identifying abandoned terraces and highlights important driving factors behind their abandonment. These findings offer valuable insights and serve as a reference for future research on abandoned terraces.
期刊介绍:
Ecological engineering has been defined as the design of ecosystems for the mutual benefit of humans and nature. The journal is meant for ecologists who, because of their research interests or occupation, are involved in designing, monitoring, or restoring ecosystems, and can serve as a bridge between ecologists and engineers.
Specific topics covered in the journal include: habitat reconstruction; ecotechnology; synthetic ecology; bioengineering; restoration ecology; ecology conservation; ecosystem rehabilitation; stream and river restoration; reclamation ecology; non-renewable resource conservation. Descriptions of specific applications of ecological engineering are acceptable only when situated within context of adding novelty to current research and emphasizing ecosystem restoration. We do not accept purely descriptive reports on ecosystem structures (such as vegetation surveys), purely physical assessment of materials that can be used for ecological restoration, small-model studies carried out in the laboratory or greenhouse with artificial (waste)water or crop studies, or case studies on conventional wastewater treatment and eutrophication that do not offer an ecosystem restoration approach within the paper.