Lars Brabyn

Modeling population access to New Zealand public hospitals

This paper demonstrates a method for estimating the geographical accessibility of public hospitals. Cost path analysis was used to determine the minimum travel time and distance to the closest hospital via a road network. This analysis was applied to 38,000 census enumeration district centroids in New Zealand allowing geographical access to be linked to local populations. Average time and distance statistics have been calculated for local populations by modeling the total travel of a population if everybody visited a hospital once. These types of statistics can be generated for different population groups and enable comparisons to be made between regions. This study has shown that the northern and southern parts of New Zealand have high average travel times to hospital services.

Spatial modelling of wetness for the Antarctic Dry Valleys

This paper describes a method used to model relative wetness for part of the Antarctic Dry Valleys using Geographic Information Systems (GIS) and remote sensing. The model produces a relative index of liquid water availability using variables that influence the volume and distribution of water. Remote sensing using Moderate Resolution Imaging Spectroradiometer (MODIS) images collected over four years is used to calculate an average index of snow cover and this is combined with other water sources such as glaciers and lakes. This water source model is then used to weight a hydrological flow accumulation model that uses slope derived from Light Detection and Ranging (LIDAR) elevation data. The resulting wetness index is validated using three-dimensional visualization and a comparison with a high-resolution Advanced Land Observing Satellite image that shows drainage channels. This research demonstrates that it is possible to produce a wetness model of Antarctica using data that are becoming widely available.

Landsat remote sensing of chlorophyll a concentrations in central North Island lakes of New Zealand

We investigated the use of Landsat Enhanced Thematic Mapper (ETM) imagery to synoptically quantify chlorophyll a (chl a) concentrations. Two adjoining pairs of images of the central North Island were acquired on two different days in summer and spring 2002. 6sv atmospheric correction was compared to the cosine of the solar zenith angle correction (COST) dark object subtraction (DOS) atmospheric correction. The highest correlation between 6sv ln(Band 3) water surface reflectance and ln(chl a) was found in the 24 January 2002 image (r 2 = 0.954). 6sv atmospheric correction was preferable to COST-DOS as it gave more realistic reflectance values at a clear-water reference site and produced the highest correlation coefficient. The results from this investigation suggest that remote sensing provides a valuable tool to assess temporal and spatial distributions of chl a in unmonitored areas within lakes and that predictions may also be extended to unmonitored lakes within the domain of satellite image capture.

Classifying Landscape Character

Abstract Landscape classification is fundamental to landscape management and research because it provides a frame of reference for communication. This paper describes a method for classifying visual landscape character in New Zealand and builds on a previously developed method using Geographical Information Systems. Theoretical issues associated with classifying landscapes are discussed and incorporated into the method. The main components of the classification are landform, landcover, water, infrastructure, dominant landcover and water view. The resulting classification is hierarchical, with six levels of generalisation. A naturalness score is also built into the classification. Applications of the classification are briefly described. These included query and identification of similar landscapes, description and landscape inventory of regions, identifying rare and unique landscapes, and combining with visibility analysis to describe landscape experience on walking tracks. The New Zealand Landscape Classification is designed to complement other landscape information such as the location of historical, spiritual, cultural, and geological landscape features.

Population access to hospital emergency departments and the impacts of health reform in New Zealand.

In the current political climate of evidence-based research, GIS has emerged as a powerful research tool as it allows spatial and social health inequality to be explored efficiently. This article explores the impact health reforms had on geographical accessibility to hospital emergency department (ED) services in New Zealand from 1991 to 2001. Travel time was calculated using least-cost path analysis, which identified the shortest travel time from each census enumeration district through a road network to the nearest ED. This research found that the population further than 60 minutes from an ED has increased with some areas being affected more than others. Some of this increase is attributed to increases in population rather than the closing of hospitals. The findings will be discussed within the context of the health policy reform era and changes to health service provision.

Comparing Three GIS Techniques for Modelling Geographical Access to General Practitioners

Accessibility to health services is a social and economic issue that has many dimensions. This article compares three techniques for measuring geographical accessibility to general practitioners (family doctors). These techniques can be described as ratio, least-cost path, and allocation methods. The ratio method is the most common and simply measures the population-to-GP ratio for a given area. The least-cost path and allocation techniques rely on GIS network analysis and provide information on travel times and distances. The allocation analysis differs from least-cost path analyses because it considers the number of GPs available and this acts as a capacity constraint. The use of GIS network analysis can produce a number of different statistics on accessibility that can be visualized as a map. This article will argue that the use of a capacity constraint is important and that the population further than 30 minutes from a GP is an effective statistic for comparing the accessibility of different managemen...

Combining QuickBird, LiDAR, and GIS topography indices to identify a single native tree species in a complex landscape using an object-based classification approach

Abstract There are now a wide range of techniques that can be combined for image analysis. These include the use of object-based classifications rather than pixel-based classifiers, the use of LiDAR to determine vegetation height and vertical structure, as well terrain variables such as topographic wetness index and slope that can be calculated using GIS. This research investigates the benefits of combining these techniques to identify individual tree species. A QuickBird image and low point density LiDAR data for a coastal region in New Zealand was used to examine the possibility of mapping Pohutukawa trees which are regarded as an iconic tree in New Zealand. The study area included a mix of buildings and vegetation types. After image and LiDAR preparation, single tree objects were identified using a range of techniques including: a threshold of above ground height to eliminate ground based objects; Normalised Difference Vegetation Index and elevation difference between the first and last return of LiDAR data to distinguish vegetation from buildings; geometric information to separate clusters of trees from single trees, and treetop identification and region growing techniques to separate tree clusters into single tree crowns. Important feature variables were identified using Random Forest, and the Support Vector Machine provided the classification. The combined techniques using LiDAR and spectral data produced an overall accuracy of 85.4% (Kappa 80.6%). Classification using just the spectral data produced an overall accuracy of 75.8% (Kappa 67.8%). The research findings demonstrate how the combining of LiDAR and spectral data improves classification for Pohutukawa trees.

Alternative solutions for determining the spectral band weights for the subtractive resolution merge technique

Data fusion using subtractive resolution merge (SRM) is limited because it currently requires fixed spectral band weights predetermined for particular sensors. This is problematic because there is an increasing availability of new and emerging sensors that have no predetermined band weights. There is also a need for fusion between sensors, which potentially requires a large number of sensor combinations and band weight calculations. This article demonstrates how the least sum of minimum absolute deviation (LAD, least absolute deviation) and ordinary least squares (OLS) regressions can calculate band weights for application in the SRM technique using QuickBird satellite and Vexcel aerial images. Both methods were effective in improving image details. The results of LAD and OLS are shown using qualitative and quantitative metrics and through unsupervised classification of freshwater habitat. OLS and LAD produce similar results; however, OLS is computationally simpler and easier to automate. The ability of the user to calculate their own scene specific band weights eliminates the dependence on predetermined sensor band weights. This research concludes that OLS band weight calculations should be integrated into the SRM technique to diversify its application.

Mapping health events - a comparison of approaches

The Waikato Regional Public Health Unit (New Zealand) collects and maintains a database of notified health events that includes communicable diseases. This database can be used by health protection officers to identify problem areas that have unusually high incidences of events. Resources can then be targeted and officers can respond accordingly. However, for this data to be useful it needs to be represented as a rate of health events per 100,000 people. This paper investigates the use of Geographic Information Systems (GIS) to visualize and represent these data. Standard vector representations that use census area units are compared with raster representations that use focal neighbourhood functions. The two techniques produce rates of health events that differ by up to 100 per cent for a given location and scale. The vector technique is the standard approach and is easily understood. The raster technique enables more control over the scale of analysis and can provide a more spatially accurate representation of health events. If two techniques can produce such large differences in analysis results, it is important that health protection officers and epidemiologists give special consideration to the representation technique used.

The use of Geographical Information Systems for analysing and visualising biodiversity data

Technology plays a key role in innovative research and it is important that environmental scientists are aware of technologies that are having an impact on their discipline. This paper demonstrates the importance of Geographical Information Systems (GIS) for biodiversity management and research using three powerful analysis functions and a species presence/absence database. The three analysis functions are: data integration, terrain analysis and data visualisation. Data integration functions enable the environmental characteristics of a species location to be known and therefore provide information on species habitat. An important environmental determinant of species habitat is terrain and the identification of three terrain indices (micro, meso and macro) is demonstrated. GIS visualisation techniques make tabular data easier to understand, highlight general trends and can be a stimulus for research. This research highlights a need for coordination and leadership in biodiversity information management so that researchers can easily access this data.