|Paper title||Quantum yield estimates from in situ spectroradiometer measurements and its implications on remote sensing of sun-induced fluorescence in lakes|
|Form of presentation||Poster|
Quantum yield of fluorescence (ϕ_F) represents the small fraction of absorbed photons in phytoplankton that is converted to sun-induced fluorescence (SIF). This fraction is typically up to 2% in optically complex waters. All other absorbed photons are either used for photochemistry in the reactions centers or dissipated as heat. When fluorescence is reduced from a maximum level due to an increase in open reaction centers, Photochemical Quenching (PQ) occurs. Other forms of fluorescence reduction lead to increased thermal dissipation and are referred to as Non-photochemical Quenching (NPQ). In cases where NPQ is minimal, ϕ_F and SIF increase with higher irradiance. However, when NPQ is present due to photo-inhibition or protective measures employed by the phytoplankton, SIF may still increase with irradiance while ϕ_F decreases. Consequently, NPQ conditions also lead to lower quantum yield of photosynthesis.
Knowing the ϕ_F is key to understanding SIF emission in phytoplankton as it enables us to interpret the dynamics of SIF in relation to PQ or NPQ. Disentangling PQ from NPQ allows us to use SIF estimates in various applications in aquatic optics and remote sensing such as accurate estimation of chlorophyll-a concentration (chl a) or modelling of primary productivity. These are essential to assess the water quality status of surface waters and to understand the dynamics of aquatic ecosystems. Retrieving and interpreting SIF becomes more plausible at the present time and in the near future with the increasing availability of in-situ, airborne and spaceborne hyperspectral sensors. However, obtaining ϕ_F is challenging due to prior data necessary for the calculations especially in inland waters.
Using the autonomous Thetis profiler from the LéXPLORE platform in Lake Geneva, we demonstrate a novel way of estimating ϕ_F based on an ensemble of in-situ profiles of Inherent Optical Properties (IOPs) and Apparent Optical Properties (AOPs) taken between October 2018 and August 2021. In particular, we exploited the profiler’s hyperspectral radiometers to obtain upwelling radiances and downwelling irradiances in the top 50 m of the water column. These AOPs were the main basis of our SIF retrieval, representing natural variations in fluorescence emission under different bio-geophysical conditions. We further used hyperspectral absorption and attenuation, and backscattering measurements at discrete wavelengths to obtain the water’s IOPs. These IOPs were used in radiative transfer model simulations assuming ϕ_F=0 to obtain a second set of AOPs without fluorescence contributions. The measured and simulated reflectances obtained outside the fluorescence emission region which satisfy the optical closure analysis were kept in the succeeding steps. By associating the difference between these measured and simulated AOPs, known chlorophyll-a concentrations and IOPs, we obtained estimates of ϕ_F.
We analysed obtained ϕ_F values to determine the conditions at which NPQ occurs. Consequently, we evaluated the vertical and temporal changes in ϕ_F. We observed diurnal changes in NPQ occurrence, particularly during clear sky conditions where downwelling irradiance changes significantly throughout the day. For instance, we observed that ϕ_F can be up to 65% lower when NPQ is activated compared to PQ stimulated conditions. While downwelling irradiance is a significant contributor to changes in ϕ_F, its role can be sometimes not easily interpreted because the threshold of radiant flux at which NPQ is activated in inland waters is not consistent. Other factors such as phytoplankton photo-adaptation and the composition of different phytoplankton communities also play significant roles in understanding phytoplankton response to incident light and therefore, quenching mechanisms. Our results contribute insight on the nature of SIF and can facilitate activities to assimilating SIF and ϕ_F estimates in remote sensing algorithms, which would aid us in monitoring not only phytoplankton biomass but also the eco-physiological state of phytoplankton cells.