{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,21]],"date-time":"2026-01-21T17:58:48Z","timestamp":1769018328143,"version":"3.49.0"},"reference-count":59,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2016,2,19]],"date-time":"2016-02-19T00:00:00Z","timestamp":1455840000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>Mapping species through classification of imaging spectroscopy data is facilitating research to understand tree species distributions at increasingly greater spatial scales. Classification requires a dataset of field observations matched to the image, which will often reflect natural species distributions, resulting in an imbalanced dataset with many samples for common species and few samples for less common species. Despite the high prevalence of imbalanced datasets in multiclass species predictions, the effect on species prediction accuracy and landscape species abundance has not yet been quantified. First, we trained and assessed the accuracy of a support vector machine (SVM) model with a highly imbalanced dataset of 20 tropical species and one mixed-species class of 24 species identified in a hyperspectral image mosaic (350\u20132500 nm) of Panamanian farmland and secondary forest fragments. The model, with an overall accuracy of 62% \u00b1 2.3% and F-score of 59% \u00b1 2.7%, was applied to the full image mosaic (23,000 ha at a 2-m resolution) to produce a species prediction map, which suggested that this tropical agricultural landscape is more diverse than what has been presented in field-based studies. Second, we quantified the effect of class imbalance on model accuracy. Model assessment showed a trend where species with more samples were consistently over predicted while species with fewer samples were under predicted. Standardizing sample size reduced model accuracy, but also reduced the level of species over- and under-prediction. This study advances operational species mapping of diverse tropical landscapes by detailing the effect of imbalanced data on classification accuracy and providing estimates of tree species abundance in an agricultural landscape. Species maps using data and methods presented here can be used in landscape analyses of species distributions to understand human or environmental effects, in addition to focusing conservation efforts in areas with high tree cover and diversity.<\/jats:p>","DOI":"10.3390\/rs8020161","type":"journal-article","created":{"date-parts":[[2016,2,19]],"date-time":"2016-02-19T11:29:39Z","timestamp":1455881379000},"page":"161","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":71,"title":["Tree Species Abundance Predictions in a Tropical Agricultural Landscape with a Supervised Classification Model and Imbalanced Data"],"prefix":"10.3390","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3805-4242","authenticated-orcid":false,"given":"Sarah","family":"Graves","sequence":"first","affiliation":[{"name":"School of Forest Resources and Conservation, University of Florida, PO Box 11041 Gainesville, FL 32611, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7893-6421","authenticated-orcid":false,"given":"Gregory","family":"Asner","sequence":"additional","affiliation":[{"name":"Department of Global Ecology, Carnegie Institution for Science, 260 Panama St., Stanford, CA 94305, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3509-8530","authenticated-orcid":false,"given":"Roberta","family":"Martin","sequence":"additional","affiliation":[{"name":"Department of Global Ecology, Carnegie Institution for Science, 260 Panama St., Stanford, CA 94305, USA"}]},{"given":"Christopher","family":"Anderson","sequence":"additional","affiliation":[{"name":"Department of Global Ecology, Carnegie Institution for Science, 260 Panama St., Stanford, CA 94305, USA"}]},{"given":"Matthew","family":"Colgan","sequence":"additional","affiliation":[{"name":"Department of Global Ecology, Carnegie Institution for Science, 260 Panama St., Stanford, CA 94305, USA"}]},{"given":"Leila","family":"Kalantari","sequence":"additional","affiliation":[{"name":"Department of Computer and Information Science and Engineering, University of Florida, PO Box 116120, Gainesville, FL 32611, USA"}]},{"given":"Stephanie","family":"Bohlman","sequence":"additional","affiliation":[{"name":"School of Forest Resources and Conservation, University of Florida, PO Box 11041 Gainesville, FL 32611, USA"},{"name":"Smithsonian Tropical Research Institute, Apartado 0843\u201303092, Balboa, Ancon, Republic of Panama"}]}],"member":"1968","published-online":{"date-parts":[[2016,2,19]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"2377","DOI":"10.1080\/01431160117096","article-title":"Using remote sensing to assess biodiversity","volume":"22","author":"Nagendra","year":"2001","journal-title":"Int. J. Remote Sens."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"306","DOI":"10.1016\/S0169-5347(03)00070-3","article-title":"Remote sensing for biodiversity science and conservation","volume":"18","author":"Turner","year":"2003","journal-title":"Trends Ecol. Evol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1579","DOI":"10.1890\/13-1160.1","article-title":"Coexistence and environmental filtering of species-specific biomass in an African savanna","volume":"95","author":"Colgan","year":"2014","journal-title":"Ecology"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1414","DOI":"10.1126\/science.288.5470.1414","article-title":"Spatial patterns in the distribution of tropical tree species","volume":"288","author":"Condit","year":"2000","journal-title":"Science"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"2088","DOI":"10.1016\/j.rse.2007.10.011","article-title":"Classification of Australian forest communities using aerial photography, CASI and HyMap data","volume":"112","author":"Lucas","year":"2008","journal-title":"Remote Sens. Environ."},{"key":"ref_6","first-page":"1658","article-title":"Hyperspectral remote sensing for invasive species detection and mapping","volume":"3","author":"Ustin","year":"2002","journal-title":"IEEE Int. Geosci. Remote Sens. Sympos."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"150","DOI":"10.1016\/S0034-4257(03)00096-8","article-title":"Mapping nonnative plants using hyperspectral imagery","volume":"86","author":"Underwood","year":"2003","journal-title":"Remote Sens. Environ."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"381","DOI":"10.1111\/j.1472-4642.2011.00761.x","article-title":"Benefits of hyperspectral remote sensing for tracking plant invasions","volume":"17","author":"He","year":"2011","journal-title":"Divers. Distrib."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"375","DOI":"10.1016\/j.rse.2005.03.009","article-title":"Hyperspectral discrimination of tropical rain forest tree species at leaf to crown scales","volume":"96","author":"Clark","year":"2005","journal-title":"Remote Sens. Environ."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"3431","DOI":"10.1007\/s10531-008-9479-0","article-title":"High resolution satellite imagery for tropical biodiversity studies: The devil is in the detail","volume":"17","author":"Nagendra","year":"2008","journal-title":"Biodivers. Conserv."},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Baldeck, C.A., Asner, G.P., Martin, R.E., Anderson, C.B., Knapp, D.E., Kellner, J.R., and Wright, S.J. (2015). Operational tree species mapping in a diverse tropical forest with airborne imaging spectroscopy. PLoS ONE, 10.","DOI":"10.1371\/journal.pone.0118403"},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1016\/j.rse.2014.03.018","article-title":"Urban tree species mapping using hyperspectral and lidar data fusion","volume":"148","author":"Alonzo","year":"2014","journal-title":"Remote Sens. Environ."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2457","DOI":"10.3390\/rs4082457","article-title":"Semi-supervised methods to identify individual crowns of lowland tropical canopy species using imaging spectroscopy and lidar","volume":"4","author":"Asner","year":"2012","journal-title":"Remote Sens."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2075","DOI":"10.1080\/01431160050021303","article-title":"Using vegetation reflectance variability for species level classification of hyperspectral data","volume":"21","author":"Cochrane","year":"2000","journal-title":"Int. J. Remote Sens."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2682","DOI":"10.3390\/rs6042682","article-title":"Improving remote species identification through efficient training data collection","volume":"6","author":"Baldeck","year":"2014","journal-title":"Remote Sens."},{"key":"ref_16","first-page":"1","article-title":"Tree species discrimination in tropical forests using airborne imaging spectroscopy","volume":"51","author":"Asner","year":"2012","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"513","DOI":"10.1080\/2150704X.2013.764027","article-title":"Identifying Santa Barbara\u2019s urban tree species from AVIRIS imagery using canonical discriminant analysis","volume":"4","author":"Alonzo","year":"2013","journal-title":"Remote Sens. Lett."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"2101","DOI":"10.1890\/0012-9658(2001)082[2101:DADOTS]2.0.CO;2","article-title":"Dominance and distribution of tree species in upper Amazonian terra firme forests","volume":"82","author":"Pitman","year":"2001","journal-title":"Ecology"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1300\/J091v20n03_03","article-title":"A survey of small-scale farmers using trees in pastures in Herrera Province, Panama","volume":"20","author":"Love","year":"2005","journal-title":"J. Sustain. For."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"37","DOI":"10.1080\/10549811.2011.571589","article-title":"The structure and composition of a tropical dry forest landscape after land clearance; Azuero peninsula, Panama","volume":"30","author":"Griscom","year":"2011","journal-title":"J. Sustain. For."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"687","DOI":"10.1142\/S0218001409007326","article-title":"Classification of imbalanced data: A review","volume":"23","author":"Sun","year":"2009","journal-title":"Int. J. Pattern Recognit. Artif. Intell."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1016\/j.rse.2004.06.017","article-title":"Toward intelligent training of supervised image classifications: Directing training data acquisition for SVM classification","volume":"93","author":"Foody","year":"2004","journal-title":"Remote Sens. Environ."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"258","DOI":"10.1016\/j.rse.2012.03.013","article-title":"Tree species classification in the Southern Alps based on the fusion of very high geometrical resolution multispectral\/hyperspectral images and lidar data","volume":"123","author":"Dalponte","year":"2012","journal-title":"Remote Sens. Environ."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"214","DOI":"10.1016\/j.rse.2012.07.010","article-title":"Mapping tree species composition in South African savannas using an integrated airborne spectral and lidar system","volume":"125","author":"Cho","year":"2012","journal-title":"Remote Sens. Environ."},{"key":"ref_25","first-page":"90","article-title":"Using multiple Landsat scenes in an ensemble classifier reduces classification error in a stable nearshore environment","volume":"28","author":"Knudby","year":"2014","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"139","DOI":"10.1016\/j.rse.2012.08.030","article-title":"Comparing endmember selection techniques for accurate mapping of plant species and land cover using imaging spectrometer data","volume":"127","author":"Roth","year":"2012","journal-title":"Remote Sens. Environ."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"2841","DOI":"10.1016\/j.rse.2010.07.002","article-title":"Assessing the utility of airborne hyperspectral and lidar data for species distribution mapping in the coastal Pacific Northwest, Canada","volume":"114","author":"Jones","year":"2010","journal-title":"Remote Sens. Environ."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"1820","DOI":"10.3390\/rs4061820","article-title":"Species-Level Differences in Hyperspectral Metrics among Tropical Rainforest Trees as Determined by a Tree-Based Classifier","volume":"4","author":"Clark","year":"2012","journal-title":"Remote Sens."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"3462","DOI":"10.3390\/rs4113462","article-title":"Mapping savanna tree species at ecosystem scales using support vector machine classification and BRDF correction on airborne hyperspectral and lidar data","volume":"4","author":"Colgan","year":"2012","journal-title":"Remote Sens."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"1778","DOI":"10.1109\/TGRS.2004.831865","article-title":"Classification of hyperspectral remote sensing images with support vector machines","volume":"42","author":"Melgani","year":"2004","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"1007","DOI":"10.1080\/01431160512331314083","article-title":"Support vector machines for classification in remote sensing","volume":"26","author":"Pal","year":"2005","journal-title":"Int. J. Remote Sens."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"78","DOI":"10.1016\/j.isprsjprs.2012.04.001","article-title":"Comparison of support vector machine, neural network, and CART algorithms for the land-cover classification using limited training data points","volume":"70","author":"Shao","year":"2012","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1263","DOI":"10.1109\/TKDE.2008.239","article-title":"Learning from Imbalanced Data","volume":"21","author":"He","year":"2009","journal-title":"IEEE Trans. Knowl. Data Eng."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1186\/1471-2105-11-523","article-title":"Class prediction for high-dimensional class-imbalanced data","volume":"11","author":"Blagus","year":"2010","journal-title":"BMC Bioinform."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"13","DOI":"10.1093\/bib\/bbs006","article-title":"Class-imbalanced classifiers for high-dimensional data","volume":"14","author":"Lin","year":"2013","journal-title":"Brief. Bioinform."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"505","DOI":"10.1007\/s10980-008-9215-x","article-title":"On the accuracy of landscape pattern analysis using remote sensing data","volume":"23","author":"Shao","year":"2008","journal-title":"Landsc. Ecol."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1325","DOI":"10.1007\/s00442-014-2986-7","article-title":"Pasture succession in the Neotropics: Extending the nucleation hypothesis into a matrix discontinuity hypothesis","volume":"175","author":"Peterson","year":"2014","journal-title":"Oecologia"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"933","DOI":"10.1641\/0006-3568(2001)051[0933:TEOTWA]2.0.CO;2","article-title":"Terrestrial Ecoregions of the World: A New Map of Life on Earth","volume":"51","author":"Olson","year":"2001","journal-title":"Bioscience"},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"454","DOI":"10.1016\/j.rse.2012.06.012","article-title":"Carnegie Airborne Observatory-2: Increasing science data dimensionality via high-fidelity multi-sensor fusion","volume":"124","author":"Asner","year":"2012","journal-title":"Remote Sens. Environ."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"15","DOI":"10.1016\/j.rse.2014.11.011","article-title":"Quantifying forest canopy traits: Imaging spectroscopy versus field survey","volume":"158","author":"Asner","year":"2015","journal-title":"Remote Sens. Environ."},{"key":"ref_41","unstructured":"Misc functions of the Department of Statistics, Probability Theory Group (Formally: E1071), TU Wein 2015. Available online: https:\/\/cran.r-project.org\/web\/packages\/e1071\/index.html."},{"key":"ref_42","unstructured":"R Foundation for Statistical Computing: A Language and Environment for Statistical Computing. Available online: https:\/\/www.R-project.org."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"273","DOI":"10.1007\/BF00994018","article-title":"Support-vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"397","DOI":"10.1016\/j.neucom.2009.08.006","article-title":"Margin calibration in SVM class-imbalanced learning","volume":"73","author":"Yang","year":"2009","journal-title":"Neurocomputing"},{"key":"ref_45","first-page":"352","article-title":"A kernel functions analysis for support vector machines for land cover classification","volume":"11","author":"Kavzoglu","year":"2009","journal-title":"Int. J. Appl. Earth Obs. Geoinf."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"3735","DOI":"10.1016\/j.csda.2009.04.009","article-title":"Estimating classification error rate: Repeated cross-validation, repeated hold-out and bootstrap","volume":"53","author":"Kim","year":"2009","journal-title":"Comput. Stat. Data Anal."},{"key":"ref_47","doi-asserted-by":"crossref","first-page":"427","DOI":"10.1016\/j.ipm.2009.03.002","article-title":"A systematic analysis of performance measures for classification tasks","volume":"45","author":"Sokolova","year":"2009","journal-title":"Inf. Process. Manag."},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"35","DOI":"10.1016\/0034-4257(91)90048-B","article-title":"A review of assessing the accuracy of classifications of remotely sensed data","volume":"37","author":"Congalton","year":"1991","journal-title":"Remote Sens. Environ."},{"key":"ref_49","doi-asserted-by":"crossref","first-page":"281","DOI":"10.1109\/TSMCB.2008.2002909","article-title":"SVMs modeling for highly imbalanced classification","volume":"39","author":"Tang","year":"2009","journal-title":"IEEE Trans. Syst. Man. Cybern. B. Cybern."},{"key":"ref_50","unstructured":"Conrad, O. Module Watershed Segmentation. Available online: http:\/\/www.saga-gis.org\/saga_module_doc\/2.1.3\/imagery_segmentation_0.html."},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"122","DOI":"10.1016\/j.rse.2012.10.031","article-title":"Making better use of accuracy data in land change studies: Estimating accuracy and area and quantifying uncertainty using strati field estimation","volume":"129","author":"Olofsson","year":"2013","journal-title":"Remote Sens. Environ."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"42","DOI":"10.1016\/j.rse.2014.02.015","article-title":"Good practices for estimating area and assessing accuracy of land change","volume":"148","author":"Olofsson","year":"2014","journal-title":"Remote Sens. Environ."},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"135","DOI":"10.1016\/j.rse.2015.05.007","article-title":"Differentiating plant species within and across diverse ecosystems with imaging spectroscopy","volume":"167","author":"Roth","year":"2015","journal-title":"Remote Sens. Environ."},{"key":"ref_54","unstructured":"Chen, X., Gerlach, B., and Casasent, D. (August, January 31). Pruning support vectors for imbalanced data classification. Proceedings of International Joint Conference on Neural Networks, Montreal, QC, Canada."},{"key":"ref_55","doi-asserted-by":"crossref","first-page":"2501","DOI":"10.1109\/JSTARS.2014.2346475","article-title":"Single-species detection with airborne imaging spectroscopy data: A comparison of support vector techniques","volume":"8","author":"Baldeck","year":"2014","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_56","first-page":"61","article-title":"Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods","volume":"10","author":"Platt","year":"1999","journal-title":"Adv. Large Margin Classif."},{"key":"ref_57","first-page":"211","article-title":"Sparse Bayesian learning and the relevance vector machine","volume":"1","author":"Tipping","year":"2001","journal-title":"J. Mach. Learn. Res."},{"key":"ref_58","unstructured":"Rasmussen, C.E. (2004). Advanced Lectures on Machine Learning, Springer Berlin Heidelberg."},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"202","DOI":"10.1016\/j.rse.2013.01.016","article-title":"V Estimating area from an accuracy assessment error matrix","volume":"132","author":"Stehman","year":"2013","journal-title":"Remote Sens. Environ."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/8\/2\/161\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T19:19:23Z","timestamp":1760210363000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/8\/2\/161"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2016,2,19]]},"references-count":59,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2016,2]]}},"alternative-id":["rs8020161"],"URL":"https:\/\/doi.org\/10.3390\/rs8020161","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2016,2,19]]}}}