TY - JOUR

T1 - Wildlife habitat association over a twelve-year period (2008-2020) in the Greater Mahale Ecosystem, western Tanzania

AU - Thomsen, Simon

AU - Loos, Jacqueline

AU - Stewart, Fiona A.

AU - Piel, Alex K.

N1 - AKP and FAS would like to thank the Tanzania Wildlife Research Institute (TAWIRI), Commission for Science and Technology (COSTECH), Tanzania National Parks (TANAPA), and Tanganyika, Uvinza, and Kigoma District governments for research permission and support during the fieldwork. A special thanks to GMERC field assistants for their assistance during data collection, especially Kiumbe Ahmedi, Matsumbuku Ahmedi, Mashaka Alimas, Justina Boniface, Patrick Hassan, Busoti Juma, Mtunda Kapanda, Joffrey and Shedrack Lucas, and Msigwa Rashid. Finally, the UCSD/ Salk Center for Academic Research and Training in Anthropogony (CARTA) provided long-term support to GMERC. We would like to acknowledge the Arcus Foundation, Lilian Pintea and the Jane Goodall Institute, Matt Brown and The Nature Conservancy, and Kathryn Doody and Magnus Mosha from Frankfurt Zoological Society for supporting, facilitating, or funding this work. JL thanks the Robert-Bosch Foundation for funding the project “Wildlife, Values, Justice: Reconciling Sustainability of African Protected Areas.” ST thanks Pascal Fust for comments on the analysis.

PY - 2023/10/1

Y1 - 2023/10/1

N2 - The Greater Mahale Ecosystem (GME) in western Tanzania hosts high faunal biodiversity, including charismatic species such as elephants and chimpanzees. Across the GME, habitat occurs in differently administered, unprotected and protected areas, including village land, forest reserves, and national parks. Areas of lower protective status are prone to deforestation and land cover changes. These land cover changes influence suitable habitat itself as well as the connectivity between habitat patches, but the impact on different mammal species in this region is unknown. In this study, we investigated changes in habitat availability by creating species distribution models for 11 medium-large mammal species based on species occurrence data collected from 2008 to 2020. We tested habitat associations for each species and evaluated the importance of the static predictor variables elevation, slope, aspect and terrain-ruggedness and the dynamic predictor variables distance to deforestation, percent forest, vegetation intensity, and vegetation variability, retrieved through remotely sensed data analysis. Our models suggest habitat declined for most, but not all species. Whilst elephants and buffalo lost more than 50 % of available habitat during this period, primates exhibited habitat stability. Habitat losses were pronounced in the southeast of the GME. Slope, elevation and median tasseled-cap wetness index (vegetation intensity) and distance to deforestation were the most important predictor variables. Our study presents a spatially and temporally explicit estimation of habitat changes in a critically important, biodiverse region undergoing drastic land cover changes. We call for consideration in land-use planning to foster connectivity and landscape integrity to protect and conserve wildlife across the ecosystem.

AB - The Greater Mahale Ecosystem (GME) in western Tanzania hosts high faunal biodiversity, including charismatic species such as elephants and chimpanzees. Across the GME, habitat occurs in differently administered, unprotected and protected areas, including village land, forest reserves, and national parks. Areas of lower protective status are prone to deforestation and land cover changes. These land cover changes influence suitable habitat itself as well as the connectivity between habitat patches, but the impact on different mammal species in this region is unknown. In this study, we investigated changes in habitat availability by creating species distribution models for 11 medium-large mammal species based on species occurrence data collected from 2008 to 2020. We tested habitat associations for each species and evaluated the importance of the static predictor variables elevation, slope, aspect and terrain-ruggedness and the dynamic predictor variables distance to deforestation, percent forest, vegetation intensity, and vegetation variability, retrieved through remotely sensed data analysis. Our models suggest habitat declined for most, but not all species. Whilst elephants and buffalo lost more than 50 % of available habitat during this period, primates exhibited habitat stability. Habitat losses were pronounced in the southeast of the GME. Slope, elevation and median tasseled-cap wetness index (vegetation intensity) and distance to deforestation were the most important predictor variables. Our study presents a spatially and temporally explicit estimation of habitat changes in a critically important, biodiverse region undergoing drastic land cover changes. We call for consideration in land-use planning to foster connectivity and landscape integrity to protect and conserve wildlife across the ecosystem.

KW - Species distribution modelling

KW - Habitat suitability dynamics

KW - Spatial modelling

KW - Mammalian conservation

KW - African fauna

KW - Biology

KW - Ecosystems Research

UR - http://www.scopus.com/inward/record.url?scp=85166316426&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/dce5bae4-5db9-3e90-a149-7909d50b4718/

U2 - 10.1016/j.jnc.2023.126464

DO - 10.1016/j.jnc.2023.126464

M3 - Journal articles

VL - 75

JO - Journal for Nature Conservation

JF - Journal for Nature Conservation

SN - 1617-1381

M1 - 126464

ER -