Integrating Phenological and Geographical Information with Artificial Intelligence Algorithm to Map Rubber Plantations in Xishuangbanna

Abstract

Most natural rubber trees (Hevea brasiliensis) are grown on plantations, making rubber an important industrial crop. Rubber plantations are also an important source of household income for over 20 million people. The accurate mapping of rubber plantations is important for both local governments and the global market. Remote sensing has been a widely used approach for mapping rubber plantations, typically using optical remote sensing data obtained at the regional scale. Improving the efficiency and accuracy of rubber plantation maps has become a research hotspot in rubber-related literature. To improve the classification efficiency, researchers have combined the phenology, geography, and texture of rubber trees with spectral information. Among these, there are three main classifiers: maximum likelihood, QUEST decision tree, and random forest methods. However, until now, no comparative studies have been conducted for the above three classifiers. Therefore, in this study, we evaluated the mapping accuracy based on these three classifiers, using four kinds of data input: Landsat spectral information, phenology–Landsat spectral information, topography–Landsat spectral information, and phenology–topography–Landsat spectral information. We found that the random forest method had the highest mapping accuracy when compared with the maximum likelihood and QUEST decision tree methods. We also found that adding either phenology or topography could improve the mapping accuracy for rubber plantations. When either phenology or topography were added as parameters within the random forest method, the kappa coefficient increased by 5.5% and 6.2%, respectively, compared to the kappa coefficient for the baseline Landsat spectral band data input. The highest accuracy was obtained from the addition of both phenology–topography–Landsat spectral bands to the random forest method, achieving a kappa coefficient of 97%. We therefore mapped rubber plantations in Xishuangbanna using the random forest method, with the addition of phenology and topography information from 1990–2020. Our results demonstrated the usefulness of integrating phenology and topography for mapping rubber plantations. The machine learning approach showed great potential for accurate regional mapping, particularly by incorporating plant habitat and ecological information. We found that during 1990–2020, the total area of rubber plantations had expanded to over three times their former area, while natural forests had lost 17.2% of their former area.

New high-resolution maps show that rubber causes significant deforestation

“…This is highly relevant to rubber as oil palm has expanded into traditional rubber growing areas 39 , with new compensatory rubber plantations being established elsewhere, e.g., in uplands 18,30 and often climatically marginal areas 16 , where they may be associated with deforestation. Thus, while the use of extrapolation 13,14 and modelbased 11,12 approaches provide some form of estimation for the extent of deforestation due to rubber plantations, we advocate caution in their interpretation. Instead, where available, we argue for the use of results from direct observations of the dynamics of crop production systems (e.g., using remotely-sensed satellite imagery), thereby greatly increasing the accuracy of deforestation estimates.…”

“…The current estimates for deforestation caused by rubber 3,4 used for policy considerations in the EU 7 and UK 6 are based on a land-balance model 11,12 . Such models typically allocate total deforestation area to different commodities based on national (or sub-national, e.g.…”

“…For the defoliation stage, we generated a composite image using 15% NDVI percentile of all images acquired during January-February in 2021 and 2022 for region-A, and during June- To run the machine learning classification, individual samples are required for each classification category. In this study, we collected a total of 2,010 rubber points and 1,816 evergreen forest points from ground-truthed points, publications 11,12 and random sampling points, which we visually interpreted using high-resolution satellite imagery through the software Collect Earth Online [13][14][15] (CEO). For the latter, we randomly sampled 1,000 forest points and 1,000 rubber points from ground-truth data collected in 2010 (World Agroforestry Centre Southeast Asia) and visually interpretated and subsequently re-labelled these points for the year 2021 through CEO and Google Earth 16 .…”

“…However, with relatively few exceptions -most notably oil palm and soy 1,10 -directly observed global or regional maps derived from satellite imagery are unavailable for most commodities. Instead, commodity-specific global deforestation is typically estimated using models 11,12 and extrapolations 13,14 with large levels of uncertainty..…”

Artificial Intelligence-Based Online Education System for University Music

“…rough the timeliness and locality of Internet technology, develop and build an online music education platform, provide learning resources, promote students' autonomous learning ability, and improve the quality of education to a certain extent [3,4].…”