Eesti Maaülikool. EMU Dspace · 2018 · 12 citations · 14 references
Natural ResourcesEarth SciencesForestryForest EcologyForest InventoryRemote SensingLand Cover Change DetectionLand Cover ClassificationRemote Sensing Image AnalysisVegetation Change DetectionVegetation IndicesVegetation MonitoringForest Cover ClassificationLand Change DetectionRemote Sensing Of DeforestationSatellite Remote Sensing DataSatellite-based Forest Mapping
Aim of this study is to evaluate different remote sensing indices to detect spatial distribution of wet soils using GIS based algorithms. Ar ea of this study represents different soil types on various quaternary deposits as well as different forest types. We analyzed 25 sites with the area of 1 km 2 each in central and western part of Latvia. Data about soil characteristics like thickness of pea t layer and presence of reductimorphic colors in soil was collected during field surveys in 228 random points within study sites. ANOVA test for comparing means of different soil wetness classes and binary logistic regression analysis for evaluating the ac curacy of different remote sensing indices to model spatial distribution of wet areas are used for analysis. Main conclusion of this study is that for different quaternary deposits and soil texture classes different algorithms for soil wetness prediction s hould be used. Data layers for predicting soil wetness in this study are various modifications and resolutions of digital elevation model like depressions, slope and SAGA wetness index as well as Sentinel - 2 multispectral satellite imagery. Accuracy of soil wetness classification of soils on moraine, fluvial and eolian sediments exceeds 94%, whereas on the clayey sediments it is close to 80%.
14
Measuring phenological variability from satellite imagery
Bradley C. Reed, Jesslyn F. Brown, Darrel VanderZee et al. · Journal of Vegetation Science · 1994 · 1.5K citations