An optimal waterline approach for studying tidal flat morphological changes using remote sensing data: A case of the northern coast of Vietnam

Abstract

Tidal flats on the north coast of Vietnam suffer diurnal tide with a tide range varying between 0.3 m and 3.5 m. Along the 350 km long coastline, the diversity of environmental conditions induces various tidal flats with different characteristics. This study applies the waterline method for multi-temporal satellite images to build Digital Elevation Models (DEMs) of tidal flats during the last 25 years. 117 Landsat images acquired with TM, ETM+, and OLI have been processed to construct tidal flat DEMs in 1989, 2000, and 2014. Waterlines extracted from single spectral bands (near-infrared [NIR], short wave infrared [SWIR]) or band ratios (normalized difference water index [NDWI], normalized difference vegetation index [NDVI], Green/SWIR) of the Landsat data have been compared with waterlines digitalized on Spot, Aster and Worldview 2 images. This experiment allows us to determine the best band (or band ratio) for extracting waterlines depending on local conditions. Consequently, the study shows that the Green/SWIR ratio image is a good solution for extracting waterlines in the black coal tidal flats of Cam Pha. However, the NDWI index appears to be a better choice for the other parts of the study area. The vertical accuracy of the tidal flat DEMs reaches 0.144 m. The change analysis of the DEMs also emphasizes the tidal flat evolution in both vertical and horizontal dimensions, i.e. erosion or accretion. The erosion of the tidal flats along the northern coast of Vietnam is particularly developed in the area extending from Yen Hung to Mong Cai, especially in Mong Cai with an amount of about 50 × 106 m3 of sediments lost between 1989 and 2014. On the contrary, the tidal flats in the south of the study area show a high rate of deposition due to the sediments fed by Red and Thai Binh rivers. About 35 × 106 m3 of sediments deposited in the tidal flat surrounding the Red River mouth between 1989 and 2014. This study represents a development of the waterline extraction method to investigate the evolution of tidal flat at a large scale and a diversified coastal environment using optical satellite images and fieldwork.