2018 ESA Annual Meeting (August 5 -- 10)

PS 70-231 - Land cover change as a determinant of house infestation with Chagas disease vectors in the Argentine Chaco

Friday, August 10, 2018
ESA Exhibit Hall, New Orleans Ernest N. Morial Convention Center
Lucía I. Rodríguez-Planes1,2, Hernán D. Argibay2, María Sol Gaspe2 and Ricardo E. Gürtler1,2, (1)Departamento de Ecología, Genética y Evolución, Universidad de Buenos Aires, CABA, Argentina, (2)Instituto de Ecología, Genética y Evolución de Buenos Aires (IEGEBA), Universidad de Buenos Aires, Consejo Nacional de Investigaciones Científicas y Técnicas, CABA, Argentina
Background/Question/Methods

The transmission of vector-borne pathogens requires the spatio-temporal coincidence of pathogens, hosts and vectors. Understanding the main determinants of vector occurrence and persistence may improve our ability to predict infestation risk and prevent pathogen transmission. Chagas disease is caused by Trypanosoma cruzi, which is transmitted between mammalian and human hosts through insects (Triatominae) in the Americas. The main vector, domestic Triatoma infestans, and a secondary peridomestic or sylvatic species (Triatoma sordida) live in sympatry and are able to transmit T. cruzi in rural houses from the Gran Chaco ecoregion. Forest fragments near houses may harbor triatomine colonies, and regional land-use change may increase or accelerate their arrival to houses. We investigated the house infestation patterns of both triatomines in a well-defined area (394 houses) of the Argentine Chaco before and after a community-wide house spraying with insecticide over 8 years. Vegetation surveys discriminated between 5 forest types, 1 short-disturbance and 1 no-vegetation class. We used satellite imagery from pre- and postspraying periods for supervised classification. The surroundings of each household were characterized at three spatial scales based on triatomine mobility. Linear models were fitted to the data to assess the relative importance of vegetation cover and its change over time.

Results/Conclusions

In total, 41% of the pixels shifted from forest fragments to short-disturbance or no vegetation over the 8-yr period. Humid forest, short-disturbance and no vegetation were the most relevant classes for house context discrimination. Vegetation near the houses at all spatial scales was intimately related to the pre-spraying pattern of house infestation with T. sordida but not T. infestans, which lacks local sylvatic foci. No-vegetation pixels within a 100 m radius around houses were associated with a 2.5-fold higher probability of postspraying house infestation with either species. The frequency of pixels classified as forests before insecticide spraying and as short-disturbance or no vegetation at 8-yr postspraying significantly increased the probability of house reinfestation with T. sordida but not T. infestans. Our findings suggest that the loss of forest cover (coupled with the increase of short-disturbance or no-vegetation areas) is associated with an increased risk of triatomine occurrence, especially for T. sordida, a species with sylvatic foci. This may be relevant to direct vector control efforts to houses with higher infestation risk: hotspots of forest loss. Whether habitat loss or the higher temperature related to barren soil facilitate the invasion and successful establishment of triatomines remains unknown and deserves further investigation.