2
Supplement S1 Additional Methods
Mosquito sampling with sweeping nets
We chose sweeping nets as a method to sample adult mosquitoes in order to make our results comparable with a previous study done at Mt Konpira in 1989 (Zea Iriarte et al. 1991; Chaves 2016). This method has the advantage of being economic, since only a sweeping net is necessary, and is appropriate to sample mosquitoes in the tribe Aedini (Culicidae: Culicinae), the tribe to which the three mosquito species in our study belong (Tanaka et al. 1979). Aedes spp mosquitoes are active during daytime, the time when we collected our samples (Silver 2008). Sweeping nets also reduce the possibility of damage to adult specimens when compared to traps that operate with fan suction, which can on occasion damage the mosquitoes as they are sampled into a sampling cup (Miyagi & Toma 1980) or when using backpack aspirators (Edman et al. 1998). We randomized the order of the sampling locations each time in order to avoid any potential systematic sampling bias (Chaves et al. 2015). From our previous experience in Okinawa (Hoshi et al. 2014), we know net sweeping might be unable to catch other mosquito species, especially when compared with automatic light traps operated over a longer time period. In Okinawa light traps were able to collect mosquito species from other tribes belonging to the sub-family Culicinae, as well as, and from the sub-family Anophelinae (Hoshi et al. 2014). Nevertheless, several studies have shown net sweeping is appropriate to sample resting mosquitoes around focal points, as well as, mosquitoes active at the time when the sampling is performed (Silver 2008). Regarding our study setting, we know that mosquito biodiversity in Nagasaki for light traps set at 1.5m above the ground and sweeping nets has shown no differences regarding the composition of species caught (Zea Iriarte et al. 1991; Tsuda et al. 2003).
Mosquito Identification
Mosquitoes were identified using the taxonomic key by Tanaka et al (1979). Damaged specimens, whose morphological identification was impossible, were identified with the cytochrome c oxidase subunit I (COI) DNA barcoding protocol described by Taira et al (2012). Voucher specimens were deposited in the Entomological Collection of Nagasaki University Institute of Tropical Medicine, Japan and in the Mosquito Collection in the Walter Reed Biosystematics Unit – Smithsonian Institute, Washington DC, USA.
Model Selection and Diagnostics
In all cases, i.e., spatial and temporal models, “full” models for each Aedes spp. were simplified by a process of backward elimination, i.e., by removing covariates whose exclusion minimized the Akaike Information Criterion (AIC) in steps where models with the same number of parameters were compared, ending model selection, with a “best” model, when further removal of covariates increased the AIC by more than 2 units (Kuhn & Johnson 2013).
For the spatial models, we tested the assumption of spatial independence by estimating Moran’s I index of spatial autocorrelation from residuals of the “best” model. The null hypothesis of this statistic is that of spatial independence, i.e., when the estimated Moran’s I index is not different from what can be expected by random (Brunsdon & Comber 2015). For the temporal models, we examined the ACF of model residuals, which is not expected to be different from random when temporal autocorrelation is adequately modeled (Shumway & Stoffer 2011).
References
Brunsdon, C. & Comber, L. (2015). An introduction to R for spatial analysis and mapping. Sage Publications LTD., London.
Chaves, L.F. (2016). Climate change and the biology of insect vectors of human pathogens. In: Invertebrates and Global Climate Change (eds. Johnson, S & Jones, H). Wiley Chichester, UK, p. In Press.
Chaves, L.F., Imanishi, N. & Hoshi, T. (2015). Population dynamics of Armigeres subalbatus (Diptera: Culicidae) across a temperate altitudinal gradient. Bulletin of Entomological Research, 105, 589-597.
Edman, J.D., Scott, T.W., Costero, A., Morrison, A.C., Harrington, L.C. & Clark, G.G. (1998). Aedes aegypti (Diptera : Culicidae) movement influenced by availability of oviposition sites. Journal of Medical Entomology, 35, 578-583.
Hoshi, T., Imanishi, N., Higa, Y. & Chaves, L.F. (2014). Mosquito Biodiversity Patterns Around Urban Environments in South-Central Okinawa Island, Japan. Journal of the American Mosquito Control Association, 30, 260-267.
Kuhn, M. & Johnson, K. (2013). Applied Predictive Modeling. Springer, New York.
Miyagi, I. & Toma, T. (1980). Studies on the mosquitoes in Yaeyama Islands, Japan : 5. Notes on the mosquitoes collected in forest areas of Iriomotejima. Japanese Journal of Sanitary Zoology, 31, 81-91.
Shumway, R.H. & Stoffer, D.S. (2011). Time series analysis and its applications. 3rd edn. New York: Springer.
Silver, J.B. (2008). Mosquito ecology: field sampling methods. 3rd edn. Springer, New York.
Taira, K., Toma, T., Tamashiro, M. & Miyagi, I. (2012). DNA barcoding for identification of mosquitoes (Diptera: Culicidae) from the Ryukyu Archipelago, Japan. Medical Entomology and Zoology, 63, 289-306.
Tanaka, K., Mizusawa, K. & Saugstad, E.S. (1979). A revision of the adult and larval mosquitoes of Japan (including the Ryukyu Archipelago and the Ogasawara Islands) and Korea (Diptera: Culicidae). Contributions of the American Entomological Institute 16, 1-987.
Tsuda, Y., Maekawa, Y., Saita, S., Hasegawa, M. & Takagi, M. (2003). Dry ice-trap collection of mosquitoes flying near a tree canopy in Nagasaki, Japan, with special reference to Aedes albopictus (Skuse) and Culex pipiens pallens Coquillett (Diptera : Culicidae). Medical Entomology and Zoology, 54, 325-330.
Zea Iriarte, W.L., Tsuda, Y., Wada, Y. & Takagi, M. (1991). Distribution of mosquitoes on a hill of Nagasaki city, with emphasis to the distance from human dwellings. Tropical Medicine, 33, 55-60.