Andrew Mellor

The Performance of Random Forests in an Operational Setting for Large Area Sclerophyll Forest Classification

Mapping and monitoring forest extent is a common requirement of regional forest inventories and public land natural resource management, including in Australia. The state of Victoria, Australia, has approximately 7.2 million hectares of mostly forested public land, comprising ecosystems that present a diverse range of forest structures, composition and condition. In this paper, we evaluate the performance of the Random Forest (RF) classifier, an ensemble learning algorithm that has recently shown promise using multi-spectral satellite sensor imagery for large area feature classification. The RF algorithm was applied using selected Landsat Thematic Mapper (TM) imagery metrics and auxiliary terrain and climatic variables, while the reference data was manually extracted from systematically distributed plots of sample aerial photography and used for training (75%) and accuracy (25%) assessment. The RF algorithm yielded an overall accuracy of 96% and a Kappa statistic of 0.91 (confidence interval (CI) 0.909–0.919) for the forest/non-forest classification model, given a Kappa maximised binary threshold value of 0.5. The area under the receiver operating characteristic plot produced a score of 0.91, also indicating high model performance. The framework described in this study contributes to the operational deployment of a robust, but affordable, program, able to collate and process large volumes of multi-sourced data using open-source software for the production of consistent and accurate forest cover maps across the full spectrum of Victorian sclerophyll forest types.

Modelling vulnerability to gambling related harm: How disadvantage predicts gambling losses

Electronic gambling machines (EGMs) are ubiquitous in social venues such as hotels and clubs in most Australian states, and account for 55% of total gambling expenditure in Australia; they are also associated with most gambling derived harm. Because of the difficulty of assessing the prevalence of problem gambling and the incidence of gambling derived harms, gambling expenditure (i.e., the losses of those gambling) is often used in gambling research as a proxy indicator of harm. This study examines the relationship between socioeconomic disadvantage (measured by the Australian Bureau of Statistics SEIFA Index of Relative Socioeconomic Disadvantage [IRSD]), and EGM losses at the suburban level across a major Australian city. It develops a predictive spatial model of gambling vulnerability and presents the output visually. The findings reveal increasing levels of loss as disadvantage increases across IRSD quintiles. The highest mean annual EGM losses of

Automated In-Situ Laser Scanner for Monitoring Forest Leaf Area Index

849 per adult (95% CI

Understanding the Effects of ALS Pulse Density for Metric Retrieval across Diverse Forest Types

AU749–963) occurred in areas classified in IRSD Quintile 1, the most disadvantaged areas; in the least disadvantaged areas, mean annual losses were

Using discrete-return airborne laser scanning to quantify number of canopy strata across diverse forest types

298 per adult (CI

Comparison of relative radiometric normalization methods using pseudo-invariant features for change detection studies in rural and urban landscapes

260 –

Using ensemble margin to explore issues of training data imbalance and mislabeling on large area land cover classification

342). The density of EGMs confounds the relationship between losses and disadvantage. In this model, 40% of the apparent effect of disadvantage is explained by the density of EGMs. The vulnerability surface reflects socioeconomic patterns across Melbourne. EGM vulnerability is clustered (Morans Index 0.52; p < 0.001). High levels of EGM density in disadvantaged areas are contributing to a disproportionate share of EGM losses in already disadvantaged neighbourhoods. Regulation of EGMs could be improved to better protect vulnerable neighbourhoods from EGM harm.

Monitoring Victoria’s public forests: implementation of the Victorian Forest Monitoring Program

An automated laser rangefinding instrument was developed to characterize overstorey and understorey vegetation dynamics over time. Design criteria were based on information needs within the statewide forest monitoring program in Victoria, Australia. The ground-based monitoring instrument captures the key vegetation structural information needed to overcome ambiguity in the estimation of forest Leaf Area Index (LAI) from satellite sensors. The scanning lidar instrument was developed primarily from low cost, commercially accessible components. While the 635 nm wavelength lidar is not ideally suited to vegetation studies, there was an acceptable trade-off between cost and performance. Tests demonstrated reliable range estimates to live foliage up to a distance of 60 m during night-time operation. Given the instruments scan angle of 57.5 degrees zenith, the instrument is an effective tool for monitoring LAI in forest canopies up to a height of 30 m. An 18 month field trial of three co-located instruments showed consistent seasonal trends and mean LAI of between 1.32 to 1.56 and a temporal LAI variation of 8 to 17% relative to the mean.

Forest attribution using K-NN methods with Landsat 8 imagery and forest field plots

Pulse density, the number of laser pulses that intercept a surface per unit area, is a key consideration when acquiring an Airborne Laser Scanning (ALS) dataset. This study compares area-based vegetation structure metrics derived from multireturn ALS simulated at six pulse densities (0.05 to 4 pl m-2) across a range of forest types: from savannah woodlands to dense rainforests. Results suggest that accurate measurement of structure metrics (canopy height, canopy cover, and vertical canopy structure) can be achieved with a pulse density of 0.5 pl m-2 across all forest types when compared to a dataset of 10 pl m-2. For pulse densities <0.5 pl m-2, two main sources of error lead to inaccuracies in estimation: the poor identification of the ground surface and sparse vegetation cover leading to under sampling of the canopy profile. This analysis provides useful information for land managers determining capture specifications for large-area ALS acquisitions.

Woody vegetation landscape feature generation from multispectral and LiDAR data (A CRCSI 2.07 woody attribution paper)

The vertical arrangement of forest canopies is a key descriptor of canopy structure, a driver of ecosystem function and indicative of forest successional stage. Yet techniques to attribute for canopy vertical structure across large and potentially heterogeneously forested areas remain elusive. This study introduces a new technique to estimate the Number of Strata (NoS) that comprise a canopy profile, using discrete-return Airborne Laser Scanning (ALS) data. Vertically resolved gap probability (P-gap) aggregated over a plot is generalized with a nonparametric cubic spline regression (P-s). Subsequently a count of the positive zero-crossings of second derivative of 1 - P-s is used to estimate NoS. Comparison with inventory derived estimates at 24 plots across three diverse study areas shows a good agreement between the two techniques (RMSE=041 strata). Furthermore, this is achieved without altering model parameters, indicating the transferability of the technique across diverse forest types. NoS values ranged from 0 to 4 at a further 239 plots, emphasizing the need for a method to quantify canopy vertical structure across forested landscapes. Comparison of NoS with other commonly derived ALS descriptors of canopy structure (canopy height, canopy cover and return height coefficient of determination) returned only a moderate correlation (r(2)<04). It is proposed the presented method provides a primary descriptor of canopy structure to complement canopy height and cover, as well as a candidate Ecological Biodiversity Variable for characterizing habitat structure.