Recent decline in suitable environmental conditions for African great apes

Abstract

Aim: To predict the distribution of suitable environmental conditions (SEC) for eight African great ape taxa for a first time period, the 1990s and then project it to a second time period, the 2000s; to assess the relative importance of factors influencing SEC distribution and to estimate rates of SEC loss, isolation and fragmentation over the last two decades. Location: Twenty-two African great ape range countries. Methods: We extracted 15,051 presence localities collected between 1995 and 2010 from 68 different areas surveyed across the African ape range. We combined a maximum entropy algorithm and logistic regression to relate ape presence information to environmental and human impact variables from the 1990s with a resolution of 5 × 5 km across the entire ape range. We then made SEC projections for the 2000s using updated human impact variables. Results: Total SEC area was approximately 2,015,480 and 1,807,653 km2 in the 1990s and 2000s, respectively. Loss of predicted SEC appeared highest for Cross River gorillas (-59%), followed by eastern gorillas (-52%), western gorillas (-32%), bonobos (-29%), central chimpanzees (-17%) and western chimpanzees (-11%). SEC for Nigeria-Cameroon chimpanzees and eastern chimpanzees was not greatly reduced. Except for Cross River and eastern gorillas, the number of SEC patches did not change significantly, suggesting that SEC loss was caused mainly by patch size reduction. Main conclusions: The first continent-wide perspective of African ape SEC distribution shows dramatic declines in recent years. The model has clear limitations for use at small geographic scales, given the quality of available data and the coarse resolution of predictions. However, at the large scale it has potential for informing international policymaking, mitigation of resource extraction and infrastructure development, as well as for spatial prioritization of conservation effort and evaluating conservation effectiveness.Additional co-authors: Head, J.,Huijbregts, B., Lindsell, J., McLennan, M., Martinez, L., Morgan, D., N'Goran K.P., Ntongho, A., Petre, C.A., Regnaut, S., Sanz, C., Tondossama, A

Landsat ETM+ and SRTM Data Provide Near Real-Time Monitoring of Chimpanzee (Pan troglodytes) Habitats in Africa

“…Data on the spatial distribution of habitat suitability for most of the study area are from Reference [25], who used location data from the IUCN/Species Survival Commission A.P.E.S. (Ape Populations, Environments and Surveys) database (http://apes.eva.mpg.de), which serves as a central repository for all ape related survey data.…”

“…Logistic regression modelling was used in Reference [25] to develop HSMs. Logistic regression requires both a positive and negative outcome.…”

“…Grid-cells with low predicted suitability from the MaxEnt model were more likely to be used as the negative outcome in the logistic regression models. Predictor variables used in Reference [25] included bioclimatic variables from the WorldClim dataset, forest cover products derived from the Moderate Resolution Imaging Spectrometer (MODIS) and the Advanced Very High Resolution Radiometer (AVHRR) sensors, human population density, proximity to roads and proximity to navigable rivers.…”

“…Location data covering the dry forests and miombo woodlands of western Tanzania were not included in the APES database; therefore these important chimpanzee habitats were not reflected in the output from Reference [25]. In order to include these habitats in our model, we incorporated a habitat suitability model created by the authors of [21] specifically for this region based on nest locations ( Figure 2B).…”

Assessment of Chimpanzee Nest Detectability in Drone-Acquired Images

“…Several studies have documented the impact of habitat loss [2][3][4], poaching [5][6][7] and infectious disease [8,9] on wild populations. In Tanzania, 75% woodland vegetation than the closed riverine forest.…”