River Sediment Database-Amazon (RivSed-Amazon)

<p>The River Sediment Database-Amazon (RivSed-Amazon) database contains surface suspended sediment concentrations (SSC) derived from Landsat 5, 7, and 8 Level 1 Collection 1 surface reflectance from all rivers in the Amazon River Basin that are ~60 meters wide or greater. SSC represent spatially integrated "reach" median concentrations over the footprint of SWOT River Database (SWORD, Altenau et al., 2021) centerlines (median reach length = 10 km) where high quality river water pixels were detected within each Landsat image from 1984-2018. </p> <p>The methods used to produce this database were initially developed in the following publications:</p> <ul> <li>Gardner, J., Pavelsky, T. M., Topp, S., Yang, X., Ross, M. R., & Cohen, S. (2023). Human activities change suspended sediment concentration along rivers. <em>Environmental Research Letters. </em><a href="https://iopscience.iop.org/article/10.1088/1748-9326/acd8d8">https://iopscience.iop.org/article/10.1088/1748-9326/acd8d8</a> <strong>and </strong></li> <li>Gardner et al. (2020). The color of rivers. Geophysical Research Letters.  <a href="https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020GL088946">https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020GL088946</a></li> </ul> <p>The publication associated with RivSed-Amazon is in review.</p> <p><strong>Files:</strong></p> <p>1) Metadata (rivSed_Amazon_metadata_v1.01.pdf): Description key data files associated with this repository. </p> <p>2) RiverSed (RiverSed_Amazon_v1.1.txt). Table of  SSC  and associated data that is joinable to SWORD based on the ""reach_id".</p> <p>3) Shapefile of river centerlines over South America to which the reflectance data can be attached (SWORD_SA.shp).</p> <p>4) Shapefile of the reach polygons associated with SWORD_SA over the Amazon Basin. (reach_polygons_amazon.shp).</p> <p>5) SSC-Landsat matchup database with extended metadata on locations and in-situ data (train_full_v1.1.csv).</p> <p>6) The final training data used to build the xgboost machine learning model (train_v1.1.csv).</p> <p>7) The xgboost model that can make SSC predictions over inland waters in USA using Landsat bands/band combinations (tssAmazon_model_v1.1.rds and .rda). The model can only be loaded and used in R at this time.</p> <p>8) The correction coefficients applied to Landsat 5 and 8 to harmonized surface reflectance across Landsat 5,7,8 and over all bands to enable time series analysis.</p> <p> </p> <p> </p&gt