Spatial scales of optical variability in the coastal ocean: Implications for remote sensing and in situ sampling

Abstract

AbstractUse of ocean color remote sensing to understand the effects of environmental changes and anthropogenic activities on estuarine and coastal waters requires the capability to measure and track optically detectable complex biogeochemical processes. An important remote sensor design consideration is the minimum spatial resolution required to resolve key ocean features of physical and biological significance. The spatial scale of variability in optical properties of coastal waters has been investigated using continuous, along‐track measurements collected using instruments deployed from ships, aircraft, and satellites. We defined the average coefficient of variance, , within an image pixel as the primary statistical measure of subpixel variability and investigated how changes as a function of the Ground Sampling Distance (GSD). In general, d /dGSD is positive, indicating that the subpixel variability increases with GSD. The relationship between and GSD is generally nonlinear and the greatest rate of change occurs at small spatial scales. Points of distinct transition in the relationship between and GSD are evident between 75 and 600 m, varying depending on the location and the optical parameter, and representing the GSD above which most of the spatial variability due to small‐scale features is subsumed within a pixel. At GSDs greater than the transition point, most of the small‐scale variability occurs at subpixel scales and, therefore, cannot be resolved. On average, the transition GSD is around 200 m. The results have application in both sensor design and in situ sampling strategy in support of coastal remote sensing operations.