Bridging Gaps in Aquatic Remote Sensing Reflectance Validation: Pixel Boundary Effect and Its Induced Errors.

Ocean color remote sensing is important for monitoring marine biogeochemical processes. The accuracy of remote sensing reflectance (Rrs), a fundamental data product, is critical yet challenged by the scale mismatch between in situ point measurements and satellite-based areal observations from pixels. This mismatch introduces uncertainty, notably from the non-uniform spatial response within a pixel-a potential error source at pixel boundaries that remains poorly quantified. To address this issue, we introduced the pixel-level spatial mismatch index (PSMI) to assess spatial representativeness errors induced by the pixel boundary effect (PBE). Using AERONET-OC (AErosol RObotic NETwork-Ocean Color) data alongside MODIS/Aqua and OLCI/Sentinel-3A observations, we showed that the PSMI effectively identified a systematic Rrs deviation peak when a site lay within a pixel's edge attenuation zone. This phenomenon, observed across sensors with different resolutions (MODIS and OLCI), exhibited sensor- and band-dependent peak characteristics. We further proposed a quantitative framework called a Riemann Stieltjes integral-based index to measure the spatial concentration of this deviation peak, and a baseline method to objectively define the PBE window. Our analyses revealed that PBE not only acts as an independent error source but also interacts with atmospheric and geometric errors, forming new multifactor interactions that significantly modulate the overall uncertainty in Rrs products. Consequently, pixel-scale effects should be incorporated into future validation protocols, and the PSMI framework can provide an intrinsic tool for this purpose.