Monitoring Crop Photosynthesis through Satellite-based Measurements of Sun-Induced Fluorescence (CropSIF)
Ecology of Land Use
Final Report Abstract
The objective of the Crop-SIF project was to study crop productivity and yield variability by making use of recent developments related to satellite vegetation observations and data mining in the context of ensuring food security. Crop-SIF activities dealt with both the machine learning and deep learning methods applied to both vegetation and meteorological data, as well as the utilization of - only recently quantified from satellite platforms - sun-induced chlorophyll fluorescence (SIF). This research contributed to the assessment of agricultural productivity and climate impacts by taking advantage using novel methodologies (interpretable deep learning), datasets (SIF) and computing platforms (Google Earth Engine). In addition, within the project, the cooperation between GFZ German Research Centre for Geosciences and the Nanjing University was strengthened, including exchange visits of the scientists from both research institutions.
Publications
- (2019) Estimating Crop Primary Productivity with Sentinel-2 and Landsat 8 using Machine Learning Methods Trained with Radiative Transfer Simulations. Remote Sensing of Environment, 225, 441-457
Wolanin A., Camps-Valls G., Gommez-Chova L., Mateo-Garcıa G., van der Tol Ch., Zhang Y., Guanter L.
(See online at https://doi.org/10.1016/j.rse.2019.03.002) - (2020) Estimating and Understanding Crop Yields with Explainable Deep Learning in the Indian Wheat Belt. Environmental Research Letters, 15, 02401
Wolanin A., Mateo-Garcıa G., Camps-Valls G., Gomez-Chova L. ,Meroni, M., Duveiller, G., You, L., Guanter L.
(See online at https://doi.org/10.1088/1748-9326/ab68ac)
