In addition, Canny is able to produce a polyline is an approximation to the original pixel curve. He concluded that each curve contains sequences of pixels is able to convert to vector layer by fitting piecewise line segments to it. Maged (2002) stated that Canny algorithm produces pixel wide skeleton curves and able to extract the sequences of pixels along this curve. Additionally, the concave coastline is detected which is obvious along the coastline of Sultan Mohamed Airport. Indeed, Canny algorithm can perform pseudo-edge appearance. Furthermore, it is also clear that small texture details have detected. It is interesting to notice that Canny algorithm is provided an automatic digitizing for shoreline. Figure 3 shows the result of Canny algorithm. This result confirms that studies of Vachon et al., (1993) and Maged (2004). On other words, it can be used to map SAR observed spectra into real ocean spectra. Indeed, it is able to solve the matter of non-linearity between SAR observed spectra and ocean wave spectra. According to Maged (2004) velocity bunching model produces precisely information of wave spectra. The implementation of the Management Plan would dependent on the co-operations of the government departments and agencies, private sector and the public.īetween velocity bunching model and in situ data shows positive correlation as r 2 values for both ERS-1 and RADARSAT-1 SAR data are 0.69 and 0.73, respectively. A total of 1963 km of coastline was evaluated and of this, 3.3% of the mapped shoreline is classified as being extreme vulnerability, 11% of Peninsular Malaysia shoreline is classified as very high vulnerability and 40 % as high vulnerability. The ranking is on a linear scale from 1 to 5 in order of increasing vulnerability value 1 represents the lowest risk ranking assigned to the coastline whereas value 5 ranks the coastline with the highest risk. These six variables consist of geomorphology, shoreline change rate, maximum current speed, maximum tidal range, significant wave height and sea level rise. This study incorporated six variables to assess the CVI for the study area. (Douglas and Crowell 2000).Ībstract-The main objective of the present study is to develop a coastal vulnerability index (CVI) for the Peninsular Malaysia coastline. In addition, causes for variation in rate of change include geomorphic features such as inlets, wave energy, engineering changes, etc. 1998), temporal and spatial bias in the estimation of shoreline rate-of- change statistics (Eliot and Clarke 1989), and the method used to calculate the rate (Dolan et al. 2010b), total time span of the shoreline data acquisition (Douglas et al. The accuracy of shoreline change rate estimation reflects actual changes and prediction of future changes depends on several factors, such as the accuracy in shoreline position data, variability of the shoreline movement, number of measured shoreline data points (Kumar et al. Calculations of accurate shoreline change rates are frequently employed to summarize historical shoreline movements and to predict the future shoreline positions through different modeling procedures (Li et al. The EPR method uses only two data points to delineate a change rate - the earliest and most- recent shoreline positions. However, they are always subjected to uncertainty because of inherent errors and deficiencies in the model used to evaluate the historical shoreline position, but apply the only one method in used shoreline changes that model is end point rate (EPR). Several methods are available for calculating the shoreline position namely, End Point Rate (EPR), Average of Rates (AOR), Linear Regression (LR), Minimum Description Length, Ordinary Least Squares and jack- knifing are being widely used to estimate and forecast the rate of change in shoreline. topographic surveys, coastal monument, and beach profiles and aerial photographs) each with its own measurement uncertainty. The analysis results are discussed in transects wise and location specific Shoreline positions are often defined from various sources (e.g.
By using statistical methods, the study area has equally divided into 80 transect lines, and each transects have 250m length of the coastline.