Phenology and Spectral Unmixing-Based Invasive Kudzu Mapping: A Case Study in Knox County, Tennessee

As an invasive plant species, kudzu has been spreading rapidly in the Southeastern United States in recent years. Accurate mapping of kudzu is critical for effective invasion control and management. However, the remote detection of kudzu distribution using multispectral images is challenging because of the mixed reflectance and potential misclassification with other vegetation. We propose a three-step classification process to map kudzu in Knox County, Tennessee, using multispectral Sentinel-2 images and the integration of spectral unmixing analysis and phenological characteristics. This classification includes an initial linear unmixing process to produce an overestimated kudzu map, a phenological-based masking to reduce misclassification, and a nonlinear unmixing process to refine the classification. The initial linear unmixing provides high producer’s accuracy (PA) but low user’s accuracy (UA) due to misclassification with grasslands. The phenological-based masking increases the accuracy of the kudzu classification and reduces the domain for further processing. The nonlinear unmixing further refines the kudzu classification via the selection of an appropriate nonlinear model. The final kudzu classification for Knox County reaches relatively high accuracy, with UA, PA, Jaccard, and Kappa index values of 0.858, 0.907, 0.789, and 0.725, respectively. Our proposed method has potential for continuous monitoring of kudzu in large areas.

[1]  Tung Fung,et al.  The Determination of Optimal Threshold Levels for Change Detection Using Various Accuracy Indices , 2008 .

[2]  Onisimo Mutanga,et al.  Advancements in satellite remote sensing for mapping and monitoring of alien invasive plant species (AIPs) , 2019, Physics and Chemistry of the Earth, Parts A/B/C.

[3]  D. Horler,et al.  The red edge of plant leaf reflectance , 1983 .

[4]  N. Tharayil,et al.  Plant litter chemistry and microbial priming regulate the accrual, composition and stability of soil carbon in invaded ecosystems. , 2014, The New phytologist.

[5]  B. Bradley.  Remote detection of invasive plants: a review of spectral, textural and phenological approaches , 2014, Biological Invasions.

[6]  Jamal Jokar Arsanjani,et al.  Employing Machine Learning for Detection of Invasive Species using Sentinel-2 and AVIRIS Data: The Case of Kudzu in the United States , 2020, Sustainability.

[7]  Le Wang,et al.  Retrieval of subpixel Tamarix canopy cover from Landsat data along the Forgotten River using linear and nonlinear spectral mixture models , 2010 .

[8]  D. Alderman.  Channing Cope and the Making of a Miracle Vine* , 2004 .

[9]  P. Pyšek,et al.  Timing Is Important: Unmanned Aircraft vs. Satellite Imagery in Plant Invasion Monitoring , 2017, Front. Plant Sci..

[10]  A. Mackay.  Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change , 2008 .

[11]  John F. Mustard,et al.  Spectral unmixing , 2002, IEEE Signal Process. Mag..

[12]  D. Roberts,et al.  Mapping understory invasive plant species with field and remotely sensed data in Chitwan, Nepal , 2020 .

[13]  L. Loope.  AN OVERVIEW OF PROBLEMS WITH INTRODUCED PLANT SPECIES IN NATIONAL PARKS AND BIOSPHERE RESERVES OF THE UNITED STATES , 2004 .

[14]  Jan Thiele,et al.  Open-Source Processing and Analysis of Aerial Imagery Acquired with a Low-Cost Unmanned Aerial System to Support Invasive Plant Management , 2017, Front. Environ. Sci..

[15]  Le Wang,et al.  An object-based SVM method incorporating optimal segmentation scale estimation using Bhattacharyya Distance for mapping salt cedar (Tamarisk spp.) with QuickBird imagery , 2015 .

[16]  B. M. Larson,et al.  Kudzu Vine, Pueraria montana , Adventive in Southern Ontario , 2012 .

[17]  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 .

[18]  Baofeng Su,et al.  Significant Remote Sensing Vegetation Indices: A Review of Developments and Applications , 2017, J. Sensors.

[19]  Su Ye,et al.  Comparison of linear and nonlinear spectral unmixing approaches: a case study with multispectral TM imagery , 2017 .

[20]  Yingkui Li,et al.  Mapping Vegetation at Species Level with High-Resolution Multispectral and Lidar Data Over a Large Spatial Area: A Case Study with Kudzu , 2020, Remote. Sens..

[21]  Laura E. Hoyos,et al.  Monitoring the invasion of an exotic tree (Ligustrum lucidum) from 1983 to 2006 with Landsat TM/ETM + satellite data and Support Vector Machines in Córdoba, Argentina , 2012 .

[22]  Tomaž Podobnikar,et al.  Recognition of the invasive species Robinia pseudacacia from combined remote sensing and GIS sources , 2012 .

[23]  Le Wang,et al.  Development of an invasive species distribution model with fine-resolution remote sensing , 2014, Int. J. Appl. Earth Obs. Geoinformation.

[24]  Heather McNairn,et al.  Evaluating multispectral remote sensing and spectral unmixing analysis for crop residue mapping , 2010 .

[25]  R. Congalton,et al.  Accuracy assessment: a user's perspective , 1986 .

[26]  Chein-I Chang,et al.  Fully constrained least squares linear spectral mixture analysis method for material quantification in hyperspectral imagery , 2001, IEEE Trans. Geosci. Remote. Sens..

[27]  John R. Miller,et al.  Comparative study between a new nonlinear model and common linear model for analysing laboratory simulated‐forest hyperspectral data , 2009 .

[28]  Susan L. Ustin,et al.  Mapping an invasive species, kudzu (Pueraria montana), using hyperspectral imagery in western Georgia , 2007 .

[29]  Timothy L. Hawthorne,et al.  Mapping non-native invasive species and accessibility in an urban forest: A case study of participatory mapping and citizen science in Atlanta, Georgia , 2015 .

[30]  Le Wang,et al.  Characterizing spatial patterns of invasive species using sub-pixel classifications , 2011 .

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

[32]  Le Wang,et al.  Phenology-guided saltcedar (Tamarix spp.) mapping using Landsat TM images in western U.S. , 2016 .

[33]  Jacob Cohen.  A Coefficient of Agreement for Nominal Scales , 1960 .

[34]  Bing Wang,et al.  Assessment of Sentinel-2 MSI Spectral Band Reflectances for Estimating Fractional Vegetation Cover , 2018, Remote. Sens..