Authors: Xingchen Chen*, Department of Geography, Texas A&M University, College Station, Texas, 77840, Anthony Filippi, Department of Geography, Texas A&M University, College Station, Texas 77840, Inci Guneralp, Department of Geography, Texas A&M University, College Station, Texas 77840
Topics: Geomorphology, Remote Sensing, Geographic Information Science and Systems
Keywords: Geomorphology, Remote Sensing, GEOBIA,Floodplain
Session Type: Poster
Start / End Time: 1:20 PM / 3:00 PM
Room: Napoleon Foyer/Common St. Corridor, Sheraton, 3rd Floor
Presentation File: No File Uploaded
Detailed classification of floodplain geomorphology and land cover is a fundamental step in understanding the process-pattern interactions between a river channel and its floodplain. It also provides important information for estimating the spatial extent of flooding and developing effective management strategies for rivers and their ecosystems. Although supervised classification (Brondízio, Eduardo, et al. 1996, Friedl, Mark A., et al. 2002, Gómez, Cristina, Joanne C. 2016, Khatami, Reza, Giorgos Mountrakis, and Stephen V. Stehman. 2016) has been used extensively to classify a range of landscapes, including floodplain landscapes, GEOgraphic-Object-Based Image Analysis (GEOBIA) has been rarely used (Iersel, W. K., et al 2016, Furtado 2015) to classify the geomorphology and vegetation of these environments. In this study, using GEOBIA, we examine the floodplain of Río Beni located in the Bolivian Amazon. Rio Beni is a highly dynamic river system, characterized by rapid planform migration and chute and neck cut-off processes that can be observable within the scales of decades. The floodplain of Rio Beni is abundant with geomorphic landforms, ranging from oxbow lakes, meander scars, to scroll bars. The floodplain vegetation is composed of forest and non-forest vegetation, entailing spatially varying patterns on the floodplain. We perform the classification using Landsat 7 images and ancillary information that we obtain from vegetation and water indices (NDVI, NDWI, respectively), topography, and canopy coverage. Comparing the results from GEOBIA and per-pixel methods, we perform a quantitative comparative assessment of classification accuracy, and we evaluate the efficacy of various input dataset combinations in improving classification accuracy.