Mapping Deforestation in North Korea Using Phenology-Based Multi-Index and Random Forest

Phenology-based multi-index with the random forest (RF) algorithm can be used to overcome the shortcomings of traditional deforestation mapping that involves pixel-based classification, such as ISODATA or decision trees, and single images. The purpose of this study was to investigate methods to identify specific types of deforestation in North Korea, and to increase the accuracy of classification, using phenological characteristics extracted with multi-index and random forest algorithms. The mapping of deforestation area based on RF was carried out by merging phenology-based multi-indices (i.e., normalized difference vegetation index (NDVI), normalized difference water index (NDWI), and normalized difference soil index (NDSI)) derived from MODIS (Moderate Resolution Imaging Spectroradiometer) products and topographical variables. Our results showed overall classification accuracy of 89.38%, with corresponding kappa coefficients of 0.87. In particular, for forest and farm land categories with similar phenological characteristic (e.g., paddy, plateau vegetation, unstocked forest, hillside field), this approach improved the classification accuracy in comparison with pixel-based methods and other classes. The deforestation types were identified by incorporating point data from high-resolution imagery, outcomes of image classification, and slope data. Our study demonstrated that the proposed methodology could be used for deciding on the restoration priority and monitoring the expansion of deforestation areas.

[1]  Satoshi Uchida,et al.  Discriminating different landuse types by using multitemporal NDXI in a rice planting area , 2010 .

[2]  Dar A. Roberts,et al.  Ten-Year Landsat Classification of Deforestation and Forest Degradation in the Brazilian Amazon , 2013, Remote. Sens..

[3]  A. Belward,et al.  The Best Index Slope Extraction ( BISE): A method for reducing noise in NDVI time-series , 1992 .

[4]  Ramakrishna R. Nemani,et al.  Real-time monitoring and short-term forecasting of land surface phenology , 2006 .

[5]  France Gerard,et al.  Exploring the Use of MODIS NDVI-Based Phenology Indicators for Classifying Forest General Habitat Categories , 2012, Remote. Sens..

[6]  Mario Chica-Olmo,et al.  An assessment of the effectiveness of a random forest classifier for land-cover classification , 2012 .

[7]  J. R. Jensen,et al.  Remote Sensing of Urban/Suburban Infrastructure and Socio‐Economic Attributes , 2011 .

[8]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[9]  B. Gao.  NDWI—A normalized difference water index for remote sensing of vegetation liquid water from space , 1996 .

[10]  Monitoring land ecosystems using the NOAA Global Vegetation Index data set , 1991 .

[11]  S. Goetz,et al.  Estimated carbon dioxide emissions from tropical deforestation improved by carbon-density maps , 2012 .

[12]  J. Tenhunen,et al.  On the relationship of NDVI with leaf area index in a deciduous forest site , 2005 .

[13]  Miao Li,et al.  RNDSI: A ratio normalized difference soil index for remote sensing of urban/suburban environments , 2015, Int. J. Appl. Earth Obs. Geoinformation.

[14]  L. Tang,et al.  Forest degradation deepens around and within protected areas in East Asia , 2010 .

[15]  S. E. Franklin,et al.  Classification of alpine vegetation using Landsat Thematic Mapper SPOT HRV and DEM data , 1994 .

[16]  Chong-hwa Park,et al.  The Utilization of Google Earth Images as Reference Data for The Multitemporal Land Cover Classification with MODIS Data of North Korea , 2007 .

[17]  Russell G. Congalton,et al.  Assessing the accuracy of remotely sensed data : principles and practices , 1998 .

[18]  S. Goward,et al.  Global Primary Production: A Remote Sensing Approach , 1995 .

[19]  I. Thompson,et al.  An Operational framework for defining and monitoring forest degradation , 2013 .

[20]  B. Holben.  Characteristics of maximum-value composite images from temporal AVHRR data , 1986 .

[21]  Simonetti Edoardo,et al.  Phenology-based land cover classification using Landsat 8 time series , 2014 .

[22]  R. Chazdon.  Beyond Deforestation: Restoring Forests and Ecosystem Services on Degraded Lands , 2008, Science.

[23]  Yoko NISHIMURA,et al.  Google Earth , 2008, Encyclopedia of GIS.

[24]  A. Huete,et al.  Overview of the radiometric and biophysical performance of the MODIS vegetation indices , 2002 .

[25]  Y. Richard,et al.  A statistical study of NDVI sensitivity to seasonal and interannual rainfall variations in Southern Africa , 1998 .

[26]  C. Alados,et al.  Improving the Accuracy of Vegetation Classifications in Mountainous Areas , 2013 .

[27]  Stephen R. Yool,et al.  Techniques for computer-assisted mapping of rangeland change. , 1997 .

[28]  Y. Setiawan,et al.  A simple method for developing near real-time nationwide forest monitoring for Indonesia using MODIS near- and shortwave infrared bands , 2016 .

[29]  Clement Atzberger,et al.  Tree Species Classification with Random Forest Using Very High Spatial Resolution 8-Band WorldView-2 Satellite Data , 2012, Remote. Sens..

[30]  Yunming Ye,et al.  Stratified sampling for feature subspace selection in random forests for high dimensional data , 2013, Pattern Recognit..

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

[32]  David O. Wallin,et al.  Rates and patterns of landscape change between 1972 and 1988 in the Changbai Mountain area of China and North Korea , 1997, Landscape Ecology.

[33]  Andrew D. Richardson,et al.  Phenological Differences Between Understory and Overstory , 2009 .

[34]  Qian Song,et al.  Exploring the Use of Google Earth Imagery and Object-Based Methods in Land Use/Cover Mapping , 2013, Remote. Sens..

[35]  D. Lloyd,et al.  A phenological classification of terrestrial vegetation cover using shortwave vegetation index imagery , 1990 .

[36]  Per Jönsson,et al.  Seasonality extraction by function fitting to time-series of satellite sensor data , 2002, IEEE Trans. Geosci. Remote. Sens..

[37]  Zhanqing Li,et al.  Estimating fire-related parameters in boreal forest using SPOT VEGETATION , 2002 .

[38]  S. Sader,et al.  Detection of forest harvest type using multiple dates of Landsat TM imagery , 2002 .

[39]  Patrick Hostert,et al.  Using MODIS time series and random forests classification for mapping land use in South-East Asia , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[40]  Steven E. Franklin,et al.  A comparison of pixel-based and object-based image analysis with selected machine learning algorithms for the classification of agricultural landscapes using SPOT-5 HRG imagery , 2012 .

[41]  Qihao Weng,et al.  A survey of image classification methods and techniques for improving classification performance , 2007 .

[42]  Per Jönsson,et al.  TIMESAT - a program for analyzing time-series of satellite sensor data , 2004, Comput. Geosci..

[43]  Jesslyn F. Brown,et al.  Measuring phenological variability from satellite imagery , 1994 .

[44]  John R. Jensen,et al.  Introductory Digital Image Processing: A Remote Sensing Perspective , 1986 .

[45]  Dong-Kun Lee,et al.  Terrace Fields Classification in North Korea Using MODIS Multi-temporal Image Data , 2016 .

[46]  H. Elsenbeer,et al.  Soil organic carbon concentrations and stocks on Barro Colorado Island — Digital soil mapping using Random Forests analysis , 2008 .

[47]  G. Vieilledent,et al.  Estimating deforestation in tropical humid and dry forests in Madagascar from 2000 to 2010 using multi-date Landsat satellite images and the random forests classifier , 2013 .

[48]  Jie Wang,et al.  Comparison of Classification Algorithms and Training Sample Sizes in Urban Land Classification with Landsat Thematic Mapper Imagery , 2014, Remote. Sens..

[49]  C. Tucker.  Remote sensing of leaf water content in the near infrared , 1980 .

[50]  T. Mitchell Aide,et al.  A scalable approach to mapping annual land cover at 250 m using MODIS time series data: A case study in the Dry Chaco ecoregion of South America , 2010 .

[51]  Dong Kun Lee,et al.  Mapping riparian habitat using a combination ofremote-sensing techniques , 2016 .

[52]  S. Tarantola,et al.  Detecting vegetation leaf water content using reflectance in the optical domain , 2001 .

[53]  Onisimo Mutanga,et al.  Land-use/cover classification in a heterogeneous coastal landscape using RapidEye imagery: evaluating the performance of random forest and support vector machines classifiers , 2014 .

[54]  A. Skidmore,et al.  Satellite-derived vegetation indices contribute significantly to the prediction of epiphyllous liverworts , 2014 .

[55]  J. Wickham,et al.  Land cover and land use change , 2018 .

[56]  Joanne C. White,et al.  Optical remotely sensed time series data for land cover classification: A review , 2016 .

[57]  Antonio Wolf,et al.  Using WorldView 2 Vis-NIR MSI Imagery to Support Land Mapping and Feature Extraction Using Normalized Difference Index Ratios , 2011 .

[58]  Z. Buřivalová,et al.  Relative Contributions of the Logging, Fiber, Oil Palm, and Mining Industries to Forest Loss in Indonesia , 2015 .

[59]  A. Thorne-Lyman,et al.  Food shortages and nutrition in North Korea , 2002, The Lancet.

[60]  R. Fensholt,et al.  Derivation of a shortwave infrared water stress index from MODIS near- and shortwave infrared data in a semiarid environment , 2003 .

[61]  Abbas Alimohammadi,et al.  Land cover mapping based on random forest classification of multitemporal spectral and thermal images , 2015, Environmental Monitoring and Assessment.

[62]  Wataru Takeuchi,et al.  Development of normalized Vegetation, soil and water indices derived from satellite remote sensing data , 2004 .

[63]  Dirk Pflugmacher,et al.  Mapping Rubber Plantations and Natural Forests in Xishuangbanna (Southwest China) Using Multi-Spectral Phenological Metrics from MODIS Time Series , 2013, Remote. Sens..

[64]  Mahesh Pal,et al.  Random forest classifier for remote sensing classification , 2005 .

[65]  V. Radeloff,et al.  Phenological differences in Tasseled Cap indices improve deciduous forest classification , 2002 .

[66]  C. Justice,et al.  The generation of global fields of terrestrial biophysical parameters from the NDVI , 1994 .

[67]  Peijun Li,et al.  Urban Built-Up Area Extraction from Landsat TM/ETM+ Images Using Spectral Information and Multivariate Texture , 2014, Remote. Sens..

[68]  Francis E. Putz,et al.  Critical need for new definitions of “forest” and “forest degradation” in global climate change agreements , 2009 .