Developing detailed age-specific thematic maps for coffee (Coffea arabica L.) in heterogeneous agricultural landscapes using random forests applied on Landsat 8 multispectral sensor

Abstract Coffee is a commodity of international trade significance, and its value chain can benefit from age-specific thematic maps. This study aimed to assess the potential of Landsat 8 OLI to develop these maps. Using field-collected samples with the random forest classifier, splitting coffee into three age classes (Scheme A) was compared with running the classification with one compound coffee class (Scheme B). Higher overall classification accuracy was obtained in Scheme B (90.3% for OLI and 86.8% for ETM+) than in Scheme A (86.2% for OLI and 81.0% for ETM+). The NIR band of OLI was the most important band in intra-class discrimination of coffee. Landsat 8 OLI mapped area closely matched farm records (R2 = 0.88) compared to that of Landsat 7 ETM+ (R2 = 0.78). It was concluded that Landsat 8 OLI data can be used to produce age-specific thematic maps in coffee production areas although disaggregating coffee classes reduces overall accuracy.

[1]  Bernardo Rudorff,et al.  Monitoring biennial bearing effect on coffee yield using modis remote sensing imagery , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[2]  M. McDonald,et al.  Impacts of Type of Fallow and Invasion by Chromolaena odorata on Weed Communities in Crop Fields in Cameroon , 2004 .

[3]  Charis Gresser,et al.  Mugged: Poverty in Your Coffee Cup , 2004 .

[4]  K. P. P. Nair,et al.  The Agronomy and Economy of Important Tree Crops of the Developing World , 2010 .

[5]  George P. Petropoulos,et al.  Land cover mapping with emphasis to burnt area delineation using co-orbital ALI and Landsat TM imagery , 2012, Int. J. Appl. Earth Obs. Geoinformation.

[6]  Dawei Han,et al.  Selection of classification techniques for land use - land cover change investigation , 2012 .

[7]  Eduardo Delgado Assad,et al.  Comparação de dados dos satélites Ikonos-II e Landsat/ETM+ no estudo de áreas cafeeiras , 2006 .

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

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

[10]  Johannes R. Sveinsson,et al.  Random Forest classification of multisource remote sensing and geographic data , 2004, IGARSS 2004. 2004 IEEE International Geoscience and Remote Sensing Symposium.

[11]  S. Hamilton,et al.  Ecosystem services and agriculture: Cultivating agricultural ecosystems for diverse benefits , 2007 .

[12]  O. Mutanga,et al.  Evaluating the utility of the medium-spatial resolution Landsat 8 multispectral sensor in quantifying aboveground biomass in uMgeni catchment, South Africa , 2015 .

[13]  Amir Hossein Alavi,et al.  Machine learning in geosciences and remote sensing , 2016 .

[14]  Hadi Memarian,et al.  Estimating oil palm yields using vegetation indices derived from quickbird , 2013 .

[15]  Lalit Kumar,et al.  Comparative assessment of the measures of thematic classification accuracy , 2007 .

[16]  Nathaniel A. Brunsell,et al.  Remotely Sensed Phenology of Coffee and Its Relationship to Yield , 2009 .

[17]  Massimiliano Pittore,et al.  Performance Evaluation of Machine Learning Algorithms for Urban Pattern Recognition from Multi-spectral Satellite Images , 2014, Remote. Sens..

[18]  José Carlos Neves Epiphanio,et al.  Spectral variability of coffee crop related to scene fractional components estimated through high resolution image , 2005 .

[19]  D. Leckie,et al.  Prediction Models for Estimating the Area, Volume, and Age of Rubber (Hevea brasiliensis) Plantations in Malaysia using Landsat TM Data , 2004 .

[20]  Jakub P. Walawender,et al.  A new ArcGIS toolset for automated mapping of land surface temperature with the use of LANDSAT satellite data , 2012, 2012 IEEE International Geoscience and Remote Sensing Symposium.

[21]  Rich Caruana,et al.  An empirical comparison of supervised learning algorithms , 2006, ICML.

[22]  K. Giller,et al.  Evaluating coffee yield gaps and important biotic, abiotic, and management factors limiting coffee production in Uganda , 2015 .

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

[24]  G. Foody.  Thematic map comparison: Evaluating the statistical significance of differences in classification accuracy , 2004 .

[25]  Osvaldo Coelho Pereira Neto,et al.  Uso da geotecnologia para caracterizar os cafezais no município de Londrina-PR, em relação à altimetria, declividade e tipo de solo , 2010 .

[26]  P. Mecocci,et al.  Random Forest ensembles for detection and prediction of Alzheimer's disease with a good between-cohort robustness , 2014, NeuroImage: Clinical.

[27]  M. Jurado-Expósito,et al.  Evaluation of pixel- and object-based approaches for mapping wild oat (Avena sterilis) weed patches in wheat fields using QuickBird imagery for site-specific management , 2014 .

[28]  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 .

[29]  Giles M. Foody,et al.  Status of land cover classification accuracy assessment , 2002 .

[30]  Stéphanie Nicolin.  Implementing environmental water requirements in Buzi River basin, Mozambique An impact analysis based on the Water Resource Yield Model , 2011 .

[31]  Md. Latifur Rahman Sarker,et al.  Exploring the Potential of High Resolution Remote Sensing Data for Mapping Vegetation and the Age Groups of Oil Palm Plantation , 2014 .

[32]  I. Ozdemir,et al.  Estimating stem volume by tree crown area and tree shadow area extracted from pan‐sharpened Quickbird imagery in open Crimean juniper forests , 2008 .

[33]  Meine van Noordwijk,et al.  Impact of cropping methods on biodiversity in coffee agroecosystems in Sumatra, Indonesia , 2004 .

[34]  J. Epiphanio,et al.  CROP PARAMETERS FOR THE EVALUATION OF THE SPECTRAL RESPONSE OF COFFEE ( Coffea Arabica ) AREAS WITHIN THE STATE OF MINAS GERAIS , BRAZIL , 2004 .

[35]  G. A. Mahdiraji,et al.  Airborne hyperspectral discrimination of tree species with different ages using discrete wavelet transform , 2015 .

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

[37]  Lovisa Lagerblad.  Assessment of environmental flow requirements in Buzi River basin, Mozambique , 2010 .

[38]  A. Klein,et al.  Linking deforestation scenarios to pollination services and economic returns in coffee agroforestry systems. , 2007, Ecological applications : a publication of the Ecological Society of America.

[39]  Asmala Ahmad,et al.  Analysis of Maximum Likelihood Classificationon Multispectral Data , 2012 .

[40]  D. Kutywayo,et al.  Bioclimatic modelling of current and projected climatic suitability of coffee (Coffea arabica) production in Zimbabwe , 2016, Regional Environmental Change.

[41]  L. G. Carvalho,et al.  Geospatial analysis of ecological vulnerability of coffee agroecosystems in Brazil , 2013 .

[42]  Marcos Adami,et al.  Análise espectral e temporal da cultura do café em imagens Landsat , 2004 .

[43]  A. Veldkamp,et al.  utomated high resolution mapping of coffee in Rwanda using an xpert Bayesian network , 2014 .

[44]  Antonella Battaglini,et al.  Climate hotspots: key vulnerable regions, climate change and limits to warming , 2011 .

[45]  Onisimo Mutanga,et al.  Exploiting machine learning algorithms for tree species classification in a semiarid woodland using RapidEye image , 2013 .

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

[47]  A. Chemura,et al.  Determination of the age of oil palm from crown projection area detected from WorldView-2 multispectral remote sensing data: The case of Ejisu-Juaben district, Ghana , 2015 .

[48]  Julia Mcmorrow,et al.  Linear regression modelling for the estimation of oil palm age from Landsat TM , 2001 .

[49]  Xingfa Gu,et al.  Land cover classification using Landsat 8 Operational Land Imager data in Beijing, China , 2014 .

[50]  K. A. Kumar,et al.  Remote sensing and GIS application for land quality assessment for coffee growing areas of Karnataka , 2008 .

[51]  John R. Schott,et al.  Leveraging EO-1 to Evaluate Capability of New Generation of Landsat Sensors for Coastal/Inland Water Studies , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[52]  Y. Dian,et al.  Comparison of the Different Classifiers in Vegetation Species Discrimination Using Hyperspectral Reflectance Data , 2014, Journal of the Indian Society of Remote Sensing.

[53]  Onisimo Mutanga,et al.  Testing the reliability and stability of the internal accuracy assessment of random forest for classifying tree defoliation levels using different validation methods , 2015 .

[54]  A. Cracknell,et al.  Use of UK-DMC 2 and ALOS PALSAR for studying the age of oil palm trees in southern peninsular Malaysia , 2013 .

[55]  J. Im,et al.  Machine learning approaches for forest classification and change analysis using multi-temporal Landsat TM images over Huntington Wildlife Forest , 2013 .