Resolving the Spectral and Directional Structure of Solar Radiation for the NASA Libera Mission: From TOA Retrieval to Terrestrial Ecosystem Impacts

Abstract

This dissertation enhances the representation of clear-sky solar (shortwave) radiation in support of the NASA Libera mission by resolving spectral and directional structure from top-of-atmosphere radiance retrievals through surface radiative forcing to land-surface impacts. It addresses three coupled challenges: developing an operational method to partition broadband shortwave radiation into visible (VIS) and near-infrared (NIR) components, establishing a physically based benchmark for evaluating clear-sky spectrally resolved surface irradiances, and quantifying the extent to which these spectral and directional refinements alter land-surface model responses.First, the dissertation develops a ratio-based radiance-to-irradiance partitioning framework for clear-sky land scenes that avoids the immediate need for dedicated spectral angular distribution models. The approach solves for the ratio between NIR and VIS anisotropic factors (β) and anchors spectral partitioning to an operational broadband shortwave irradiance product. Sensitivity analyses show that β is controlled primarily by viewing-illumination geometry and aerosol loading, enabling a low-order polynomial parameterization that remains physically interpretable. Internal validation against radiative-transfer simulations demonstrates high accuracy, with root-mean-square errors in retrieved VIS and NIR irradiances consistently below 3.0 W m-2 across diverse land surface types and viewing geometries. Second, the dissertation establishes the physical credibility of clear-sky spectrally resolved surface shortwave irradiances using a hierarchical radiative-transfer framework and demonstrates that broadband agreement can mask substantial errors in the internal VIS-NIR partitioning due to spectral cancellation. Across Clouds and the Earth’s Radiant Energy System (CERES) SYN1deg, MODTRAN 6.0, and a modified CCCma model, NIR irradiance sensitivities to precipitable water vapor exhibit consistent sign, magnitude, and spatial structure. Broadband clear-sky surface shortwave irradiance decreases by approximately 0.71 W m-2 per kg m-2 of precipitable water, with more than 90% of this attenuation attributable to NIR absorption. Third, the dissertation uses the Community Land Model at the Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) site to perform orthogonal forcing experiments that independently correct (i) VIS-NIR partitioning (VIS-ONLY), (ii) direct-diffuse partitioning (DIF-ONLY), and (iii) both together (ALL). Canopy-integrated impacts are regime-dependent and crop-specific. Relative to the control forcing, summer crops exhibit net productivity gains dominated by diffuse corrections: Gross Primary Production (GPP) increases by +0.13% (VIS-ONLY), +2.84% (DIF-ONLY), and +3.06% (ALL) for corn, and by +0.57% (VIS-ONLY), +3.64% (DIF-ONLY), and +4.41% (ALL) for soybean. Winter wheat exhibits net reductions, with GPP changes of -1.08% (VIS-ONLY), -1.72% (DIF-ONLY), and -2.93% (ALL). Consistent structural responses occur in leaf area index, with corn changing by +0.61% (VIS-ONLY), +0.76% (DIF-ONLY), and +0.92% (ALL), soybean changing by +0.43% (VIS-ONLY), +0.43% (DIF-ONLY), and +0.63% (ALL), and winter wheat changing by -1.21% (VIS-ONLY), -2.28% (DIF-ONLY), and -3.18% (ALL). These carbon and structural responses propagate to surface coupling through albedo, net radiation, and turbulent flux partitioning, and they redistribute the growing-season hydrologic budget by shifting evapotranspiration components and soil-water storage, even under fixed incoming shortwave. Overall, this dissertation shows that resolving spectral and directional structure improves both the physical interpretation of Earth radiation budget observations and the realism of land-surface modeling, providing an end-to-end foundation for connecting Libera spectral measurements to surface radiative forcing and terrestrial energy, water, and carbon responses.