Digimap Case Study:
Measurement of Historic Coastal Erosion Rates using multi-temporal data
Author
Katja Haberland
Title
Measurement of Historic Coastal Erosion Rates using multi-temporal data
Date
October 2007 – April 2008
Subject Area
Geomatics, GIS/Remote Sending, Long-term Monitoring
Relevance to other subject areas
Geography
Project Type
Undergraduate dissertation
Summary
The project aims to measure coastal erosion rates at Marsden Bay, South Tyneside, Tyne and Wear, UK. The methodology has been chosen on the basis of previous studies in this or a similar study area. Marsden Bay consists of Magnesian Limestone which is prone to erosion, especially on parts of the shore which are westwards-exposed. Mechanical wave erosion, wind and physical erosion are the main creators of the bizarre and unique shoreline. The focus is on estimating the coastal erosion rate by applying a remote sensing and GIS approach. The information on the shoreline was extracted from aerial imagery and historic map data and brought into a specially-designed GIS environment, which allows measurement of multi-temporal data simultaneously. Two different types of erosion rates were obtained, but the result used here, a major erosion rate of approximately 15 cm per year, underpins the general estimation for Marsden Bay.
Datasets used
Historic Map Series: County and National Grid Series
Edition: First Revision, First Revision, Third Revision, National Grid (NatSurvey)
Scale: 1:2,500
Names of other data used: Aerial photography (from South Tyneside Council) for the period of 1999-2005
Source: Historic Digimap
Aims and Objectives
The main aim was to find out which changes can be recorded using different types of data - aerial imagery and historical maps. Additionally the quality of the maps and images was estimated due to the fact that these accuracy values are needed for the later calculations.
Methodology
After the data was collected it had to be processed and prepared for further analyses. The processing consisted of merging, georeferencing and digitising (digitising requires a huge amount of time and patience!).After these steps were successfully finished the digitised vector data (polyline) was brought into the GIS (ESRI ArcGIS 9.2) with a special extension called Digital Shoreline Analysis System (DSAS, provided by the USGS). The shorelines from all years were appended to one layer and a baseline, separately created and stored, and transects were calculated. The spacing and length of them need to be manually specified depending on the characteristics of the site. Due to the fact that the images and maps had been evaluated previously the accuracy was used in the calculations using linear weighted regression, i.e. less accurate data get a lower weight than high accuracy data with a low RMSE).
Results/Outcome
The results can be variously interpreted statistically, but the main differences of interest are those between oldest/youngest and closest/furthest shorelines. From these results annual average rates can be obtained. From these a maximum erosion rate of approximately 15 cm per year has been calculated.
Additional Information
This graph shows the quality of the maps, taking control points from Digimap (Mastermap) to calculate the deviation from the real coordinates. The trend line shows the obvious trend that the older data is the less accurate it is (compared to present aerial photography), although the map from 1943 was very accurate.
There are major not correctable gaps and shifts. Re-georeferencing could help. Besides, there was a huge gap within one tile so that this map could not be used.
Reference and/or acknowledgements
Thanks to my supervisor for any assistance, motivation and encouragement. Thanks also to the South Tyneside Council which supplied aerial imagery from 1999-2005.
EDINA is a JISC National Data Centre based at the University of Edinburgh
