Identifying core driving factors of urban land use change from global land cover products and POI data using the random forest method

Abstract Rapid urbanization at the expense of the environment led to a reduction in vegetation cover, and consequently aggravated land degradation, urban water logging, heat island effect and other effects. Revealing the driving mechanism behind urban land use change facilitates deeper insight into the human and biophysical effects in the urbanization process and thereby supports the sustainable urban development. This work proposed a margin-based measure of random forest for core driving factor identification of urban land use change, which mainly included urban vegetation change, constructed land, water bodies, etc., using multitemporal global land cover products and point-of-interest (POI) data. Taking the land use change in Wuhan from 2010 to 2020 as the case study, the proposed method was employed to measure and sort the driving forces of 24 biophysical and human factors. The results suggested that the margin-based method was more reliable and sensitive than the traditional importance measure of random forest when detecting the driving mechanism behind land use change. Meanwhile, both the importance values and the ranking orders of driving factors measured by the margin-based method were stable regardless of which similarity measure was chosen and applied. The findings also showed that topographic conditions persistently affected urban land use change, while transportation factors, instead of business services, gradually became the most important human driving factors in Wuhan in the last 10 years.

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

[2]  Zhong Lu,et al.  Ground Deformation of Wuhan, China, Revealed by Multi-Temporal InSAR Analysis , 2020, Remote. Sens..

[3]  F. van den Bergh,et al.  Limits to detectability of land degradation by trend analysis of vegetation index data , 2012 .

[4]  Xianglong Tang,et al.  Mapping Urban Spatial Structure Based on POI (Point of Interest) Data: A Case Study of the Central City of Lanzhou, China , 2020, ISPRS Int. J. Geo Inf..

[5]  Development of a global 30 m impervious surface map using multisource and multitemporal remote sensing datasets with the Google Earth Engine platform , 2020, Earth System Science Data.

[6]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[7]  Michael Batty,et al.  Modeling urban growth with GIS based cellular automata and least squares SVM rules: a case study in Qingpu–Songjiang area of Shanghai, China , 2016, Stochastic Environmental Research and Risk Assessment.

[8]  Yuei-An Liou,et al.  Land-Use Land-Cover Classification by Machine Learning Classifiers for Satellite Observations - A Review , 2020, Remote. Sens..

[9]  Xia Wu,et al.  A big data-driven dynamic estimation model of relief supplies demand in urban flood disaster , 2020 .

[10]  Liangyun Liu,et al.  Consistency Analysis and Accuracy Assessment of Three Global 30-m Land-Cover Products over the European Union using the LUCAS Dataset , 2020, Remote. Sens..

[11]  Daniel Neagu,et al.  Interpreting random forest models using a feature contribution method , 2013, 2013 IEEE 14th International Conference on Information Reuse & Integration (IRI).

[12]  Gilles Louppe,et al.  Ensembles on Random Patches , 2012, ECML/PKDD.

[13]  P. Sedgwick.  Spearman’s rank correlation coefficient , 2018, British Medical Journal.

[14]  Xin Pan,et al.  Joint Deep Learning for land cover and land use classification , 2019, Remote Sensing of Environment.

[15]  Michel Dekking,et al.  A Modern Introduction to Probability and Statistics: Understanding Why and How , 2007 .

[16]  Zhen Li,et al.  Examining the sensitivity of spatial scale in cellular automata Markov chain simulation of land use change , 2019, Int. J. Geogr. Inf. Sci..

[17]  GLC_FCS30: global land-cover product with fine classification system at 30 m using time-series Landsat imagery , 2021 .

[18]  David Sneath,et al.  State Policy and Pasture Degradation in Inner Asia , 1998, Science.

[19]  Gaofeng Zhu,et al.  Land Use/Land Cover Changes and Their Driving Factors in the Northeastern Tibetan Plateau Based on Geographical Detectors and Google Earth Engine: A Case Study in Gannan Prefecture , 2020, Remote. Sens..

[20]  Achim Zeileis,et al.  Bias in random forest variable importance measures: Illustrations, sources and a solution , 2007, BMC Bioinformatics.

[21]  J. Watson,et al.  The Effect of Urban Density and Vegetation Cover on the Heat Island of a Subtropical City , 2018, Journal of Applied Meteorology and Climatology.

[22]  Bhre Wijaya Aroengbinang,et al.  Driving Force Analysis of Landuse and Cover Changes in Cimandiri and Cibuni Watersheds , 2015 .

[23]  Michael J. Oimoen,et al.  ASTER Global Digital Elevation Model Version 2 - summary of validation results , 2011 .

[24]  Yanling Zhu,et al.  Driving forces behind land use and cover change in the Qinghai-Tibetan Plateau: a case study of the source region of the Yellow River, Qinghai Province, China , 2009 .

[25]  Dalton Lunga,et al.  Large-Scale Classification of Urban Structural Units From Remote Sensing Imagery , 2021, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[26]  Zhifeng Wu,et al.  Investigating the Role of Green Infrastructure on Urban WaterLogging: Evidence from Metropolitan Coastal Cities , 2021, Remote. Sens..

[27]  P. Thakuriah,et al.  The geography of human activity and land use: A big data approach , 2020 .

[28]  Bing Wang,et al.  Land use changes and its driving forces in hilly ecological restoration area based on gis and rs of northern china , 2015, Scientific Reports.

[29]  Anne Gobin,et al.  Logistic modelling to derive agricultural land use determinants: A case study from southeastern Nigeria , 2002 .

[30]  Samia Boukir,et al.  Relevance of airborne lidar and multispectral image data for urban scene classification using Random Forests , 2011 .

[31]  Johannes R. Sveinsson,et al.  Random Forests for land cover classification , 2006, Pattern Recognit. Lett..

[32]  M. Goodchild.  Citizens as sensors: the world of volunteered geography , 2007 .

[33]  S. Carpenter,et al.  Global Consequences of Land Use , 2005, Science.

[34]  Jin Chen,et al.  Global land cover mapping at 30 m resolution: A POK-based operational approach , 2015 .

[35]  Yao Yao,et al.  Multiple intra-urban land use simulations and driving factors analysis: a case study in Huicheng, China , 2018, GIScience & Remote Sensing.

[36]  Ruifeng Zhao,et al.  Land use and land cover change and driving mechanism in the arid inland river basin: a case study of Tarim River, Xinjiang, China , 2012, Environmental Earth Sciences.

[37]  Hongzhang Xu,et al.  Deep learning in environmental remote sensing: Achievements and challenges , 2020, Remote Sensing of Environment.

[38]  Yatao Zhang,et al.  Mapping fine-scale population distributions at the building level by integrating multisource geospatial big data , 2017, Int. J. Geogr. Inf. Sci..

[39]  E. Moran.  Deforestation and land use in the Brazilian Amazon , 1993 .

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

[41]  Peter Nijkamp,et al.  Multidimensional urban sprawl in Europe: A self-organizing map approach , 2011, Comput. Environ. Urban Syst..

[42]  Xia Li,et al.  Simulating urban land use change by integrating a convolutional neural network with vector-based cellular automata , 2020, Int. J. Geogr. Inf. Sci..

[43]  Jay Wright Forrester,et al.  Urban Dynamics , 1969 .

[44]  L. Sheng,et al.  Land Use Change Trends and Their Driving Forces in the Kilombero Valley Floodplain, Southeastern Tanzania , 2019, Sustainability.

[45]  Yoav Freund,et al.  Boosting the margin: A new explanation for the effectiveness of voting methods , 1997, ICML.

[46]  J. W. Bruce,et al.  The causes of land-use and land-cover change: moving beyond the myths , 2001 .

[47]  David Mease,et al.  Explaining the Success of AdaBoost and Random Forests as Interpolating Classifiers , 2015, J. Mach. Learn. Res..

[48]  C. Nunes,et al.  Land-Use and Land-Cover Change (LUCC): Implementation Strategy , 1999 .

[49]  Bryan C. Pijanowski,et al.  A big data urban growth simulation at a national scale: Configuring the GIS and neural network based Land Transformation Model to run in a High Performance Computing (HPC) environment , 2014, Environ. Model. Softw..

[50]  Melanie Hilario,et al.  Knowledge and Information Systems , 2007 .

[51]  Chunyang He,et al.  How much of the world’s land has been urbanized, really? A hierarchical framework for avoiding confusion , 2014, Landscape Ecology.

[52]  Xiaoping Liu,et al.  A future land use simulation model (FLUS) for simulating multiple land use scenarios by coupling human and natural effects , 2017 .

[53]  Rolf H H Groenwold,et al.  Overestimation of risk ratios by odds ratios in trials and cohort studies: alternatives to logistic regression , 2012, Canadian Medical Association Journal.

[54]  Fan Yang,et al.  Margin based permutation variable importance: A stable importance measure for random forest , 2017, 2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE).

[55]  Xavier Pons,et al.  Land-cover and land-use change in a Mediterranean landscape: A spatial analysis of driving forces integrating biophysical and human factors , 2008 .

[56]  A. Pfaff.  What drives deforestation in the Brazilian Amazon? Evidence from satellite and socioeconomic data , 1997 .