Using Machine Learning to Map Western Australian Landscapes

Landscapes evolve due to climatic conditions, tectonic activity, geological features, biological activity, and sedimentary dynamics. These processes link geological processes at depth to surface features. Consequently, the study of landscapes can reveal essential information about the geochemical footprint of ore deposits at depth. Advances in satellite imaging and computing power have enabled the creation of large geospatial datasets, the sheer size of which necessitates automated processing. This study describes a methodology to enable the automated mapping of landscape pattern domains using machine learning (ML) algorithms. From a freely available Digital Elevation Model, derived data, and sample landclass boundaries provided by domain experts, our algorithm produces a dense map of the model region in Western Australia. Both random forest and support vector machine classification achieve about 98% classification accuracy with reasonable runtime of 48 minutes on a single core. We discuss computational resources and study the effect of grid resolution. Larger tiles result in a more contiguous map, while smaller tiles result in a more detailed, and at some point, noisy map. Diversity and distribution of landscapes mapped in this study support previous results. In addition, our results are consistent with the geological trends and main basement features in the region.

[1]  Chih-Jen Lin,et al.  LIBSVM: A library for support vector machines , 2011, TIST.

[2]  B. Pillans.  Old, flat and red-Australia's distinctive landscape , 2012 .

[3]  John F. Roddick,et al.  Geographic Data Mining and Knowledge Discovery , 2001 .

[4]  I. González-Álvarez,et al.  Mineral exploration in regolith-dominated terrains: Global considerations and challenges , 2016 .

[5]  T. Stepinski,et al.  Landscape similarity, retrieval, and machine mapping of physiographic units , 2014 .

[7]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[8]  M. Nilsson,et al.  Using machine learning to generate high-resolution wet area maps for planning forest management: A study in a boreal forest landscape , 2019, Ambio.

[9]  Jürgen Böhner,et al.  Land-Surface Parameters Specific to Topo-Climatology , 2009 .

[10]  G. Prasicek,et al.  GIS Applications in Geomorphology , 2018 .

[11]  R. Anand,et al.  A guide for mineral exploration through the regolith in the Yilgarn Craton, Western Australia , 2010 .

[12]  J. Gallant,et al.  A multiresolution index of valley bottom flatness for mapping depositional areas , 2003 .

[13]  Tomasz F. Stepinski,et al.  Multi-scale segmentation algorithm for pattern-based partitioning of large categorical rasters , 2018, Comput. Geosci..

[14]  Eric J. Gustafson,et al.  How has the state-of-the-art for quantification of landscape pattern advanced in the twenty-first century? , 2018, Landscape Ecology.

[15]  M. Summerfield.  Neotectonics and landform genesis , 1987 .

[16]  R. Anand,et al.  Evolution of regoliths and landscapes in deeply weathered terrain — implications for geochemical exploration , 2000 .

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

[18]  Timothy A. Warner,et al.  Implementation of machine-learning classification in remote sensing: an applied review , 2018 .

[19]  I. González-Álvarez,et al.  Sea-level changes and buried islands in a complex coastal palaeolandscape in the South of Western Australia: Implications for greenfield mineral exploration , 2016 .

[20]  Chih-Jen Lin,et al.  LIBLINEAR: A Library for Large Linear Classification , 2008, J. Mach. Learn. Res..

[21]  Peter Vogt,et al.  Describing and analyzing landscape patterns: where are we now, and where are we going? , 2019, Landscape Ecology.

[22]  Thomas Blaschke,et al.  Understanding and quantifying landscape structure – A review on relevant process characteristics, data models and landscape metrics , 2015 .

[23]  Amy E. Frazier,et al.  Landscape Metrics: Past Progress and Future Directions , 2017, Current Landscape Ecology Reports.

[24]  C. Ollier.  Evolution of the Australian landscape , 2001 .

[25]  R. Craig,et al.  Applied Geomorphology , 2020 .

[26]  E. Grunsky,et al.  Recognition of geochemical footprints of mineral systems in the regolith at regional to continental scales , 2017 .

[27]  C. Twidale.  Ancient Australian Landscapes , 2007 .

[28]  M. Lewis,et al.  Objective Regolith-Landform Mapping in a Regolith Dominated Terrain to Inform Mineral Exploration , 2018, Geosciences.

[29]  J. Mercader,et al.  Archaeological Site Formation in Rain Forests: Insights From the Ituri Rock Shelters, Congo , 2003 .

[30]  Xin Du,et al.  The Combined Use of Remote Sensing and Social Sensing Data in Fine-Grained Urban Land Use Mapping: A Case Study in Beijing, China , 2017, Remote. Sens..

[31]  Tomasz F. Stepinski,et al.  GeoPAT: A toolbox for pattern-based information retrieval from large geospatial databases , 2015, Comput. Geosci..

[32]  D. E. Pagendam,et al.  A regolith depth map of the Australian continent , 2016 .

[33]  Jiawei Han,et al.  Geographic Data Mining and Knowledge Discovery , 2001 .

[34]  Abdulhakim M. Abdi,et al.  Land cover and land use classification performance of machine learning algorithms in a boreal landscape using Sentinel-2 data , 2019, GIScience & Remote Sensing.

[35]  Olaf Conrad,et al.  Large-scale atmospheric forcing and topographic modification of precipitation rates over High Asia – a neural-network-based approach , 2014 .

[36]  Max Kuhn,et al.  Applied Predictive Modeling , 2013 .