Simulation study on improving spectrally-resolved solar-induced fluorescence satellite retrievals under uncertain atmospheric states
Kukkurainen, Antti | Lipponen, Antti | Sabater, Neus | Arola, Antti | Kolehmainen, Ville | Ilmatieteen laitos | Finnish Meteorological Institute | 0000-0002-3371-7337 | 0000-0002-6902-9974 | 0000-0002-0296-4378 | 0000-0002-9220-0194
Over recent decades, advancements in satellite remote sensing for solar-induced fluorescence (SIF) have accelerated, driven by the link between SIF and vegetation photosynthesis. Many algorithms focus on SIF at specific spectral regions, such as Solar Fraunhofer lines or oxygen absorption bands at the top-of-atmosphere (TOA) and top-of-canopy (TOC) levels, respectively. However, few algorithms capture SIF across the entire red-to-near-infrared range. The primary challenge in estimating total emitted SIF at the TOA level is accurately disentangling atmospheric disturbances. This article introduces the approximation error (AE) method to model and compensate for the modeling uncertainties associated with the atmospheric scattering and absorption effects. Coupled with the Bayesian retrieval algorithm called SIFFI, developed to estimate spectrally resolved SIF, the AE technique, tested on simulated data, can be generalized and integrated with other SIF retrieval approaches. This integration significantly improves SIF estimates, particularly when atmospheric conditions are not well-characterized.
Показать больше [+] Меньше [-]Ключевые слова АГРОВОК
Библиографическая информация
Эту запись предоставил Finnish Food Authority