Abstract
Abstract. Solar-induced fluorescence (SIF) data from satellites are increasingly used as a proxy for photosynthetic activity by vegetation and as a constraint on gross primary production. Here we report on improvements in the algorithm to retrieve mid-morning (09:30 LT) SIF estimates on the global scale from the GOME-2 sensor on the MetOp-A satellite (GOME-2A) for the period 2007–2019. Our new SIFTER (Sun-Induced Fluorescence of Terrestrial Ecosystems Retrieval) v2 algorithm improves over a previous version by using a narrower spectral window that avoids strong oxygen absorption and being less sensitive to water vapour absorption, by constructing stable reference spectra from a 6-year period (2007–2012) of atmospheric spectra over the Sahara and by applying a latitude-dependent zero-level adjustment that accounts for biases in the data product. We generated stable, good-quality SIF retrievals between January 2007 and June 2013, when GOME-2A degradation in the near infrared was still limited. After the narrowing of the GOME-2A swath in July 2013, we characterised the throughput degradation of the level-1 data in order to derive reflectance corrections and apply these for the SIF retrievals between July 2013 and December 2018. SIFTER v2 data compare well with the independent NASA v2.8 data product. Especially in the evergreen tropics, SIFTER v2 no longer shows the underestimates against other satellite products that were seen in SIFTER v1. The new data product includes uncertainty estimates for individual observations and is best used for mostly clear-sky scenes and when spectral residuals remain below a certain spectral autocorrelation threshold. Our results support the use of SIFTER v2 data being used as an independent constraint on photosynthetic activity on regional to global scales.
Highlights
Solar-induced fluorescence (SIF) by vegetation is directly related to light absorption by the chlorophyll complex during photosynthesis (Porcar-Castell et al, 2014; Mohammed et al, 2019)
Our new SIFTER (Sun-Induced Fluorescence of Terrestrial Ecosystems Retrieval) v2 algorithm improves over a previous version by using a narrower spectral window that avoids strong oxygen absorption and being less sensitive to water vapour absorption, by constructing stable reference spectra from a 6-year period (2007–2012) of atmospheric spectra over the Sahara and by applying a latitude-dependent zero-level adjustment that accounts for biases in the data product
In SIFTER v1, principal components (PCs) were calculated from all top-ofatmosphere spectra taken over the non-vegetated parts of the Saharan desert in the 12 months preceding the GOME2A measurement of interest
Summary
Solar-induced fluorescence (SIF) by vegetation is directly related to light absorption by the chlorophyll complex during photosynthesis (Porcar-Castell et al, 2014; Mohammed et al, 2019). The modelled spectrum consists of contributions from surface reflectance, atmospheric transmittance, and fluorescence, where the latter fills in the Fraunhofer lines present in the incoming sunlight This technique was pioneered by Joiner et al (2013) and Köhler et al (2015), and explored further by Sanders et al (2016), who established the Sun-Induced Fluorescence of Terrestrial Ecosystems Retrieval (SIFTER) algorithm. We propose a number of improvements to the SIFTER approach based on radiative transfer modelling tests These improvements focus on optimising the spectral fitting window, and on calculating atmospheric transmittance terms from satellite spectra over a reference region without vegetation. We compare the results of our new SIFTER v2 algorithm to SIFTER v1 and the data generated by NASA and discuss the uncertainty budget and limitations
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
