{"title":"通过相互作用网络揭示土地竞争:一个基于一致性的挖掘和模拟模型,整合了土地利用的抑制效应","authors":"Xun Liang , Jun-Long Huang , Qingfeng Guan","doi":"10.1016/j.landurbplan.2025.105458","DOIUrl":null,"url":null,"abstract":"<div><div>Exploration of the competition among multiple land uses can reveal the fundamental mechanism of the evolution process of land system. However, quantification of the competition among land uses remains a challenge. Because most land use simulation studies do not consider the amplitude differences resulting from the influences of the spatial suitability map, neighborhood aggregation effect, and stochastic effect of multiple land uses, the driving and inhibiting effects among land uses have not yet been thoroughly discovered. To address this problem, we propose an interaction network discovery model via consistency-based simulation, called intPLUS (available for download at <span><span>https://github.com/HPSCIL/intPLUS</span><svg><path></path></svg></span>), to find the interaction relationships among land uses and to improve the projections of future land use changes. This model uses the logarithm transformation to embed weights into multiple effects, including the inter-land use inhibiting effects, which drive the evolution of land use. The correctly projected land use change (i.e., consistency) is analyzed with a random forest (RF) model to explore the weights of the driving and inhibiting effects between land uses. This model is applied to Wuhan, China. The results showed that ‘cultivated field’ was greatly restrained and was restrained by other land uses. The application of the interaction network obtained accuracy enhancements of 30% and 13% in the calibration and future allocation processes, respectively. This model takes full advantage of the consistency information of the process of spatial simulation; the interaction network among land uses derived by the proposed model provides an insightful means to advance our understanding of spatial competition.</div></div>","PeriodicalId":54744,"journal":{"name":"Landscape and Urban Planning","volume":"263 ","pages":"Article 105458"},"PeriodicalIF":9.2000,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Unveiling land competition through interaction networks: A consistency-based mining and simulation model that integrates inhibiting effects of land uses\",\"authors\":\"Xun Liang , Jun-Long Huang , Qingfeng Guan\",\"doi\":\"10.1016/j.landurbplan.2025.105458\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Exploration of the competition among multiple land uses can reveal the fundamental mechanism of the evolution process of land system. However, quantification of the competition among land uses remains a challenge. Because most land use simulation studies do not consider the amplitude differences resulting from the influences of the spatial suitability map, neighborhood aggregation effect, and stochastic effect of multiple land uses, the driving and inhibiting effects among land uses have not yet been thoroughly discovered. To address this problem, we propose an interaction network discovery model via consistency-based simulation, called intPLUS (available for download at <span><span>https://github.com/HPSCIL/intPLUS</span><svg><path></path></svg></span>), to find the interaction relationships among land uses and to improve the projections of future land use changes. This model uses the logarithm transformation to embed weights into multiple effects, including the inter-land use inhibiting effects, which drive the evolution of land use. The correctly projected land use change (i.e., consistency) is analyzed with a random forest (RF) model to explore the weights of the driving and inhibiting effects between land uses. This model is applied to Wuhan, China. The results showed that ‘cultivated field’ was greatly restrained and was restrained by other land uses. The application of the interaction network obtained accuracy enhancements of 30% and 13% in the calibration and future allocation processes, respectively. This model takes full advantage of the consistency information of the process of spatial simulation; the interaction network among land uses derived by the proposed model provides an insightful means to advance our understanding of spatial competition.</div></div>\",\"PeriodicalId\":54744,\"journal\":{\"name\":\"Landscape and Urban Planning\",\"volume\":\"263 \",\"pages\":\"Article 105458\"},\"PeriodicalIF\":9.2000,\"publicationDate\":\"2025-08-05\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Landscape and Urban Planning\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0169204625001653\",\"RegionNum\":1,\"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":"Landscape and Urban Planning","FirstCategoryId":"93","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0169204625001653","RegionNum":1,"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}
Unveiling land competition through interaction networks: A consistency-based mining and simulation model that integrates inhibiting effects of land uses
Exploration of the competition among multiple land uses can reveal the fundamental mechanism of the evolution process of land system. However, quantification of the competition among land uses remains a challenge. Because most land use simulation studies do not consider the amplitude differences resulting from the influences of the spatial suitability map, neighborhood aggregation effect, and stochastic effect of multiple land uses, the driving and inhibiting effects among land uses have not yet been thoroughly discovered. To address this problem, we propose an interaction network discovery model via consistency-based simulation, called intPLUS (available for download at https://github.com/HPSCIL/intPLUS), to find the interaction relationships among land uses and to improve the projections of future land use changes. This model uses the logarithm transformation to embed weights into multiple effects, including the inter-land use inhibiting effects, which drive the evolution of land use. The correctly projected land use change (i.e., consistency) is analyzed with a random forest (RF) model to explore the weights of the driving and inhibiting effects between land uses. This model is applied to Wuhan, China. The results showed that ‘cultivated field’ was greatly restrained and was restrained by other land uses. The application of the interaction network obtained accuracy enhancements of 30% and 13% in the calibration and future allocation processes, respectively. This model takes full advantage of the consistency information of the process of spatial simulation; the interaction network among land uses derived by the proposed model provides an insightful means to advance our understanding of spatial competition.
期刊介绍:
Landscape and Urban Planning is an international journal that aims to enhance our understanding of landscapes and promote sustainable solutions for landscape change. The journal focuses on landscapes as complex social-ecological systems that encompass various spatial and temporal dimensions. These landscapes possess aesthetic, natural, and cultural qualities that are valued by individuals in different ways, leading to actions that alter the landscape. With increasing urbanization and the need for ecological and cultural sensitivity at various scales, a multidisciplinary approach is necessary to comprehend and align social and ecological values for landscape sustainability. The journal believes that combining landscape science with planning and design can yield positive outcomes for both people and nature.