Abstract <p>Accurately retrieving Sun-Induced Fluorescence (SIF) is critical for monitoring plant physiological status, yet the signal is significantly distorted by light reabsorption and scattering within the canopy. While empirical models exist for the far-red region of the spectrum, accurately accounting for the photon escape fraction in the complete Chlorophyll Fluorescence (ChlF) emission range remains challenging. Based on our previous work under monochromatic conditions, in this work we present a photophysical framework to estimate the chlorophyll fluorescence escape fraction (<i>f</i>_<i>esc</i><i>)</i> across the full chlorophyll emission spectrum (600–800&#xa0;nm) under polychromatic excitation. The methodology integrates experimental radiance measurements of <i>Bistorta amplexicaulis</i> with an algorithm to decouple reflectance from emission. To evaluate the model’s robustness in the field, we conducted a global sensitivity analysis using a synthetic dataset generated by coupling the SMARTS atmospheric radiative transfer model with the PROSAIL canopy model. Our results demonstrate that failing to account for canopy light reabsorption and scattering can underestimate fluorescence yields by approximately 25%. We identified distinct drivers for <i>f</i>_<i>esc</i> in SIF-relevant bands: <i>f</i>_<i>esc</i> in the red region (687&#xa0;nm) is primarily governed by chlorophyll content and Leaf Area Index (LAI) due to intense fluorescence reabsorption, while <i>f</i>_<i>esc</i> in the far-red region (760&#xa0;nm) is dominated by canopy structure and leaf inclination (LIDFa). This study provides a practical and robust estimation method for <i>f</i>_<i>esc</i> at the canopy level, offering a key tool for improving the accuracy of SIF-based photosynthetic efficiency assessments in both environmental and agronomic remote sensing applications.</p> Graphical abstract <p></p>