Mapping the Expansion of Boom Crops in Mainland Southeast Asia Using Dense Time Stacks of Landsat Data

Abstract

Abstract:We performed a multi-date composite change detection technique using a dense-time stack of Landsat data to map land-use and land-cover change (LCLUC) in Mainland Southeast Asia (MSEA) with a focus on the expansion of boom crops, primarily tree crops. The supervised classification was performed using Support Vector Machines (SVM), which are supervised non-parametric statistical learning techniques. To select the most suitable SMV classifier and the related parameter settings, we used the training data and performed a two-dimensional grid search with a three-fold internal cross-validation. We worked in seven Landsat footprints and found the linear kernel to be the most suitable for all footprints, but the most suitable regularization parameter C varied across the footprints. We distinguished a total of 41 LCLUCs (13 to 31 classes per footprint) in very dynamic and heterogeneous landscapes. The approach proved useful for distinguishing subtle changes over time and to map a variety of land covers, tree crops, and transformations as long as sufficient training points could be collected for each class. While to date, this approach has only been applied to mapping urban extent and expansion, this study shows that it is also useful for mapping change in rural settings, especially when images from phenologically relevant acquisition dates are included.

The expansion of tree-based boom crops in mainland Southeast Asia: 2001 to 2014

“…We classified the MODIS time-series following an 'upscaling' approach and extracted training 9 and verification points from seven Landsat-based land-cover change classifications in the region (Hurni et al 2017). Due to the differences in pixel sizes, this operation is usually performed by defining a threshold of a majority cover of the 30m resolution land-cover classes within a MODIS pixel (circa 231m resolution), extracting the MODIS pixels that meet the threshold criteria, and classifying the MODIS data with these samples (Clark et al, 2010;DeFries, Townshend, & Hansen, 1999;Hurni et al, 2013).…”

“…We needed to adjust for this imbalance of sample points across the classes. Various scholars have shown that the appropriate amount of sample points for an accurate supervised classification of land-cover change does not only depend on the class area (spatial heterogeneity), but also on the temporal heterogeneity and the classification scheme (spectral-temporal similarity of classes) (Foody & Arora, 1997;Heydari & Mountrakis, 2018;Hurni et al, 2017). To represent temporal heterogeneity, change classes consequently need relatively more sample points despite the small areas they cover.…”

Accelerating forest loss in Southeast Asian Massif in the 21st century: A case study in Nan Province, Thailand

“…In addition to the expansion of tree crops (particularly rubber) in SAM (Hurni, Schneider, & Heinimann, 2017), many local studies have also reported substantial nontree crop (particularly corn) expansion throughout SAM in the 21st century (e.g. Schmidt-Vogt et al, 2009;Viau, Keophosay, & Castella, 2009;Castella, Jobard, Lestrelin, Nanthavong, & Lienhard, 2012;Heinimann et al, 2013;Kitchaicharoen et al, 2015;Pongkijvorasin & Teerasuwannajak, 2015).…”

Political transition and emergent forest‐conservation issues in Myanmar

“…Other crops, including cashew, coffee, and pineapple, are grown successfully in mainland Southeast Asia (Hurni et al. ) and could expand into Myanmar. Future agricultural commodity prices, economic liberalization, and suitable climates will determine which crops will expand from 2016 to 2026 and affect large areas of remaining forest.…”

“…Although southern Myanmar does not have an optimal climate for oil-palm cultivation (Saxon & Sheppard 2014), it has become a significant driver of deforestation there (Donald et al 2015). Other crops, including cashew, coffee, and pineapple, are grown successfully in mainland Southeast Asia (Hurni et al 2017) and could expand into Myanmar. Future agricultural commodity prices, economic liberalization, and suitable climates will determine which crops will expand from 2016 to 2026 and affect large areas of remaining forest.…”