Labeled GOES-16 Satellite Images for River Inflow and Discharge Estimation

DESCRIPTION This dataset contains labeled satellite images acquired from the GOES-16 satellite and associated with river inflow and discharge measurements in the Tietê River basin, Brazil. The dataset was developed to support research in hydrology, remote sensing, and machine learning. Each satellite image is linked to hydrological observations obtained from streamflow gauging stations, enabling supervised learning approaches for river inflow and discharge estimation. The dataset includes preprocessed satellite images, metadata describing monitoring stations, and labels containing inflow and discharge values. DATA CONTENTS GOES-16 satellite images Hydrological labels (river inflow and discharge) Metadata describing streamflow gauging stations Station information file (stations_info.csv) STATION METADATA The file stations_info.csv provides metadata for the streamflow gauging stations used in this dataset, including station identifiers, river basin and sub-basin, associated rivers, geographic location (latitude and longitude), altitude, and drainage area. Geographic coordinates are provided in degrees, minutes, and seconds (DMS), following the original records from the Brazilian National Water Agency (ANA). Accented characters were removed from CSV files to ensure compatibility across different operating systems and data processing tools. INTENDED USE This dataset is intended for research and academic purposes, including: Machine learning and deep learning applications Hydrological forecasting Remote sensing studies River inflow and discharge estimation Users are encouraged to cite the dataset when used in scientific publications. SOURCE CODE The Python source code used for satellite image acquisition, preprocessing, and dataset generation is available at: GitHub repository: https://github.com/thaisrochaBR/atlasUnicampZenodo (Software): https://doi.org/10.5281/zenodo.18471522 LICENSE This dataset is distributed under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. AUTHORS Thais Rocha (University of Campinas – UNICAMP)Andre Franceschi de Angelis (University of Campinas – UNICAMP) Corresponding author: [e-mail]