Berita terupdate dan terpercaya

Detection of shoreline changes using Geospatial tools and Automatic Computation

Detection of shoreline changes using Geospatial tools and Automatic Computation: Case of Kuchaveli DS Division, the East coast of Sri Lanka
aDepartment of Geography, Eastern University, Sri Lanka

Coastal erosion is a natural phenomenon affecting a large number of coastal areas. The coastal zone is an area with immense geological, geomorphological, and ecological interest. The shoreline change has its impact, but which is not so visible. To observe this, we need a long and continuous set of data. This study focuses on the detection and analysis of the shoreline changes by automatic image analysis techniques using multi-temporal Landsat images (1991, 2000, 2010, and 2019) and Digital Shoreline Analysis System (DSAS) along the coastal of Kuchaveli DS division (East Coast of Sri Lanka) occurred between 1991 and 2019. Landsat images were geometrically and radiometrically corrected for the quantitative coastline delineation analysis. DSAS (Digital Shoreline Analysis System) was used as a reliable statistical approach for the rate of coastline change. The rate of shoreline change was demarcated based on End Point Rate (EPR) and Linear Regression Rate of change (LRR). As a result of the analysis, in some parts of the research area, remarkable shoreline changes (1.151m/yr withdrawal and -0.800 m/yr erosion) were observed for four periods (1991, 2000, 2010, and 2019). This study provides a synoptic outlook of the degree of potential threat to the coastal system, to prioritize actions, and to develop suitable adaptation measures.
Keywords: Shoreline change, DSAS, Erosion, Accretion, Geospatial technique

Coastal zones consist of essential ecosystem services and ecological productivity, which have traditionally been the source of wealth for many communities involved in the fishing and aquaculture industry. The beaches were great places for leisure and recreation. Since then, demand for coastal tourism increased that led to intense exploitation of the coast (Rodríguez et al., 2009). Detection of shoreline changes in time is essential for monitoring and management of coastal zones. Li et al. (2001) Defined shoreline as the line of contact between land and a body of water (Aedla et al., 2015). Further, Genz et al. (2007) defined shoreline as the position of the land-water interface at one point in time (Hakkou et al., 2018). The area, such as Geographical exploration, coastal erosion monitoring, coastal resource management, seeks essential information about coastal line position, orientation, and geometric shapes (Liu & Jezek, 2004). Shoreline change is considered as one of the most dynamic processes in the coastal area (Castelle et al., 2018), and observed changes indicate either coastal erosion or accretion. Shoreline changes occur over a wide range of time scales from geological to short-lived extreme events (Natesan et al., 2015). The shift in shoreline is mainly associated with waves, tides, winds, periodic storms, sea-level change, and the geomorphic processes of erosion and accretion and human activities (Salghuna & Bharathvaj, 2015). Shoreline also depicts the recent formations and destructions that have happened along the shore. Waves change the coastline morphology and form the distinctive coastal landforms (Salghuna & Bharathvaj, 2015). Marfai et al. (2008) mention that mapping shoreline changes became important as input data for coastal hazard assessment (Kuleli et al., 2011). Shoreline position changes have become, in recent years, one of the significant environmental problems affecting the coastal zones worldwide. Indeed, nearly 80% of the world's coasts are eroding, with rates ranging from 1 cm/year to 10 m/year (Pilkey & Hume, 2001). The spatial and temporal analysis of shoreline changes has been subject of several studies worldwide (Abu Zed et al., 2018; Aedla et al., 2015; Alesheikh et al., 2007; Castelle et al., 2018; Ford, 2013; Gopikrishna & Deo, 2018; Guneroglu, 2015; Hakkou et al., 2018; Harley et al., 2019; Kaliraj et al., 2017; Kankara et al., 2015; Kermani et al., 2016; Klein & Lichter, 2006; Kuleli et al., 2011; Li & Damen, 2010; Liu & Jezek, 2004; Maiti & Bhattacharya, 2009; Misra & Balaji, 2015; Moussaid et al., 2015; Ozturk & Sesli, 2015; Pilkey & Hume, 2001; Ramírez-Cuesta et al., 2016; Rodríguez et al., 2009; Salghuna & Bharathvaj, 2015; Shetty et al., 2015; Toure et al., 2019). 

Coastal areas continually modify due to a wide variety of phenomena, such as sea level variation, storm surge, tidal inundations, tectonic and land subsidence, sediment budget change, and human activities that play fundamental roles in coastal erosion (Hakkou et al., 2018). Coastal erosion is often influencing hundreds of kilometers of shoreline and represents a severe socio-economic problem at both the local and regional level. An understanding of long-term shoreline changes is an essential step for the management of the coastal area (Aiello et al., 2013; Ozturk & Sesli, 2015). GIS and Remote sensing-based approaches provide a better platform for collecting shoreline positional information, which enables estimating shoreline changes. This study detects and analyses shoreline changes for a period of 30 years (1990–2020), using geographic information systems (GIS) and Digital Shoreline Analysis System (DSAS). The main objective of this paper is to detect and analyze the shoreline changes along the Kuchaveli DS Division coast between 1991 and 2019, in a view to identify and quantify the erosion and accretion areas. 
Study Area
The study area corresponds to the Eastern part coast of Sri Lanka. It lies between 8030'0"- 8056'0" N Latitude and 8102'0"- 81014'0" E Longitude. The total length of the coast understudy is 44.11 kilometers. The study area climate is characterized by an irregular sub-tropical humid with two distinct seasons: a cold and rainy winter and a hot and dry summer. However, the proximity effect of the sea confers to this area a temperate climate. Indeed, the summer season ranging from April to October, is characterized by relatively moderate temperatures 27.80C. The average annual rainfall is greater than 1631mm. The study area comprises of some settlements. Generally, fishing on the shore, aquacultural by the nearshore area, and agricultural cultivation in the land is being mainly adapted.
Figure 1: Location of the Study Area
Materials and Methodology
Orthorectified and geodetically accurate global land dataset of Landsat Multi-spectral Scanner (MSS), Thematic Mapper (TM) and Enhanced Thematic Mapper (ETM+) data have been widely used in coastal research and environmental studies for many years (Kuleli et al., 2011). multi-spectral features and easy availability make Landsat suitable for monitoring coastline change (Kuleli et al., 2011). In this study, TM (Thematic Mapper), ETM+ (Enhanced Thematic Mapper Plus), and OLI (Operational Land Imager) datasets acquired for the period 1991-2019 were used as primary data sources to demarcate the shoreline changes along Kuchaveli DS division coast. The transformation of the coast in the study area was analyzed for a period of 28 years (1991 to 2019), which is considered as medium-term analysis (Hegde & Akshaya, 2015). Satellite imageries from 1991-2019 were obtained from the Earth Explorer (USGS), and they are collected in multiple periods (1991,2000,2010,2019). 

The data used for the study were selected based on the criteria of Cloud cover less than 10%, the season in which the data was acquired. The datasets were obtained predominantly from June to October to receive cloud-free images. Data has been processed in Arc GIS software and projected in UTM projection with zone no 44N and WGS 1984 datum. In this study, shoreline changes in spatial and temporal aspects were analyzed using the integrated remote sensing, GIS, and DSAS technology. Further information about the specifications of satellite data used in the study is given in Table 1.

  Acq. date
  File Type
  Data Source

 Various techniques for coastline extraction and change detection from satellite imagery have been developed. Manual, write function memory insertion, image enhancement, multi-date data classification and comparison of two independent land cover classifications, density slice using single or multiple bands and multi-spectral classification, both supervised and unsupervised (e.g., PCA, Tasseled Cap) is the most common change detection techniques (Kuleli et al., 2011).

The Landsat satellite images obtained were taken as an input for spectral pre-processing. Using layer stacking method, the individual bands of the satellite data were converted into a False Colour Composite image. The geometric transformation was done using a first-order polynomial equation. The shorelines datasets from the multi-date satellite data from 1991 to 2019 were extracted using Arc GIS 10.4.1 software. 

The determination of a shoreline position in satellite data is a very subjective one. In the past, researchers had used various proxies for shoreline positions such as high tide line, high water line (HWL), wet-dry line, vegetation line, dune line, toe, or berm of the beach, cliff base or top, mean high water (MHW) line. 
After considering all these factors, high water line (HWL) mark i.e., the effective shoreline, is equivalent "wet/dry line" of the previous tide, which is clearly identifiable from all images is found most appropriate to monitor the changes (Ozturk & Sesli, 2015). High waterline boundary (HWL) was considered as the shoreline proxy and demarcated using Arc GIS 10.4.1 software.

The shoreline positions were compiled in ArcGIS 10.4.1 with five attribute fields, which include Object ID (a unique number assigned to each transect), shape (polyline), shape length, ID, date (original survey year) and uncertainty values. All different shoreline features were merged as a single feature on the attribute table, which enabled the multiple coastline files to be appended together into a single shapefile. Digital Shoreline Analysis System (DSAS) version 5.0, an extension to ArcGIS developed by the USGS, was used to calculate the shoreline change rate. 

The vector layers of shorelines were overlaid, and erosion/accretion rates were calculated at 50 m intervals along the Kuchaveli DS division coast using Digital Shoreline Analysis System (DSAS). Several statistical methods are used to calculate the shoreline change rates with the most commonly used being End Point Rate (EPR) calculations or Linear Regression Rate (LRR) of change statistic. LRR is the most widely applied statistical technique for expressing shoreline movement and estimating rates of change. EPR is calculated by dividing the distance of shoreline movement by the time elapsed between the earliest and latest measurements. The following equation is used for EPR calculation.
        EPR (m/y)
         Distance (A-B) in m
        The time between youngest and 
oldest shoreline
In the present study, shoreline changes were estimated using the Linear Regression Rate (LRR). The LRR can be determined by fitting a least-squares regression line to all shoreline points for a particular transect obtained from the analysis. In order to assess the shoreline trend, an on-shore baseline was created with a position of approximately 2 km distance behind the shorelines. Based on the baseline and shoreline, the erosion/accretion regimes were calculated for 50 m orthogonal transects along the coastline. The variation in beach widths (m) was calculated for each transects to obtain rates of shoreline change (m/y). Positive LRR values represent shoreline movement towards the sea (i.e., the rate of accretion), and negative values indicate movement towards the land (i.e., the rate of erosion). Transect shoreline intersections along this baseline are then used to calculate the rate of change statistics. Finally, the obtained erosion and accretion rates for the Kuchaveli DS division coast were divided into seven categories (Table 2) (Raj et al., 2019).
Shoreline Classification based on EPR and LRR
  Rate of Shoreline
Change (m/yr)
  Shoreline Classification
  Very High Erosion
  >-1 to <-2
  High Erosion
  >-1 to<0
  Moderate Erosion
  >0 to <1
  Moderate Accretion
  >1 to <2
  High Accretion
  Very High Accretion

 The methodology flowchart, adopted for the entire study is given in figure 1
          Process begins
         Satellite imageries (1991,2000,2010,2019)
        Shoreline extraction
        Geometric Correction
(Digital Shoreline Analysis System)
        Statistical Results
        Hazard line Demarcation
        Polynomial Georectification using ArcGIS 10.4.1
                  1991,2000,2010,2019: High Water Level Line
             Output: Transect 
Metadata file
             Shoreline change analysis
        DSAS workflow
1: Reference line input
2: Transect setting
3: Statistical calculation


 Result and Discussion 
The study of the shoreline change of the Kuchaveli DS division coast has been delineated from Landsat (30 m resolution) imageries of different years 1991,2000,2010, and 2019. 

Figure 2: shoreline change of Kuchaveli DS division coast (1991-2019)

 The coastal erosion and accretion were both found on the coast of the Kuchaveli DS division. The direction of shoreline change was quantified from erosion and accretion of the study area for a period of 28 years from 1991 to 2019 using Remote Sensing and GIS.

Figure 3: Detail shoreline change of Kuchaveli DS division coast (1991-2019)

 DSAS generated 799 transects for the coast of the Kuchaveli DS division (Figure 2). These transects oriented perpendiculars to the baseline at 50 m spacing along 44.11 km length of Thiriyai GN division coast to Periyakulam GN division coast. 
The transect map of LRR and EPR for the Kuchaveli DS division coast is shown in Figures 6, and This map depicts the pattern of erosion and accretion. The legends in the map are classified according to the range of LRR and EPR of erosion and accretion value, as shown in Figures 4 & 5.

Figure 4:EPR and LRR Maps of Kuchaveli DS Division Coast

 This range classification was carried out in ArcGIS 10.4.1 tool. The process of movements of sediments drives the significant shoreline change in different parts of the coastal area due to the action of physical parameters. The coast has been affected by sea waves, winds, sea currents, tidal activities, etc. Results of the analysis of Landsat satellite data and DSAS are discussed in detail as follows.

An overall average of 1.151m/yr (LRR) accretion and about -0.800 m/yr (LRR) erosion was noticed along the Kuchaveli DS division coast. Maximum accretion/erosion rates of 6.350 and -5.360 m/yr are observed along the Kuchaveli DS division coast based on EPR. LRR shows 9.27 and -4.21 m/yr accretion and erosion rates on both the locations. Figure 4 & 5 depicts the rate of shoreline changes based on EPR and LRR. Eroding shorelines are observed at transect from 30-44, 56-78, 133-150, 211-224, 306-310, 509-519, 566-568, 575-580, 671-671, 695-697, 732-741, 762-763, and 773-775 (particularly coast of Thiriyai GN, Senthur GN, Jayanagar GN, Cassim Nagar GN, Kuchaveli GN, Kumpurupiddi North GN, Velor GN, and Iqbalnagar GN). In contrast, prograding shorelines are observed at transect form 1-9, 45-56, 101-104, 117-121, 162-170, 232-238, 243-250, 254-280, 292-300, 322-338, 344-351, 366-383, 403-455, 473-482, 521-536, 551-564, 582-657, 677-691, 700-728, 742-750, and 790-799 (particularly coast of Veerancholai GN, Iranaikerni GN, Kumpurupiddi East GN, Irakkanday GN, Valaiyoothu GN, Kopalapuram GN, Nilaveli GN, Periyakulam GN).

Figure 5:Detail EPR and LRR Maps of Kuchaveli DS Division Coast

 Transect from 01 to 230 predominantly falls under the erosion category. Very high erosion rate is noticed at transect 42 (Thiriyai GN ) at the rate of -4.15 (LRR) and -5.19 (EPR) m/yr whereas very high accretion rate is observed at transect 631 (Valaiyoothu GN) at the rate of 
9.27 (LRR) and 4.63 (EPR) m/yr  
Figure 6: Transect IDs of Kuchaveli DS Division Coast
The area of accretion is dominant than the rate of erosion. The negative value in the table indicates erosion, whereas positive value represents accretion. The table shows the details of EPR and LRR with an uncertainty of (+/-) 4.4m, a default setting of DSAS. 

Details of EPR and LRR
  No of transect
  Mean rate
  Standard Deviation
  Highest accretion
  Highest erosion
Figure 7: Rate of Shoreline Changes (1991-2019)

Remote sensing and geospatial techniques together with DSAS, will be useful for shoreline change monitoring and provide a comprehensive view of erosion and accretion patterns of the coastal areas. This study focuses on shoreline change detection analysis from 1991 to 2019 using Landsat-satellite imageries, and the results are encouraging. Maximum accretion/erosion rates of 6.350, -5.360 m/yr and 9.270, -4.210 m/yr are observed along the Kuchaveli DS division coast based on EPR and LRR respectively. Even though a large part of the study area show accretion, eroding shorelines are found in Thiriyai GN, Senthur GN, Jayanagar GN, Cassim Nagar GN, Kuchaveli GN, Kumpurupiddi North GN, Velor GN, and Iqbalnagar GN divisions. Continuous monitoring of the shoreline is essential for the coastal areas to observe the changes in the future. The results of this study can be useful for the stakeholders, policymakers, coastal managers, scientists, and coastal livelihoods to sustainably manage the Kuchaveli DS division coast to protect its natural integrity and coastal resources.


Post a Comment

Blog Archive

Definition List

Unordered List