Show simple item record

Development of a species distribution model using machine learning methods

dc.creatorPavlović, Lazar
dc.creatorStojanović, Dejan B.
dc.creatorKresoja, Milena
dc.creatorStjepanović, Stefan
dc.creatorOrlović, Saša
dc.creatorBojović, Mirjana
dc.date.accessioned2023-04-07T14:12:49Z
dc.date.available2023-04-07T14:12:49Z
dc.date.issued2017
dc.identifier.issn0563-9034
dc.identifier.urihttps://redun.educons.edu.rs/handle/123456789/306
dc.description.abstractKlimatske promene koje se intenzivno dešavaju u poslednjih nekoliko decenija imaju globalni efekat na vegetaciju i šumski pokrivač, što dovodi do velikih transformacija u prirodnim resursima i strukturi pejzaža. Uticaj klimatskih promena na vrste često se procenjuje korištenjem modela distribucije vrste (SDMs). Ovi modeli koriste podatke o životnoj sredini i prisustvu/odsustvu neke vrste, utvrđuju njihov međusobni odnos, te na drugim lokacijama pokazuju da li su uslovi sredine pogodni ili ne za postojanje te vrste. Pošto se modeli lako implementiraju, oni se danas koriste u velikoj meri za razmatranje različitih pitanja u istraživanju životne sredine, kao i za pružanje smernica za primenjena istraživanja. Cilj ovog rada je razviti i oceniti Random Forest (RF) model zasnovan na trenutnim podacima o rasprostranjenju šuma evropske bukve, ekoloških i klimatskih karakteristika na teritoriji Srbije. Dobijeni model će poslužiti kao osnova za izgradnju modela koji će predvideti distribucije vrste u budućnosti.Tačnost modela je ispitana upotrebom adekvatnih statističkih metoda. Analiza True Skill Statistic (TSS) ukazuje na veliku tačnost modela (TSS = 0.87, specifičnost =87.81, senzitivnost =99.44). Tačnost je potvrđena analizom površine ispod ROC (Receiver Operating Characteristic) krive (AUC) (AUC=0.97, specifičnost =88.01, senzitivnost=99.27). Takođe, rezultati ukazuju na potrebu za uključivanjem više ekološki relevantnih topografskih varijabli prilikom projektovanja modela distribucije vrsta u odnosu na klimatske promene, naročito za vrste koje su u korelaciji sa topografijom, odnosno visinskom raspodelom.sr
dc.description.abstractClimate change that has been intensively occurring in the last few decades has a global effect on vegetation and forest cover, leading to major transformations in natural resources and the landscape structure. The impact of climate changes on species is often estimated using a species distribution models (SDMs). These models use environmental data and presence/absence of a species, determine their mutual relationship, in order to show on other locations whether environmental conditions are suitable or not for the existence of this species. Since models are easy to implement, they are now widely used to consider various issues in environmental research, as well as providing guidance for applied research. The aim of this paper is to develop and evaluate the Random Forest (RF) model based on current data on existence of European beech, ecological and climatic characteristics in the territory of Serbia. The model obtained will serve as the basis for building a model that will foresee the distribution of species in the future. The accuracy of the model was tested using adequate statistical methods. The True Skill Statistic (TSS) analysis indicates a high accuracy of the model (TSS = 0.87, specificity = 87.81, sensitivity = 99.44). The accuracy was confirmed by the analysis of the area under the ROC (Receiver Operating Characteristic) curve (AUC) (AUC = 0.97, specificity = 88.01, sensitivity = 99.27). Also, the results pointed to the need to include more environmentally relevant topographic variables when designing a SDM in relation to climate change, especially for species that are correlated with topography.en
dc.publisherUniverzitet u Novom Sadu - Institut za nizijsko šumarstvo i životnu sredinu, Novi Sad
dc.relationinfo:eu-repo/grantAgreement/MESTD/Integrated and Interdisciplinary Research (IIR or III)/43007/RS//
dc.rightsopenAccess
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.sourceTopola
dc.subjectmodeli distribucije vrstasr
dc.subjectmašinsko učenjesr
dc.subjectEvropska bukvasr
dc.subjectBIOMOD2sr
dc.subjectspecies distribution models - SDMsen
dc.subjectmachine learningen
dc.subjectEuropean beechen
dc.subjectBIOMOD2en
dc.titleRazvoj modela potencijalne distribucije vrsta pomoću metoda mašinskog učenjasr
dc.titleDevelopment of a species distribution model using machine learning methodsen
dc.typearticle
dc.rights.licenseBY
dc.citation.epage175
dc.citation.issue199-200
dc.citation.other(199-200): 167-175
dc.citation.rankM51
dc.citation.spage167
dc.identifier.fulltexthttp://redun.educons.edu.rs/bitstream/id/122/303.pdf
dc.identifier.rcubconv_303
dc.type.versionpublishedVersion


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record