Vector-borne pathogen transmission cycles minimally consist of an arthropod vector, a vertebrate host, and a pathogen, but many are zoonotic and transmitted among a complex array of vectors and vertebrate hosts (e.g., West Nile virus; see Figure 3-2). For most zoonotic arboviruses, transmission to humans and to

9Center for Vectorborne Diseases, School of Veterinary Medicine, University of California, Old Davis Road, Davis, CA 95616, Reisen E-mail: [email protected]; Barker E-mail: [email protected]



FIGURE 3-2 West Nile virus transmission cycle. SOURCE: CDC (2005).

some extent domestic animals causes disease but is a "dead end" for the virus. Surveillance data on arthropod vectors or infection in reservoir hosts typically are skeletal, often leaving the passive detection of human or veterinary illness as the only consistent measure of pathogen activity. However, the diverse spectrum of clinical symptoms frequently makes syndromic surveillance difficult, and for many zoonoses, symptomatic individuals represent only a small proportion of the infected population, making them an insensitive measure of pathogen activity.

Regardless of the intensity of surveillance or transmission cycle complexity, pathogen dynamics are directly affected by climate at a variety of spatial and temporal scales. Long-term surveillance programs by control agencies provide one of the few measures of vector populations suitable for assessing the impact of climate variation on vector-pathogen-host systems. A detailed understanding of these climate-health relationships is the first step toward developing models and forecasting risk, which then can be assessed by measuring ecosystem and pathogen transmission dynamics. Risk forecasts are extremely useful in intervention programs charged with mitigating pathogen amplification and protecting the public using preventive methods, whereas measures of risk in real time form an integral part of decision support systems.

In this paper, we explore how climate variation impacts the transmission dynamics of vector-borne disease using California's mosquito-borne encephalitis virus surveillance and control program as an example. The California program


provides an excellent model because (1) the state encompasses multiple biomes that vary markedly across north-south latitudinal and east-west elevational gradients; (2) an intensive surveillance program has been consistently monitoring mosquito abundance and encephalitis virus activity for more than 50 years; and (3) there is a statewide decision support system, including a response plan, that uses surveillance data to estimate risk and recommend appropriate levels of control.

Defining Climate Variation: Importance of Scale

Climate encompasses a variety of meteorological parameters, including temperature and wetness, which normally are averaged over a defined time period to delineate "average" conditions for a specific geographic region. Climate variation describes deviations about these long-term means that may be measured at a variety of scales from days to years, whereas climate change is directional and consists of long-term shifts in means over decades to centuries. Carefully monitoring climate variation and understanding its potential impact on ecosystem dynamics provides an important tool for forecasting vector-borne pathogen transmission. Models capturing several climate parameters have provided estimates of hydro-logic conditions that were related to outbreaks of mosquito-borne encephalitis in Florida (Shaman et al., 2002, 2004). These models are less useful, however, when vectors exploit anthropogenic water sources in urban or agricultural ecosystems. Other indices measure biological parameters directly, such as the Normalized Difference Vegetation Index, which uses remotely sensed reflectance to estimate the vigor and density of live green vegetation (Tucker, 1979) as a surrogate for other biotic factors influencing vector populations. The value of raw data from satellite and ground sensors is enhanced through additional processing such as NASA's Terrestrial Observation and Prediction System (TOPS; see Figure 3-3), a modeling framework that integrates and preprocesses data so that land surface models can be run in near real time (Nemani et al., 2003). These models use ground and satellite instruments to measure various water (evaporation, transpiration, stream flows, and soil moisture), carbon (net photosynthesis, plant growth), and nutrient (uptake and mineralization) processes at a variety of spatial scales, from global net primary productivity (NPP) anomalies at 0.5 x 0.5-degree resolution to local estimates of ecosystem parameters at resolutions as fine as 250 m. At each spatial resolution, TOPS uses different sources of satellite data (Moderate Resolution Imaging Spectroradiometer [MODIS] to Ikonos) and meteorological data (single weather station to global atmospheric model outputs).

Once average climate conditions have been established, deviations or anomalies can be tracked at varying scales. Short-term changes (weather) can be forecast days to weeks in advance and predict events such as rainstorms or heat waves that may immediately affect vector-borne pathogen transmission. Interannual variation—driven by global cycles such as El Niño—may be used to forecast ecosystem change seasons in advance and therefore forecast changes in vector

Was this article helpful?

0 0

Post a comment