Spatial variability in almond yield reflects complex interactions between canopy structure, physiological status, and environmental conditions. Reliable, objective yield characterization is essential for precision orchard management. Remote sensing offers a data-driven alternative to traditional assessments, with recent advances highlighting the potential to better understand yield variability through solar-induced chlorophyll fluorescence (SIF) as a direct proxy of photosynthetic activity. This study evaluates the contribution of SIF, quantified at different spectral resolutions, to explaining yield variability in a commercial almond orchard in South Australia. Airborne hyperspectral data was acquired to retrieve SIF at different resolutions and wavelengths and to retrieve structural and biochemical canopy traits. In addition, thermal images were used to retrieve the Crop Water Stress Index (CWSI). These variables were analysed using a Gaussian Process Regression (GPR) model to assess how spectral resolution and the wavelength selection used for SIF quantification influence the ability to track variability in almond yield. Results demonstrate that SIF derived from high-resolution (0.1 nm FWHM) imagery significantly improves yield-variability estimates compared to mid-resolution imagery (5.8 nm FWHM). SIF quantified in the O₂A absorption band (~ 760 nm) contributed more strongly than O₂B (~ 687 nm), while the inclusion of O₂B provided only marginal additional benefit. These findings highlight the importance of spectral resolution for exploiting SIF in precision orchard management.