Persistence of residual Anopheles gambiae populations associated with frontier human settlements and nocturnal livelihood-related activities at the fringes of a large conservation area in southern Tanzania

Following the rapid scale-up of long-lasting insecticidal nets (LLINs) in 2008 across the Kilombero Valley of southern Tanzania, Anopheles gambiae Giles, a highly efficient and human-specialized malaria vector, essentially disappeared within two years and has rarely been detected since. However, an ecological study of its sibling species within the An. gambiae complex, namely An. arabiensis Patton and An. quadriannulatus Theobald, in an area spanning human settlements and conserved wilderness, nevertheless detected sparse, highly focal populations of An. gambiae. Out of 4,704 An. gambiae complex specimens collected, only 10 were identified as nominate An. gambiae by polymerase chain reaction. Seven of these were captured at three surveyed fishing camps inside a Wildlife Management Area (WMA), while none were caught in five established villages west of the WMA or the national park to the east. Two of the other three were found among scattered farming settlements along the edge of the WMA. The final individual was caught in a near-natural area deep inside the WMA, but even this site was near small transient homesteads. Poisson regression modelling confirmed that capture rates at fishing camps were higher than in the peripheral settlements (Relative Rate [RR] [95% Confidence Interval (CI)] = 0.088 [0.011, 0.695], P = 0.021) or conserved natural areas (RR [95% CI] = 0.043 [0.003, 0.495], P = 0.012). Conservative estimates put the mean standing populations at each of the three fishing camps between 26 and 41 females per settlement. The absence of An. gambiae from well-conserved areas without resident humans aligns with its anthropophagic nature and dependency on human hosts. Residual An. gambiae populations persisting among frontier communities appear linked to locally common, livelihood-related nocturnal activities that preclude the practical use of LLINs, specifically fishing and guarding crops against wildlife. The continued presence of this previously important species in these ecological niches, where local livelihoods apparently undermine high LLIN coverage, shows persistence under sustained, partial insecticide pressure. This could lead to the re-emergence of this vector through the gradual evolution of resistance. However, their sparse and focal distribution suggests that new supplementary interventions, such as transfluthrin emanators, might effectively eliminate these remaining refuge populations.