Open Access Repository

Object-based random forest classification of Landsat ETM+ and WorldView2 satellite imagery for mapping lowland native grassland communities in Tasmania, Australia

Melville, B ORCID: 0000-0003-3040-6071, Lucieer, A ORCID: 0000-0002-9468-4516 and Aryal, J ORCID: 0000-0002-4875-2127 2018 , 'Object-based random forest classification of Landsat ETM+ and WorldView2 satellite imagery for mapping lowland native grassland communities in Tasmania, Australia' , International Journal of Applied Earth Observation and Geoinformation, vol. 66 , pp. 46-55 , doi: 10.1016/j.jag.2017.11.006 Rece.

Full text not available from this repository.


This paper presents a random forest classification approach for identifying and mapping three types of lowland native grassland communities found in the Tasmanian Midlands region. Due to the high conservation priority assigned to these communities, there has been an increasing need to identify appropriate datasets that can be used to derive accurate and frequently updateable maps of community extent. Therefore, this paper proposes a method employing repeat classification and statistical significance testing as a means of identifying the most appropriate dataset for mapping these communities. Two datasets were acquired and analysed; a Landsat ETM+ scene, and a WorldView-2 scene, both from 2010. Training and validation data were randomly subset using a k-fold (k = 50) approach from a pre-existing field dataset. Poa labillardierei, Themeda triandra and lowland native grassland complex communities were identified in addition to dry woodland and agriculture. For each subset of randomly allocated points, a random forest model was trained based on each dataset, and then used to classify the corresponding imagery. Validation was performed using the reciprocal points from the independent subset that had not been used to train the model. Final training and classification accuracies were reported as per class means for each satellite dataset. Analysis of Variance (ANOVA) was undertaken to determine whether classification accuracy differed between the two datasets, as well as between classifications. Results showed mean class accuracies between 54% and 87%. Class accuracy only differed significantly between datasets for the dry woodland and Themeda grassland classes, with the WorldView-2 dataset showing higher mean classification accuracies. The results of this study indicate that remote sensing is a viable method for the identification of lowland native grassland communities in the Tasmanian Midlands, and that repeat classification and statistical significant testing can be used to identify optimal datasets for vegetation community mapping.

Item Type: Article
Authors/Creators:Melville, B and Lucieer, A and Aryal, J
Keywords: native grasslands, random forest, k-fold cross-validation, object-based image analysis
Journal or Publication Title: International Journal of Applied Earth Observation and Geoinformation
Publisher: Elsevier Bv
ISSN: 0303-2434
DOI / ID Number: 10.1016/j.jag.2017.11.006 Rece
Copyright Information:

© 2017 Elsevier B.V. All rights reserved.

Related URLs:
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page