{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T04:03:59Z","timestamp":1774584239363,"version":"3.50.1"},"reference-count":68,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2021,4,13]],"date-time":"2021-04-13T00:00:00Z","timestamp":1618272000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Brain Korea 21 Project","award":["4199990114242"],"award-info":[{"award-number":["4199990114242"]}]},{"DOI":"10.13039\/501100003654","name":"Korea Environmental Industry and Technology Institute","doi-asserted-by":"publisher","award":["2017000210001"],"award-info":[{"award-number":["2017000210001"]}],"id":[{"id":"10.13039\/501100003654","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>Species distribution models have been used for various purposes, such as conserving species, discovering potential habitats, and obtaining evolutionary insights by predicting species occurrence. Many statistical and machine-learning-based approaches have been proposed to construct effective species distribution models, but with limited success due to spatial biases in presences and imbalanced presence-absences. We propose a novel species distribution model to address these problems based on bootstrap aggregating (bagging) ensembles of deep neural networks (DNNs). We first generate bootstraps considering presence-absence data on spatial balance to alleviate the bias problem. Then we construct DNNs using environmental data from presence and absence locations, and finally combine these into an ensemble model using three voting methods to improve prediction accuracy. Extensive experiments verified the proposed model\u2019s effectiveness for species in South Korea using crowdsourced observations that have spatial biases. The proposed model achieved more accurate and robust prediction results than the current best practice models.<\/jats:p>","DOI":"10.3390\/rs13081495","type":"journal-article","created":{"date-parts":[[2021,4,13]],"date-time":"2021-04-13T22:55:09Z","timestamp":1618354509000},"page":"1495","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":36,"title":["A Robust Prediction Model for Species Distribution Using Bagging Ensembles with Deep Neural Networks"],"prefix":"10.3390","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3952-7582","authenticated-orcid":false,"given":"Jehyeok","family":"Rew","sequence":"first","affiliation":[{"name":"School of Electrical Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, Korea"}]},{"given":"Yongjang","family":"Cho","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, Korea"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0418-4092","authenticated-orcid":false,"given":"Eenjun","family":"Hwang","sequence":"additional","affiliation":[{"name":"School of Electrical Engineering, Korea University, 145 Anam-ro, Seongbuk-gu, Seoul 02841, Korea"}]}],"member":"1968","published-online":{"date-parts":[[2021,4,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1016\/j.biocon.2012.01.068","article-title":"Global food security, biodiversity conservation and the future of agricultural intensification","volume":"151","author":"Tscharntke","year":"2012","journal-title":"Biol. Conserv."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"345","DOI":"10.1016\/j.tree.2010.01.007","article-title":"Global pollinator declines: Trends, impacts and drivers","volume":"25","author":"Potts","year":"2010","journal-title":"Trends Ecol. Evol."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"89","DOI":"10.1046\/j.1472-4642.2003.00012.x","article-title":"Global amphibian declines: Sorting the hypotheses","volume":"9","author":"Collins","year":"2003","journal-title":"Divers. Distrib."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"251","DOI":"10.3197\/096327197776679077","article-title":"Biodiversity as the source of biological resources: A new look at biodiversity values","volume":"6","author":"Wood","year":"1997","journal-title":"Environ. Values"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"163","DOI":"10.1086\/262021","article-title":"Valuing biodiversity for use in pharmaceutical research","volume":"104","author":"Simpson","year":"1996","journal-title":"J. Political Econ."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1164","DOI":"10.1126\/science.1187512","article-title":"Global biodiversity: Indicators of recent declines","volume":"328","author":"Butchart","year":"2010","journal-title":"Science"},{"key":"ref_7","unstructured":"Almond, R., Grooten, M., and Peterson, T. (2020). Living Planet Report 2020\u2014Bending the Curve of Biodiversity Loss, World Wildlife Fund."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"607","DOI":"10.2307\/1313420","article-title":"Quantifying threats to imperiled species in the United States","volume":"48","author":"Wilcove","year":"1998","journal-title":"BioScience"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.jeem.2012.02.002","article-title":"Endangered species conservation on private land: Assessing the effectiveness of habitat conservation plans","volume":"64","author":"Langpap","year":"2012","journal-title":"J. Environ. Econ. Manag."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/S0048-9697(99)00315-0","article-title":"Endangered species mitigation banking: Promoting recovery through habitat conservation planning under the Endangered Species Act","volume":"240","author":"Bonnie","year":"1999","journal-title":"Sci. Total Environ."},{"key":"ref_11","unstructured":"Elith, J. (2006). Quantitative Methods for Modeling Species Habitat: Comparative Performance and an Application to Australian Plants, Springer."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"21","DOI":"10.2981\/0909-6396(2007)13[21:AMFETH]2.0.CO;2","article-title":"A model for evaluating the \u2018habitat potential\u2019 of a landscape for capercaillie Tetrao urogallus: A tool for conservation planning","volume":"13","author":"Braunisch","year":"2007","journal-title":"Wildl. Biol."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"11","DOI":"10.1016\/S0169-2046(00)00095-5","article-title":"Multiple-scale habitat modeling approach for rare plant conservation","volume":"51","author":"Wu","year":"2000","journal-title":"Landsc. Urban Plan."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"59","DOI":"10.3391\/ai.2012.7.1.007","article-title":"Ensemble forecasting of potential habitat for three invasive fishes","volume":"7","author":"Poulos","year":"2012","journal-title":"Aquat. Invasions"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"694","DOI":"10.1111\/2041-210X.12200","article-title":"SDMtoolbox: A python-based GIS toolkit for landscape genetic, biogeographic and species distribution model analyses","volume":"5","author":"Brown","year":"2014","journal-title":"Methods Ecol. Evol."},{"key":"ref_16","first-page":"73","article-title":"Equilibrium or not? Modelling potential distribution of invasive species in different stages of invasion","volume":"18","author":"Meentemeyer","year":"2011","journal-title":"Divers. Distrib."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"2261","DOI":"10.1016\/j.ecolmodel.2010.04.018","article-title":"Comparison of alternative strategies for invasive species distribution modeling","volume":"221","author":"Robinson","year":"2010","journal-title":"Ecol. Model."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"727","DOI":"10.1111\/j.2007.0906-7590.05041.x","article-title":"A null-model for significance testing of presence-only species distribution models","volume":"30","author":"Raes","year":"2007","journal-title":"Ecography"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1016\/S0304-3800(02)00199-0","article-title":"Predicting species spatial distributions using presence-only data: A case study of native New Zealand ferns","volume":"157","author":"Zaniewski","year":"2002","journal-title":"Ecol. Model."},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"410","DOI":"10.1111\/j.1365-2664.2009.01765.x","article-title":"Ground validation of presence-only modelling with rare species: A case study on barbastelles Barbastella barbastellus (Chiroptera: Vespertilionidae)","volume":"47","author":"Rebelo","year":"2010","journal-title":"J. Appl. Ecol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1111\/j.2041-210X.2011.00141.x","article-title":"Comparative interpretation of count, presence-absence and point methods for species distribution models","volume":"3","author":"Aarts","year":"2012","journal-title":"Methods Ecol. Evol."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"66","DOI":"10.1111\/j.1600-0587.2008.05505.x","article-title":"Do they? How do they? WHY do they differ? On finding reasons for differing performances of species distribution models","volume":"32","author":"Elith","year":"2009","journal-title":"Ecography"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"921","DOI":"10.1046\/j.1365-2664.2001.00647.x","article-title":"Evaluating presence-absence models in ecology: The need to account for prevalence","volume":"38","author":"Manel","year":"2001","journal-title":"J. Appl. Ecol."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Duan, R.-Y., Kong, X.-Q., Huang, M.-Y., Fan, W.-Y., and Wang, Z.-G. (2014). The predictive performance and stability of six species distribution models. PLoS ONE, 9.","DOI":"10.1371\/journal.pone.0112764"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Mungu\u00eda, M., Rahbek, C., Rangel, T.F., Diniz-Filho, J.A.F., and Ara\u00fajo, M.B. (2012). Equilibrium of global amphibian species distributions with climate. PLoS ONE, 7.","DOI":"10.1371\/journal.pone.0034420"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"469","DOI":"10.1111\/j.1466-822X.2004.00090.x","article-title":"Bioclimate envelope models: What they detect and what they hide","volume":"13","author":"Hampe","year":"2004","journal-title":"Glob. Ecol. Biogeogr."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"1343","DOI":"10.1016\/j.ecolmodel.2011.01.018","article-title":"What does ecological modelling model? A proposed classification of ecological niche models based on their under-lying methods","volume":"222","author":"Sillero","year":"2011","journal-title":"Ecol. Model."},{"key":"ref_28","doi-asserted-by":"crossref","first-page":"1810","DOI":"10.1016\/j.ecolmodel.2011.02.011","article-title":"The crucial role of the accessible area in ecological niche modeling and species distribution modeling","volume":"222","author":"Barve","year":"2011","journal-title":"Ecol. Model."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"2176","DOI":"10.1111\/jbi.13033","article-title":"Applying species distribution modelling to a data poor, pelagic fish complex: The ocean sunfishes","volume":"44","author":"Phillips","year":"2017","journal-title":"J. Biogeogr."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"71","DOI":"10.3354\/meps09391","article-title":"Species distribution modelling of marine benthos: A North Sea case study","volume":"442","author":"Reiss","year":"2011","journal-title":"Mar. Ecol. Prog. Ser."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"993","DOI":"10.1111\/j.1461-0248.2005.00792.x","article-title":"Predicting species distribution: Offering more than simple habitat models","volume":"8","author":"Guisan","year":"2005","journal-title":"Ecol. Lett."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"1400","DOI":"10.1016\/j.biocon.2008.03.018","article-title":"Applying species distribution modelling for the conservation of the threatened saproxylic Stag Beetle (Lucanus cervus)","volume":"141","author":"Thomaes","year":"2008","journal-title":"Biol. Conserv."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"3178","DOI":"10.1890\/0012-9658(2000)081[3178:CARTAP]2.0.CO;2","article-title":"Classification and regression trees: A powerful yet simple technique for ecological data analysis","volume":"81","author":"Fabricius","year":"2000","journal-title":"Ecology"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"243","DOI":"10.1890\/0012-9658(2007)88[243:BTFEMA]2.0.CO;2","article-title":"Boosted trees for ecological modeling and prediction","volume":"88","year":"2007","journal-title":"Ecology"},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1016\/j.ecolmodel.2005.11.005","article-title":"Genetic algorithms for optimisation of predictive ecosystems models based on decision trees and neural networks","volume":"195","author":"Goethals","year":"2006","journal-title":"Ecol. Model."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"144","DOI":"10.1016\/j.biocon.2013.07.037","article-title":"Statistical solutions for error and bias in global citizen science datasets","volume":"173","author":"Bird","year":"2014","journal-title":"Biol. Conserv."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1139","DOI":"10.1111\/ddi.12477","article-title":"What determines spatial bias in citizen science? Exploring four recording schemes with different proficiency requirements","volume":"22","author":"Geldmann","year":"2016","journal-title":"Divers. Distrib."},{"key":"ref_38","doi-asserted-by":"crossref","unstructured":"Rademaker, M., Hogeweg, L., and Vos, R. (2019). Modelling the niches of wild and domesticated Ungulate species using deep learning. bioRxiv, 744441.","DOI":"10.1101\/744441"},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Botella, C., Joly, A., Bonnet, P., Monestiez, P., and Munoz, F. (2018). A Deep Learning Approach to Species Distribution Modelling, Springer.","DOI":"10.1007\/978-3-319-76445-0_10"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"101137","DOI":"10.1016\/j.ecoinf.2020.101137","article-title":"Effects of sample size and network depth on a deep learning approach to species distribution modeling","volume":"60","author":"Benkendorf","year":"2020","journal-title":"Ecol. Inform."},{"key":"ref_41","unstructured":"(2020, November 22). GBIF Homepage. Available online: https:\/\/www.gbif.org."},{"key":"ref_42","unstructured":"(2020, November 22). VertNet Homepage. Available online: http:\/\/vertnet.org."},{"key":"ref_43","unstructured":"(2020, November 22). BISON Homepage, Available online: https:\/\/bison.usgs.gov."},{"key":"ref_44","unstructured":"(2020, November 22). Naturing Homepage. Available online: https:\/\/www.naturing.net."},{"key":"ref_45","unstructured":"GBIF.org (2021, April 12). GBIF Occurrence Download. Available online: https:\/\/bit.ly\/3a0rwZ2."},{"key":"ref_46","unstructured":"GBIF.org (2021, April 12). GBIF Occurrence Download. Available online: https:\/\/bit.ly\/3sjPW6l."},{"key":"ref_47","unstructured":"GBIF.org (2021, April 12). GBIF Occurrence Download. Available online: https:\/\/bit.ly\/3s8726R."},{"key":"ref_48","unstructured":"GBIF.org (2021, April 12). GBIF Occurrence Download. Available online: https:\/\/bit.ly\/2PV798Q."},{"key":"ref_49","unstructured":"GBIF.org (2021, April 12). GBIF Occurrence Download. Available online: https:\/\/bit.ly\/3wOD6jO."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"773","DOI":"10.1111\/j.0906-7590.2006.04700.x","article-title":"The effect of sample size and species characteristics on performance of different species distribution modeling methods","volume":"29","author":"Hernandez","year":"2006","journal-title":"Ecography"},{"key":"ref_51","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/S0304-3800(01)00388-X","article-title":"Effects of sample size on accuracy of species distribution models","volume":"148","author":"Stockwell","year":"2002","journal-title":"Ecol. Model."},{"key":"ref_52","doi-asserted-by":"crossref","first-page":"541","DOI":"10.1111\/ecog.01132","article-title":"spThin: An R package for spatial thinning of species occurrence records for use in ecological niche models","volume":"38","author":"Boria","year":"2015","journal-title":"Ecography"},{"key":"ref_53","doi-asserted-by":"crossref","first-page":"4302","DOI":"10.1002\/joc.5086","article-title":"WorldClim 2: New 1-km spatial resolution climate surfaces for global land areas","volume":"37","author":"Fick","year":"2017","journal-title":"Int. J. Climatol."},{"key":"ref_54","unstructured":"Arino, O., Perez, J.R., Kalogirou, V., Bontemps, S., Defourny, P., and van Bogaert, E. (2012). Global Land Cover Map for 2009 (GlobCover 2009), Universit\u00e9 Catholique de Louvain (UCL). European Space Agency (ESA)."},{"key":"ref_55","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1111\/j.1600-0587.2013.00205.x","article-title":"Where is positional uncertainty a problem for species distribution modelling?","volume":"37","author":"Naimi","year":"2013","journal-title":"Ecography"},{"key":"ref_56","doi-asserted-by":"crossref","first-page":"327","DOI":"10.1111\/j.2041-210X.2011.00172.x","article-title":"Selecting pseudo-absences for species distribution models: How, where and how many?","volume":"3","author":"Jiguet","year":"2012","journal-title":"Methods Ecol. Evol."},{"key":"ref_57","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1016\/j.ecolmodel.2015.05.018","article-title":"A framework for species distribution modelling with improved pseudo-absence generation","volume":"312","author":"Iturbide","year":"2015","journal-title":"Ecol. Model."},{"key":"ref_58","doi-asserted-by":"crossref","first-page":"478","DOI":"10.1016\/j.ecolmodel.2007.08.010","article-title":"Assessing the effects of pseudo-absences on predictive distribution model performance","volume":"210","author":"Chefaoui","year":"2008","journal-title":"Ecol. Model."},{"key":"ref_59","doi-asserted-by":"crossref","first-page":"122","DOI":"10.32614\/RJ-2018-019","article-title":"Tackling Uncertainties of Species Distribution Model Projections with Package mopa","volume":"10","author":"Iturbide","year":"2018","journal-title":"R J."},{"key":"ref_60","doi-asserted-by":"crossref","unstructured":"Chernick, M. (2007). Bootstrap Methods: A Guide for Researchers and Practitioners, Wiley.","DOI":"10.1002\/9780470192573"},{"key":"ref_61","doi-asserted-by":"crossref","unstructured":"Jung, S., Moon, J., Park, S., Rho, S., Baik, S.W., and Hwang, E. (2020). Bagging ensemble of multilayer perceptrons for missing electricity consumption data imputation. Sensors, 20.","DOI":"10.3390\/s20061772"},{"key":"ref_62","unstructured":"Canty, A.J. (2021, April 12). Resampling Methods in R: The Boot Package. The Newsletter of the R Project, December 2002, Volume 2\/3. Available online: http:\/\/cran.fhcrc.org\/doc\/Rnews\/Rnews_2002-3.pdf."},{"key":"ref_63","doi-asserted-by":"crossref","unstructured":"Rew, J., Cho, Y., Moon, J., and Hwang, E. (2020). Habitat Suitability Estimation Using a Two-Stage Ensemble Approach. Remote Sens., 12.","DOI":"10.3390\/rs12091475"},{"key":"ref_64","doi-asserted-by":"crossref","first-page":"369","DOI":"10.1111\/j.1600-0587.2008.05742.x","article-title":"BIOMOD\u2014A platform for ensemble forecasting of species distributions","volume":"32","author":"Thuiller","year":"2009","journal-title":"Ecography"},{"key":"ref_65","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"ref_66","first-page":"162","article-title":"Distribution, breeding status, and conservation of the black-faced spoonbill (Platalea minor) in South Korea","volume":"12","author":"Kang","year":"2016","journal-title":"For. Sci. Technol."},{"key":"ref_67","first-page":"1","article-title":"Home range and movement of juvenile black-faced spoonbill Platalea minor in South Korea","volume":"41","author":"Kang","year":"2017","journal-title":"J. Ecol. Environ."},{"key":"ref_68","doi-asserted-by":"crossref","first-page":"263","DOI":"10.1111\/j.0021-8901.2004.00881.x","article-title":"An improved approach for predicting the distribution of rare and endangered species from occurrence and pseudo-absence data","volume":"41","author":"Engler","year":"2004","journal-title":"J. Appl. Ecol."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/13\/8\/1495\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T05:47:41Z","timestamp":1760161661000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/13\/8\/1495"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,4,13]]},"references-count":68,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2021,4]]}},"alternative-id":["rs13081495"],"URL":"https:\/\/doi.org\/10.3390\/rs13081495","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,4,13]]}}}