William De Genst

HIGH-RESOLUTION VEGETATION DATA FOR MANGROVE RESEARCH AS OBTAINED FROM AERIAL PHOTOGRAPHY

In this methodological study, the applicability of aerial photographs for monitoring mangrove vegetation dynamics at high resolution was investigated. Vegetation maps of three mangrove forests in Sri Lanka (Galle, Rekawa and Pambala) were produced based on visual analysis of aerial photographs. The visual analysis was aided by applying an interpretation key constructed during a first fieldwork mission. Image attributes used for the identification of individual trees included: gray values, texture, form and size of the crowns and the presence or absence of a shaded side. For the identification of species assemblages, the vegetation structure (i.e. the distribution of individual trees) appeared to be an important attribute. The accuracy and reliability of the vegetation maps were investigated during a second fieldwork mission. The aerial photographs proved to be very useful for the production of genus-based vegetation maps. The error analysis showed that density estimations (quantitative identification) based on aerial photography was not sufficiently accurate for the objectives of the study, but that the overall identification of vegetation assemblages (qualitative identification) coincided most satisfactory with the ground-truth data. In addition to the applicability of aerial photography in monitoring mangroves, the importance of aerial photography in the management of the mangrove ecosystem is clearly highlighted.

Assessing effects of input uncertainty in structural landscape classification

This paper presents the results of a study aimed at assessing the effects of input uncertainty on the outcome of a raster-based model for structural landscape classification. The model uses a DEM and a land-cover map as input, and calculates four structural indices from these data. The first two indices determine the openness of the landscape, the other two determine the degree of landscape homogeneity. By combining both aspects, nine different landscape types are defined. Applying Monte Carlo simulation, the effect of DEM error, uncertainty in land-cover classification, and the combined effect of both sources of uncertainty on the outcome of the landscape model are assessed. Special attention is paid to the spatial structure of uncertainty in both data sources.

Uncertainty Modeling in Buffer Operations Applied to Connectivity Analysis

In this paper we will study the potential connectivity of red squirrels in a fragmented landscape, using a buffer operation that takes into account the difficulty of moving through the landscape. The outcome of such an analysis is greatly influenced by the various sources of uncertainty that are introduced in the model. Two main sources of uncertainty can be identified: source layer uncertainty and model uncertainty. In this paper the propagation of source layer uncertainty resulting from a multivariate statistical classification of remotely sensed data is studied using Monte Carlo simulation, taking the spatial structure of uncertainty into account. Model uncertainty results from the adoption of deterministic model parameters regarding the dispersal capacity and the landscape effect, and is examined using fuzzy set theory. Comparing the outcome of error sensitized models to the observed dispersal activity of squirrels, demonstrates how modeling of uncertainty can help to explain the dispersal activity of red squirrels.

The influence of transformations of an ordinal scale of a floristic gradient, applied on a TWINSPAN classification

Summary TWINSPAN is a widely applied method for hierarchical classification. Particular options such as the minimum group size, the maximum split levels and both the amount and score of cut levels to be used are left to the judgement of the researcher. When using small and homogeneous data sets, the score assigned to the cut levels can be derived from intermediate transformations of the cover-abundance values, rather than transformations obtained with a strong weighting on presence, or an emphasis on dominance. Our research question is to know whether these coefficients have the same robustness for more heterogeneous and larger data sets, with a large amount of transition communities. A comparison was made between five different sets of cut levels in order to determine the most effective one for the classification of data on a floristic gradient. The analysed cut level sets differed in the number of cut levels and in the weight provided to the cover-abundance coefficients. The classifications resulting from the different sets of cut levels were compared, by determining the amount of identical associations that were clustered together, by visualising the results of the different cut level types through DCA, by comparing the correlations between the median values of the abiotic factors in each TWINSPAN group and DCA-values of the first axis of the different cut level types. The analysis showed that the cut level types in particular derived from the original values of the Braun-Blanquet cover-abundance coefficients, reflect best the floristic gradient and the underlying structure of the data. Results on heterogeneous data sets are complementary to the previous findings in literature where better results were achieved with a combined transformation for more homogeneous data sets.

Development and Implementation of a Procedure for Automated Map Projection Selection

The increased accessibility of geographic information systems (GIS) and mapping software to inexperienced users brings with it the risk that the suitability of a map may be diminished through the choice of an inadequate map projection. Clearly some guidelines on choosing the right map projection are necessary in order to assist the user to portray data in such a way as not to hamper the message the map is meant to transmit. A review of some of the better known existing selection schemes for small-scale map projections emphasizes the need for the development of a procedure for automated map projection selection. This article presents a procedure for an application-dependent choice of map projection suitable for implementation in a digital environment. Selection is based on the specification of map projection features relevant to the application. Optimization of the graticule of the selected projection guarantees minimum visual distortion.

Interactivity with the urban information users in the land-use/land-cover information extraction from VHR data

Very High Resolution (VHR) satellite imagery offers a great potential for extracting land-use and land-cover related information for urban areas, but do they meet the requirements of present day urban planners? Assessing user needs for urban land use/land cover data, and investigating the potential of VHR data to better meet these needs is therefore essential. These two parts lead to an interactive definition of remote sensing products in Belgium. This paper presents the background of our analysis (previous surveys at European and French level), the methods that we use to assess the urban users needs (questionnaire and survey), how these can be met by VHR data (classification results) and some preliminary results of the Belgian survey obtained for both the Walloon and Brussels region. Among these results, the survey reports the preference on ortho-rectified aerial photographs when this product is available, a scarce use of remote sensing data explained by spatial resolution and cost reasons, and the lack of awareness of the new VHR images capabilities. As results for the ongoing survey become complete, we hope to better understand what data products derived from VHR imagery can potentially be of interest to users of LU/LC data in Belgium. This will enable us to propose image processing methods that better fulfil the needs of local and regional authorities in Belgium.

Development of a technique to decompose heterogeneous mapping units in a categorical map of the biotic environment using correlated categorical data

In this article a technique is presented to estimate the proportions of different map categories in a series of heterogeneous mapping units, using information on the degree of spatial correlation with other categorical data. The technique has been applied to decompose ecotope complexes in a categorical map of the biotic environment in Flanders, using secondary information on land cover and soil type. Because the conditional probability of an ecotope occurring given a certain soil type depends on the frequency with which the ecotope occurs in an area, determining the probability of occurrence of an ecotope from the conditional probabilities can lead to predictions that contradict prior knowledge about the composition of the different mapping units. A measure expressing the affinity of an ecotope for a soil type is proposed and is used as an alternative to conditional probability in the estimation procedure. The proposed method has been tested in a study area for which detailed field observations were colle...