No abstract

T o date there has been little Australian research on repeat victimisation. This is a study of repeat burglary in an area of Brisbane using police calls for service data. We demonstrate: (a) the prevalence of residential repeat victim addresses ('hot dots') is of a similar magnitude to that found in studies in the United Kingdom; (b) the time distributions of revictimisation are identical with those found in studies in the UK and elsewhere; (c) 'hot spots' (small areas with high crime density) can be identified by statistical analyses of spatial concentrations of incidents; (d) unstable hot spots tend to be temporary aggregations of hot dots, whereas stable hot spots seem to reflect more the social and physical characteristics of certain localities; and (e) the overall incidence of burglary could be reduced by at least 25 per cent if all repeat victimisation could be eliminated. There are a number of areas where concepts and techniques for repeat victim research could potentially be strengthened: (a) clarifying the connections between hot dots and hot spots, particularly through exploration of the concept of a 'near repeat address'; (b) applying survival analysis to the data on the time periods between victimisations; and (c) using moving average techniques to examine changes in the spatial distributions of burglary over time.Burglary has been a growing problem, both in Queensland and in Australia as a whole, for the past 20 years, although police statistics and crime victim surveys indicate that the Queensland rate is close to the national average (Criminal Justice Commission, 1996). However, there are indications that the rate may have peaked, with recent marked declines in the burglary rate in Queensland (Queensland Police

Changes in rainfall pattern have been suggested as a mechanism for the landward incursion of mangrove into salt marsh. The aim of the research was to assess the relationship between rainfall patterns and the spatial distribution of mangrove forests at study sites in Moreton Bay, southeast Queensland, Australia, over a 32-year period from 1972 to 2004. To identify periods of relatively consistent rainfall patterns points at which rainfall patterns changed (change-points) were identified using the non-parametric Pettitt-Mann-Whitney-Statistic and the cumulative sum technique. The change-points were then used to define the temporal periods over which changes to mangrove area were assessed. Both mangrove and salt marsh area were measured by digitizing aerial photographs acquired in 1972, 1990 (the year with the most significant change-point), and 2004.The rates of change in mangrove area pre-1990 (a wetter period) and post-1990 (a drier period) were estimated. A significant positive relationship was demonstrated between rainfall variables and landward mangrove expansion, but not for seaward expansion. We concluded that rainfall variability is one of the principal factors influencing the rate of upslope encroachment of mangrove. However, the rate of expansion may vary from site to site due to site-specific geomorphological and hydrological characteristics and the level of disturbance in the catchment.

No abstract

Land use/cover and mangrove spatial changes were assessed for ten sites and their sub-catchments in Southeast Queensland, Australia. Two time periods were involved: 1972 to 1990, a period of relatively high rainfall, and 1990 to 2004, which was significantly drier. Aerial photographs and Landsat satellite imagery were used to map the inter-tidal wetlands and classify the land use/cover in the sub-catchments. A Maximum Likelihood Classification was used to map three types of land cover: agriculture, built-up and plantation forest. Mangroves (mainly Avicennia marina) were the focus as they have been recorded over recent decades encroaching into salt marsh. The Mangrove-Salt marsh Interface (MSI) Index was developed to quantify the relative opportunity for mangroves to expand into salt marshes, based on the shared boundary between them. The index showed a consistent relationship with mangrove expansion and change. To address problems of high dimensionality and multi-collinearity of predictor variables, a Partial Least Squares Regression (PLSR) model was used. A key finding of this research was that the contribution of environmental variables to spatial changes in the mangroves was altered following a reduction in rainfall. For example, agriculture had more influence on mangrove expansion and change during the wet period than during the dry period.