Conservation of the endemic species of the Albertine Rift under future climate change
Introduction
The Albertine Rift region, an ecoregion bordering the Democratic Republic of Congo, Uganda, Rwanda, Burundi and Tanzania, contains more endemic and globally threatened vertebrates than any other ecoregion on the African continent (Plumptre et al., 2007), and is part of the eastern Afromontane hotspot (Plumptre et al., 2004). Many of these species are found only on the mountains and massifs bordering Lake Albert in the north to Lake Tanganyika in the south. The Albertine Rift ranges in altitude from 600 m to 5100 m a.s.l and contains a wide variety of habitats from lowland rainforest, through medium altitude semi-deciduous rainforest, savanna grasslands and woodlands, Miombo woodland, Oxytenanthera bamboo, papyrus wetlands, Carex wetlands, montane forest, Sinarundinaria bamboo, Hagenia-Hypericum woodland, giant heather, giant Senecio and Lobelia, alpine moorland, up to bare rock at the summits of some mountains. It is this diversity of habitats which has contributed to the diversity and endemism of this region.
The conservation of the Albertine Rift started in 1925 with the establishment of Africa's first national park, the Virunga National Park in eastern Democratic Republic of Congo (Languy and de Merode, 2009). Subsequent protected areas were established in Uganda, Rwanda, Burundi and Tanzania in the 1930s and several of these were upgraded from reserves or hunting concessions to national parks in the 1950s and 1960s in Uganda and Tanzania and later in Rwanda and Burundi (Olupot et al., 2010). Much of the creation of the protected areas was aimed at protecting large mammals such as elephants (Loxodonta africana), hippopotamuses (Hippopotamus amphibius), mountain gorillas (Gorilla beringei beringei) and lions (Panthera leo), following observations of declining numbers due to human hunting (Willock, 1965). Recognition of the rich biodiversity of the region became apparent following surveys of Virunga National Park by Belgian scientists in the 1930s and 1940s (Languy and de Merode, 2009), and subsequent surveys of other sites from the 1960s onwards. More recently new forested national parks have been created such as Nyungwe National Park in Rwanda, and Bwindi Impenetrable National Park and Rwenzori Mountains National Park in Uganda, established from existing forest reserves.
The impacts of climate change on species and ecosystems are already evident in many ecosystems (Tingley et al., 2009; Pearce-Higgins et al., 2015). Even if CO2 emissions were curtailed today, it would take centuries to reverse the impacts of the accumulated greenhouse gases (Solomon et al., 2009). It is estimated that climate change could surpass habitat destruction as a leading cause of extinction in species (Leadley et al., 2010). High latitude and high elevation ecosystems, and those with abrupt land use boundaries, are particularly vulnerable to predicted climate change (Parmesan, 2006; Sala et al., 2000). Temperature decreases by 5.2–6.5 °C with each 1000 m increase in altitude on mountains in the Tropics (Colwell et al., 2008) and consequently species can move upslope to maintain the same climate envelope with global warming (Anderson et al., 2013; Root et al., 2003). However, cases have already been documented where species were pushed up to a point where they could not move higher (Pounds et al., 1999).
Current climate forecasts suggest the Albertine Rift will become warmer and wetter in the future, with greater differences between wet and dry seasons and increasing likelihood of flash floods and landslides in the September–November wet period (Seimon et al., 2011; Seimon and Picton Phillipps, 2010; Picton Phillipps and Seimon, 2010). This information, together with advances in methods to undertake species distribution modelling (Elith et al., 2006, Elith et al., 2011; Phillips et al., 2004; Phillips et al., 2006) over the past 10 years, now makes an assessment of predicted climate change impacts on species in the Albertine Rift possible.
A preliminary trait-based assessment of species likely to be impacted by climate change in the Albertine Rift was made by Carr et al. (2013). They assessed 2358 species for traits that would likely make them susceptible to climate change and found 31 amphibians (28% of those assessed), 199 birds (20%), 31 freshwater fish (6%), 107 mammals (30%), 79 plants (39%) and 70 reptiles (42%) were likely to be most vulnerable to climate change. However, their study did not model species distributions to make predictions of the impacts of climate change from changes in niche suitability. Instead they used the IUCN Red List Extent of Occurrence polygons/range maps for estimating species distributions.
Conservation planning should incorporate the impacts of future climate change if species are to be conserved in the long term (Groves et al., 2012). The narrow range of habitats on mountains in the Albertine Rift is likely to make them vulnerable to climate change and here we use species distribution modelling to predict the probable impacts on endemic mammals, birds, reptiles, amphibians and plants. We assess how well the existing protected area network conserves species in the future and also assess the impact of the recent creation of three new protected areas in eastern Democratic Republic of Congo (DR Congo).
Section snippets
Species records
We estimated the current and future distributions for 162 endemic species using location data. Species occurrence records for 119 species across 5 taxa: birds (27), mammals (18), plants (49), reptiles (11) and amphibians (14) were obtained from Wildlife Conservation Society biodiversity survey data, the Tanzanian mammal atlas, Global Biodiversity Information Facility (GBIF 2012: http://www.gbif.org), records of amphibian collections made by Mathias Behangana (Makerere University), Michele
Results of tests of models
The 162 species distribution models were sent to experts for the various taxa for examination and in a few cases were re-modelled following these assessments by using masking to produce a final set of habitat suitability models. Tests of these models in the North Balala region indicated that for the 18 endemic birds and 14 endemic plant species that were detected (with several individuals across the altitude gradient), the accuracy of the models in predicting presence were 98% sensitivity and
Modelling constraints
Species distribution modelling under future climate change scenarios has its critics (Faye et al., 2014) because of unknown changes in species realized niches that may occur with climate change (Guisan and Thuiller, 2005). Niche models try to predict how species change but depend on coarse climate models that have been downscaled to a 1km2 resolution using altitude data primarily (Hijmans et al., 2005). Care must be taken when applying such modelling techniques and we accept that our models
Data accessibility information
Data will be made available on the WCS web site-www.albertinerift.org and accessible by request at [email protected] once the paper is publishedand is also available as part of the data in brief article associated with this paper (Ayebare et al., 2018).
Acknowledgements
We are grateful to the John D. and Catherine T. MacArthur Foundation (Grant: 97201) for funding this work together with the US Fish and Wildlife Service (Grant: F13AP00680), US Agency for International Development/CAFEC program (AID-660-A-13-00010-64788) and Wildlife Conservation Society (WCS). We are also grateful to Eli Greenbaum (University of Texas at El Paso), Julian Kerbis Peterhans and John Bates (Chicago Field Museum) and the Charles Foley and Tim Davenport (Tanzania Mammal Atlas) for
References (51)
- et al.
Conservation in areas of high population density in sub-saharan Africa
Biol. Conserv.
(2007) - et al.
A comparison of the performance of threshold criteria for binary classification in terms of predicted prevalence and kappa
Ecol. Model.
(2008) - et al.
Incorporating climate change into spatial conservation prioritisation: a review
Biol. Conserv.
(2016) - et al.
Maximum entropy modeling of species geographic distributions
Ecol. Model.
(2006) - et al.
The biodiversity of the Albertine Rift
Biol. Conserv.
(2007) - et al.
Use and misuse of the IUCN red list criteria in projecting climate change impacts on biodiversity
Glob. Chang. Biol.
(2006) - et al.
Preventing species extinctions resulting from climate change
Nat. Clim. Chang.
(2014) - et al.
Current analogues of future climate indicate the likely response of a sensitive montane tropical avifauna to a warming world
PLoS One
(2013) - et al.
Extent of biodiversity surveys and ranges for endemic species in the Albertine rift
Biological Conservation
(2018) - et al.
Vital But Vulnerable: Climate Change Vulnerability and Human Use of Wildlife in Africa's Albertine Rift. Occasional Paper of the IUCN species Survival Commission No. 48
(2013)
Global warming, elevational range shifts, and lowland biotic attrition in the wet tropics
Science
Novel methods improve prediction of species' distributions from occurrence data
Ecography
A statistical explanation of MaxEnt for ecologists
Divers. Distrib.
GlobCover Land Cover (v2.3). European Space Agency (ESA)
Strong discrepancies between local temperature mapping and interpolated climatic grids in tropical mountainous agricultural landscapes
PLoS One
Upslope migration of Andean trees
J. Biogeogr.
A review of methods for the assessment of prediction errors in conservation presence/absence models
Environ. Conserv.
A comparison of methods for mapping species ranges and species richness
Glob. Ecol. Biogeogr.
Incorporating climate change into systematic conservation planning
Biodivers. Conserv.
Predicting species distribution: offering more than simple habitat models
Ecol. Lett.
Predicting species distributions for conservation decisions
Ecol. Lett.
Very high resolution interpolated climate surfaces for global land areas
Int. J. Climatol.
Emissions Scenarios. A Special Report of Working Group III of the Intergovernmental Panel on Climate Change
Climate Change 2007: The Physical Science Basis
Cited by (32)
Seed fate, seedling establishment and the role of propagule size in forest regeneration under climate change conditions
2022, Forest Ecology and ManagementCitation Excerpt :To test the former hypotheses, we focused on two Mexican oak species (Quercus, Fagaceae) that produce recalcitrant seeds with elevated intraspecific variability in their biomass and performed field experiments at sites where air temperature is predicted to increase and rainfall is predicted to decrease over the next 20 years (period 2021–2040). We considered this time period for our experiments because studies modeling the future distribution of plant species predict that their distribution ranges will strongly decline in the long term (e.g., 2080–2100) under dramatical changes in temperature and rainfall (Ayebare et al., 2018; Schipper et al., 2020), but little is known about the near-term consequences of climate change on their recruitment. We sowed different sized oak seeds under the current climate and simulated climate change conditions to answer four main questions, including (1) will climate change alter post-dispersal fates of oak seeds (i.e., their chances of getting attacked by parasites, desiccating or germinating)?
Impact of Pteridium aquilinum on vegetation in Nyungwe Forest, Rwanda
2020, HeliyonCitation Excerpt :Bracken fern dominance seen as a perturbation, similar to secondary forests, leads to a change in plant distributions patterns (Schneider, 2006). Nyungwe Forest belonging to the afromontane forest and the Albertine Rift region (Plumptre et al., 2002, 2006; Ayebare et al., 2018), the afromontane (AME) and Albertine Rift (ARE) species were supposed to dominate in areas sampled but in this study, those species were decreased gradually from the primary forest to the natural regenerations after removal of fern and after fires. In the primary forest, 40 % and 25 % of total species belonged to AME and ARE respectively while in the secondary forest of lowland of Mubuga site, they were 16% and 14% of total species respectively (Table 6 and Figure 7).