Assessing Tidal Hydrodynamics in a Tropical Seascape Using Structure‐from‐Motion Photogrammetry and 2D Flow Modelling

Abstract

AbstractTidal wetlands continue to be threatened by changes in seascape hydrological regime and connectivity resulting from human activities (e.g. urbanisation, engineered barriers) and climate change. Reliable and parsimonious models that can be used by managers and practitioners to simulation tidal wetland hydroperiod dynamics (duration, depth, and frequency of tidal inundation) at high-resolution are limited presumably because these ecosystems have very low elevation across their flooding plain. Here, we developed a two-dimensional hydrodynamic model parameterised using a high-resolution (3 cm) and accurate (8-cm RMSE elevation error) digital elevation model (DEM) and land cover map (2-cm resolution) derived from unoccupied aerial vehicles (UAVs) structure from motion photogrammetry (SfM) to assist in the understanding of tidal wetland hydroperiod and hydrological connectivity of an upper tidal Australian tropical seascape. Ground-based water level datasets were used to calibrate and validate the model with higher accuracy (RMSE = 7 cm between maximum observed and simulated depth). The high-resolution approach demonstrates how small changes in topography such as vehicle tracks can interfere with hydrological connectivity. Centimetre-changes in tidal height resulted in important variations (10 ha) in the total area of the wetland being inundated, suggesting that small anthropogenic modifications of tidal inputs (e.g. culverts and sea-level rise) might have important implications on tidal wetland inundation patterns. Despite challenges related to reconstructing topography in densely vegetated areas and obtaining bathymetric data, the method developed here represents an accurate and cost-effective approach to quantify tidal wetland hydroperiod. This approach assists in planning, defining, and implementing effective and measurable restoration and protection projects of tidal wetland ecosystems. Graphical Abstract