Seasonal and spatial variation in sediment and macrozoobenthic properties of tidal flats and their prediction using remote sensing

Tidal flats play a vital role in sustaining biodiversity and supporting ecological processes. They are characterized by environmental variability, including seasonal and spatial heterogeneity, which influence both physical processes and biological communities. This study investigated the seasonal and spatial variability of environmental and ecological variables of a tidal flat in the Dutch Wadden Sea. It also explored the temporal alignment of remote-sensing data (PlanetScope, Dove-C) relative to field samples, for predicting the variables through random forest. Over two years (2022 & 2023), samples were collected during three monthly field campaigns (April, July, and August), focusing on sediment properties (median grain size, silt content), macrozoobenthic characteristics (biomass, abundance, species richness), and chlorophyll-a concentrations (a proxy for microphytobenthic biomass). For chlorophyll-a, seasonality explained 17% of observed variance, whilst sediment and macrozoobenthic properties did not exhibit consistent seasonal variance ('2% variance explained). Sediment properties showed variability between sampling locations (explaining '80% of variance). In macrozoobenthic characteristics and chlorophyll-a concentrations, sampling location explained 50% and 30% of observed variance, respectively. This highlights the need for both temporal and location-specific sampling for chlorophyll-a, while sediment and macrozoobenthic properties require mainly location-specific sampling. We evaluated four image-matching schemes to assess how the timing of satellite imagery relative to field sampling influences prediction accuracy. Median grain size (R² = 0.11), silt content (R² = 0.53), and chlorophyll-a (R² = 0.26) were best predicted when imagery was aligned with sampling moments, whereas species distribution (F1 = 0.46–0.81) was not sensitive to timing. These results enhance the sampling design of tidal-flat variables, providing a framework for targeted monitoring and improved ecological assessment.