Ecological Modelling 222 (2011) 1174–1184 Contents lists available at ScienceDirect Ecological Modelling journal homepage: www.elsevier.com/locate/ecolmodel Mapping eucalypt forest susceptible to dieback associated with bell miners (Manorina melanophys) using laser scanning, SPOT 5 and ancillary topographical data Andrew Haywood a,∗ , Christine Stone b a Victorian Department of Sustainability and Environment, PO Box 500, East Melbourne, Vic., 3002, Australia b Forest Resources Research, NSW Department of Primary Industries, PO Box 100, Beecroft, NSW, 2119, Australia article info Article history: Received 2 November 2009 Received in revised form 14 December 2010 Accepted 19 December 2010 Available online 14 January 2011 Keywords: Eucalypt forest dieback Bell miners Laser scanning SPOT 5 Spatial modelling Predictive mapping abstract Mapping the location and extent of forest at risk from damaging agents or processes assists forest managers in prioritizing their planning and operational mitigation activities. In Australia, Bell Miner Associated Dieback (BMAD) refers to a form of canopy decline observed in eucalypt crowns occupied by colonies of bell miners (Manorina melanophrys). High densities of bell miners are associated with decreased avian abundance and diversity and an increase in psyllid abundance on crown foliage. BMAD has recently been nominated as a key threatening process in New South Wales (NSW). Consequently, a modelling system for predicting bell miner distribution in coastal eucalypt forests of NSW has been developed. The presence or absence of bell miners was recorded in 130 plots located within a 12,800 ha catchment study area containing a range of eucalypt forest types. The modelling system was produced by integrating a machine learning software suite (WEKA), and the statistical software R within the geo- graphic resources analysis support system (GRASS) geographical information system (GIS). The variable modelled was the binary variable: presence or absence of bell minors. Six modelling techniques (Logis- tic regression; generalised additive models; two tree-based ensemble classification algorithms, random forest and Adaboost and Neural Networks) were integrated with airborne laser scanning; SPOT 5 and topo- graphic derived variables. Model evaluation and parameter selection were measured by three threshold dependent measures (sensitivity, specificity and kappa) and the threshold independent Receiver Opera- tor Curve (ROC) analysis. The final presence and absence maps were obtained through maximisation of the kappa statistic and applied at a resolution of 10 m across the entire catchment study area. For this data set, the most accurate algorithm for predicting the distribution of bell miner colonies was random forest (kappa = 0.84; ROC area under curve = 0.97). Variables most commonly selected in the six models were the laser scanning metrics; coefficient of variation, skewness, and the 10th and 90th percentiles derived from the shape of the height frequency distribution which, in turn, is directly influenced by ver- tical structure of the forest. An image textural statistic based on the shortwave infrared (SWIR) band of SPOT 5 was also commonly selected by the models. The SWIR band is sensitive to vegetation and soil moisture content. These models predicted that forest stands with a sparse eucalypt canopy over a moist, dense understorey were susceptible to being colonised by bell miners and hence BMAD. Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved. 1. Introduction Identifying the location and extent of forest at risk from dam- aging agents and processes assists forest managers in prioritizing their planning and operational mitigation activities. The risk of canopy damage is a function of the abundance of the damaging agent and the susceptibility of the forest stand to damage (Fettig et al., 2007). While assessment of population levels will continue ∗ Corresponding author. Tel.: +61 3 9637 9680. E-mail address: Andrew.Haywood@dse.vic.gov.au (A. Haywood). to rely on ground based diagnosis and survey of the causal agents, numerous studies have demonstrated the capacity of remotely acquired image data to assess both the physiological condition (e.g. Blackburn, 2003; Sampson et al., 2003; Suárez et al., 2008) and structure of forest stands (e.g. Zimble et al., 2003; Maltamo et al., 2005; Næsset and Gobakken, 2005; Goodwin et al., 2006; Coops et al., 2007). For individual damaging agents, a preliminary task is to identify the relationships between stand factors associated with the presence of the damaging agent or process and metrics derived from remotely sensed and ancillary geographical information sys- tem (GIS) data (Coops et al., 2006). The second step is to develop a predictive model that most accurately maps the actual as well 0304-3800/$ – see front matter. Crown Copyright © 2011 Published by Elsevier B.V. All rights reserved. doi:10.1016/j.ecolmodel.2010.12.012