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.

[1]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[2]  Michael J. Falkowski,et al.  Classification of Landsat images based on spectral and topographic variables for land-cover change detection in Zagros forests , 2012 .

[3]  France Gerard,et al.  Exploring the Use of MODIS NDVI-Based Phenology Indicators for Classifying Forest General Habitat Categories , 2012, Remote. Sens..

[4]  Robert M. Hawlick Statistical and Structural Approaches to Texture , 1979 .

[5]  Mario Chica-Olmo,et al.  An assessment of the effectiveness of a random forest classifier for land-cover classification , 2012 .

[6]  P. Caccetta,et al.  Mapping forest cover, Kimberley Region of Western Australia , 2001 .

[7]  John R. G. Townshend,et al.  Strategies for monitoring tropical deforestation using satellite data , 2000 .

[8]  Peter Scarth,et al.  Prediction and validation of foliage projective cover from Landsat-5 TM and Landsat-7 ETM+ imagery , 2009 .

[9]  Sandra Eckert,et al.  Improved Forest Biomass and Carbon Estimations Using Texture Measures from WorldView-2 Satellite Data , 2012, Remote. Sens..

[10]  D. Stow,et al.  THE EFFECT OF TRAINING STRATEGIES ON SUPERVISED CLASSIFICATION AT DIFFERENT SPATIAL RESOLUTIONS , 2002 .

[11]  David L.B. Jupp,et al.  Detecting Structural and Growth Changes in Woodlands and Forests: The Challenge for Remote Sensing and the Role of Geometric-Optical Modelling , 1997 .

[12]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[13]  Johannes R. Sveinsson,et al.  Random Forests for land cover classification , 2006, Pattern Recognit. Lett..

[14]  Volker C. Radeloff,et al.  The Impact of Phenological Variation on Texture Measures of Remotely Sensed Imagery , 2009, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[15]  M. Austin,et al.  Current approaches to modelling the environmental niche of eucalypts: implication for management of forest biodiversity , 1996 .

[16]  G. Moisen,et al.  PresenceAbsence: An R Package for Presence Absence Analysis , 2008 .

[17]  Jonathan Cheung-Wai Chan,et al.  Evaluation of random forest and adaboost tree-based ensemble classification and spectral band selection for ecotope mapping using airborne hyperspectral imagery , 2008 .

[18]  C. Howell,et al.  Sustainable forest management reporting in Australia , 2008 .

[19]  G. Green,et al.  Deforestation History of the Eastern Rain Forests of Madagascar from Satellite Images , 1990, Science.

[20]  F. M. Danson,et al.  Satellite remote sensing of forest resources: three decades of research development , 2005 .

[21]  Markus Neteler,et al.  Calculating landscape diversity with information-theory based indices: A GRASS GIS solution , 2013, Ecol. Informatics.

[22]  K. Shadan,et al.  Available online: , 2012 .

[23]  Patrick Hostert,et al.  Evaluating the Remote Sensing and Inventory-Based Estimation of Biomass in the Western Carpathians , 2011, Remote. Sens..

[24]  Fabio Maselli,et al.  Use of MODIS NDVI data to improve forest-area estimation , 2011 .

[25]  Maxwell R. Jacobs,et al.  Growth habits of the Eucalypts. , 1955 .

[26]  D. Roberts,et al.  A comparison of methods for monitoring multitemporal vegetation change using Thematic Mapper imagery , 2002 .

[27]  T. Tsegaye,et al.  Incorporation of digital elevation models with Landsat-TM data to improve land cover classification accuracy , 2000 .

[28]  P. Atkinson,et al.  Incorporating Spatial Variability Measures in Land-cover Classification using Random Forest , 2011 .

[29]  Mariela Soto-Berelov,et al.  Creating a large area landcover dataset for public land monitoring and reporting , 2013 .

[30]  F. Deppe,et al.  Forest Area Estimation Using Sample Surveys and Landsat MSS and TM Data , 1998 .

[31]  Víctor Urrea,et al.  Letter to the Editor: Stability of Random Forest importance measures , 2011, Briefings Bioinform..

[32]  Paul E. Gessler,et al.  Integrating Landsat TM and SRTM-DEM derived variables with decision trees for habitat classification and change detection in complex neotropical environments , 2008 .

[33]  Antoine Guisan,et al.  Predictive habitat distribution models in ecology , 2000 .

[34]  D. Lu Aboveground biomass estimation using Landsat TM data in the Brazilian Amazon , 2005 .

[35]  P. Defourny,et al.  Retrieving forest structure variables based on image texture analysis and IKONOS-2 imagery , 2006 .

[36]  G. Shao,et al.  On the accuracy of landscape pattern analysis using remote sensing data , 2008, Landscape Ecology.

[37]  L. Beaumont,et al.  Predicting species distributions: use of climatic parameters in BIOCLIM and its impact on predictions of species’ current and future distributions , 2005 .

[38]  Joanne C. White,et al.  Multiscale satellite and spatial information and analysis framework in support of a large-area forest monitoring and inventory update , 2010, Environmental monitoring and assessment.

[39]  D. R. Cutler,et al.  Utah State University From the SelectedWorks of , 2017 .

[40]  R.M. Haralick,et al.  Statistical and structural approaches to texture , 1979, Proceedings of the IEEE.

[41]  Ronald E. McRoberts,et al.  Probability- and model-based approaches to inference for proportion forest using satellite imagery as ancillary data , 2010 .

[42]  Neil Flood,et al.  An Operational Scheme for Deriving Standardised Surface Reflectance from Landsat TM/ETM+ and SPOT HRG Imagery for Eastern Australia , 2013, Remote. Sens..

[43]  J. Franklin Predictive vegetation mapping: geographic modelling of biospatial patterns in relation to environmental gradients , 1995 .

[44]  Craig A. Coburn,et al.  A multiscale texture analysis procedure for improved forest stand classification , 2004 .

[45]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[46]  Santosh Bhandari,et al.  Monitoring Forest Dynamics using Time Series of Satellite Image Data in Queensland, Australia , 2011 .

[47]  Simon Ferrier,et al.  Evaluating the predictive performance of habitat models developed using logistic regression , 2000 .

[48]  R. Reynolds,et al.  A non-parametric, supervised classification of vegetation types on the Kaibab National Forest using decision trees , 2003 .

[49]  Nicholas C. Coops,et al.  Landscape Controls on Structural Variation in Eucalypt Vegetation Communities: Woronora Plateau, Australia , 2011 .