Abstract. The spatial distribution of Dermacentor variabilis, the most commonly identified vector of the bacterium Rickettsia rickettsii which causes Rocky Mountain spotted fever (RMSF) in humans, and the spatial distribution of RMSF, have not been previously studied in the south central United States of America, particularly in Texas. From an epidemiological perspective, one would tend to hypothesise that there would be a high degree of spatial concordance between the habitat suitability for the tick and the incidence of the disease. Both maximum-entropy modelling of the tick's habitat suitability and spatially adaptive filters modelling of the human incidence of RMSF disease provide reliable portrayals of the spatial distributions of these phenomenons. Even though rates of human cases of RMSF in Texas and rates of Dermacentor ticks infected with Rickettsia bacteria are both relatively low in Texas, the best data currently available allows a preliminary indication that the assumption of high levels of spatial concordance would not be correct in Texas (Kappa coefficient of agreement = 0.17). It will take substantially more data to provide conclusive findings, and to understand the results reported here, but this study provides an approach to begin understanding the discrepancy.

The spatial distribution of Ixodes scapularis, the most common tick vector of the bacterium Borrelia burgdorferi, the cause of Lyme disease in humans, has not been studied previously in Texas, United States of America. It has only rarely been reported in this state, so its local, spatial relationship to the distribution of this disease is unknown. From an epidemiological perspective, one would tend to hypothesise that there should be a high degree of spatial concordance between habitat suitability for the tick and incidence of the disease. Both maximum-entropy modelling of the tick's habitat probability and modelling of human incidence of Lyme disease using spatially adaptive filters provide reliable portrayals of the spatial distributions of these phenomena. Even though rates of human cases of Lyme disease as well as rates of Ixodes ticks infected with Borrelia bacteria are both relatively low in Texas, the best data currently available indicate that the assumption of high levels of spatial concordance would not be correct in Texas (Kappa coefficient of agreement = 0.039). It will take substantially more data to provide conclusive findings and to understand the results reported here, but this study provides an approach to begin understanding the discrepancy.

No abstract