Exploiting the MODIS albedos with the Two-stream Inversion Package (JRC-TIP): 2. Fractions of transmitted and absorbed fluxes in the vegetation and soil layers

Abstract

[1] The two-stream model parameters and associated uncertainties retrieved by inversion against MODIS broadband visible and near-infrared white sky surface albedos were discussed in a companion paper. The present paper concentrates on the partitioning of the solar radiation fluxes delivered by the Joint Research Centre Two-stream Inversion Package (JRC-TIP). The estimation of the various flux fractions related to the vegetation and the background layers separately capitalizes on the probability density functions of the model parameters discussed in the companion paper. The propagation of uncertainties from the observations to the model parameters is achieved via the Hessian of the cost function and yields a covariance matrix of posterior parameter uncertainties. This matrix is propagated to the radiation fluxes via the model's Jacobian matrix of first derivatives. Results exhibit a rather good spatiotemporal consistency given that the prior values on the model parameters are not specified as a function of land cover type and/or vegetation phenological states. A specific investigation based on a scenario imposing stringent conditions of leaf absorbing and scattering properties highlights the impact of such constraints that are, as a matter of fact, currently adopted in vegetation index approaches. Special attention is also given to snow-covered and snow-contaminated areas since these regions encompass significant reflectance changes that strongly affect land surface processes. A definite asset of the JRC-TIP lies in its capability to control and ultimately relax a number of assumptions that are often implicit in traditional approaches. These features greatly help us understand the discrepancies between the different data sets of land surface properties and fluxes that are currently available. Through a series of selected examples, the inverse procedure implemented in the JRC-TIP is shown to be robust, reliable, and compliant with large-scale processing requirements. Furthermore, this package ensures the physical consistency between the set of observations, the two-stream model parameters, and radiation fluxes. It also documents the retrieval of associated uncertainties.

Estimation of FAPAR over Croplands Using MISR Data and the Earth Observation Land Data Assimilation System (EO-LDAS)

“…where d is the MISR white sky albedo in visible and near-infrared broadbands; x t is the state vector, which is comprised of the inputs of the two stream model H t ( x t ) [35]. The state parameters are: the effective Leaf Area Index (LAI); the background albedos (BA) r g (λ); the Single Scattering Albedo of leaf (SSA):…”

“…An efficient minimisation of Equation (3) is achieved by exploiting the adjoint code (the adjoint provides an efficient estimate of the gradient of a function and can be obtained using automatic differentiation tools [38]) of the two stream model [35], which allows the use of gradient descent methods. These methods exploit the availability of the gradient of the cost function provided by the adjoint to propose the search direction of the optimisation.…”

“…In addition, this application provides information about theoretical uncertainties for both output state parameters and output fluxes. JRC-TIP output was tested against independent parameter estimates over a range of different areas [17,35].…”

Consistent retrieval of land surface radiation products from EO, including traceable uncertainty estimates

“…On the line of constant WSA NIR, one peak is located at WSA VIS around 0.03 and the other peak around 0.13. As pointed out by Pinty et al (2011b), this reflects the influence of the soil background, which for LAI values in the range from 0.3 to 0.5 exhibits an equally complex uncertainty structure (Fig. 4d).…”

“…Thus, additional, complementary constraints are required to further reduce uncertainties in the system. Such complementarity has been demonstrated for Earth observa-tion (EO) products Pinty et al, 2011b) of the fraction of absorbed photosynthetically active radiation (FAPAR), which provide information on the vegetation phenology and colour, for example. The effect on carbon and water fluxes of assimilating FAPAR in addition to atmospheric carbon dioxide samples is, for example, quantified by Kaminski et al (2012) and Schürmann et al (2016).…”