Integrating remote sensing and machine learning into environmental monitoring and assessment of land use change

Abstract Addressing the increasing burden on land use requires effective policy for sustainable land use along with economic development. Analysis of local and global indicators based on land use maps could reveal information on the progress of sustainable development. This study proposes a method that reduces the time and cost of creating land use maps applicable for many purposes of environmental protection. Freely accessible existing data, Sentinel-2 satellite images, together with a machine learning algorithm, Random Forest, are integrated to generate an annual map, sufficient to meet the intended needs. The method is illustrated by a case study of Phuket in Thailand. An annual map for Phuket created using the proposed method was compared to the official map released by the Thai government for the year 2018. The two maps did not differ significantly, validating the efficacy of the proposed method. Annual maps were then produced for several years to assess the effect of land use change in the past 19 years on the environmental and sustainable management in Phuket. Although there was evidence of the efforts to develop Phuket island as a sustainable province such as the government policy to conserve green areas, land use change based analytical results indicated Phuket's urban development was not going in an environmentally sustainable direction.

[1]  John F. Mustard,et al.  LAND-USE AND LAND-COVER CHANGE PATHWAYS AND IMPACTS , 2004 .

[2]  David Lin,et al.  Decentralization & local food: Japan's regional Ecological Footprints indicate localized sustainability strategies , 2021 .

[3]  Impacts of interpolation techniques on groundwater potential modeling using GIS in Phuket Province, Thailand , 2012 .

[4]  Xiaoling Zhang,et al.  Production- and consumption-based convergence analyses of global CO2 emissions , 2020 .

[5]  Tim Appelhans,et al.  Improving the accuracy of rainfall rates from optical satellite sensors with machine learning — A random forests-based approach applied to MSG SEVIRI , 2014 .

[6]  Hanyue Chen,et al.  High-resolution mapping of forest canopy height using machine learning by coupling ICESat-2 LiDAR with Sentinel-1, Sentinel-2 and Landsat-8 data , 2020, Int. J. Appl. Earth Obs. Geoinformation.

[7]  H. Balzter,et al.  Tropical forest canopy height estimation from combined polarimetric SAR and LiDAR using machine-learning , 2021, ISPRS Journal of Photogrammetry and Remote Sensing.

[8]  C. Ranasinghe,et al.  Estimation of carbon sequestration potential in coconut plantations under different agro-ecological regions and land suitability classes , 2012 .

[9]  Hong Anh Thi Nguyen,et al.  Comparative carbon footprint assessment of agricultural and tourist locations in Thailand , 2020 .

[10]  M. Wackernagel,et al.  Our ecological footprint , 1996 .

[11]  J. Ometto,et al.  Land cover changes and greenhouse gas emissions in two different soil covers in the Brazilian Caatinga. , 2016, The Science of the total environment.

[12]  Yanbo Huang,et al.  Agricultural remote sensing big data: Management and applications , 2018, Journal of Integrative Agriculture.

[13]  André Stumpf,et al.  Combining Random Forests and object-oriented analysis for landslide mapping from very high resolution imagery , 2011 .

[14]  Changhui Peng,et al.  Quantitative assessment of the importance of bio-physical drivers of land cover change based on a random forest method , 2021, Ecol. Informatics.

[15]  Zeinab Shirvani,et al.  A Holistic Analysis for Landslide Susceptibility Mapping Applying Geographic Object-Based Random Forest: A Comparison between Protected and Non-Protected Forests , 2020, Remote. Sens..

[16]  Anka Lisec,et al.  Estimating the Performance of Random Forest versus Multiple Regression for Predicting Prices of the Apartments , 2018, ISPRS Int. J. Geo Inf..

[17]  Mathis Wackernagel,et al.  Accounting for demand and supply of the biosphere's regenerative capacity: The National Footprint Accounts’ underlying methodology and framework , 2013 .

[18]  Michele Dalponte,et al.  Tree Species Classification in Boreal Forests With Hyperspectral Data , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[19]  R. Prăvălie.  Major perturbations in the Earth's forest ecosystems. Possible implications for global warming , 2018, Earth-Science Reviews.

[20]  Vinicius Picanco Rodrigues,et al.  Process-related key performance indicators for measuring sustainability performance of ecodesign implementation into product development , 2016 .

[21]  Bhabani Shankar Prasad Mishra,et al.  Image Change Detection Using Particle Swarm Optimization , 2015 .

[22]  S. Abd-Elmabod,et al.  Rapid urbanisation threatens fertile agricultural land and soil carbon in the Nile delta. , 2019, Journal of environmental management.

[23]  M. Sperow.  Updated potential soil carbon sequestration rates on U.S. agricultural land based on the 2019 IPCC guidelines , 2020 .

[24]  D. Lu,et al.  Change detection techniques , 2004 .

[25]  Tyler C. McCandless,et al.  Examining the Potential of a Random Forest Derived Cloud Mask from GOES-R Satellites to Improve Solar Irradiance Forecasting , 2020, Energies.

[26]  Uwe Sörgel,et al.  Machine learning methods for remote sensing applications: an overview , 2018, Remote Sensing.

[27]  V. Mendes,et al.  Comparison between Inflexible and Flexible Charging of Electric Vehicles—A Study from the Perspective of an Aggregator , 2020 .

[28]  Clement Atzberger,et al.  First Experience with Sentinel-2 Data for Crop and Tree Species Classifications in Central Europe , 2016, Remote. Sens..

[29]  Xavier Pons,et al.  Automatic and improved radiometric correction of Landsat imagery using reference values from MODIS surface reflectance images , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[30]  C. Gomez,et al.  Joint use of Sentinel-1 and Sentinel-2 for land cover classification : a machine learning approach , 2017 .

[31]  L. Chu,et al.  Carbon emission and sequestration of urban turfgrass systems in Hong Kong. , 2014, The Science of the total environment.

[32]  Ferran Gascon,et al.  Sen2Cor for Sentinel-2 , 2017, Remote Sensing.

[33]  Marco Heurich,et al.  Linking the Remote Sensing of Geodiversity and Traits Relevant to Biodiversity - Part II: Geomorphology, Terrain and Surfaces , 2020, Remote. Sens..

[34]  D. Guan,et al.  The global CO2 emissions growth after international crisis and the role of international trade , 2017 .

[35]  Dietrich Schmidt-Vogt,et al.  Carbon stock assessment and soil carbon management in agricultural land‐uses in Thailand , 2008 .

[36]  X. Tong,et al.  Modeling changes in China’s 2000–2030 carbon stock caused by land use change , 2020 .

[37]  Hongyun Han,et al.  Sectoral changing patterns of China’s green GDP considering climate change: An investigation based on the economic input-output life cycle assessment model , 2020 .

[38]  V. Masson‐Delmotte,et al.  Climate Change and Land: an IPCC special report on climate change, desertification, land degradation, sustainable land management, food security, and greenhouse gas fluxes in terrestrial ecosystems , 2019 .

[39]  T. Mungcharoen,et al.  Thailand Green GDP assessment based on environmentally extended input-output model , 2017 .

[40]  F. Müller,et al.  Mapping ecosystem service supply, demand and budgets , 2012 .

[41]  Michael A. Wulder,et al.  Characterizing stand-level forest canopy cover and height using Landsat time series, samples of airborne LiDAR, and the Random Forest algorithm , 2015 .

[42]  C. Yaman,et al.  Potential for greenhouse gas reduction and energy recovery from MSW through different waste management technologies , 2020 .

[43]  Martin Kappas,et al.  Comparison of Random Forest, k-Nearest Neighbor, and Support Vector Machine Classifiers for Land Cover Classification Using Sentinel-2 Imagery , 2017, Sensors.

[44]  Xiaohui Yang,et al.  Land use/cover classification in an arid desert-oasis mosaic landscape of China using remote sensed imagery: Performance assessment of four machine learning algorithms , 2020, Global Ecology and Conservation.

[45]  M. A. Haq,et al.  Spatial prediction of permafrost occurrence in Sikkim Himalayas using logistic regression, random forests, support vector machines and neural networks , 2020 .

[46]  Hankui K. Zhang,et al.  Using the 500 m MODIS Land Cover Product to Derive a Consistent Continental Scale 30 m Landsat Land Cover Classification , 2017 .

[47]  Qiang Wang,et al.  Preventing carbon emission retaliatory rebound post-COVID-19 requires expanding free trade and improving energy efficiency , 2020, Science of The Total Environment.

[48]  P. Schröder,et al.  The effect of financial crises on air pollutant emissions: An assessment of the short vs. medium-term effects. , 2020, The Science of the total environment.

[49]  Feng Wu,et al.  Effects of land use and land cover change on carbon sequestration and adaptive management in Shanghai, China , 2020 .

[50]  Ferran Gascon,et al.  Next updates of atmospheric correction processor Sen2Cor , 2020, Remote Sensing.

[51]  J. Keson,et al.  Estimation of Carbon Stocks from Land-use Change Due to Tourism in Phuket Island, Thailand , 2017 .

[52]  Manjula Ranagalage,et al.  Sentinel-2 Data for Land Cover/Use Mapping: A Review , 2020, Remote. Sens..

[53]  Rajendra P. Shrestha,et al.  Role of land tenure security and farm household characteristics on land use change in the Prasae Watershed, Thailand , 2008 .

[54]  Mariana Belgiu,et al.  Random forest in remote sensing: A review of applications and future directions , 2016 .

[55]  N. Volakakis,et al.  Developing a decision support tool for evaluating the environmental performance of olive production in terms of energy use and greenhouse gas emissions , 2020 .