R Packages for Field Spectroscopy: From System Characterization to Sun-Induced Chlorophyll Fluorescence Retrieval.
We are happy to announce that 2 R Packages for Field Spectroscopy are now available in Github. The packages have been developed in collaboration between Unimib, Max Planck Institute and Julich Research center and have been presented at the ESA workshop: “REMOTE SENSING OF FLUORESCENCE, PHOTOSYNTHESIS AND VEGETATION STATUS” , 17 – 19 January 2017 at ESA-ESRIN, Frascati, Italy.
Below you find the overview of the poster presented at ESRIN:
An R Package for Field Spectroscopy: From System Characterization to Sun-Induced Chlorophyll Fluorescence Retrieval.
Tommaso Julitta1,5, Thomas Wutzler2, Micol Rossini1, Roberto Colombo1, Sergio Cogliati1, Michele Meroni3, Andreas Burkart4,5, Mirco Migliavacca2
1University of Milano-Bicocca, LTDA, Department of Earth and Environmental Sciences, Milano (MI), Italy; 2Max Planck Institute for Biogeochemistry; 3Join Research Center, Institute for Environment and Sustainability; 4Forschungszentrum Jülich, Institute of Biogeosciences; 5JB Hyperspectral Devices UG
In recent literature several retrieval methods for sun-induced Chlorophyll Fluorescence (SIF) have been proposed. Each of them relies on some assumption to decouple the fluorescence emission from the reflected radiance. However, the application of these methods is not trivial, and no detailed information for standardized implementation is available. Moreover, the technical characteristics of the device used to collect the spectra in the field (e.g. spectral resolution, spectral sampling interval, signal to noise ratio) can prevent the application of some of these methods. As a result, slight differences in the algorithm implementation applied to an unsuited measurement setups can provide misleading outputs, hampering the comparability between data collected in different experimental studies. In order to overcome this problem of standardization we developed two open source R packages that can be used for: i) characterization and calibration of spectrometers (FieldspectroscopyCC); ii) field spectra data processing (FieldspectroscopyDP). The characterization and calibration of the spectrometers are preparatory steps needed to obtain knowledge on the systems properties (e.g. signal to noise ratio, actual spectral resolution) and to make periodical checks of spectral and radiometric drift. Some program functions are designed to process measurements acquired on known laboratory standards (e.g. calibration lamps), some are designed to process outdoor measurements using the sun-light as illumination source. Once the characteristics of the systems are known and the field spectrometer is spectrally and radiometrically calibrated, the FiledpsctroscopyDP package can be used to retrieve SIF and other relevant vegetation indexes. For instance, data processing functions allow: the conversion of data acquired in digital number to radiance units, the computation of the spectral reflectance factor, and the computation of a variety of reflectance-based vegetation indices. Finally, SIF is computed using multiple published methods for the fluorescence retrieval, including the Fraunhofer Line Discriminator (in its various versions: FLD, 3FLD and iFLD); the Spectral Fitting Methods (SFM).