Preprint FZJ-2024-06812

http://join2-wiki.gsi.de/foswiki/pub/Main/Artwork/join2_logo100x88.png
A Multi-Layer Perceptron Approach for SIF Retrieval in the O2-A Absorption band from Hyperspectral Imagery of the HyPlant Airborne Sensor System

 ;  ;  ;  ;  ;  ;  ;  ;  ;  ;

2024

Abstract: Accurate estimation of solar-induced fluorescence (SIF) from passively sensedhyperspectral remote sensing data has been identified as instrumental in assessingthe photosynthetic activity of plants for various scientific and ecologicalapplications at various spatial scales. Different techniques to derive SIFhave been developed over the last decades. In view of ESA’s upcoming EarthExplorer satellite mission FLEX aiming to provide high-quality global imageryfor SIF retrieval an increased interest is placed in physical approaches.We present a novel method to retrieve SIF in the O2-A absorption band ofhyperspectral imagery acquired by the HyPlant sensor system. It aims at atight integration of physical radiative transfer principles and self-supervisedneural network training. To this end, a set of spatial and spectral constraintsand a specific loss formulation are adopted. In a validation study we find goodagreement between our approach and established retrieval methods as wellas with in-situ top-of-canopy SIF measurements. In two application studies,we additionally find evidence that the estimated SIF (i) satisfies a first-ordermodel of diurnal SIF variation and (ii) locally adapts the estimated opticaldepth in topographically variable terrain.


Contributing Institute(s):
  1. Datenanalyse und Maschinenlernen (IAS-8)
  2. Pflanzenwissenschaften (IBG-2)
Research Program(s):
  1. 5112 - Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups (POF4-511) (POF4-511)
  2. 2173 - Agro-biogeosystems: controls, feedbacks and impact (POF4-217) (POF4-217)

Appears in the scientific report 2024
Click to display QR Code for this record

The record appears in these collections:
Institute Collections > IAS > IAS-8
Document types > Reports > Preprints
Institute Collections > IBG > IBG-2
Workflow collections > Public records
Publications database

 Record created 2024-12-06, last modified 2024-12-14



Rate this document:

Rate this document:
1
2
3
 
(Not yet reviewed)