CONCEPTS & SYNTHESIS
EMPHASIZING NEW IDEAS TO STIMULATE RESEARCH IN ECOLOGY
Ecology, 90(4), 2009, pp. 888–900
? 2009 by the Ecological Society of America
The ecology of climate change and infectious diseases
KEVIN D. LAFFERTY1
Western Ecological Research Center, U.S. Geological Survey, Marine Science Institute,
University of California, Santa Barbara, California 93106 USA
diseases with climate change suggests a pending societal crisis. The subject is increasingly
attracting the attention of health professionals and climate-change scientists, particularly with
respect to malaria and other vector-transmitted human diseases. The result has been the
emergence of a crisis discipline, reminiscent of the early phases of conservation biology.
Latitudinal, altitudinal, seasonal, and interannual associations between climate and disease
along with historical and experimental evidence suggest that climate, along with many other
factors, can affect infectious diseases in a nonlinear fashion. However, although the globe is
significantly warmer than it was a century ago, there is little evidence that climate change has
already favored infectious diseases. While initial projections suggested dramatic future
increases in the geographic range of infectious diseases, recent models predict range shifts in
disease distributions, with little net increase in area. Many factors can affect infectious disease,
and some may overshadow the effects of climate.
The projected global increase in the distribution and prevalence of infectious
Key words:climate; ENSO; global warming; malaria; vector; yellow fever.
In balmy medieval times, exports from English
vintners economically threatened competing French
wineries (Pfister 1988). Although English wine country
seems a quaint historical oddity, a recent headline
cheered: ‘‘Global Warming Spawns Wine in U.K.:
Changing Climate Brings New Crops to the Land of
Rain and Clouds!’’ Different grape varietals have
specific physical requirements and a string of warm
summers in southern England has again cracked open
the climate window to produce a drinkable Chardonnay,
Pinot Noir, and Pinot Meunier.
Is there a catch? Like grapes, mosquitoes and the
infectious diseases they transmit like it warm. The warm
period that favored medieval English vintners was also
malarious. For instance, Chaucer’s (1385) Canterbury
Tales tells of bouts of deadly fever, indicating endemic
malaria. It is easy, then, to imagine how climate change
might return infectious diseases like malaria to England.
The British Chief Medical Officer predicts that by 2050
the climate of England will again be suitable for endemic
malaria (Department of Health 2002). And not just
England: the conventional wisdom is that global climate
change will result in an expansion of tropical diseases,
particularly vector-transmitted diseases, throughout tem-
perate areas (Epstein 2000). Examples include schistoso-
miasis (bilharzia or snail fever), onchocerciasis (river
blindness), dengue fever, lymphatic filariasis (elephanti-
asis), African trypanosomiasis (sleeping sickness), leish-
maniasis, American trypanosomiasis (Chagas disease),
yellow fever, and many less common mosquito and tick-
transmitted diseases of humans. Obviously, nonhuman
hosts are also subject to infectious diseases, and the
concern about climate change and disease extends to
agriculture, conservation biology, and fisheries.
The extent and generality of the prediction that climate
change will increase infectious disease is an important
topic for ecologists to consider. I start this review by
introducing how temperature drives important biological
processes. I then consider how climate might affect
spatial (latitude and altitude) and temporal (seasonal,
interannual, historical) patterns in disease. Given that the
climate has warmed during the last century, I ask
whether current warming has affected infectious disease.
Finally, I review models that project how climate change
might affect infectious diseases in the future. I conclude
that while climate has affected and will continue to affect
habitat suitability for infectious diseases, climate change
seems more likely to shift than to expand the geographic
Manuscript received 15 January 2008; accepted 31 January
2008. Corresponding Editor: K. Wilson.
ranges of infectious diseases. Furthermore, many other
factors affect the distribution of infectious disease,
dampening the proposed role of climate. Finally, in
parallel with predictions for biodiversity loss, shifts in
climate suitability might actually reduce the geographic
distribution of some infectious diseases.
Warm temperatures speed up biochemical reactions
(catabolism and anabolism) that expend energy, permit-
ting increased activity, growth, development, and
reproduction. However, faster metabolism comes at a
cost because it requires higher food consumption rates
to maintain a positive energy balance. This can decrease
survivorship as temperature increases, particularly for
non-feeding free-living stages (e.g., eggs, cysts, larvae;
King and Monis 2007). For instance, the survival of
Cryptosporidium parvum cysts declines with prolonged
exposure to warm temperatures because increased
metabolism drains the energy reserves of cysts (Fayer
et al. 1998). For these reasons, the relationship between
temperature and an organism’s performance (e.g.,
growth, fitness, lifespan, reproductive output) should
follow a convex function. Climate varies on multiple
time scales (daily, seasonally, annually, and longer) and
to buffer against such changes, many ectotherms,
including most disease vectors, use behavioral thermo-
regulation. In addition, molecular mechanisms, such as
heat-shock proteins, help other proteins keep their shape
across a range of temperatures (Feder and Hofmann
1999). Furthermore, organisms might adapt to changing
climates (Bradshaw and Holzapfel 2001). For instance,
the mosquito Wyeomyia smithii uses a shortening day-
length cue to enter diapause in advance of cooling
weather; at higher latitudes, where temperature is cooler,
the shift comes earlier (at a longer day length; Bradshaw
and Holzapfel 2001). Pathogens can also locally adapt to
their environment. The lineage Ia West Nile Virus strain
from New York requires warmer temperatures for
transmission than does the lineage II strain from South
Africa (Reisen et al. 2006). So, while all species have
lower and upper temperature limits, organisms often can
deal with substantial variation in climate.
For cases where temperature extremes set boundaries
on species distributions, global warming might alter the
range (in altitude or latitude) of suitable habitat for a
species. As a result, species should be closer to their
physiological limits near the edges of their distribution
and this is where effects of climate change should be
most apparent. Global warming should initially expand
the area of potential habitat for species that presently
thrive in very warm places. In addition, polar and high
altitude species (marmots, polar bears, penguins and
their parasites) are likely to experience a net decrease in
habitat suitability with global warming. However, for
most species, global warming should result in a pole-
ward shift in both the upper and lower ranges of habitat
suitability. For example, while climate change may have
made Sweden more suitable for western-type tick-borne
encephalitis, a parallel degradation of climatic suitability
south of Sweden corresponds to a predicted Europe-
wide reduction in the range of TBE (Randolph and
Rogers 2000). Range limits are not the only aspect of
infectious disease affected by temperature. Climate may
affect the severity of disease. Most notably, climate
affects the length of the transmission season for malaria.
As the climate increases in temperature and humidity
malaria transmission can increase from zero to epidem-
ic, hypoendemic, mesoendemic, hyperendemic, and
holoendemic (Hay et al. 2004).
Indirect effects may complicate the physiological
response of a species to climate change. For instance,
a pollinator may shift in distribution directly in response
to climate and indirectly in response to the physiological
response of flowering plants to climate (Memmott et al.
2007). Because infectious diseases, like pollinators, are
parts of complex networks of species, (Lafferty et al.
2006), each with its own thermal tolerance, indirect
effects of climate on disease distribution are likely. For
this reason, much of the research on the effects of
climate change on infectious diseases concerns the
thermal tolerance of insect vectors such as mosquitoes
EVIDENCE FOR EFFECTS OF CLIMATE
IN INFECTIOUS DISEASES
The tendency for higher species diversity in the tropics
is one of the clearest patterns in biogeography. Higher
species diversity at low latitudes could result from
warmer temperatures due to energy (warmth/sunlight)
increasing the raw material on which speciation can act,
physiological tolerances biased toward warm climates,
or faster rates of speciation where high temperatures
lead to shorter generation times (Currie et al. 2004).
Because most host taxa decrease in diversity with
latitude (MacArthur 1972), a decline in parasite diversity
with latitude might occur if parasite diversity depends on
host diversity (Hechinger and Lafferty 2005). An
exception that proves the rule is that the diversity of
helminths in marine mammals does not decline with
latitude, presumably because marine mammal diversity
is low in the tropics (Rohde 1982). While these patterns
suggest that communities of parasites should track
communities of hosts, scientists are often interested in
the parasite community of a single host species,
particularly humans. The diversity of infectious diseases
of humans is higher in countries near the equator than in
countries at higher latitude (Guernier et al. 2004). Is this
a result of climate? The diversity of all disease categories
increases with the maximum range of precipitation, and
most disease categories increase with monthly temper-
ature range; however, independent of latitude, there is
surprisingly little effect of mean annual temperature or
precipitation on the diversity of human infectious
diseases (Guernier et al. 2004).
April 2009889CLIMATE CHANGE AND INFECTIOUS DISEASES
CONCEPTS & SYNTHESIS
The higher diversity of important infectious diseases
of humans in the tropics does not appear to be a result
of higher rates of evolution in the tropics. Wolfe et al.
(2007) found that infectious diseases of humans (exclud-
ing helminth parasites) were equally likely to have
originated in tropical or temperate regions. The early
humans that migrated out of Africa and into temperate
latitudes initially left several infectious diseases behind:
only one of the 10 major tropical diseases, cholera,
followed into temperate latitudes. However, 11000 years
ago, several infectious diseases of newly domesticated
temperate animals jumped to humans and most of these
novel infectious diseases subsequently spread into the
tropics (Wolfe et al. 2007).
The high diversity of infectious diseases in the tropics
could result from a high diversity of vectors. For
instance, mosquito diversity declines with latitude
(Schafer and Lundstrom 2001) and this should affect
the diversity of mosquito-transmitted diseases. Likewise,
the diversity of fleas, which are vectors for bacterial
diseases (e.g., plague, murine typhus), declines with
altitude and latitude (Marshall 1981). Blood parasites
are less prevalent in birds from the Arctic than in birds
from lower latitudes (Piersma 1997), perhaps due to
differences in vector diversity. The inability of human
tropical diseases to spread from the tropics to temperate
regions may be due to the higher fraction of tropical
diseases that have a specific vector (80% tropical vs. 13%
temperate) and/or a wild animal reservoir (80% tropical
vs. 20% temperate; Wolfe et al. 2007). In primates, the
diversity of vector-borne diseases is higher in the tropics,
though this pattern does not apply to viruses or parasitic
worms (Nunn et al. 2005). Thus, ecological understand-
ing of vectors and reservoirs may be the key to predicting
the effect of climate change on infectious diseases.
In parallel with broad patterns of diversity, many
infectious diseases show clines in prevalence with
climate. The observation that malaria is more prevalent
in lowland areas (Lindsay and Martens 1998) may be the
oldest observed spatial pattern in infectious diseases.
For instance, in Tanzania, the annual number of
mosquito bites per person drops precipitously at high
elevations because the cool climate at high altitude
impairs mosquito development rates (Bodker et al.
2003). In addition, cooler temperatures at high altitudes
slow the development of infectious agents so that they
cannot complete their life cycles. As a result, in Africa,
human settlements at high altitude are relatively free of
malaria (Lindsay and Martens 1998). Similarly, endemic
forest birds in Hawaii are able to escape introduced
avian malaria only at cool, high-altitude refuges (van
Riper et al. 2003).
The present restriction of malaria to the tropics
suggests a strong effect of climate on this disease. While
climate does affect malaria transmission, other factors
probably enforce the current distribution. Most notable
is a strong increase in per-capita gross national product
with latitude. This results in both greater surveillance
and increased funds for control and treatment in
temperate areas. Today, malarious countries have GDPs
one fifth that of non-malarious countries (Gallup and
Sachs 2001), suggesting that economic forces, particu-
larly environmental destruction, have pushed malaria
out of temperate zones. Malaria is harder to control
under the climatic conditions where it is holoendemic
(transmitted year-round) and malaria might depress
economic development in a positive feedback loop
(Bruce-Chwatt and de Zulueta 1980). In other words,
tropical climate might increase infectious diseases such
as malaria, which then depress economic growth
required for disease control (Sachs and Malaney 2002).
As early as Homer’s Iliad (800–900 BC), people were
aware of seasonal patterns in malaria (harvest fever) that
might relate to climate. Because temperature and
precipitation change with the season, seasonal variabil-
ity in disease transmission suggests an effect of climate
on diseases. Hence, we might expect that climate change
is most likely to affect diseases with seasonal patterns
(Hay et al. 1998; Fig. 1). The incidence of black spot (a
trematode metacercaria) in salmon is highest in the
warmer days of the summer season (Cairns et al. 2005),
consistent with the observation that trematodes shed
more cercarial stages from snail vectors when the water
is warm (Poulin 2006). Similarly, endemic cholera occurs
in the season when water temperature is high (Lipp et al.
Not all seasonal patterns suggest an increase in disease
with global warming. Some infectious diseases decline
during warm, wet periods. For instance, meningeal
meningitis peaks in the dry season (Moore 1992) and
houseflies infected with the Enthomophthora muscae
fungus are more common during cool periods of the
year because the fungus dies at higher temperatures
(Mullens et al. 1987). A strong link between winter and
flu in temperate climates has resulted in numerous
hypotheses, including crowding during cold weather,
physiological stress due to cold, indoor heating, and
atmospheric dispersion during cold fronts (Lofgren et al.
2007). Recent experimental evidence (see Experiments)
and outbreaks in H5N1 influenza virus in birds following
temperature drops suggest a direct positive effect of cold
temperature on the spread of influenza (Liu et al. 2007).
Seasonality in disease does not necessarily indicate an
effect of climate on disease. For instance, seasonal
variation corresponds with day length, which is impor-
tant for many biological processes, but climate change
will not alter day length (Fig. 1). Other patterns, such as
the spring rise in egg production and transmission of
trichostrongyle nematodes in domestic animals (Field et
al. 1960), relate to the seasonality of host reproduction,
not climate. The seasonal pattern of common human
diseases (measles, pertussis, chickenpox) may occur
independent of climate because the start of the school
year aggregates susceptible children (Stone et al. 2007).
KEVIN D. LAFFERTY890Ecology, Vol. 90, No. 4
CONCEPTS & SYNTHESIS
Time lags between climate and species abundances
create statistical challenges for investigating seasonal
effects on infectious disease. For instance, conditions
that provide good habitat for mosquito larvae do not
instantly lead to infectious disease transmission because
mosquito and pathogen development take time. Re-
searchers may test a variety of lags to determine the best
fit to the data. For example, Tong and Hu (2001) found
the best match between climate and Ross River virus
epidemics in Cairns, Australia occurred with a two-
month lag for rainfall and a five-month lag for
humidity. Teklehaimanot et al. (2004) explored how
the fit between malaria epidemics and climate varied
considerably with the lag chosen, and found that the lag
with the best fit was different for different climate
variables. While lag fitting may be necessary for
revealing patterns, testing many lags also increases the
chance of finding spurious patterns. For this reason,
statistical models optimized with a variety of measures
should either be corrected for multiple comparisons or,
even better, be subject to independent validation
(Peterson 2003) before concluding that climate affects
Like the weather, epidemics can vary from year to
year. Several authors have linked epidemics of mosquito-
borne diseases to the strong interannual variation in
weather associated with the Southern Oscillation, or El
Nin ˜ o (Hay et al. 2000). For instance, malaria cases
increase after El Nin ˜ o events in Venezuela (Bouma and
Dye 1997). Most interannual associations with disease
relate to precipitation. The aquatic larvae of mosquitoes
require aquatic habitats and several studies show a
positive association between heavy rain and subsequent
outbreaks of mosquito-transmitted diseases (Landesman
et al. 2007). In particular, precipitation may drive much
of the observed variation in malaria in Africa (Small et
al. 2003). However, the effect is complex; in areas with
little standing water, drought may eliminate mosquito
habitat, while in areas with flowing water (which is
unsuitable for mosquitoes), drought may create isolated
pools suitable for breeding (Landesman et al. 2007). In
the eastern United States, container-breeding mosquitoes
transmit West Nile virus, and outbreaks follow the
expected pattern, increasing after wet winters. However,
in the western United States, wetland-breeding mosqui-
toes transmit West Nile virus and outbreaks follow
drought years (Landesman et al. 2007). Why the opposite
patterns? Mosquitoes are susceptible to a range of
vertebrate and invertebrate predators found in perma-
nent water bodies. While natural and artificial containers
(important in the east) generally lack predators, wetlands
(important in the west) normally contain predators.
Periods of drought can eliminate predators from
wetlands, resulting in safe havens for mosquitoes. For
these reasons, the effect of precipitation on mosquito
populations may be strongly context dependent.
variation in maximum temperature (solid line) and rainfall (dashed line). In the four years following World War II, Poland
regularly suffered thousands of cases of malaria per year with a clear spring–summer peak (Reiter 2001). The log number of malaria
cases is almost a perfect fit to day length (R2¼ 0.97). Long-term (1961–1990) average measures of temperature (mean, minimum,
maximum) and rainfall also increase with day length, but with a one-month time lag, so that climate lags behind malaria cases. If
climate directly affected malaria transmission, one would expect human cases to lag behind the optimal temperature for
transmission (because of time needed for development of the parasite in humans and mosquitoes), not the reverse.
Seasonality in the number of cases of malaria (bars; mean 6 SD). The histogram is superimposed on plots of seasonal
April 2009 891 CLIMATE CHANGE AND INFECTIOUS DISEASES
CONCEPTS & SYNTHESIS
Historical records of disease can provide long time
series for investigating interannual associations between
climate and disease. In 1878, a deadly yellow fever
epidemic hit several cities in the United States after a
large El Nin ˜ o event. Because precipitation favors the
container-breeding yellow fever vector, Aedes aegypti,
these mosquitoes probably thrived under the unusually
warm, wet El Nin ˜ o conditions (Diaz and McCabe 1999).
Furthermore, between 1870 and 1880, the number of
cities affected by a yellow fever epidemic in a particular
year increased with the global El Nin ˜ o index (explaining
34% of the variation in epidemic frequency; Fig. 2 inset
and large solid circles). Diaz and McCabe (1999) noted
that seven of the nine major yellow fever outbreaks
(.1000 deaths) in the United States during the 19th
century coincided with an El Nin ˜ o event.
While this seems like very solid evidence for an effect
of climate on yellow fever (and, by extension, an effect of
climate change on the future of yellow fever), one cannot
establish a statistical linkage without knowledge of the
frequency of El Nin ˜ o in years without epidemics. To
determine the proportion of the variation in yellow fever
epidemics attributable to El Nin ˜ o, I analyzed data on
historical yellow fever epidemics in all U.S. cities
between 1668 and the last United States yellow fever
epidemic in 1905 (Reiter 2001). For the strength of El
Nin ˜ o, I used the estimated year’s (t) and preceding year’s
(t ? 1) cold-season NINO3 index (Mann et al. 2000). I
included date as a covariate. I compared the climate in
the years of Diaz and McCabe’s (1999) nine deadly U.S.
epidemics with non-epidemic years as controls. For the
non-epidemic controls, I considered only dates at least
three years distant from a large epidemic. I ran a similar
analysis for the number of cities experiencing an
epidemic in a particular year, and separate analyses for
epidemics in three cities: New Orleans, Philadelphia, and
Charleston, South Carolina. Finally, I plotted the
association between El Nin ˜ o and the number of cities
experiencing an epidemic from 1668 to 1905.
The analyses confirmed that the nine deadly yellow
fever epidemics were more likely to follow an El Nin ˜ o
event (t ? 1) in comparison to years not linked to an
epidemic (v2¼ 4.0, df ¼ 1, P ¼ 0.047), with El Nin ˜ o
explaining 4% of annual variation in deadly epidemics.
The frequency of epidemics also increased slightly over
time (v2¼ 6.1, df ¼ 1, P ¼ 0.014), perhaps because of
population growth. After considering the effect of El
Nin ˜ o on the number of cities experiencing an epidemic, a
similar pattern arose, except that, in this case, it was the
El Nin ˜ o of the present year (t) (T ratio2,235¼ 2.46, P ¼
0.015), not the prior year (t ? 1), that correlated with
epidemics, explaining about 2% of the variation in the
frequency of deadly epidemics (Fig. 2, small open
circles). Again, the frequency of epidemics increased
slightly over time (T ratio2,235¼ 2.48, P ¼ 0.014). For
separate analyses of New Orleans, Philadelphia, and
Charleston, there was no indication that El Nin ˜ o led to
epidemics. In summary, while some comparisons indi-
cated that climate was statistically associated with
historical yellow fever epidemics in the United States,
El Nin ˜ o did not explain a substantial proportion in the
variance in epidemics.
1870 to 1880 (solid line) and the NINO-3 index (dashed line). In the main figure, large solid circles and the dashed regression line
show the number of cities experiencing epidemics vs. the NINO-3 index from 1870 to 1880. Open circles and solid regression line
show the number of cities experiencing epidemics vs. the NINO-3 index from 1688 to 1905. The NINO-3 index goes from?1 to 1,
with positive values indicating El Nin ˜ o years.
Yellow fever outbreaks in the United States and El Nin ˜ o. The inset shows the number of cities reporting epidemics from
KEVIN D. LAFFERTY892Ecology, Vol. 90, No. 4
CONCEPTS & SYNTHESIS
Other factors besides El Nin ˜ o likely affected yellow
fever outbreaks in the United States. Susceptible hosts
build up during non-epidemic years, accumulating
enough susceptibles for an epidemic after three years
(Hay et al. 2000). The movement of infected hosts
(people or other primates) and the abundance of
previously unexposed hosts in a population probably
contributed to the timing of outbreaks (Reiter 2001).
Furthermore, after 1905, public health measures ended
the transmission of yellow fever in the United States,
despite continuing El Nin ˜ os, a pattern repeated in
history for other diseases in developed nations. The near
elimination of yellow fever from the United States under
apparently favorable climatic conditions is an example
of one of the most striking aspects of the long-term
historical record of human infectious diseases.
Hay et al. (2000) conducted a similar historical
retrospective analysis of monthly data on periodic
outbreaks of dengue fever in Bangkok, Thailand and
malaria in Kericho, Kenya. The data from Kenya are
particularly interesting given the relative lack of control
efforts in that region (permitting a clearer evaluation of
the role of climate). A clear seasonal signature and a
three-year periodicity emerged in both data sets, but this
interannual variation in incidence among years did not
correspond to changes in rainfall, temperature, or El
Nin ˜ o events.
Malaria in England also has an instructive history,
particularly with respect to understanding the relative
importance of various environmental factors on epide-
miology. While malaria appears to have been common
during the medieval warm period (1200) in Britain, and
outbreaks may have corresponded to unusually warm
years, malaria did not disappear during the little ice
age beginning in the mid-1560s (Bruce-Chwatt and de
Zulueta 1980). Declines in malaria deaths correspond to
land-use changes more than to climate. Ironically, as
malaria declined, the climate warmed and got slightly
wetter. Underscoring the old saying: ‘‘malaria flees before
the plough’’ (Najera-Morrondo 1991), malaria primarily
decreased from 1840 to 1910 with the increase in wetland
destruction, and, to a lesser extent, with the size of the
cattle population (cattle might draw mosquito bites from
humans); in comparison, warm years and rainfall only
marginally increased malaria (Kuhn et al. 2003).
While malaria decreased as the globe warmed in the
northern hemisphere, some indications suggest malaria
increased in Africa. Was this because recent climate
change increased habitat suitability for malaria in
Africa? A retrospective analysis on a causal link between
climate change and malaria in Africa from 1911 to 1995
suggests that most areas in Africa did not change in
suitability for malaria and that those areas where
suitability for malaria increased (Mozambique) were
offset by areas where suitability for malaria decreased
(e.g., the Sahel) Small et al. (2003). Therefore, other
factors besides climate change may have led to the recent
increases in malaria in Africa.
It is possible to experimentally investigate the effects
of climate variables (primarily temperature) on vital
rates of some infectious diseases and their vectors. For
instance, schistosomiasis is a tropical disease caused by
trematode parasites. Humans become infected at water
contact sites when infective cercarial stages leave the
snail and penetrate the host’s skin. Several authors have
held infected snails at different water temperatures and
counted the cercariae that emerge. Poulin (2006)
conducted a meta-analysis of how trematode cercarial
shedding rates changed with temperature, finding strong
evidence that the production of cercariae increases
dramatically with temperature. Similarly, the develop-
ment rate of angiostrongyloid nematodes (lung worms)
increases with warmer temperature (Lv et al. 2006).
However, only if increases in development rates and
productivity of parasites can outpace increases in
mortality rates will warmer temperatures lead to net
increases in transmission.
Experiments have indicated how temperature may
determine whether a potential pathogen will or will not
cause disease. Mass mortalities in Mediterranean
gorgonians correspond with warmer than average
temperatures and with bacterial infections (Romano et
al. 2000). Experiments find that increasing the water
temperature in aquaria to the level experienced by
gorgonians during die offs leads to increased tissue
mortality in infected and uninfected gorgonians, and
that tissue damage is faster when bacteria are present,
indicating that temperature both stresses gorgonians
and improves conditions for pathogenic bacteria (Bally
and Garrabou 2007). Such temperature-related impacts
of bacterial pathogens on marine invertebrates seem to
derive from a higher temperature optima for the bacteria
than for the host (Harvell et al. 2002). Another example
of temperature-dependent pathogenicity derives from
Strongyloid nematodes, which are intestinal parasites of
humans and other mammals. These worms have a
plastic life cycle; larvae (found in contaminated soil) can
either adopt an infectious or free-living mode of life.
Laboratory experiments with Strongyloides ratti dem-
onstrate that the temperature experienced by newly
hatched larvae is a key determinant in this plastic life
history; cool temperatures induce infectious larvae while
warm temperatures induce free-living worms (Minato et
al. 2008). Such a strategy seems remarkably adaptive for
the parasite because higher metabolism at warmer
temperatures would more rapidly use the energy reserves
of an infectious larva.
Recent experimental work suggests that climate
might directly affect influenza epidemics. By manipu-
lating temperature and humidity, Lowen et al. (2007)
found that cold, dry conditions favor transmission of
influenza from infected to naı¨ve guinea pigs, an effect
not related to host immune response. Although the
mechanisms are not yet clear, these experiments seem to
April 2009 893 CLIMATE CHANGE AND INFECTIOUS DISEASES
CONCEPTS & SYNTHESIS
suggest that global warming could directly reduce
An advantage of experiments is that they can
manipulate variables beyond normal ranges. A partic-
ularly insightful example concerns blowflies. Blowfly
maggots suck blood from nestlings then detach and
become subject to ambient temperatures in the nest.
Dawson et al. (2005) noted that the activity and number
of maggots in tree swallow nests increases linearly with
ambient temperature. This positive association between
temperature and blowfly abundance provides a mecha-
nism by which climate change could reduce bird
populations. However, instead of extrapolating beyond
the observed natural variation in temperature, the
authors conducted a notable field experiment. They
placed heating pads in some Tree Swallow nest boxes. In
the nests heated above the natural range of temperature
variation, the blowflies did poorly, suggesting that
sufficiently high increases in temperature could actually
decrease parasitism, a prediction that would not have
been possible without such an experiment.
Although it is not tractable to manipulate air
temperature in the field across large areas, some power
plants discharge thermal effluent into lakes and streams.
Marcogliese (2001) reviewed studies that found thermal
effluent allows fish parasites to complete their life cycles
in cold months but impairs transmission during the
warmest months. In addition, fish hosts may live shorter
lives, decreasing the extent to which parasites may
accumulate in warm sites. Sometimes, declines in host
biodiversity at very warm sites lead to a reduction in
The effects of warming on infectious disease will
depend on the extent to which climate becomes wetter or
more arid. Manipulations of precipitation are difficult,
but Chase and Knight (2003) were able to investigate
experimentally the effect of rainfall on mosquitoes using
wetland mesocosms. The number of mosquitoes that
emerged from previously dried mesocosms was 20 times
higher than in consistently inundated mesocosms, which
had developed a high biomass of mosquito predators.
This experiment helps explain why, in the western
United States, St. Louis encephalitis virus (Shaman et
al. 2002) and West Nile virus (Epstein 2001b) sometimes
increase after droughts.
EXISTING EFFECTS OF CLIMATE CHANGE
Because recent climate appears to be warmer than
does past climate, we may already be experiencing
changes in infectious diseases. Around 3.3% of the
earth’s surface changed from one climate category to
another between 1951 and 2000. For the categorical
shifts in climate that have occurred, the trend has been
for a loss of polar and boreal climates and an increase in
arid climates (Becker et al. 2006). In the Arctic,
conditions may have improved for some disease vectors.
For instance, lung worm of muskoxen can now develop
in their intermediate host slugs in just one year, instead
of the two years historically required, suggesting that
this infectious disease in the Arctic may increase in
severity (Kutz et al. 2005).
We know more about shifts in the distribution of
animals and plants than we do about infectious diseases.
Each decade, species ranges have shifted 6 km toward
the poles (and away from the equator) and 6 m in higher
in altitude; the results of 41% of studies suggest that
climate change has altered species distributions (62%
found climate alters phenology, e.g., flowering date;
Parmesan and Yohe 2003). However, a strict economic
assessment (such as advocated by the Intergovernmental
Panel on Climate Change) finds that climate is the
primary determinant of few species ranges and that the
main factor responsible for shifts in distribution is
probably habitat destruction (Parmesan and Yohe 2003).
Recent increases in infectious diseases may or may not
indicate an effect of climate change. Habitat alteration,
invasive species, agriculture, travel and migration, drug
and pesticide resistance, malnutrition, urban heat
islands, population density, health service, the distribu-
tion of wealth, and education, all of which are in flux,
may affect diseases. For instance, in the last three
decades, as global ocean temperatures have warmed,
there has been an increase in the reports of diseases for
six of nine studied marine taxa; of those six taxa with
increasing reports of diseases, warming was a likely
factor for corals and sea turtles, while different factors,
such as indirect effects of fishing, may explain the other
increases (Ward and Lafferty 2004).
Increases in tick-borne encephalitis correlate signifi-
cantly with increased temperature (Lindgren et al. 2000),
but the correlation is weak and the increases in
temperature actually followed the increases in disease,
making a causal link suspect (Randolph and Rogers
2000). Furthermore, in the Baltics, a sudden jump in
climate suitability for TBE in 1989 did not result in a
landscape-wide increase in incidence (Sumilo et al.
2007). Statistical analyses considering a wide-range of
climatic and socioeconomic data find that the best
predictor of changes in TBE from 1970 to 2005 is
poverty; presumably economic collapse in some areas
led to a decrease in vaccinations, an increase in hosts for
ticks, and an increased need for humans to forage in
tick-infected areas (Sumilo et al. 2008).
Malaria in highland regions since the 1980s has been
linked to global warming (Epstein et al. 1998, Epstein
2001a). Altitudinal expansion of malaria exposes pop-
ulations with little resistance, leading to substantial
mortality in humans (Lindsay and Martens 1998). A
widely cited example is an increase in malaria in Rwanda
in the warm, wet year of 1987 (Loevinsohn 1994).
Analysis of a longer time series suggests that climate has
warmed at four highlands sites to temperatures suitable
for increased mosquito abundance (Pascual et al. 2006).
This is not necessarily a distributional change; some
have argued that none of the ‘‘new’’ reports are above
the historical altitudinal limits for malaria (Reiter 2001).
KEVIN D. LAFFERTY894 Ecology, Vol. 90, No. 4
CONCEPTS & SYNTHESIS
For most examples of climate-induced increases in
highland malaria, logical alternative explanations exist.
Control efforts, clearing for agriculture (Lindblade et al.
2000), drug resistance (Ndymugyenyi and Magnussen
2004), and changes in surveillance (Reiter 2001) also
correspond with the changes in malaria, making it
difficult to determine the contribution of climate change.
Season, epidemiology and interannual climate variation
all contribute to variation in the number of malaria
cases reported at highland sites, with climate explaining
12–63% of the variation in malaria, suggesting that, at
some high altitude locations, recent small increases in
temperature may explain some of the increase in malaria
(Zhou et al. 2004).
It is easy to overlook places where infectious diseases
may have declined with climate change. For instance, at
the time malaria was emerging in highland regions, less
attention was paid to reductions in malaria prevalence in
the Sahel. Here, the brief rainy season creates a window
for seasonal transmission of malaria, but an increasing
frequency of drought, perhaps associated with current
climate change, corresponded with a sharp reduction in
malaria prevalence (Mouchet et al. 1996). Plotting
published estimates of the percentage of the globe’s
land mass affected by malaria during the last century
(Hay et al. 2004) against yearly climate records shows
that as the globe warmed, malaria shrank toward the
tropics (Fig. 3). This presumably occurred because
warming in the last century coincided with major land-
use changes, including efforts to reduce the burden of
malaria in developed countries. Malaria has declined
appreciably in wealthy, temperate zones and relatively
less in poor, tropical locations (Hay et al. 2004). One
way to isolate the effect of climate is to detrend the data
(thereby removing the temporal trend of malaria
control) and look for residual associations between
climate and infectious diseases (Kuhn et al. 2003). The
detrended global malaria data in Fig. 3 has a slope of 0,
indicating no residual effect of temperature on the
historical area of the globe affected by malaria (Fig. 3).
Since the 1970s, malaria has rebounded in many areas
(see Plate 1), and present increases in global warming are
a possible explanation for this change (Epstein et al.
1998). However, recent events other than climate change
are alternative causes. Migration and population growth
can increase the potential for epidemics. Some types of
economic development, particularly roads and dams,
can increase habitat for mosquitoes. In other cases,
economic upheavals that undermined public health
systems (e.g., the 1990s post-USSR economic decline
in Armenia, Azerbaijan, Tajikistan, and Turkmenistan
[Sabatinelli 1998]) caused some local rebounds in
malaria. Finally, the famous evolution of resistance of
mosquitoes to DDT and malaria to anti-malarial drugs
are potential reasons for malaria rebound (Reiter 2001).
FUTURE EFFECTS OF CLIMATE CHANGE
Process-based (also known as mechanistic or biolog-
ical) models estimate how habitat suitability for a species
changes with the environment. Parameterizing process-
based models requires knowing relationships between
climate variables and vital rates. Most process-based
reported from the U.S. National Climate Data Center Global Historical Climatology Network Land Surface Data) increases since
1900 while the percentage of the globe with malaria transmission decreases (data from Hay et al. ). The main figure shows the
data from the inset as a scatter plot (solid squares, solid line) and after detrending the decline in malaria (open squares, dashed line).
A century of malaria and climate: 1900–2000. The inset shows global mean temperature (represented as a deviation, as
April 2009 895CLIMATE CHANGE AND INFECTIOUS DISEASES
CONCEPTS & SYNTHESIS
models of infectious diseases are for vector-transmitted
diseases and they consider the likelihood that the habitat
will be suitable for perennial, seasonal, or epidemic
transmission (Craig et al. 1999). The more sophisticated
process-based models consider R0, or the average
number of infected individuals that result from the
entry of a single infected host into a population of
susceptible hosts (Rogers and Randolph 2006). If R0is
greater than one, the infectious disease can spread
through the population. Because it is not always possible
to estimate R0, components of R0, such as vectorial
capacity (Garrett-Jones 1964), or ‘‘transmission poten-
tial,’’ are sometimes used instead. As input, the models
use mathematical relationships between climate vari-
ables (usually temperature), the vital rates of vectors
(larval development rate, biting rate, adult survivorship)
and, sometimes, the vital rates of pathogens (develop-
ment rate; Garrett-Jones 1964).
For instance, because metabolism increases with
temperature, mosquitoes must feed more to maintain
positive energy balance, and increased biting rates can
increase the chance of transmission of mosquito-
vectored diseases. The time needed for the malaria
organism to develop inside the mosquito decreases with
temperature but a simultaneous increase in mosquito
mortality leads to a convex-shaped association between
temperature and the proportion of infected mosquitoes
that can transmit disease. If the temperature is too cold
(,188C), the parasite cannot develop, whereas if the
temperature is too warm (.328C), infected mosquitoes
die before the infection is transmissible. For these
reasons, the better process-based models indicate a
convex relationship between climate and habitat suit-
ability for some infectious diseases.
When parameterized with climate projections, early
process-based models captured public attention by
predicting a 60% increase in malaria by 2020 (Martens
1999). Comparing the ratio of transmission potential
before and after climate projections resulted in maps
indicating the extent to which climate change would
improve conditions for malaria. These maps suggested
very large increases in transmission potential in some
areas (Epstein 2000). However, increases in vectorial
capacity or transmission potential may not indicate
actual transmission if R0remains less than 1; in other
words, an increase in suitability will not result in malaria
if the new climate is still unsuitable for transmission
(Rogers and Randolph 2006).
Even if process-based models could accurately define
habitat suitability, they do not project actual transmis-
sion because the fundamental niche is always larger than
realized niche. Barriers to dispersal and biotic interac-
transmission sites know for schistosomiasis (a blood fluke transmitted by aquatic snails). In addition, nearly everyone gets malaria
during the wet season. Climate change in this region is likely to increase temperatures and reduce precipitation, impacting local
agriculture, while potentially decreasing transmission of schistosomiasis and malaria. But climate is not the major issue affecting
disease at Richard Toll. In 1986, damming of the Senegal River for irrigation led to spectacular increases in habitat for intermediate
host snails and mosquito vectors. These changes led to an explosion of disease among an immigrant population drawn to work the
new sugar cane fields. Photo credit: K. D. Lafferty.
Senegalese gather water and do laundry in the Senegal River at Richard Toll. This is one of the most intense
KEVIN D. LAFFERTY 896 Ecology, Vol. 90, No. 4
CONCEPTS & SYNTHESIS
tions can exclude species from parts of the fundamental
niche. For instance, in Memphis, the apparent replace-
ment of the dengue/yellow fever vector Aedes aegypti by
the newly invasive Aedes albopictus suggests that
competitive exclusion may have occurred (O’Meara et
al. 1995). Furthermore, projected increases in malaria
transmission potential in climate change scenarios might
not result in epidemics if the kinds of human activities
that led to the historical decline of malaria in temperate
regions (wetlands destruction, vector control, chemo-
therapy) prevent malaria from re-emerging.
To assess the accuracy of a process-based model
requires comparing predictions with actual data. For
instance, before using a process-based model to project
effects of climate change on dengue fever transmission,
Jetten and Focks (1997) confirmed that their model
corresponded with actual data from five cities. The
MARA/ARMA (mapping malaria risk in Africa)
process-based models of habitat suitability correspond
to interpolated maps of current observations relatively
well, but the correspondence breaks down at fine scales
(Craig et al. 1999). Tanser et al. (2003) validated their
process-based model with independent observations.
Their ‘‘threshold’’ model of malaria transmission in
Africa matched 63% of the spatial data and 90–96% of
the temporal data, though this validation is controver-
sial (Reiter et al. 2004). A goal for future maps of
malaria incidence, such as the Malaria Atlas Project
(Hay and Snow 2007) is to provide a means to validate
process-based models to allow estimates of the uncer-
tainty of malaria projections.
Recent process-based models predict decreases in
habitat suitability offsetting increases in habitat suit-
ability. For instance, process-based projections for
malaria in India show a shift in suitability for malaria,
with some higher altitude areas becoming favorable for
malaria and some central areas becoming unfavorable
(Bhattacharya et al. 2006). Some process-based models
show little change in suitability for infectious diseases
(Tanser et al. 2003, Urashima et al. 2003, Thomas et al.
2004) and several even show decreases (de Gruijl et al.
2003, Peterson 2003, Peterson and Shaw 2003).
Other attempts to predict the biological effects of
climate change make use of climate mapping, or niche
models. These statistical (or pattern matching) models
relate field observations of a species (presence, presence–
absence, or abundance) to environmental factors. This
approach dates back to Johnston’s (1924) correlational
model of the climatic requirements of an invasive cactus
in Australia. Statistical models differ from process-based
models in that they do not assume known functional
relationships between vital rates and environmental
variables, allowing a wider range of environmental
variables. The availability of remotely sensed data and
spatial datasets allows complex analyses at global scales.
Statistical models estimate habitat suitability, or the
fundamental niche. To approximate the fundamental
niche, the statistical model often only includes positive
records of disease occurrence. For instance, Peterson
and Shaw (2003) modeled the current and projected
distribution of leishmaniasis in Brazil, using a niche
model of vectors of cutaneous leishmaniasis, three
sandflies (Lutzomyia spp.). A genetic algorithm helped
determine associations between environmental data and
records of sand flies. The authors withheld half of the
records for 100 validations of the projected map,
choosing a model that correctly predicted all the
presences in the validation set (no omission error) and
minimized the prediction area (thereby minimizing
commission errors under the no omission constraint).
When applied to climate projections, the predicted
habitat suitability of the sandflies expands to the higher
southern latitudes, but retracts substantially from the
lower northern latitudes, resulting in an overall reduc-
tion in the potential geographic range of each species.
For malaria, apparent habitat suitability also increases
at higher latitudes and contracts at lower latitudes. As a
result, models estimate that by 2050, falciparum malaria
will gain 23 million human hosts in previously uninfect-
ed locations, but lose 25 million human hosts in areas no
longer suitable for transmission, the net result being
little change in the total number of people exposed to
malaria (Rogers and Randolph 2000).
Like process-based models, statistical models have
weaknesses. For one, most consider simple linear
relationships between climate and species distributions.
As described earlier, linear projections can lead to
substantial error if the actual relationship is non-linear
and the projection extends beyond the range of the
measured independent variable. Furthermore, using
multiple independent variables with multiple possible
time lags can lead to over fitting. Finally, data quality
will always constrain statistical models. Although the
quality of spatial climate data has improved, there is
need for improved mapping of disease incidence in
lesser-developed nations (Hay and Snow 2007).
Few statistical models yet include potentially impor-
tant non-climate variables (limiting resources, biotic
interactions, physical disturbance, dispersal barriers,
human intervention) that might lead to absences of
infectious diseases in areas with suitable climate (Guisan
and Thuiller 2005). Statistical models that also include
negative records may yield insight into the realized
niche, which provides a better estimate of actual
transmission. Such models could include a wide range
of non-climate variables (control, predation, competi-
tion) that influence disease distributions. Future statis-
tical models might attempt to fit observed data to the
nonlinear functions known from process-based models.
In addition, they could include non-climate variables,
including control, but minimize over fitting through
validation with independent data sets. In all cases, it is
helpful when statistical models explicitly present the
April 2009 897CLIMATE CHANGE AND INFECTIOUS DISEASES
CONCEPTS & SYNTHESIS
unexplained variance and the errors of the estimates
reported so that the confidence in projections is clear.
Because temperature and precipitation affect physiol-
ogy, climate can affect species distributions. Climate and
infectious diseases sometimes covary geographically and
over time, suggesting that climate change should lead to
changes in the geographic distribution of infectious
diseases and their vectors. In some places, climate
change will likely lead to increases in some infectious
diseases. For instance, recent evidence indicates an
association between thermal stress and disease in corals
(Lafferty et al. 2004). Many insect vectors prefer the
warm, wet conditions predicted under some climate-
change scenarios. This alarms public health officials who
rightly worry that climate change will broadly increase
important infectious diseases such as malaria. Predic-
tions of tropical diseases spreading into wealthy
temperate nations have understandably captured the
public’s attention. This developing societal concern is
also a challenging ecological research question.
As for most aspects of ecology, the link between
climate and infectious disease is complex, requiring
careful study and rigorous evaluation. There are several
reasons climate change may not always lead to a net
increase in the geographic distribution infectious dis-
eases. Firstly, most species, including infectious diseases,
have upper and lower limits to their temperature
tolerance. This means that changes in climate should
often lead to shifts, not expansions, in habitat suitabil-
ity. Furthermore, while a reduction in habitat suitability
should reduce a species’ range, an increase in habitat
suitability does not necessarily result in an increase in
geographic distribution. This is because other factors
besides climate, such as barriers to dispersal, competi-
tion, and predation, affect the realized niche. For
infectious diseases that depend on other species for
vectoring or as intermediate hosts, habitat degradation
can prevent transmission even if climate is suitable. In
particular, because disease control efforts have success-
fully reduced or eliminated the transmission of previ-
ously widespread infectious diseases from developed
countries, human activities will prevent the expansion of
some infectious diseases even if climate becomes more
suitable. For these reasons, it seems plausible that the
geographic distribution of some infectious diseases may
actually experience a net decline with climate change.
While this is the reverse of the conventional wisdom, it is
consistent with the increasingly accepted view that
climate change will reduce biodiversity. Infectious
diseases, as important and sensitive components of
biodiversity, may be some of the earliest casualties. This
does not mean that we should welcome climate change.
Reductions in disease that result from the loss of vectors
and intermediate hosts will likely correspond to the loss
of biodiversity in general. Furthermore, changes to
climate that reduce infectious disease (e.g., drought, very
hot weather) will likely have their own set of impacts.
Finally, while some areas may see reductions in
infectious disease, populations in areas where infectious
disease will expand have genuine cause for concern.
R. Kirby, A. Kuris, D. Marcogliese, J. McLaughlin, E.
Mordecai, A. Ngyuen, S. Randolph, and T. Suchanek provided
helpful comments on an earlier draft of the manuscript. J. Yee
reviewed the statistical procedures. This manuscript benefited
from support received from the National Science Foundation
through the NIH/NSF Ecology of Infectious Disease Program
(DEB-0224565). Any use of trade, product, or firm names in
this publication is for descriptive purposes only and does not
imply endorsement by the U.S. government.
Bally, M., and J. Garrabou. 2007. Thermodependent bacterial
pathogens and mass mortalities in temperate benthic
communities: a new case of emerging disease linked to
climate change. Global Change Biology 13:2078–2088.
Becker, C., J. Grieser, M. Kottek, F. Rubel, and B. Rudolf.
2006. Characterizing global climate change by means of
Koppen climate classification. Pages 139–149 in Annual
report. Duetscher Wetterdienst, Hamburg, Germany.
Bhattacharya, S. C., C. Sharma, R. C. Dhiman, and A. P.
Mitra. 2006. Climate change and malaria. India Current
Bodker, R., J. Akida, D. Shayo, W. Kisinza, H. A. Msangeni,
E. M. Pedersen, and S. W. Lindsay. 2003. Relationship
between altitude and intensity of malaria transmission in the
Usambara Mountains, Tanzania. Journal of Medical Ento-
Bouma, M. J., and C. Dye. 1997. Cycles of malaria associated
with El Nin ˜ o in Venezuela. Journal of the American Medical
Bradshaw, W. E., and C. M. Holzapfel. 2001. Genetic shift in
photoperiodic response correlated with global warming.
Proceedings of the National Academy of Sciences (USA)
Bruce-Chwatt, L., and J. de Zulueta. 1980. The rise and fall of
malaria in Europe: a historico-epidemiological study. Oxford
University Press, Oxford, UK.
Cairns, M. A., J. L. Ebersole, J. P. Baker, P. J. Wigington,
H. R. Lavigne, and S. M. Davis. 2005. Influence of summer
stream temperatures on black spot infestation of juvenile
coho salmon in the Oregon Coast Range. Transactions of the
American Fisheries Society 134:1471–1479.
Chase, J. M., and T. M. Knight. 2003. Drought-induced
mosquito outbreaks in wetlands. Ecology Letters 6:1017–
Craig, M. H., R. W. Snow, and D. le Sueur. 1999. A climate-
based distribution model of malaria transmission in sub-
Saharan Africa. Parasitology Today 15:105–111.
Currie, D. J., G. G. Mittelbach, H. V. Cornell, R. Field, J. F.
Guegan, B. A. Hawkins, D. M. Kaufman, J. T. Kerr, T.
Oberdorff, E. O’Brien, and J. R. G. Turner. 2004. Predictions
and tests of climate-based hypotheses of broad-scale varia-
tion in taxonomic richness. Ecology Letters 7:1121–1134.
Dawson, R. D., K. K. Hillen, and T. L. Whitworth. 2005.
Effects of experimental variation in temperature on larval
densities of parasitic Protocalliphora (Diptera: Calliphoridae)
in nests of Tree Swallows (Passeriformes: Hirundinidae).
Environmental Entomology 34:563–568.
de Gruijl, F. R., J. Longstreth, M. Norval, A. P. Cullen, H.
Slaper, M. L. Kripke, Y. Takizawa, and J. C. van der Leun.
2003. Health effects from stratospheric ozone depletion and
interactions with climate change. Photochemical and Photo-
biological Sciences 2:16–28.
KEVIN D. LAFFERTY 898Ecology, Vol. 90, No. 4
CONCEPTS & SYNTHESIS
Department of Health. 2002. . Health effects of climate change
in the UK. Department of Health, London, UK.
Diaz, H. F., and G. McCabe. 1999. A possible connection
between the 1878 yellow fever epidemic in the southern
United States and the 1877–78 El Nino Episode. Bulletin of
the American Meteorological Society 80:21–27.
Epstein, P. R. 2000. Is global warming harmful to health?
Scientific American 283(August):50–57.
Epstein, P. R. 2001a. Climate change and emerging infectious
diseases. Microbes and Infection 3:747–754.
Epstein, P. R. 2001b. West Nile virus and the climate. Journal
of Urban Health, Bulletin of the New York Academy of
Epstein, P. R., H. F. Diaz, S. Elias, G. Grabherr, N. E.
Graham, W. J. M. Martens, E. Mosley-Thompson, and J.
Susskind. 1998. Biological and physical signs of climate
change: focus on mosquito-borne diseases. Bulletin of the
American Meteorological Society 79:409–417.
Fayer, R., J. M. Trout, and M. C. Jenkins. 1998. Infectivity of
Cryptosporidium parvum oocysts stored in water at environ-
mental temperatures. Journal of Parasitology 84:1165–1169.
Feder, M. E., and G. E. Hofmann. 1999. Heat-shock proteins,
molecular chaperones, and the stress response: evolutionary
and ecological physiology. Annual Review of Physiology 61:
Field, A. C., M. R. Brambell, and J. A. Campbell. 1960. Spring
Rise in faecal worm-egg counts of housed sheep, and its
importance in nutritional experiments. Parasitology 50:387–
Gallup, J. L., and J. D. Sachs. 2001. The economic burden of
malaria. American Journal of Tropical Medicine and
Garrett-Jones, C. 1964. Prognosis for interruption of malaria
transmission through assessment of mosquitos vectorial
capacity. Nature 204:1173.
Guernier, V., M. E. Hochberg, and J. F. O. Guegan. 2004.
Ecology drives the worldwide distribution of human diseases.
PLoS Biology 2:740–746.
Guisan, A., and W. Thuiller. 2005. Predicting species distribu-
tion: offering more than simple habitat models. Ecology
Harvell, C. D., C. E. Mitchell, J. R. Ward, S. Altizer, A. P.
Dobson, R. S. Ostfeld, and M. D. Samuel. 2002. Climate
warming and disease risks for terrestrial and marine biota.
Hay, S. I., C. A. Guerra, A. J. Tatem, A. M. Noor, and R. W.
Snow. 2004. The global distribution and population at risk of
malaria: past, present, and future. Lancet Infectious Diseases
Hay, S. I., M. F. Myers, D. S. Burke, D. W. Vaughn, T. Endy,
N. Ananda, G. D. Shanks, R. W. Snow, and D. J. Rogers.
2000. Etiology of interepidemic periods of mosquito-borne
disease. Proceedings of the National Academy of Sciences
Hay, S. I., and R. W. Snow. 2007. The malaria atlas project:
developing global maps of malaria risk. PLoS Medicine 3:
Hay, S. I., R. W. Snow, and D. J. Rogers. 1998. Predicting
malaria seasons in Kenya using multitemporal meteorolog-
ical satellite sensor data. Transactions of the Royal Society of
Tropical Medicine and Hygiene 92:12–20.
Hechinger, R. F., and K. D. Lafferty. 2005. Host diversity
begets parasite diversity: bird final hosts and trematodes in
snail intermediate hosts. Proceedings of the Royal Society B
Jetten, T. H., and D. A. Focks. 1997. Potential changes in the
distribution of dengue transmission under climate warming.
American Journal of Tropical Medicine and Hygiene 57:285–
Johnston, T. H. 1924. The relation of climate to the spread of
prickly pear. Transactions of the Royal Society of South
King, B. J., and P. T. Monis. 2007. Critical processes affecting
Cryptosporidium oocyst survival in the environment. Parasi-
Kuhn, K. G., D. H. Campbell-Lendrum, B. Armstrong, and
C. R. Davies. 2003. Malaria in Britain: past, present, and
future. Proceedings of the National Academy of Sciences
Kutz, S. J., E. P. Hoberg, L. Polley, and E. J. Jenkins. 2005.
Global warming is changing the dynamics of Arctic host–
parasite systems. Proceedings of the Royal Society B 272:
Lafferty, K. D., A. P. Dobson, and A. M. Kuris. 2006. Parasites
dominate food web links. Proceedings of the National
Academy of Sciences (USA) 103:11211–11216.
Lafferty, K. D., J. W. Porter, and S. E. Ford. 2004. Are diseases
increasing in the ocean? Annual Review of Ecology
Evolution and Systematics 35:31–54.
Landesman, W. J., B. F. Allan, R. B. Langerhans, T. M.
Knight, and J. M. Chase. 2007. Inter-annual associations
between precipitation and human incidence of West Nile
virus in the United States. Vector-Borne and Zoonotic
Lindblade, K. A., E. D. Walker, and M. L. Wilson. 2000. Early
warning of malaria epidemics in African highlands using
Anopheles (Diptera: Culicidae) indoor resting density. Jour-
nal of Medical Entomology 37:664–674.
Lindgren, E., L. Talleklint, and T. Polfeldt. 2000. Impact of
climatic change on the northern latitude limit and population
density of the disease-transmitting European tick Ixodes
ricinus. Environmental Health Perspectives 108:119–123.
Lindsay, S. W., and W. J. M. Martens. 1998. Malaria in the
African highlands: past, present and future. Bulletin of the
World Health Organization 76:33–45.
Lipp, E. K., A. Huq, and R. R. Colwell. 2002. Effects of global
climate on infectious disease: the cholera model. Clinical
Microbiology Reviews 15:757–770.
Liu, C., S. Lin, Y. Chen, K. C. Lin, T. J. Wu, and C. King.
2007. Temperature drops and the onset of severe avian
influenza A H5N1 virus outbreaks. PLoS one 2:e191.
Loevinsohn, M. E. 1994. Climatic warming and increased
malaria incidence in Rwanda. Lancet 343:714–718.
Lofgren, E., N. H. Fefferman, Y. N. Naumov, J. Gorski, and
E. N. Naumova. 2007. Influenza seasonality: underlying
causes and modeling theories. Journal of Virology 81:5429–
Lowen, A. C., S. Mubareka, J. Steel, and P. Palese. 2007.
Influenza virus transmission is dependent on relative
humidity and temperature. PLoS Pathogens 3:e151.
Lv, S., X. N. Zhou, Y. Zhang, H. X. Liu, D. Zhu, W. G. Yin, P.
Steinmann, X. H. Wang, and T. W. Jia. 2006. The effect of
temperature on the development of Angiostrongylus canto-
nensis (Chen 1935) in Pomacea canaliculata (Lamarck 1822).
Parasitology Research 99:583–587.
MacArthur, R. H. 1972. Geographical ecology. Patterns in the
distribution of species. Harper and Row, New York, New
Mann, M. E., E. P. Gille, R. S. Bradley, M. K. Hughes, J.
Overpeck, F. T. Keimig, and W. S. Gross. 2000. Global
temperature patterns in past centuries: an interactive
presentation. IGBP Pages/World Data Center for Paleocli-
matology. NOAA/NCDC Paleoclimatology Program, Boul-
der, Colorado, USA.
Marcogliese, D. J. 2001. Implications of climate change for
parasitism of animals in the aquatic environment. Canadian
Journal of Zoology 79:1331–1352.
Marshall, A. G. 1981. The ecology of ectoparasitic insects.
Academic Press, London, UK.
Martens, P. 1999. Climate change impacts on vector-borne
disease transmission in Europe. Pages 45–54 in A. Haines and
A. J. McMichael, editors. Climate change and human health.
The Royal Society, London, UK.
April 2009899CLIMATE CHANGE AND INFECTIOUS DISEASES
CONCEPTS & SYNTHESIS
Memmott, J., P. G. Craze, N. M. Waser, and M. V. Price. 2007.
Global warming and the disruption of plant–pollinator
interactions. Ecology Letters 10:710–717.
Minato, K., E. Kimura, Y. Shintoku, and S. Uga. 2008. Effect
of temperature on the development of free-living stages of
Strongyloides ratti. Parastology Research 102:315–319.
Moore, P. S. 1992. Meningococcal meningitis in sub-Saharan
Africa: a model for the epidemic process. Clinical Infectious
Mouchet, J., O. Faye, J. Julvez, and S. Manguin. 1996. Drought
and malaria retreat in the Sahel, West Africa. Lancet 348:
Mullens, B. A., J. L. Rodriguez, and J. A. Meyer. 1987. An
epizootiological study of Entomophthora muscae in muscoid
fly populations on Southern California poultry facilities, with
emphasis on Musca domestica. Hilgardia 55:1–41.
Najera-Morrondo, J. 1991. Malaria control: history shows it’s
possible. World Health September/October:3–4.
Ndymugyenyi, R., and P. Magnussen. 2004. Trends in malaria-
attributable morbidity and mortality among young children
admitted to Ugandan hospitals, for the period 1990–2001.
Annals of Tropical Medicine and Parasitology 98:315–327.
Nunn, C. L., S. M. Altizer, W. Sechrest, and A. A. Cunning-
ham. 2005. Latitudinal gradients of parasite species richness
in primates. Diversity and Distributions 11:249–256.
O’Meara, G. F., L. F. Evans, A. D. Gettman, and J. P. Cuda.
1995. Spread of Aedes albopictus and decline of Aedes aegypti
(Diptera, Culicidae) in Florida. Journal of Medical Ento-
Parmesan, C., and G. Yohe. 2003. A globally coherent
fingerprint of climate change impacts across natural systems.
Pascual, M., J. A. Ahumada, L. F. Chaves, X. Rodo, and M.
Bouma. 2006. Malaria resurgence in the East African
highlands: temperature trends revisited. Proceedings of the
National Academy of Sciences (USA) 103:5829–5834.
Peterson, A. T. 2003. Predicting the geography of species’
invasions via ecological niche modeling. Quarterly Review of
Peterson, A. T., and J. Shaw. 2003. Lutzomyia vectors for
cutaneous leishmaniasis in Southern Brazil: ecological niche
models, predicted geographic distributions, and climate
change effects. International Journal for Parasitology 33:
Pfister, C. 1988. Variations in the spring–summer climate of
central Europe from the High Middle Ages to 1850. Pages
57–82 in H. Wanner and U. Siegenthaler, editors. Long and
short term variability of climate. Springer-Verlag, Berlin,
Piersma, T. 1997. Do global patterns of habitat use and
migration strategics co-evolve with relative investments in
immunocompetence due to spatial variation in parasite
pressure? Oikos 80:623–631.
Poulin, R. 2006. Global warming and temperature-mediated
increases in cercarial emergence in trematode parasites.
Randolph, S. E., and D. J. Rogers. 2000. Fragile transmission
cycles of tick-borne encephalitis virus may be disrupted by
predicted climate change. Proceedings of the Royal Society B
Reisen, W. K., Y. Fang, and V. M. Martinez. 2006. Effects of
temperature on the transmission of West Nile virus by Culex
tarsalis (Diptera: Culicidae). Journal of Medical Entomology
Reiter, P. 2001. Climate change and mosquito-borne disease.
Environmental Health Perspectives 109S:141–161.
Reiter, P., C. J. Thomas, P. M. Atkinson, S. I. Hay, S. E.
Randolph, D. J. Rogers, G. D. Shanks, R. W. Snow, and A.
Spielman. 2004. Global warming and malaria: a call for
accuracy. Lancet Infectious Diseases 4:323–324.
Rogers, D. J., and S. E. Randolph. 2000. The global spread of
malaria in a future, warmer world. Science 289:1763–1766.
Rogers, D. J., and S. E. Randolph. 2006. Climate change and
vector-borne diseases. Advances in Parasitology 62:345–381.
Rohde, K. 1982. Ecology of marine parasites. University of
Queensland Press, St. Lucia, Australia.
Romano, J. C., N. Bensoussan, W. A. N. Younes, and D.
Arlhac. 2000. Thermal anomaly in the waters of the Gulf of
Marseille during summer 1999. A partial explanation of the
mortality of certain fixed invertebrates? Comptes Rendus de
l’Academie des Sciences Serie III (Life Sciences) 323:415–427.
Sabatinelli, G. 1998. Malaria situation and implementation of
the global malaria control strategy in the WHO European
region. A report by the World Health Organization Expert
Committee on Malaria. World Health Organization, Geneva,
Sachs, J., and P. Malaney. 2002. The economic and social
burden of malaria. Nature 415:680–685.
Schafer, M., and J. O. Lundstrom. 2001. Comparison of
mosquito (Diptera: Culicidae) fauna characteristics of
forested wetlands in Sweden. Annals of the Entomological
Society of America 94:576–582.
Shaman, J., M. Stieglitz, C. Stark, S. Le Blancq, and M. Cane.
2002. Using a dynamic hydrology model to predict mosquito
abundances in flood and swamp water. Emerging Infectious
Small, J., S. J. Goetz, and S. I. Hay. 2003. Climatic suitability
for malaria transmission in Africa, 1911–1995. Proceedings
of the National Academy of Sciences (USA) 100:15341–
Stone, L., R. Olinky, and A. Huppert. 2007. Seasonal dynamics
of recurrent epidemics. Nature 446:533–536.
Sumilo, D., L. Asokliene, A. Bormane, V. Vasilenko, I.
Golovijova, and S. Randolph. 2007. Climate change cannot
explain the upsurge of tick-borne encephalitis in the Baltics.
PLoS One 6:e500.
Sumilo, D., A. Bormane, L. Asokliene, V. Vasilenko, I.
Golovijova, T. Avsic-Zupanc, Z. Hubalek, and S. Randolph.
2008. Socio-economic factors in the differential upsurge of
tick-borne encephalitis in Central and Eastern Europe.
Reviews in Medical Virology 18:81–95.
Tanser, F. C., B. Sharp, and D. le Sueur. 2003. Potential effect
of climate change on malaria transmission in Africa. Lancet
Teklehaimanot, H. D., M. Lipsitch, A. Teklehaimanot, and J.
Schwartz. 2004. Weather-based prediction of Plasmodium
falciparum malaria in epidemic-prone regions of Ethiopia I.
Patterns of lagged weather effects reflect biological mecha-
nisms. Malaria Journal 3:41–43.
Thomas, C. D., et al. 2004. Extinction risk from climate change.
Tong, S. L., and W. B. Hu. 2001. Climate variation and
incidence of Ross River virus in Cairns, Australia: a time-
series analysis. Environmental Health Perspectives 109:1271–
Urashima, M., N. Shindo, and N. Okabe. 2003. A seasonal
model to simulate influenza oscillation in Tokyo. Japanese
Journal of Infectious Diseases 56:43–47.
van Riper, C., S. G. van Riper, M. L. Goff, and M. Laird. 2003.
The epizootiology and ecological significance of malaria in
Hawaiian land birds. Ecological Monographs 56:327–344.
Ward, J. R., and K. D. Lafferty. 2004. The elusive baseline of
marine disease: are diseases in ocean ecosystems increasing?
PLoS Biology 2:542–547.
Wolfe, N. D., C. P. Dunavan, and J. Diamond. 2007. Origins of
major human infectious diseases. Nature 447:279–283.
Zhou, G., N. Minakawa, A. K. Githeko, and G. Y. Yan. 2004.
Association between climate variability and malaria epidem-
ics in the East African highlands. Proceedings of the National
Academy of Sciences (USA) 101:2375–2380.
KEVIN D. LAFFERTY 900Ecology, Vol. 90, No. 4
CONCEPTS & SYNTHESIS