Multi-scale habitat modelling and predicting change in the distribution of tiger and leopard using random forest algorithm.

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
Read More Add to Saved list
  • Additional Information
    • Source:
      Publisher: Nature Publishing Group Country of Publication: England NLM ID: 101563288 Publication Model: Electronic Cited Medium: Internet ISSN: 2045-2322 (Electronic) Linking ISSN: 20452322 NLM ISO Abbreviation: Sci Rep Subsets: MEDLINE
    • Publication Information:
      Original Publication: London : Nature Publishing Group, copyright 2011-
    • Subject Terms:
      Conservation of Natural Resources* ; Ecology*; Panthera/*physiology ; Tigers/*physiology; Algorithms ; Animals ; Cats ; Ecosystem ; Humans ; Models, Biological
    • Abstract:
      Tigers and leopards have experienced considerable declines in their population due to habitat loss and fragmentation across their historical ranges. Multi-scale habitat suitability models (HSM) can inform forest managers to aim their conservation efforts at increasing the suitable habitat for tigers by providing information regarding the scale-dependent habitat-species relationships. However the current gap of knowledge about ecological relationships driving species distribution reduces the applicability of traditional and classical statistical approaches such as generalized linear models (GLMs), or occupancy surveys to produce accurate predictive maps. This study investigates the multi-scale habitat relationships of tigers and leopards and the impacts of future climate change on their distribution using a machine-learning algorithm random forest (RF). The recent advancements in the machine-learning algorithms provide a powerful tool for building accurate predictive models of species distribution and their habitat relationships even when little ecological knowledge is available about the species. We collected species occurrence data using camera traps and indirect evidence of animal presences (scats) in the field over 2 years of rigorous sampling and used a machine-learning algorithm random forest (RF) to predict the habitat suitability maps of tiger and leopard under current and future climatic scenarios. We developed niche overlap models based on the recently developed statistical approaches to assess the patterns of niche similarity between tigers and leopards. Tiger and leopard utilized habitat resources at the broadest spatial scales (28,000 m). Our model predicted a 23% loss in the suitable habitat of tigers under the RCP 8.5 Scenario (2050). Our study of multi-scale habitat suitability modeling provides valuable information on the species habitat relationships in disturbed and human-dominated landscapes concerning two large felid species of conservation importance. These areas may act as refugee habitats for large carnivores in the future and thus should be the focus of conservation importance. This study may also provide a methodological framework for similar multi-scale and multi-species monitoring programs using robust and more accurate machine learning algorithms such as random forest.
    • References:
      PLoS Biol. 2010 Sep 14;8(9):. (PMID: 20856904)
      Oecologia. 1999 Oct;121(1):138-148. (PMID: 28307883)
      BMC Bioinformatics. 2008 Jul 11;9:307. (PMID: 18620558)
      Science. 1974 Jul 5;185(4145):27-39. (PMID: 17779277)
      Ecol Lett. 2012 Apr;15(4):365-377. (PMID: 22257223)
      BMC Bioinformatics. 2006 Jan 06;7:3. (PMID: 16398926)
      Mol Ecol. 2008 Jan;17(1):167-78. (PMID: 18173499)
      Ecol Lett. 2005 Sep;8(9):993-1009. (PMID: 34517687)
      Sci Rep. 2017 Apr 27;7(1):1213. (PMID: 28450747)
      PeerJ. 2017 Dec 5;5:e4095. (PMID: 29230356)
      Ecol Evol. 2017 Feb 08;7(5):1541-1552. (PMID: 28261463)
      Science. 1999 Aug 20;285(5431):1265-7. (PMID: 10455053)
      Proc Natl Acad Sci U S A. 2004 Apr 6;101(14):4854-8. (PMID: 15041746)
      Proc Biol Sci. 2015 Jan 22;282(1799):20141857. (PMID: 25621330)
      Ecol Lett. 2008 Nov;11(11):1135-1146. (PMID: 18713269)
      Science. 2008 Oct 10;322(5899):261-4. (PMID: 18845755)
      Trends Ecol Evol. 2008 Mar;23(3):149-58. (PMID: 18289716)
      Evolution. 2008 Nov;62(11):2868-83. (PMID: 18752605)
      Proc Natl Acad Sci U S A. 2003 Sep 30;100(20):11474-7. (PMID: 14500914)
      Evolution. 2004 Aug;58(8):1781-93. (PMID: 15446430)
      Ecol Appl. 2006 Jun;16(3):1062-75. (PMID: 16827003)
      Proc Biol Sci. 2013 Jan 7;280(1750):20121890. (PMID: 23075836)
      Syst Biol. 2018 May 01;67(3):428-438. (PMID: 29088474)
    • Publication Date:
      Date Created: 20200712 Date Completed: 20201208 Latest Revision: 20240329
    • Publication Date:
      20240329
    • Accession Number:
      PMC7351791
    • Accession Number:
      10.1038/s41598-020-68167-z
    • Accession Number:
      32651414