TROPOMI SIF high resolution data at 0.005° for CONUS as estimated by the convolutional neural network SIFnet

We develop a Convolutional Neural Network, named SIFnet, that increases the spatial resolution of SIF from the TROPOMI by a factor of 10 to a spatial resolution of 0.005°. SIFnet utilizes coarse SIF observations together with a broad range high resolution auxiliary data. The insights gained from interpretable machine learning techniques allow us to make quantitative claims about the relationships between SIF and other common parameters related to photosynthesis.

Temporal coverage: April 2018 until March 2021, 16 day time steps

Data for other regions can be requested and produced by the authors. Please refer for further information to: 

Gensheimer, J., Turner, A. J., Köhler, P., Frankenberg, C., & Chen, J. (2022). A Convolutional Neural Network for Spatial Downscaling of Satellite-Based Solar-Induced Chlorophyll Fluorescence (SIFnet). A convolutional neural network for spatial downscaling of satellite-based solar-induced chlorophyll fluorescence (SIFnet). Biogeosciences, 19(6), 1777-1793. DOI: https://doi.org/10.5194/bg-19-1777-2022