Classification of shallow sea beds - a pilot study east of Gräsö
Background and objectives
With respect to the Swedish environmental goals, consideration should be given, in connection with fishing, shipping and other uses of seas and water areas, as well as construction and other development in coastal and archipelago areas, to the productive capacity, biological diversity, natural and cultural assets and outdoor recreation assets of the water areas. These considerations require knowledge and information about the status of the subsurface environment and the need for mapping of these areas has increased. During many years, information about these areas has been collected through field investigations. These field surveys generate point wise data with high level of detail, but only for a very limited area. Satellite data that covers larger areas could be a good complement to these field investigations and a support for planing activities.
The general goal during 2006 and 2007 for this project is to investigate the possibilities to develop a method for classification and mapping of hard- and soft bottoms and different vegetation types in shallow coastal areas. The purpose is to derive information that will facilitate the environmental work and decision making for a number of authorities. The analysis will be based on high resolution QuickBird data collected in the coastal area outside Arkö-Gränsö in the Municipality of Norrköping.
The collected QuickBird data are presently being evaluated using field data collected in the end of August 2006 and additional field data from the archive. The dataset collected during 2006 consists of approximately 120 visited locations with information about substrate, vetetation type and cover and depth. A new set of images and additional field data will be collected during 2007. In general, the developments will be focused on analysis of the spectral information representing different objects. We will also make use of the advantage of having two to three images from different seasons, i.e. different vegetations status. Methods for sun glint and depth correction can be found in the research literature and have been/will be implemented and tested.
If the results turn out to be useful with respect to the end user needs, the produced maps will serve as an important basis in the environmental work for the local authorities. They will be used for planning activities in order to find interesting field locations that can be further investigated in field surveys. Repeated analysis will also make it possible to identify areas that have changed between years. This will increase the knowledge of the annual and seasonal fluctuations and if the observed status is an ordinary or unusual state. There is of course a significant cost involved in using this type of imagery and the necessary image analysis work and it will be important to clearly define the cost-benefit situation.