Samuel Hislop, Mariela Soto-Berelov, Sacha Jellinek, Yung En Chee, Simon Jones
{"title":"利用Sentinel-2卫星影像监测城市河岸植被","authors":"Samuel Hislop, Mariela Soto-Berelov, Sacha Jellinek, Yung En Chee, Simon Jones","doi":"10.1111/emr.12624","DOIUrl":null,"url":null,"abstract":"<p>Riparian vegetation is important for biodiversity, creates habitat for various fauna and provides vital ecosystem services such as flow regulation, water filtration, uptake, transformation and cycling of minerals, nutrients and organic matter. These ecological functions and processes contribute to the health of waterways. In many urban and semiurban areas, riparian vegetation is subject to degradation pressures such as land clearing for urban development, altered hydrological regimes due to modified catchment land cover and runoff patterns and other human-induced disturbances. Monitoring riparian vegetation cover and condition is vital for informing effective management. This can be achieved via detailed field assessment; however, this is often costly, time and labour intensive and extremely difficult to undertake across large regions such as an entire stream network. This study examines the utility of medium-resolution Sentinel-2 satellite imagery for monitoring riparian vegetation across the Greater Melbourne region, Australia. The paper demonstrates several examples of how spectral trends from Sentinel-2 imagery (2019–2023) can be used to provide insights, which are expected to improve over time as more data become available. We also assessed the potential of combining Sentinel-2 imagery with field observations to model riparian vegetation conditions. Using 425 field observations and a series of Sentinel-2–derived predictor variables, a Random Forest regression model was developed to model riparian vegetation condition. The model explained 53% of the variance in the data. Although the spatial resolution of Sentinel-2 (10 m) is a limitation when working with narrow riparian vegetation corridors, our methods demonstrate it can be used to help prioritise areas for protection and/or enhancement and identify areas that may be at risk of further degradation. This can assist waterway managers to prioritise field activities such as restoration and revegetation, and the control of key weeds and browsing animals.</p>","PeriodicalId":54325,"journal":{"name":"Ecological Management & Restoration","volume":"26 1","pages":""},"PeriodicalIF":2.2000,"publicationDate":"2025-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/emr.12624","citationCount":"0","resultStr":"{\"title\":\"Monitoring Riparian Vegetation in Urban Areas With Sentinel-2 Satellite Imagery\",\"authors\":\"Samuel Hislop, Mariela Soto-Berelov, Sacha Jellinek, Yung En Chee, Simon Jones\",\"doi\":\"10.1111/emr.12624\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Riparian vegetation is important for biodiversity, creates habitat for various fauna and provides vital ecosystem services such as flow regulation, water filtration, uptake, transformation and cycling of minerals, nutrients and organic matter. These ecological functions and processes contribute to the health of waterways. In many urban and semiurban areas, riparian vegetation is subject to degradation pressures such as land clearing for urban development, altered hydrological regimes due to modified catchment land cover and runoff patterns and other human-induced disturbances. Monitoring riparian vegetation cover and condition is vital for informing effective management. This can be achieved via detailed field assessment; however, this is often costly, time and labour intensive and extremely difficult to undertake across large regions such as an entire stream network. This study examines the utility of medium-resolution Sentinel-2 satellite imagery for monitoring riparian vegetation across the Greater Melbourne region, Australia. The paper demonstrates several examples of how spectral trends from Sentinel-2 imagery (2019–2023) can be used to provide insights, which are expected to improve over time as more data become available. We also assessed the potential of combining Sentinel-2 imagery with field observations to model riparian vegetation conditions. Using 425 field observations and a series of Sentinel-2–derived predictor variables, a Random Forest regression model was developed to model riparian vegetation condition. The model explained 53% of the variance in the data. Although the spatial resolution of Sentinel-2 (10 m) is a limitation when working with narrow riparian vegetation corridors, our methods demonstrate it can be used to help prioritise areas for protection and/or enhancement and identify areas that may be at risk of further degradation. This can assist waterway managers to prioritise field activities such as restoration and revegetation, and the control of key weeds and browsing animals.</p>\",\"PeriodicalId\":54325,\"journal\":{\"name\":\"Ecological Management & Restoration\",\"volume\":\"26 1\",\"pages\":\"\"},\"PeriodicalIF\":2.2000,\"publicationDate\":\"2025-01-23\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://onlinelibrary.wiley.com/doi/epdf/10.1111/emr.12624\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Ecological Management & Restoration\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1111/emr.12624\",\"RegionNum\":4,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ECOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ecological Management & Restoration","FirstCategoryId":"93","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/emr.12624","RegionNum":4,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ECOLOGY","Score":null,"Total":0}
Monitoring Riparian Vegetation in Urban Areas With Sentinel-2 Satellite Imagery
Riparian vegetation is important for biodiversity, creates habitat for various fauna and provides vital ecosystem services such as flow regulation, water filtration, uptake, transformation and cycling of minerals, nutrients and organic matter. These ecological functions and processes contribute to the health of waterways. In many urban and semiurban areas, riparian vegetation is subject to degradation pressures such as land clearing for urban development, altered hydrological regimes due to modified catchment land cover and runoff patterns and other human-induced disturbances. Monitoring riparian vegetation cover and condition is vital for informing effective management. This can be achieved via detailed field assessment; however, this is often costly, time and labour intensive and extremely difficult to undertake across large regions such as an entire stream network. This study examines the utility of medium-resolution Sentinel-2 satellite imagery for monitoring riparian vegetation across the Greater Melbourne region, Australia. The paper demonstrates several examples of how spectral trends from Sentinel-2 imagery (2019–2023) can be used to provide insights, which are expected to improve over time as more data become available. We also assessed the potential of combining Sentinel-2 imagery with field observations to model riparian vegetation conditions. Using 425 field observations and a series of Sentinel-2–derived predictor variables, a Random Forest regression model was developed to model riparian vegetation condition. The model explained 53% of the variance in the data. Although the spatial resolution of Sentinel-2 (10 m) is a limitation when working with narrow riparian vegetation corridors, our methods demonstrate it can be used to help prioritise areas for protection and/or enhancement and identify areas that may be at risk of further degradation. This can assist waterway managers to prioritise field activities such as restoration and revegetation, and the control of key weeds and browsing animals.
期刊介绍:
Ecological Management & Restoration is a peer-reviewed journal with the dual aims of (i) reporting the latest science to assist ecologically appropriate management and restoration actions and (ii) providing a forum for reporting on these actions. Guided by an editorial board made up of researchers and practitioners, EMR seeks features, topical opinion pieces, research reports, short notes and project summaries applicable to Australasian ecosystems to encourage more regionally-appropriate management. Where relevant, contributions should draw on international science and practice and highlight any relevance to the global challenge of integrating biodiversity conservation in a rapidly changing world.
Topic areas:
Improved management and restoration of plant communities, fauna and habitat; coastal, marine and riparian zones; restoration ethics and philosophy; planning; monitoring and assessment; policy and legislation; landscape pattern and design; integrated ecosystems management; socio-economic issues and solutions; techniques and methodology; threatened species; genetic issues; indigenous land management; weeds and feral animal control; landscape arts and aesthetics; education and communication; community involvement.