Spatial variation at the Suriname coast: Deriving alongshore and cross-shore patterns in tidal channel geometry, elevation and vegetation from UAV imagery

Abstract

Globally, 40% of the population lives in coastal areas, m, often using satellite data. However, little efforts have been undertaken to study detailed observations of channel morphology on the Amazon coast, and its interplay with erosion and accretion is therefore not well understood. Unmanned aerial vehicles (UAV’s) combined with photogrammetric procedures have proven to be a valuable method for observing small-scale (changes in) coastal morphology, as it provides high resolution imagery and elevation data. However, research into the application of this technique in observing coastal elevation is scarce. This thesis focusses specifically on the application of UAV- photogrammetry in analysis of changes and spatial patterns in coastal erosion and accretion, and channel geometry. This research comprises a case study on coastal change and spatial variations in morphological features at Weg naar Zee, a coastal area in Paramaribo, Suriname, through the use of UAV imagery. The objectives were as follows. The potential of using UAV imagery for the assessment of coastal morphology is examined (1). Then, spatial patterns (2) and temporal changes (3) in channel geometry, vegetation cover and elevation were analysed. Orthophotos and DEM’s were created from the UAV imagery for two locations that were 2 km apart, and for one of these locations a second dataset was made for 8 months later in time. Morphological units were mapped manually based on the orthophoto and DEM. For each site, channels width, depth, W:D ratio, area and channel bed elevation and slope were derived from transects with the DEM elevation data. These geometries were compared for alongshore and cross-shore patterns and differences. Lastly, elevation differences between the 8-month period were derived from the DEM’s. Significant differences were visible despite the small spatial scale and short time period. The site closest to the river is bounded by dikes on the coastal side and a cross-shore levee. It showed signs of erosion, namely an immature parallel channel pattern, low elevations and a small range of elevations on the mudflat, lower width-to-depth ratios, less vegetation and dying vegetation, and an average high-water line that reached all the way to the dike. The site further west of the river is more stable, with growing mangrove vegetation and a large buffer zone between the mainland dike and the average high-water line. The channel network pattern was meandering dendritic. At this site, mangroves are planted to regenerate the vegetation, and permeable dams are constructed in the intertidal area as sediment trapping units. After the 8-month period, the elevation of the intertidal mudflat had increased with ±0.5 meters. This research has sketched a wide variety of spatial variations in morphology and coastal behaviour that can occur on a small scale in cross-shore and alongshore direction. As this local variation is critical in understanding how and where to implement erosion mitigation strategies, it is necessary to upscale this research with longer time-series and larger along shore coverage to determine which processes are at the base of this variation. Furthermore, UAV imagery has proven to be successful in determining both medium-scale and small-scale morphologies and elevations. It provides a diversity of datatypes, and it has the potential to facilitate direct erosion and accretion, and automated morphology classification.