Predicting land cover using GIS, Bayesian and evolutionary algorithm methods

Modelling land cover change from existing land cover maps is a vital requirement for anyone wishing to understand how the landscape may change in the future. In order to test any land cover change model, existing data must be used. However, often it is not known which data should be applied to the problem, or whether relationships exist within and between complex datasets. Here we have developed and tested a model that applied evolutionary processes to Bayesian networks. The model was developed and tested on a dataset containing land cover information and environmental data, in order to show that decisions about which datasets should be used could be made automatically. Bayesian networks are amenable to evolutionary methods as they can be easily described using a binary string to which crossover and mutation operations can be applied. The method, developed to allow comparison with standard Bayesian network development software, was proved capable of carrying out a rapid and effective search of the space of possible networks in order to find an optimal or near-optimal solution for the selection of datasets that have causal links with one another. Comparison of land cover mapping in the North-East of Scotland was made with a commercial Bayesian software package, with the evolutionary method being shown to provide greater flexibility in its ability to adapt to incorporate/utilise available evidence/knowledge and develop effective and accurate network structures, at the cost of requiring additional computer programming skills. The dataset used to develop the models included GIS-based data taken from the Land Cover for Scotland 1988 (LCS88), Land Capability for Forestry (LCF), Land Capability for Agriculture (LCA), the soil map of Scotland and additional climatic variables.     

Improving the effectiveness of angular dispersion in plant neighbourhood models

Spatial arrangement can be an important factor affecting competition among plants. We evaluated three ways to improve the effectiveness of angular dispersion (AD) for describing spatial arrangement in plant neighbourhood models. First, we modified Zar's (1974) AD formula by weighting each neighbour by its competitive influence. We calculated this using two different competition indices to derive an AD of competitive influence, rather than of equally weighted plant locations, around a subject plant. Secondly, we constrained the effect of AD on the neighbourhood model using an optimised parameter that defines the minimum value that AD can adopt. Thirdly, we included the direction in which competition is concentrated (the mean azimuth of the weighted AD) in the growth models. These developments were evaluated within a radial growth model of Scots pine and birch growing in semi-natural, spatially heterogeneous forest. Weighted AD resulted in significant improvements in predicted radial growth of target trees over the traditional measure of AD. The optimised parameter that defines the minimum value of AD consistently evolved values significantly higher than zero. This suggests that clumped and dispersed neighbourhoods do not differ in their negative effects on a subject tree as much as expected. The inclusion of directional components of the weighted AD did not improve the accuracy of the growth models. Weighting of the angular dispersion of neighbours improved the performance of local competition models.     

Remote sensing-based neural network mapping of tsunami damage in Aceh, Indonesia

In addition to the loss of human life, the tsunami event of 26 December 2004 caused extensive damage to coastal areas. The scale of the disaster was such that remote sensing may be the only way to determine its effects on the landscape. This paper presents the results of a neural network-based mapping of part of the region of Aceh, Sumatra. Before-and-after satellite imagery, combined with a novel neural network methodology, enabled a characterisation of landscape change. The neural network technique used a threshold of acceptance for identification, in combination with a bootstrapped identification method for identifying problem pixels. Map analysis allowed identification of urban areas that were inaccessible by road, and which aid agencies could therefore only reach by air or sea. The methods used provide a rapid and effective mapping ability and would be a useful tool for aid agencies, insurance underwriters and environmental monitoring.     

Predicting Sample Source Location from Soil Analysis Using Neural Networks

A system combining a national soils database with a neural network was developed for prediction of source location for soil samples. The neural network was trained to predict environmental characteristics, which can be of crucial importance to investigating officers in a police operation or to environmental agencies attempting to locate the source of a pollutant. When coupled with maps of environmental conditions and a generalized opinion pool approach, the system was used to produce weighted maps of source location. The system was capable of reducing search areas of a sample source to less than 0.1% of the total area.     

Sustainable local land use policy: rhetoric and reality

Abstract

This study examined trends in population redistribution and residential land use changes in northeast Scotland during 1988 to 2003. We utilised a geographical information system (GIS) tool to bring together data from interrelated sources and to analyse population settlement and land use changes at detailed spatial scale. This analysis revealed that substantial land conversion had taken place in the region; particularly conversion of agricultural lands to built-up and residential areas. It also drew attention to policy conflicts, discrepancies between policy rhetoric and policy implementations. More specifically, it was shown that substantial farm land conversion had taken place in suburban areas and rural Aberdeenshire with little change in the size of derelict land in Aberdeen City. The reason given for this was that it was costly to rehabilitate and reuse derelict lands and this conflicted with the objective of providing affordable housing in the City region. The implication of this study is that if the declared policy objectives of integrated and sustainable land use and transport policies are to be achieved, then local authorities may need to cooperate to minimise conflicts between economic and environmental objectives.