Ecological niche and potential distribution of Anopheles arabiensis in Africa in 2050

Abstract

Background: The future distribution of malaria in Africa is likely to be much more dependent on environmental conditions than the current distribution due to the effectiveness of indoor and therapeutic anti-malarial interventions, such as insecticide-treated nets (ITNs), indoor residual spraying for mosquitoes (IRS), artemisinin-combination therapy (ACT), and intermittent presumptive treatment (IPT). Future malaria epidemiology is therefore expected to be increasingly dominated by Anopheles arabiensis, which is the most abundant exophagic mosquito competent to transmit Plasmodium falciparum and exhibits a wide geographic range.

Methods: To map the potential distribution of An. arabiensis in Africa, ecological niche models were fit to 20th century collection records. Many common species distribution modelling techniques aim to discriminate species habitat from the background distribution of environments. Since these methods arguably result in unnecessarily large Type I and Type II errors, LOBAG-OC was used to identify the niche boundary using only data on An. arabiensis occurrences. The future distribution of An. arabiensis in Africa was forecasted by projecting the fit model onto maps of simulated climate change following three climate change scenarios.

Results: Ecological niche modelling revealed An. arabiensis to be a climate generalist in the sense that it can occur in most of Africa's contemporary environmental range. Under three climate change scenarios, the future distribution of An. arabiensis is expected to be reduced by 48%-61%. Map differences between baseline and projected climate suggest that habitat reductions will be especially extensive in Western and Central Africa; portions of Botswana, Namibia, and Angola in Southern Africa; and portions of Sudan, South Sudan, Somalia, and Kenya in East Africa. The East African Rift Valley and Eastern Coast of Africa are expected to remain habitable. Some modest gains in habitat are predicted at the margins of the current range in South Sudan, South Africa, and Angola.

Conclusion: In summary, these results suggest that the future potential distribution of An. arabiensis in Africa is likely to be smaller than the contemporary distribution by approximately half as a result of climate change. Agreement among the three modelling scenarios suggests that this outcome is robust to a wide range of potential climate futures.

Figure 1

Boundary estimation versus discrimative methods for ecological niche modelling. Simulated data illustrates why modelling a species’potential distribution is a problem for boundary estimation not classification. A. The left most panel represents the habitat in two environmental dimensions (e.g., precipitation and temperature) in locations at which a species is known to occur. The heavy curve depicts the true niche of the species. The dashed line is the convex hull of the sample, a naive estimate of the species niche. The black cross represents the center of the species niche, which is the most probable set of environmental conditions at which the species occurs. B. The center panel represents samples of environmental conditions at locations taken at random from the background distribution of environments. The green cross indicates the mean environment. The arrow is a vector of “niche displacement”. C. The right most panel depicts both occurrence and background data. The dashed line is the estimated optimal classification boundary between occurrence and background points. The blue-green color gradient depicts the conditional probability that a given instance is an occurrence points (blue: P (occurrence) = 1; gray: P (occurrence) = 0.5; green: P (occurrence) = 0). Inset plots illustrate the region of environmental space in which each fit model makes Type I (α) or Type II (β) errors.

Figure 2

Spatial distribution of Anopheles arabiensis. Distribution of sampling points and a balanced random sample of background points.

Figure 3

Anopheles arabiensis is found across a broad range of environments. A. Scree plot of the first ten principal components shows that a majority of the environmental variation (≈55%) may be summarized by the first two principal components. B. Points where Anopheles arabiensis has been collected represented in the space of the first two principal components of the environmental features shows that this species occupies a very large environmental range.

Figure 4

Potential distribution of Anopheles arabiensis under contemporary conditions and three global climate change scenarios. A. Modelled potential distribution of Anopheles arabiensis habitat in Africa given the current global climate. B. Future potential distribution of Anopheles arabiensis in Africa under IPCC Scenario A1B. C. Future potential distribution of Anopheles arabiensis in Africa under IPCC Scenario A2A. D. Future potential distribution of Anopheles arabiensis in Africa under IPCC Scenario B2A.

Figure 5

Summary of the difference in current and projected total habitable area of Anopheles arabiensis. Current distribution of Anopheles arabiensis habitat in Africa compared with the total land area of Africa and potential distribution under three climate change scenarios. Overplotted quantities are percent habitat loss from baseline.

Figure 6

Differences between the current and projected distribution of Anopheles arabiensis. A. Losses and gains of Anopheles arabiensis habitat in Africa under future climate scenario A1B compared with the current distribution. B. Losses and gains of Anopheles arabiensis habitat in Africa under future climate scenario A2A compared with the current distribution. C. Losses and gains of Anopheles arabiensis habitat in Africa under future climate scenario B2A compared with the current distribution.

Figure 7

Projected distribution of Anopheles arabiensis is robust to variations in climate change scenario. A. Number of scenarios in which Anopheles arabiensis habitat is predicted to be lost or gained (grey: no scenario predicts habitat; orange: habitat predicted under one climate change scenario; light green: habitat predicted under two climate change scenarios; green: habitat predicted under all three climate change scenarios). B. Universal agreement among three climate change scenarios that An. arabiensis habitat will be gained. C. Universal agreement among three climate change scenarios that An. arabiensis habitat will be lost.