Effects of refuge availability on the population dynamics of Triatoma infestans in central Argentina

1. As part of research on the control and mathematical modelling of transmission of Chagas disease, we tested the joint effects of refuge quantity and quality on the population dynamics of Triatoma infestans in 12 experimental huts under natural climatic conditions in central Argentina. 2. The 2-year trial involved three types of huts that mimicked rural houses, with a gradient of refuges determined by the type of thatched roof and whether the walls were plastered or not. Each treatment comprised three replicate huts and one control hut, each of which was a closed population that housed a chicken at night and was colonized with five virgin female and three male T. infestans in mid-summer. The replicate huts were dismantled monthly to count the number of live and dead T. infestans by stage and to mark the emerged insects. Control huts were dismantled only at the end of the first and second year to count the bugs. 3. The mean abundance of live T. infestans was significantly greater in huts with maximum (final mean 1834; range 337-3630 bugs) rather than intermediate (final mean 383; range 71-692 bugs) or minimum refuges (final mean 103; range 0-103 bugs). The bug populations under minimum refuges returned to baseline levels at 1 year post-colonization and two populations went extinct. 4. Apparent female fecundity was affected significantly by refuge availability and time. Survival of the founder females until 1 year post-colonization successively decreased from 20% to 7% and 0% in huts ranging from maximum to minimum refuges. The basic reproduction number R (o) steadily decreased from 3.91 to 2.36-0.25 bugs per female bug in maximum to minimum refuge huts, respectively. 5. Synthesis and applications . Population persistence, growth rate and the maximum number of T. infestans that a hut may support depend on the physical structure for a fixed number of chickens, in agreement with the available experimental and field evidence. Environmental management measures, such as plastering of walls and wire-fencing corrals, aimed at reducing the availability of refuges and oviposition sites may exert a strong adverse impact on the population dynamics of T. infestans.