Unmanned aerial vehicle large scale mapping for coastal erosion assessment

Abstract

Most of the time, many countries in tropical region are covered by cloudwhich obstruct the acquisition of high resolution optical image from satellite. Thus,the application of large scale mapping for coastal erosion assessment is difficult to becarried out. The unmanned aerial vehicle (UAV) can be used for acquisition of highresolution image due to its ability of flying at low altitudes and under cloud cover.This study aims to improve the methodology of data acquisition for assessing erodedcoastal area by using UAV system designed for large scale mapping. The objectivesof this study are (i) to improve the methodology of data acquisition for large scalecoastal erosion assessment using UAV with rapid and low cost, (ii) to assess theaccuracy of the digital photogrammetric products obtained with different groundcontrol points (GCP) configurations and flying altitude, and (iii) to identify andassess coastal areas due to erosion based on the large-scale orthophoto produced.Two improved methods were introduced in this study: (i) rapid data acquisition andprocessing, low cost and accurate mapping using UAV, and (ii) modification of theend point rate (EPR) and linear regression rate (LRR) calculation methods byincluding two new parameters namely the sea water and wave level to detectcoastline changes. The modified EPR and LRR methods in detecting the coastlinechanges of Crystal Bay, Alai, Melaka and Kampung Seri Pantai, Mersing, Johorwere evaluated through statistical model, namely t-test and compared with otherstudies of similar area for validation and verification. Several configurations of GCP,check points (CP) and flying altitude were used to produce the digital elevationmodel (DEM) and orthophoto which were then processed photogrammetrically.Subsequently, the coastal erosion assessment was performed and the coastlinechanges of private properties, buildings and residential areas was identified. Theresults show that the best GCP configuration to produce coastal erosion mappingscale of 1:14000 is 25 points. Meanwhile, the best flying altitude is 300m withaccuracies of ±0.002m, ±0.004m and ±0.389m in Northing, Easting and Heightrespectively. Furthermore, this study shows that most of the coastline, privateproperties, buildings and residential areas are affected by the coastal erosion. Basedon the modified EPR and LRR calculation methods in determining the coastlinechanges, it is evident that the coastline change rate is significant. In conclusion, thisstudy shows that the UAV system offers many advantages such as its ability to fly atlow altitude, low cost, rapid data acquisition and processing in detecting coastlinechanges and accurate results.