Coastal and estuarine wetlands provide a range of important ecosystem services, but are currently being damaged and degraded due to human activities, reduced sediment supply and sea level rise. Managed realignment (MR) is one approach used to compensate for the loss of intertidal habitat, however saltmarshes in MR sites have been recognised to have lower biodiversity than natural environments. This has been associated with differences in the physical functioning including the sediment structure, reduced hydraulic connectivity, and lower topographic variability such as the abundance of intertidal creek networks. Intertidal morphology, including creek networks, play an important role in supporting and regulating saltmarsh environments through the supply of sediment, nutrients and water, and in draining intertidal marshes. However, there is a lack of empirical data on the formation and evolution of topographic features and variability in saltmarsh environments. This is likely to be due to creek networks in natural marshes already being in a state of quasi-equilibrium, making MR sites an ideal environment to investigate creek development. However, traditional remote sensing techniques (such as LiDAR) tend to be relatively expensive, infrequent and at a coarse resolution meaning small, but important (cm-scale), changes are often missed. This study advances the ability to detect these small scale changes by demonstrating the suitability and potential applications of using the emerging photogrammetric method Structure-from-Motion (SfM) on images taken using a small-Unmanned Aerial System (sUAS). Three surveys from a rapidly changing, near-breach site were taken at the Medmerry Managed Realignment Site in July 2016, September 2017 and July 2018. A suitable degree of confidence was found between the modelled surface and independent check points (vertical root-mean-square-errors of 0.0245, 0.0704 and 0.1571 for 2016, 2017 and 2018 respectively). DSMs of Difference (DoD) analysis was performed to evaluate elevation change, with areas experiencing up to 85 cm of accretion between 2016 and 2018. However, when considering the error associated with both surveys, between 2016 and 2017, only 34.39% of the survey area experienced change above the level of detection (LoD). In contrast, 76.97% experienced change greater than the LoD between 2017 and 2018. Stream order analysis classified the creek networks into five orders in 2016 and four orders in 2017 and 2018, with 2016 having a higher abundance (291 in 2016 compared to 117 (2017) and 112 (2018)) and density (0.44 m/m2 in 2016 compared to 0.27 m/m2 in both 2017 and 2018) of creek networks. These results provide an innovative high resolution insight into the evolution of restored intertidal wetlands, and suggest that SfM analysis of images taken using a sUAS can be a useful tool with the potential to be incorporated into studies of MR and natural saltmarsh sites. sUAS analysis can, therefore, advance the management of these environments to ensure the provision of ecosystem services and to protect against future anthropogenic activity, sea level rise and climate change.
Read full abstract