Advancing earth observation and its integration with hydrodynamic modelling to understanding transitional waters

Abstract

Transitional water systems are dynamic land–sea environments where river inflow, tidal exchange, atmospheric forcing and shallow bathymetry interact across multiple scales. Earth Observation (EO) and hydrodynamic modelling can improve understanding of these systems, but the performance of these approaches remains uncertain in shallow, optically complex and morphologically heterogeneous environments. Well-monitored transitional systems such as the Venice Lagoon provide an opportunity to evaluate, further develop and integrate these approaches, and to assess their transferability to less-monitored, data-scarce transitional waters. This PhD thesis follows this framework to investigate surface water temperature (SWT) and surface water salinity (SWS) variability across contrasting transitional environments. Three studies address this aim. Chapter 2 evaluates and integrates satellite- and model-derived SWT products in the Venice Lagoon. Chapter 3 develops the first satellite-based framework for retrieving coloured dissolved organic matter (CDOM) and SWS in the Venice Lagoon, using the former as an optical proxy for freshwater–marine mixing. In parallel, hydrodynamic modelling represents broader spatiotemporal SWS patterns. Chapter 4 assesses the transferability of these approaches to the less-monitored Danube Delta-Razelm-Sinoie-western Black Sea continuum. In Venice Lagoon, the System of Hydrodynamic Finite Element Modules (SHYFEM), driven by real-time meteorological and hydrological observations, reproduced seasonal SWT variability with continuous coverage. Landsat Thermal Infrared Sensor (TIRS) observations processed with the Thermal Atmospheric Correction Tool (TACT) resolved fine-scale variability and industrial-discharge anomalies. Their integration improved spatiotemporal SWT representation. Multi-year in situ observations confirmed an inverse CDOM–SWS relationship, enabling Sentinel-3 Ocean and Land Colour Instrument (OLCI) to map freshwater influence, tidal redistribution and lagoon-wide gradients. In the Danube region, where local meteorological observations were unavailable, SHYFEM driven by globally available atmospheric reanalysis data, together with Landsat SWT and Sentinel-3-derived SWS, captured seasonal contrasts, river–lagoon–sea gradients and freshwater-plume pathways. The observed discrepancies between satellite-derived and modelled SWT and SWS estimates were linked to water-constituent variability, atmospheric-correction performance, shallow-water structure and environmental forcing. Collectively, this thesis advances Earth Observation and hydrodynamic modelling to improve understanding of SWT and SWS dynamics across the study sites. The developed framework is transferable to comparable environments and strengthens monitoring capacity, supports ecosystem assessment and improves the scientific basis for coastal and lagoon management.