Risk maps for range expansion of the Lyme disease vector, Ixodes scapularis, in Canada now and with climate change

Abstract

Background: Lyme disease is the commonest vector-borne zoonosis in the temperate world, and an emerging infectious disease in Canada due to expansion of the geographic range of the tick vector Ixodes scapularis. Studies suggest that climate change will accelerate Lyme disease emergence by enhancing climatic suitability for I. scapularis. Risk maps will help to meet the public health challenge of Lyme disease by allowing targeting of surveillance and intervention activities.

Results: A risk map for possible Lyme endemicity was created using a simple risk algorithm for occurrence of I. scapularis populations. The algorithm was calculated for each census sub-division in central and eastern Canada from interpolated output of a temperature-driven simulation model of I. scapularis populations and an index of tick immigration. The latter was calculated from estimates of tick dispersion distances by migratory birds and recent knowledge of the current geographic range of endemic I. scapularis populations. The index of tick immigration closely predicted passive surveillance data on I. scapularis occurrence, and the risk algorithm was a significant predictor of the occurrence of I. scapularis populations in a prospective field study. Risk maps for I. scapularis occurrence in Canada under future projected climate (in the 2020s, 2050s and 2080s) were produced using temperature output from the Canadian Coupled Global Climate Model 2 with greenhouse gas emission scenario enforcing 'A2' of the Intergovernmental Panel on Climate Change.

Conclusion: We have prepared risk maps for the occurrence of I. scapularis in eastern and central Canada under current and future projected climate. Validation of the risk maps provides some confidence that they provide a useful first step in predicting the occurrence of I. scapularis populations, and directing public health objectives in minimizing risk from Lyme disease. Further field studies are needed, however, to continue validation and refinement of the risk maps.

Figure 1

ROC analysis of risk algorithms. ROC graph of the relationship between sensitivity and 1-specificity for detection of known or suspected I. scapularis populations in Canada using Algorithm 6 (graph a: risk index = number of ticks at model equilibrium × number of tick populations within 425 km) and Algorithm 12 (graph b: risk index = number of ticks at model equilibrium × number of tick populations within 425 km × percentage forest cover).

Figure 2

Risk maps for the occurrence of the Lyme disease vector Ixodes scapularis. Expansion of I. scapularis-affected CSDs in Canada from the present (using 1971–2000 temperature normals) to the 2080s (using the temperature conditions predicted by the CGCM2 climate model under emissions scenario A2). In Figs a to d, the 'slow' scenario, the model assumes that by the end of each time period, only risk CSDs with an algorithm value in the top 10% will contain an I. scapularis population. In Figs e to h, the 'fast' scenario, the model assumes that by the end of each time period, all CSDs within the 'moderate' risk zone for I. scapularis establishment (risk CSDs) contain an I. scapularis population. For both scenarios, the time steps are 2000 to 2019, 2020 to 2049, 2050 to 2079 and 2080 to 2109. The 'high' risk regions for I. scapularis population establishment are indicated in red, the 'moderate' risk regions are in orange, the 'low' risk regions are in yellow, regions with no risk of established populations but some risk from bird-borne 'adventitious' ticks are in green, and regions with no predicted risk of either are colourless.

Figure 3

The outcome of field validation of risk maps. A map of southern Quebec showing the locations (unfilled or blue-filled circles) of CSDs in which field study sites were visited. The 'index of certainty' for the presence of an I. scapularis population was calculated from the abundance of ticks and the numbers of instars discovered during the field visit. The value 0 indicated that no I. scapularis ticks were found. The 'high' risk regions for I. scapularis population establishment are indicated in red, the 'moderate' risk regions are in orange, the 'low' risk regions are in yellow, and regions with no risk of established populations but some risk from bird-borne 'adventitious' ticks are in green

Figure 4

The relationship between the risk algorithm on which the risk maps were based, and the index of certainty that a site contained a reproducing I. scapularis population.