{"title":"揭示部分植被覆盖动态:利用 MODIS NDVI 和机器学习进行时空分析","authors":"","doi":"10.1016/j.indic.2024.100485","DOIUrl":null,"url":null,"abstract":"<div><p>Understanding the dynamics of Fractional Vegetation Cover (FVC) is crucial for effective environmental monitoring and management, especially in regions like Pakistan that are sensitive to climate change. This study employs an innovative approach using MODIS NDVI data and the Pixel Dichotomy Model (PDM) to analyze the spatiotemporal dynamics of FVC across Pakistan from 2003 to 2020. Our findings reveal an overall increasing trend in FVC, with the highest value recorded in 2017 (0.37) and the lowest in 2004 (0.26). The Hurst exponent analysis (R/S ratio = 0.718) indicates a degree of long-term memory in the FVC time series. Rainfall was found to positively correlate with FVC (r = 0.6), while Land Surface Temperature (LST) and the Compounded Night Light Index (CNLI) exhibited negative correlations (r = −0.59 and r = −0.43, respectively). The Random Forest regression model highlighted CNLI as the most influential predictor (importance = 62.4%), emphasizing the need to consider human-induced factors in environmental management. These results provide critical insights for sustainable land management and contribute to understanding vegetation-climate interactions in arid and semi-arid environments.\"</p></div>","PeriodicalId":36171,"journal":{"name":"Environmental and Sustainability Indicators","volume":null,"pages":null},"PeriodicalIF":5.4000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2665972724001533/pdfft?md5=b6636c03d511f8612631a5975f300255&pid=1-s2.0-S2665972724001533-main.pdf","citationCount":"0","resultStr":"{\"title\":\"Unveiling fractional vegetation cover dynamics: A spatiotemporal analysis using MODIS NDVI and machine learning\",\"authors\":\"\",\"doi\":\"10.1016/j.indic.2024.100485\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>Understanding the dynamics of Fractional Vegetation Cover (FVC) is crucial for effective environmental monitoring and management, especially in regions like Pakistan that are sensitive to climate change. This study employs an innovative approach using MODIS NDVI data and the Pixel Dichotomy Model (PDM) to analyze the spatiotemporal dynamics of FVC across Pakistan from 2003 to 2020. Our findings reveal an overall increasing trend in FVC, with the highest value recorded in 2017 (0.37) and the lowest in 2004 (0.26). The Hurst exponent analysis (R/S ratio = 0.718) indicates a degree of long-term memory in the FVC time series. Rainfall was found to positively correlate with FVC (r = 0.6), while Land Surface Temperature (LST) and the Compounded Night Light Index (CNLI) exhibited negative correlations (r = −0.59 and r = −0.43, respectively). The Random Forest regression model highlighted CNLI as the most influential predictor (importance = 62.4%), emphasizing the need to consider human-induced factors in environmental management. These results provide critical insights for sustainable land management and contribute to understanding vegetation-climate interactions in arid and semi-arid environments.\\\"</p></div>\",\"PeriodicalId\":36171,\"journal\":{\"name\":\"Environmental and Sustainability Indicators\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":5.4000,\"publicationDate\":\"2024-09-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2665972724001533/pdfft?md5=b6636c03d511f8612631a5975f300255&pid=1-s2.0-S2665972724001533-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Environmental and Sustainability Indicators\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2665972724001533\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENVIRONMENTAL SCIENCES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Environmental and Sustainability Indicators","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2665972724001533","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENVIRONMENTAL SCIENCES","Score":null,"Total":0}
引用次数: 0
摘要
了解部分植被覆盖度(FVC)的动态对于有效的环境监测和管理至关重要,尤其是在巴基斯坦这样对气候变化敏感的地区。本研究采用一种创新方法,利用 MODIS NDVI 数据和像素二分法模型 (PDM) 分析了 2003 年至 2020 年巴基斯坦各地的分植被覆盖率时空动态。我们的研究结果表明,森林覆盖率总体呈上升趋势,最高值出现在 2017 年(0.37),最低值出现在 2004 年(0.26)。赫斯特指数分析(R/S 比率 = 0.718)表明,降雨量时间序列具有一定程度的长期记忆。降雨量与森林覆盖率呈正相关(r = 0.6),而地表温度(LST)和复合夜光指数(CNLI)呈负相关(分别为 r = -0.59 和 r = -0.43)。随机森林回归模型强调 CNLI 是最有影响力的预测因子(重要性 = 62.4%),强调了在环境管理中考虑人为因素的必要性。这些结果为可持续土地管理提供了重要启示,有助于理解干旱和半干旱环境中植被与气候之间的相互作用"。
Unveiling fractional vegetation cover dynamics: A spatiotemporal analysis using MODIS NDVI and machine learning
Understanding the dynamics of Fractional Vegetation Cover (FVC) is crucial for effective environmental monitoring and management, especially in regions like Pakistan that are sensitive to climate change. This study employs an innovative approach using MODIS NDVI data and the Pixel Dichotomy Model (PDM) to analyze the spatiotemporal dynamics of FVC across Pakistan from 2003 to 2020. Our findings reveal an overall increasing trend in FVC, with the highest value recorded in 2017 (0.37) and the lowest in 2004 (0.26). The Hurst exponent analysis (R/S ratio = 0.718) indicates a degree of long-term memory in the FVC time series. Rainfall was found to positively correlate with FVC (r = 0.6), while Land Surface Temperature (LST) and the Compounded Night Light Index (CNLI) exhibited negative correlations (r = −0.59 and r = −0.43, respectively). The Random Forest regression model highlighted CNLI as the most influential predictor (importance = 62.4%), emphasizing the need to consider human-induced factors in environmental management. These results provide critical insights for sustainable land management and contribute to understanding vegetation-climate interactions in arid and semi-arid environments."