The boreal forest plays an important role in the global carbon cycle but has remained a significant source of uncertainty. Remote sensing can help us better understand the boreal forest’s role in the global carbon cycle. A faint light signal emitted by plant’s photosynthetic machinery, known as solar-induced chlorophyll fluorescence (SIF), is a promising remotely sensed proxy for carbon uptake, also known as gross primary productivity (GPP), due to its connection to photosynthesis and its strong relationship with GPP when observed by satellite. However, SIF and GPP are fundamentally different quantities that describe distinct, but related, physiological processes. The relationship between SIF and GPP is therefore complicated by both physical and ecophysiological controls. In particular, the dynamics of the SIF/GPP relationship are poorly understood under varying viewing directions and light conditions. This is further complicated in evergreen systems where canopy clumping and the presence of needles create a unique radiative environment. We use a combination of tower-based SIF and GPP measurements from a boreal forest field site compared with a coupled biochemical-radiative transfer model to understand illumination effects on the SIF/GPP relationship. We find that GPP is amplified under cloudy sky conditions in both measurements and model results. SIF on the other hand, shows no significant difference between sunny or cloudy sky conditions in modeled results, but does show a difference in measurements. We suggest that these differences may be due to viewing geometry effects that are important for SIF under sunny sky conditions or the presence of clumping. Accounting for the differences in the SIF/GPP relationship therefore is critical for the utility of SIF as a proxy for GPP. In summation, our results provide insight into how we can use remote sensing as a tool to understand photosynthesis in the boreal forest.

Solar-Induced Chlorophyll Fluorescence (SIF) is a powerful proxy for gross primary productivity (GPP) in Boreal ecosystems. However, SIF and GPP are fundamentally different quantities that describe distinct, but related, physiological processes. Recent work has highlighted non-linearities between SIF and GPP at finer spatial (leaf- to canopy- level) and temporal (half-hourly) scales. Therefore, questions have arisen about when, where, and why SIF is a good proxy for GPP and what the potential sources for divergence between the two are. The goal of this study is to answer two specific questions: 1) At what temporal scale is SIF a good proxy for GPP and 2) What are the predominant physical and ecophysiological drivers of nonlinearity between SIF and GPP in boreal ecosystems? We collected tower-based measurements of SIF (and other common vegetation indices) with PhotoSpec (a custom spectrometer system) and eddy-covariance GPP data at a 30-minute resolution at the Southern Old Black Spruce Site (SOBS) in Saskatchewan, CA. We applied a combination of statistical and machine learning approaches to disentangle the influence of structural/illumination effects and ecophysiological variations on the SIF signal. Our results show that at a high temporal resolution (half-hourly), SIF and GPP are predominantly dependent on photosynthetically active radiation (PAR). Therefore, the non-linear light response of GPP drives non-linearity between SIF and GPP. Additionally, canopy structure and illumination effects become important to the SIF signal at high temporal resolutions. At the seasonal timescale, SIF and GPP exhibit co-varying responses to PAR, even when accounting for changes in canopy structure. We attribute changes in the light responses of SIF and GPP to sustained photoprotection over winter which co-varies with changes in temperature. Finally, we show that the relationship between SIF and GPP has a seasonal dependence caused by small differences between the light use efficiencies of fluorescence and photosynthesis. Accounting for this seasonally variable relationship will improve the use of SIF as a proxy for GPP.

Solar-Induced chlorophyll Fluorescence (SIF) provides a powerful proxy for determining forest gross primary production (GPP), particularly in evergreen ecosystems where traditional measures of greenness fail. The dynamics of the SIF/GPP relationship, however, are poorly understood under varying viewing directions and light conditions. This is, in large part, due to challenges in measuring SIF at the spatiotemporal scale that is necessary to understand these effects. Therefore, the aim of this work is to utilize high-temporal and spatial resolution SIF measurements to better constrain the response of SIF to ambient canopy illumination and viewing geometry. We use a PhotoSpec instrument and eddy covariance measurements to explore the SIF/GPP relationship under various viewing directions and light conditions during the 2019 and 2020 growing seasons at the Old Black Spruce site in Saskatchewan, Canada. PhotoSpec is a tower-based 2-D scanning spectrometer system capable of taking Fraunhofer-line based SIF retrievals in the red and far-red wavelength ranges with a 0.7 degree field of view at a ~30 second time resolution. Measured SIF and GPP are combined with SCOPE modelling results to provide a mechanistic understanding of the physical and ecophysiological drivers for the SIF/GPP relationship in the Boreal Forest. Our results show that viewing direction and solar zenith/azimuth angles are important for the SIF signal under direct light conditions, but not under diffuse. Furthermore, the SIF/GPP relationship changes under direct and diffuse light conditions at a 30 minute, daily, and monthly resolution. Our ability to use SIF as a proxy for GPP depends on a quantitative understanding of radiative transfer within the canopy and how scanning geometry impacts SIF measurements. These results provide an important insight into these relationships in the Boreal forest, a region where GPP has been traditionally difficult to track using remote sensing.

Using data from five long-term field sites measuring soil moisture, we show the limitations of using soil moisture observations alone to constrain modelled hydrological fluxes. We test a land surface model, MESH/CLASS, with two configurations: one where the soil hydraulic properties are determined using a pedotransfer function (the texture-based calibration) and one where they are assigned directly (the hydraulic properties-based calibration). The hydraulic properties-based calibration outperforms the texture-based calibration in terms of reproducing changes in soil moisture storage within a 1.6 m deep profile at each site, but both perform reasonably well, especially in the summer months. When the models are constrained using observations of changes in soil moisture, the predicted hydrological fluxes are subject to very large uncertainties associated with equifinality. The uncertainty is larger for the hydraulic properties-based calibration, even though the performance was better. We argue that since the pedotransfer functions constrain the model parameters in the texture-based calibrations in an unrealistic way, the texture-based calibration underestimates the uncertainty in the fluxes. We recommend that reproducing observed cumulative changes in soil moisture storage should be considered a necessary but insufficient criterion of model success. Additional sources of information are needed to reduce uncertainties, and these could include improved estimation of the soil hydraulic properties and direct observations of fluxes, particularly evapotranspiration.