Estimated effects of projected climate change on the basic reproductive number of the Lyme disease vector Ixodes scapularis

Abstract

Background: The extent to which climate change may affect human health by increasing risk from vector-borne diseases has been under considerable debate.

Objectives: We quantified potential effects of future climate change on the basic reproduction number (R0) of the tick vector of Lyme disease, Ixodes scapularis, and explored their importance for Lyme disease risk, and for vector-borne diseases in general.

Methods: We applied observed temperature data for North America and projected temperatures using regional climate models to drive an I. scapularis population model to hindcast recent, and project future, effects of climate warming on R0. Modeled R0 increases were compared with R0 ranges for pathogens and parasites associated with variations in key ecological and epidemiological factors (obtained by literature review) to assess their epidemiological importance.

Results: R0 for I. scapularis in North America increased during the years 1971-2010 in spatio-temporal patterns consistent with observations. Increased temperatures due to projected climate change increased R0 by factors (2-5 times in Canada and 1.5-2 times in the United States), comparable to observed ranges of R0 for pathogens and parasites due to variations in strains, geographic locations, epidemics, host and vector densities, and control efforts.

Conclusions: Climate warming may have co-driven the emergence of Lyme disease in northeastern North America, and in the future may drive substantial disease spread into new geographic regions and increase tick-borne disease risk where climate is currently suitable. Our findings highlight the potential for climate change to have profound effects on vectors and vector-borne diseases, and the need to refocus efforts to understand these effects.

Figure 2

Maps of values of R₀ estimated from ANUSPLIN observations (1971–2000; A) and projected climate obtained from the CRCM4.2.3 driven by CGCM3.1 T47, following the SRES A2 GHG emission scenario for 2011–2040 (B) and 2041–2070 (C). The color scale indicates R₀ values. Within the zones where R₀ of I. scapularis is > 1, geographic occurrence of Lyme disease risk is also limited by other environmental variables (Diuk-Wasser et al. 2012).

Figure 1

Mean values for R₀ of the tick I. scapularis obtained in tick model simulations using observed temperature data (ANUSPLIN: 1971-2010), and projected temperature data obtained from the RCM CRCM4.2.3 [according to Special Report on Emissions Scenarios (SRES) A2 emissions scenario] for (A) Southern Ontario, (B) Huron Ontario, (C) Upper Southern Ontario, (D) South-Western Quebec, (E) the Boreal region of central Ontario and Quebec, (F) Old Lyme (Connecticut), and (G) Fort McCoy (Wisconsin). The black arrows in each panel reference the first identification of Lyme disease in the United States (Wood and Lafferty 2013). The green arrows indicate the year of first field detection of I. scapularis populations within the Canadian clusters. (A) In Southern Ontario, these dates are 1976 for Long Point (Watson and Anderson 1976), 1996 for Point Pelee (Lindsay et al. 1999a), 1999 for Rondeau Park (Morshed et al. 2003), and 2001 for Turkey Point (Scott et al. 2004). (D) The date is 2007 for a number of sites in South-Western Quebec (Ogden et al. 2008); the estimated numbers of Census Subdivisions (CSDs) with established I. scapularis populations in South-Western Quebec, based on passive surveillance data (Leighton et al. 2012), is shown as the green dashed line. The range of R₀ values produced in simulations for 2020–2069 of CRCM4.2.3 and five other GCMs and RCMs is indicated by the error bar to the right of each panel except for the U.S. sites (F,G), for which only output from CRCM4.2.3 was available. Full details of all simulations are presented in Supplemental Material, Figure S6.