Alumni meeting

Towards a framework to characterise shoreface behaviour; A study of a barrier island’s shoreface slope dynamics

Organised by Laboratory of Geo-information Science and Remote Sensing

Tue 6 June 2017 09:30 to 10:00

Venue Gaia, gebouwnummer 101
Room 1

By Samantha Krawczyk (Poland)

The shoreface is a part of the coastal zone which, from a hazard mitigation point of view, plays a crit-ical role in protecting the mainland. The run-up along the shoreface helps buffer the power of the waves, reducing their height and energy (Cowell 2000, Masselink, Hughes et al. 2011). However, the shoreface is subject to change in response to external forces such as changes in sea levels, changes in sediment availability (both natural and anthropogenic, e.g. sand nourishments or extraction), and storm events. Studying the shoreface behaviour is important to understand its response to these forces to improve our ability to model and predict them, and, ultimately, better protect the main-land.

To study the behaviour of the shoreface spatially, GIS analysis can be employed. Although coastal dynamics on small spatial (i.e. cm to m) and time (seconds to months) scales are relatively well stud-ied, due to the difficulties in obtaining data, as well as the difficulties in upscaling coastal processes, shoreface behaviour at larger spatial and temporal scales is less understood. Currently, studies on these scales are based on sparse measured bathymetric data usually supplemented with data ob-tained through remote sensing or sediment analysis, or on models. However, whatever the data source, when designing a GIS analysis of the shoreface, concepts in the field of coastal morphody-namics, GIS, and time series analysis, have to be considered as they will shape and affect the results of the analysis as well as their interpretation. A framework for the design of such studies would help structure the analysis at different scales and provide a frame of reference for issues likely affecting the analysis.

Developing such a framework is beyond the scope of this study, since this would require detailed re-search into the sensitivity of the various methods and parameters for different scenarios. The objec-tive of this study was to provide a first step towards developing such a framework, by developing a method to analyse the long-term (decadal) behaviour of the shoreface slope of a Dutch barrier island (Ameland) using the JARKUS data (bathymetry measurements collected along cross-shore transects since 1963). The method was developed based on an initial discussion of key morphodynamics, GIS and time series analysis concepts applicable to the focus of this case study. The case study area, and the focus on the long-term behaviour of the shoreface slope, was chosen to continue a series of stud-ies on the impact of coastal management on natural coastal development carried out at Wageningen University & Research.

The study addressed the following research questions:

  • What are the main concepts from a coastal morphodynamics perspective, GIS and time se-ries analysis perspective that should be considered when preparing an analysis of the shoreface slope?
  • How can these concepts be used to devise a methodology for studying the shoreface slope of Ameland using the JARKUS data?
  • What can be learned from the implementation of the methodology about the original con-cepts?

The processes affecting the shoreface are complex and act at different spatial and temporal scales. As they are hard to upscale, the scale of the data used must reflect the scale of the process being investigated. Methods for generalizing data in both the spatial and temporal dimensions are investi-gated (for raster representations of the terrain). Additionally, when looking at devising a GIS method for the analysis, consideration should be given to the different interpretations of change in GIS and statistics, as well as the impact of different methods for deriving Digital Bathymetry Models (DBMs) and DBM derivatives.

Based on the discussion above, two methods were developed for identifying shoreface dynamics that take into account the coastal morphodynamics, GIS concepts and time series analysis. They differ in their approach to the analysis of the shoreface slope. The first focuses on the analysis of the change in geometry of contour lines to detect changes in the slope gradient, while the second one focuses on the change of the slope gradient attribute to study the variety and magnitude of change at given locations.

The methods were applied to a study of the shoreface of Ameland using the JARKUS data. The im-plementation of the methods showed the limitations introduced by the data and the study area. The data is collected on an annual basis and thus better suited to studies of decadal (or longer) processes, while the seaward extent of the data mainly covered the part of the shoreface which is affected by processes acting on shorter timescales. However, the results of both methods clearly indicated the known patterns of slope behaviour (a dynamic shoreface around the island’s inlets, and more stable shoreface towards the centre of the island).

The concepts in the areas of coastal morphodynamics, GIS and time series analysis were useful in the design of the methods, and enable a methodical comparison of the results. However, to facilitate future research into the shoreface and interpretation of the results, the concepts have to be further examined and their interaction analysed, before a valid framework will become available. Specifical-ly, further research is needed into the issue of dealing with missing data both in the spatial and tem-poral domain. In the spatial domain research could be conducted into deriving bathymetry data from sources such as photos and satellite images, and combining these with existing point measurements. Additionally, the current research focused only on the analysis of the shoreface slope, however re-search is needed into other slope characteristics such as gradient and aspect and how they can pro-vide an insight into the shoreface behaviour. Finally, the current discussion of concepts could be ex-panded to include a more in depth discussion of time series analysis to facilitate the discovery of pat-terns in the temporal domain.