Novel techniques for coastal monitoring

Abstract

There are currently three active projects under development for the Delfland coast, all aiming at reinforcing the coast. The main issue is where to put the sand such that it arrives at the coast in the best place, without causing any damage to e.g. the new nature development. To answer this question, knowledge about sand transport is necessary, and for this purpose Boskalis is designing a measurement campaign within the framework of the so-called Building with Nature project. This synthesis project contributes to this plan by investigating the capabilities of different measurement techniques for usage in coastal monitoring. Using these techniques, the team has gathered information about the coastal topography, bathymetry and grain size distribution of the coast near Monster-Ter Heijde, from the sub-aqueous zone to the first dunes behind the beach. This area is considered a hot spot on the Dutch shore because of its relative vulnerability to a break of the coastal defence. Five acquisition methods have been used in this project: • Photogrammetry • Terrestrial laser scanning • Single beam echo sounding • Radioactivity probing • GPS Photogrammetry has been used to measure the topography, using two types of platforms, kite aerial photogrammetry and close range photogrammetry. The captured images from the kite aerial photogrammetry have been processed using specific photogrammetric software, leading to two results, an orthophoto and a digital terrain model, both with a resolution of 30 by 30 cm. These first results of the kite aerial photogrammetry are promising, but the achieved accuracy is not high enough yet for precise monitoring. Therefore both the acquisition and the rocessing method should be improved. For the acquisition method, the stability of the platform should be improved and bigger ground control points should be used. For the processing, more professional software, such as Leica Photogrammetry Suite, should be used. Another use of the acquired photograph which has been demonstrated in this project is for land cover classification, by means of a simple classification based on the RGB value of each pixel. Although it has not been validated, this first result is promising and indicates future potential. Another method to measure the topography is terrestrial laser scanning, using two different types of laser scanner, namely the FARO LS880 laser scanner and the Leica ScanStation, which are a phase and pulse based scanner, respectively. The FARO laser scanner has been found to have some restrictions for the purpose of this project, since the dataset contained too much noise to easily extract topography. A night time test scan resulted in data with less noise, which indicates that sunlight could be a factor contributing to the noise. With the restricted time available, further processing of the Faro scanner data was not pursued. Instead, the results of the Leica ScanStation 2 showed to be useful and a digital terrain model of a part of the survey area has been produced. Having problems with the data acquisition, at this point it is not possible to have a fair validation of the results of terrestrial laser scanning, therefore further study is recommended. Next to the topography, it is also found useful to gather the grain size distribution information over the entire area, which has been done with the Medusa radioactivity probing method. The measurements have been done on both the onshore and offshore part of the area. Due to lack of correlation between the grain size and the radiation properties of the samples, it was decided to work with the K (potassium) concentration that has a 1 : 1 negative correlation with the grain size. As the result, K-concentration maps for the offshore and onshore survey area have been produced. The result of the offshore area shows that the sediment in general becomes finer with increasing distance from the coastline. In all cases, more local variations in the K-concentration were found as well, which should be further investigated. On the onshore part, the tidal area shows a relatively low K-concentration, the beach has a rather constant K-concentration, while behind the dunes the K-concentration is relatively high. The single beam echo sounding method has been used to measure the near shore bathymethry of the area, with a jet ski used as surveying platform. As a result, a bathymetric map has been produced. In this map, interesting features are visible, such as the groynes and their scour holes. The sounding measurements have been found to have a standard deviation of 14.5 cm. The opening angle of the SBES transducer and the platform attitude contribute significantly to the standard deviation, mainly due to the fact that the jet ski is not equipped with an attitude or orientation sensor. To relate the standard deviation of the SBES measurements to several parameters (e.g. sailing speed, sailing direction, etc.), a different dataset is needed, i.e. one in which the same area is sailed many times and where only one parameter is changed at a time. These results show that the jet ski single beam echo sounding method is nearly operational. That being said, some improvements in terms of platform and surveying is still needed to achieve a higher quality dataset. All of the aforementioned methods have used GPS for georeferencing purposes, either by measuring control points or by measuring positions of the moving platforms. Another use of GPS in this project is to measure cross-shore sections of the land area which remain wet during low tide and thus are inaccessible to the photogrammetry and terrestrial laser scanning methods. Next to the individual results, data merging has been done for the onshore and offshore datasets. For the onshore datasets, the results of terrestrial laser scanning and kite aerial photogrammetry have been merged into a single digital terrain model of the beach. The result shows that the photogrammetry datasets has distortions near the edges of the photoblock, but it also shows that the differences between the photogrammetry, terrestrial laser scanning and the GPS measurements are small enough. However for a statement on the actual accuracy, a validation with more GPS data is needed. From this result it can be recommended that when a large area of beach is to be measured, the area near the edges of the photogrammetry block can also be measured with terrestrial laser scanning to eliminate distortions. Further, a validation of both onshore datasets has been done by comparison with an existing airborne laser scanning dataset of the area. The comparison shows that the differences of both datasets with the airborne laser scanning dataset show the same deformation. It is important to note that there is no overlap between the onshore and offshore datasets. During the project it was found that the jet ski was unable take valid measurements past the low water line, which means there is no sounding data available for the intertidal area. Therefore, several other acquisition techniques has been recommended to measure the intertidal area, such as videogrammetry. On the offshore side, the K-concentration map has been overlaid with the bathymetric map, which gives a good visualisation of the spread of the K-concentration and shows the correlation between the water depth and the grain size. The general trend is that the grain size decreases with the water depth, however there is a sudden inverse relation visible between 6 m and 3 m water depth. One possible cause for this disturbance might be the sand bar present at this depth interval, which contains foreign sand as a result of earlier beach nourishment, that may have disturbed the natural gradient. The relation between grain size and topography has also been looked into, which showed a negative correlation. These hypotheses could be analyzed when a larger dataset is available in order to increase the certainty and make the correlation stronger. At the end of this synthesis project a strong basis has been initialized for monitoring plan of the coastal reinforcement projects. The suggested monitoring plan is presented for different zones and features of the coastal area where different changes of morphology are expected. The monitoring strategy takes into consideration the capabilities and specifications of the different acquisition techniques as investigated and recommended in this project. According to the parameters of each zone, the most suitable acquisition method is described, together with its spatial and temporal resolution, together with approaches for data analysis. While there is not a single optimum acquisition method to monitor the various events affecting the coastal morphology, it is possible to adapt different types of measurements plan to every event. Therefore the frequency and locations of measurements are relative to the different events taken into account, namely storm, (pre) beach nourishment, and annual changes.