Project description

Coastal zone management is an important aspect of EU environmental policy and improved means of assessing the environmental quality of coastal waters would facilitate the implementation, management and assessment of EU measures aimed at protecting the marine environment.
Although submerged vegetation plays a major key role in coastal ecosystems, and acts as an indicator of their environmental quality, adequate methods for surveying and quantifying submerged vegetation on a large scale are lacking. Remote sensing has proved efficient for mapping terrestrial vegetation. Adaptation of the method to the mapping of coastal submerged vegetation would provide a useful means of assessing the environmental quality of coastal waters as a tool for coastal zone management.

Look for more details concerning:

The challenges of remote sensing in the coastal zone
Ecology of submerged vegetation in coastal areas
Remote sensing of submerged vegetation as a monitoring tool for coastal zone management

The development of methods has been based on investigations in 3 main test areas characterized by different habitats and vegetation types.

 The main test areas are:

 "Bay of Løgten"

 "Inlet of Norsminde"

 "Øresund"

The main product is an operational and widely applicable remote sensing system for the assessment of coastal water environmental quality based on the optimal detection and mapping of submerged vegetation in coastal waters from aerial photographs, airborne scanner data and/or satellite images. The remote sensing system is accompanied by a guiding system for advising on selection of input data and analysis methods in order to tailor the system for the assessment of environmental quality in different coastal waters and for different environmental management purposes.

The system combines remotely sensed data with data sets from the Geographic Information System GIS (e.g. water depth, sediment quality) and background knowledge on vegetation characteristics, thereby considerably improving the interpretation of submerged vegetation data obtained by remote sensing.

The remote sensing system is applicable to different geographical areas by adjusting the background information.  The system can also be applied to the mapping of other sedentary coastal water biota, for example mussel beds.

The sub-products of RESCOMAN are:

Developed Methods 
The process of acquiring and analyzing the remotely sensed data has required the development of various procedures and tools. These methods comprise methods for acquisition and analysis of  background information,  planning and accomplishment of aerial photo surveys, geo-coding of aerial photography and airborne scanner data and analysis of  the acquired remote sensing data. The methods are described in the respective parts of the detailed guiding system. Some of the developed methods has also been implemented as software tools to be used in connection with image processing and GIS systems like ENVI, CHIPS for Windows and ArcView.  

Guiding system
A major product of this project is the development of a step by step guiding system allowing persons interested in eelgrass mapping but not necessarily experts in remote sensing to prepare and monitor a eelgrass mapping campaign. The guiding system has been intended to cover the full width of aspects in selection of sensor type, planning, accomplishment and analysis of imagery from a remote sensing campaign.

The guiding system is implemented as a hypertext document available on the internet at NERI, Department of Lake and Estuarine Ecology's website, allowing updating whenever new information and experience is available. 

Another important aspect of the hypertext structure is that it allows the user to select his own level of details from the brief overall description of necessary considerations down to the very technical and in-depth level mainly oriented towards skilled remote sensing experts.  

Information regarding eelgrass distribution and variability.

• Factors governing eelgrass distribution - evaluation of appropriate supporting parameters
• Sampling of vegetation for remote sensing analyses.

• Database structure for supporting parameters

The RESCOMAN project has gained experience in a number of directions. When considering mapping of underwater vegetation no doubt exist that both aerial photography and airborne scanner data has both their potential, but also limitations and difficulties in the analysis. 

Aerial photography has been used utilized for numerous years in mapping of both terrestrial and aquatic environment. The primary advantage of aerial photography, is that a high spatial resolution can be obtained without scarifying anything else than maybe the spatial coverage of each image. The potential for optimizing the procedures has however also been substantial, since the acquisition, digitizing, geo-coding and analysis of many individual can be rather labor intensive if dedicated system setup and automated procedures are not used. The potentially high spatial resolution is particular useful for studies of vegetation dynamics, but aerial photography has also been successfully implemented in major mapping projects. The major constraints are primarily in the image model, where the wide viewing angle in all directions lead to varying degree of "noise" from surface reflected light. Another limitation is in the ability to detect finer contrast-changes between vegetation and bottom at depth near and above the Secchi depth.

The future perspectives of aerial photography is in the direction of direct digital acquisition. This will lead to cheaper acquisition and a faster analysis of the data. The digital cameras record the images with a dynamic range of 10 bit per color, which is higher than the analogue to digital scanners currently used for converting films to digital images. The disadvantage is though that the geometric resolution is linked to a fixed image dimension and thus set at the time of acquisition.

The acquisition and analysis of airborne scanner data has in the project period been in a phase where the systems still have a character of being research systems rather than fully operational systems. This has influenced the project with respect to availability of scanners and which kind of processing levels the scanner operators could provide. The experience of the RESCOMAN project is that some of the problems for remote sensing over water related to the low radiance levels, potential significant noise and the reflection of the water can be handled in a reasonably manner by exploiting the options of selecting the multiple spectral channels in characteristic bands and adjusting the acquisition factors. Is this way it has also been possible to clearly distinguish vegetation features even at water depth close to Secchi depth. There is however still efforts to be done in order to fully understand and model how the transfer of radiance in the case of multi-layered media with varying water depth and bottom characteristics in order to directly use spectral signatures for the mapping of bottom features.

Regarding the data acquisition and availability, the trend is however very much in favor of new customers. Systems become more abundant and reliable. The scanner systems are gradually improving with respect to the number of spectral bands which can be acquired at one time, and the efficiency of the detector to provide a low signal to noise ratio. Integrated with attitude sensors and differential GPS systems this will lead to faster delivery of well calibrated data ready to be analyzed. The potential of the airborne scanners and especially imaging spectrometers with a high number of spectral channels, lies in the fact that spectral signal from the water potentially carry specific information of both the water composition and the bottom features being observed.