Search for a command to run...
Precise characterization of highly dynamic coastal features, such as tidal inlets, is critical because they frequently act as navigation channels and govern water exchange in productive systems such as coastal lagoons and estuaries. Moreover, their morphological evolution is complex, being controlled by multiple interacting factors including tidal regime, wave conditions, and sediment budgets. Advances in spaceborne technologies can significantly transform their geomorphological monitoring by enhancing coastal mapping. Satellite-Derived Bathymetry (SDB) offers a cost-effective and scalable alternative to traditional survey methods for generating bathymetric data. However, the applicability of SDB is often constrained in optically complex waters, particularly in regions affected by turbidity. In this study, we present an SDB approach utilizing imagery from the twin Sentinel-2 A/B satellites, integrating satellite-derived water quality indicators with a multi-temporal compositing framework. This automated technique mitigates the effects of water turbidity by aggregating observations over a short time period (a variable window of up to two months), thereby improving depth retrieval while masking optically deep or noisy pixels. The approach is tested for three different years (2018, 2019, and 2020) and tidal stages (high and low tide) in two inlets of Ria Formosa (Faro-Olhão and Armona), a protected wetland enclosed by a multi-inlet barrier island system. The recurrent bathymetric products achieved spatial resolutions of 10 m and successfully resolved seafloor features to depths of up to 15 m. When benchmarked against high-resolution in-situ data at Armona Inlet, the method produces median errors ranging from 0.71 to 1.05 m and bias ranging from −0.9 to 0.53 m. In addition, an intercalibration procedure at Faro-Olhão Inlet yielded consistent validation results, with median errors ranging from 0.59 to 1.36 m. Subsequently, temporal comparison of SDB maps accurately captured the hotspots of erosion and sedimentation, closely matching high-resolution in-situ surveys. This demonstrates the potential of SDB to enhance change detection and support recurrent coastal monitoring critical to hazard assessment and climate change impact studies in vulnerable coastal areas. These outcomes also underscore the approach's robustness in less-than-ideal optical conditions and highlight its feasibility across diverse coastal systems, making it suitable for scalable use at various spatial coverages. • The twin Sentinel-2 satellites enable high-resolution mapping of shallow coastal depth. • Two tidal inlets of Ria Formosa (Portugal) are analyzed during 2018, 2019, and 2020. • A multi-temporal compositing approach automatically mitigates water quality variability. • The model delivers comprehensive and recurrent operational bathymetric monitoring. • Satellite-derived maps reveal erosion and accretion patterns consistent with in-situ measurements.