Forschungsdaten
Colourbox.de/Tashatuvango
Permanent URI for this communityhttps://data.uni-marburg.de/handle/dataumr/19
Browse
Research Data Open Access Synthetic Meteosat First Generation MVIRI Visible Band Images (Level 1.5)(Philipps-Universität Marburg) Jung, Ivo; Gaurav, Sheetabh; Bendix, JörgThe data comprises synthetic and harmonized Meteosat First Generation (MFG) MVIRI VIS reflectance Image data from 2006 - 2020. The dataset was created with the intent to extent the existing MFG MVIRI VIS image time series (https://navigator.eumetsat.int/product/EO:EUM:DAT:0080) for long-term climate research. The temporal resolution of the data is 30 minutes per image. Each image covers the WMO VI region up to 62°N. The images are stored as .nc files with reflectance values and geographic and temporal information. Due to the size of the dataset, it is compressed into annual archives. The dataset was created using a trained Random Forest Regressor Model and Meteosat Second Generation VIS channel data (https://data.eumetsat.int/product/EO:EUM:DAT:MSG:HRSEVIRI?query=SEVIRI&s=extended#). No data was created for pixels with Solar-Zenith Angles above 80° due to parallax effects and atmospheric distortion. Additional metadata can be found in the readme file. The data can be used like MFG MvIRI VIS image data i.e. using programming languages like python. The scripts for preprocessing, model training and synthesizing can be found at https://gitlab.uni-marburg.de/fb19/ag-bendix/harmonizing_synthesizing_mfg_vis.