Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
3 Historical, Scientific, and Technological Approaches to Studying the Climate-Disease Connection OVERVIEW A variety of methods are employed to identify, measure, evaluate, and predict the direct and indirect effects of climate change on infectious diseases. As illus- trated in the contributions to this chapter, these include the following: â¢ Analyses of historical records to discern long-term or ancient patterns of climate and infectious disease â¢ Monitoring programs that track disease in wild animals, which are espe- cially sensitive environmental sentinels â¢ Comparisons of environmental measurements obtained from satellite imagery with epidemiological data â¢ Climate-driven predictive models of infectious disease transmission Each of these approaches contributes to an expanding interdisciplinary effort to understand the influence of climate change and extreme weather events on infectious disease distribution and transmission dynamics. Historical analysis provides a perspective on climate and infectious disease far more sweeping than can be obtained from scientific monitoring, as demon- strated in this chapterâs first paper, which chronicles the association between drought and epidemic disease and its influence on Mexican civilizations over the past millennium. Searching the historical record of the Valley of Mexico for evidence of famines and epidemics, speaker Rodolfo AcuÃ±a-Soto of the Universidad Nacional AutÃ³noma de MÃ©xico, and coauthors identified several drought-associated epidemics of hemorrhagic fevers that had swept the region, 179
180 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS causing massive mortality. Among these, the authors describe four especially destructive epidemics that appear to have killed between 20 and 90 percent of the entire population, leading to societal collapse: the epidemics of 1003-1011, 1545-1548, 1576-1578, and 1736-1737. The authors also compare circumstances in contemporary Mexico with those associated with apparent past episodes of infectious disease emergence, when increasing human connectivity (roads then, globalization today), and the emergence of new pathogens transmitted by aerosols (smallpox and measles in the past, severe acute respiratory syndrome [SARS] and influenza today), proved to be a very dangerous combination. Emerging infectious diseases of wildlife arise when the delicate balance of host, pathogen, and environment is disturbed. Therefore, these events represent a critical target for infectious disease monitoring efforts of all sorts, including those that seek to track the influence of climate change, according to speaker William Karesh of the Wildlife Conservation Society. In the chapterâs second paper, he and coauthors provide several examples of studies that illustrate the direct and indirect influences of climate on infectious diseases of wildlife. They make the case that such interactions can serve as the basis for monitoring the ecological effects of climate change on emergent diseases that threaten not only wildlife, but also livestock and humans, because wild animals often serve as reservoirs for microbes that may cause pathogenic diseases in humans; these microbes are not necessarily pathogenic in their animal hosts. Moreover, the authors note, wild animals offer a number of advantages for disease monitoring programs: their comparatively short generation times reflect environmental changes more quickly than do humans; the great variety of wild species offers an equally wide range of life histories for the observation of disease dynamics; and they provide sensitive sentinels for changes in the environments to which they are specifically adapted. As discussed by Chretien and coauthors in Chapter 2 and as first described in Linthicum et al. (1999), efforts to predict risk for Rift Valley fever (RVF) demon- strated that trends in environmental variables detected from satellite imagery can be compared with epidemiological data to reveal relationships between climate and infectious disease transmission and geographic distribution. In his workshop presentation, speaker Compton Tucker of the National Aeronautics and Space Administration (NASA)âwho coauthored both of the previously mentioned papersâdescribed how remote sensing data are collected and analyzed, and pre- sented two additional examples of the use of this tool in examining links between climate and infectious disease. The first involved a search for significant environmental factors common to sporadic outbreaks of Ebola hemorrhagic fever (EHF). Ebola virus also affects nonhuman primates, which have been implicated as the source of severalâbut not allâhuman outbreaks through contact with the meat of infected apes (Pinzon et al., 2004). Tucker and colleagues chose to investigate the possibility that Ebola outbreaks occur independently of human cases, in nonhuman primates, and to
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 181 identify environmental factors that precipitate these outbreaks, which can then spread to humans. Analyzing satellite dataâthe monthly Normalized Difference Vegetation Index (NDVI), a proxy for precipitationâthat had been collected continuously in tropical Africa since 1981, Tucker and coworkers found that the majority of documented EHF outbreaks (in humans) occurred toward the end of rainy seasons, when a sharply drier period was followed by a sudden return to very wet conditions. They hypothesize that these apparent âtrigger eventsâ enhance viral transmission from reservoir speciesâwhich remain unknown âto nonhuman primates and humans. Today, satellite data from eastern equatorial Africa are screened routinely for the Ebola-triggering weather pattern, Tucker said. The results guide targeted testing for the virus in local primates, which may provide an early warning of future human outbreaks. Tucker also described the use of satellite imagery to investigate an unusual outbreak of RVF in southwestern Arabia, the first ever recorded there. Records of a satellite-derived index of photosynthetic capacity, the NDVI, showed that significant precipitation had fallen in the region prior to the outbreak and that the RVF outbreak was thus due to natural causes.Â The origin of the outbreak has since been attributed to infected cattle that were imported into southwestern Arabia from the Horn of Africa (Tucker et al., in press). As these examples and others in Chapter 2 illustrate, relatively simple cor- relations between remotely-sensed measurements of climatic variables (e.g., precipitation; sea surface temperature, height) and disease incidence have proven to be useful indicators of risk for a variety of infectious diseases. However, as speaker William Reisen noted, such correlations are not universally applicable and may have to be interpreted in light of other important environmental influ- ences on infectious disease transmission (see also Summary and Assessment section âPredictive Modelsâ). In the chapterâs final paper, Reisen and Christopher Barker (both at the University of California, Davis) describe the design, imple- mentation, and limitations of climate-driven predictive models of mosquito-borne encephalitis transmission used by the State of California to estimate disease risk and inform public health interventions. Under the auspices of this disease surveillance and control program, mos- quito abundance and encephalitis virus activity have been actively monitored for more than 50 years throughout many of Californiaâs diverse biomes and across wide gradients of latitude (north-south) and elevation (east-west). Early in the â In a personal communication on June 20, 2008, Dr. William Karesh stated that LeRoy et al. (2005) is probably the best work done to demonstrate (1) the presence of Ebola Zaire strain viral particles or viral fragments in 3 species of fruit bats, and (2) serological evidence of immune response to filovirus in those same species of bats. Earlier work by Swanepoel et al. showed viral shedding with no pathol- ogy for up to 28 days after fruit bats were experimentally infected with Ebola Zaire in a laboratory setting. To date, there is nothing published on live Ebola virus being isolated from naturally occurring, free-ranging bats. That, in addition to showing that those bats can serve to infect other animals, would help determine if one or more species serve as an effective reservoir.
182 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Human cases Equine cases Amplification Avian infection Mosquito infection Climate Mosquito abundance Time [months] FIGURE 3-1â Sequence of surveillance data collected during seasonal virus amplification. SOURCE: Reisen and Barker (2008). Figure 3-1 also season, before insect and wildlife testing become feasible, climate-based fore- casts inform disease control measures (Figure 3-1). Surveillance activities begin Figure 3-5 in the spring, with the goal of arresting viral amplification and avoiding the need for adult mosquito control. In the case of West Nile virus (WNV), early-season temperature measurements are paramount, because the effects of precipitation on viral transmission have been found to vary among regions (Reisen et al., in press) and vector species (e.g., urban Culex pipiens mosquitoes do well under hot, dry conditions, whereas rural Culex tarsalis do well under wet conditions in many areas). Although these early-season predictions enable response activities (such as equine vaccination, larval mosquito control, and public education) that can reduce the public health consequences of mosquito-borne disease, Reisen and Barker note that the rationale for applying insecticides in advance of an epidemic is not always understood by the public. They also warn that while WNV âprovided a wake-up call for public health agencies and clearly delineated the inability of cur- rent control programs to contain an invading, mosquito-borne, zoonosis,â waning of the epidemic has led to a loss of funding for WNV research and surveillance and, more importantly, for more general surveillance and detection programs capable of spotting âthe next invading pathogen.â
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 183 DROUGHT, EPIDEMIC DISEASE, AND MASSIVE POPULATION LOSS: 1,000 YEARS OF RECORD IN MEXICO Rodolfo AcuÃ±a-Soto, M.D., M.Sc., D.Sc. Universidad Nacional AutÃ³noma de MÃ©xico David W. Stahle, Ph.D. University of Arkansas Matthew D. Therrell, Ph.D. Southern Illinois University JosÃ© Villanueva Diaz, Ph.D. Centro Nacional de InvestigaciÃ³n Disciplinaria Introduction The Valley of Mexico with its benign climate, rich soil, and once abundant water has been a preferred population center for centuries. Today, with 20 mil- lion inhabitants, Mexico Cityâs metropolitan area is one of the largest human conglomerates in history (Yu-ping and Heligman, 1994). While this has been the result of constant growth for the last 85 years, history has not always been this benign. Over the past 1,000 years the Valley of Mexico went through three periods of catastrophic population losses (Clavijero, 1945; Cook and Simpson, 1948; Hugh, 1993). Founded only 675 years ago, Mexico City is located in the same region where the once magnificent cities TeotihuacÃ¡n and Tula collapsed 1,255 and 1,000 years ago, respectively. Similar catastrophic events occurred during the sixteenth cen- tury, when the Valley of Mexico, as well as the whole country, lost 80 to 90 per- cent of its inhabitants due to highly lethal epidemics. During the seventeenth to twentieth centuries, the population again went through several calamitous periods of high mortality, droughts, famines, and epidemics (Gerhard, 1986; LeÃ³n, 1982; Ocaranza, 1933; Therrell et al., 2004; Yu-ping and Heligman, 1994). â Corresponding author. Departamento de MicrobiologÃa y ParasitologÃa, Facultad de Medicina, Universidad Nacional AutÃ³noma de MÃ©xico, Edificio âAâ de InvestigaciÃ³n, 2o. Piso, Facultad de Medicina, Ciudad Universitaria, C.P. 04510, MÃ©xico, D.F. Phone (55) 56 23 23 81; Fax: (55) 56 23 23 82; E-mail: firstname.lastname@example.org. â Tree-Ring Laboratory, Department of Geosciences, Ozark Hall 113, Fayetteville, AR 72701. â Geograpy and Environmental Resources, Carbondale, Illinois. â RelaciÃ³n Agua-Suelo-Planta-AtmÃ³sfera CENID-RASPA, Km. 6.5 Margen Derecha Canal de Sacramento, C. P. 35140, GÃ³mez Palacio, Durango, MÃ©xico.
184 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS In spite of the importance of this topic, the formal study of famines and epi- demics in Mexico has been primarily descriptive and remains largely incomplete. The aims of this work are to present a chronology of famines and epidemics and to review some of the major events of massive population loss in the Valley of Mexico over the last 1,000 years. For this study, previously published chronologies of epidemics and famines in Mexico were reviewed. This was complemented with an exhaustive multiyear review of epidemiological, environmental, and demographic information available in archives and libraries in Mexico and the United States. The search included chronicles, old medical books, diaries, newspapers, and official documentation. Quantitative data from censuses and burial records were also obtained. In addition, events indicative of social distress, such as special religious acts, urgent govern- ment measures, or local officials asking for help, were recorded. For all documents, priority was given to descriptions written by eye witnesses. Drought was consid- ered as such when firsthand witnesses indicated the absence or drastic reduction of rain, normally associated with crop failure. Information was complemented with data from a previous publication (Therrell et al., 2004). Only Âmaterials related to the Valley of Mexico were taken into consideration for this study. The Chronology of Famines and Epidemics Over the last 1,500 years, a total of 119 major epidemics and 38 famines were identified (see Table 3-1). Drought was the main cause of 28 (73 percent) famines. The epidemics of smallpox of 1520-1521 and 1538-1539 induced fam- ine by generalized social disruption. Other historic famines were caused by the particularly disastrous combination of summer frost followed by drought. Such was the case of the legendary famine of 1542-1545, when early frost killed all the corn plants in 1542 and was followed by prolonged drought during 1543-1544 when no rain was registered for 20 months. With no new harvest, reserves ran out, creating a very stressful situation that paved the way to a major famine (Therrell et al., 2004). This series of events recurred in 1784-1786, the infamous âyear of the hungerâ (Cook and Sherburne, 1985). After the Conquest in 1520, famines were recorded with decreasing frequency in the following centuries: 10 in the sixteenth century, 8 in the seventeenth century, and 5 in the eighteenth century; no major famines were recorded during the nineteenth or twentieth centuries. For the 1,500-year period, a total of 119 epidemics were identified (see Table 3-2). Of these, viral diseases caused 55 (46.2 percent) and bacteria were â For representative documents, see AcuÃ±a, 1982; Anonymous, 1978, 1980a,b, 1981, 1995, 1998; Carmago, 1999; Chimalpahin, 1998, 2001; Cook and Sherburne, 1985; de Alva IxtlixÃ³chitl, 1975; de CÃ¡rdenas, 1945; de Grijalva, 1924; de Lorenzana, 1992; de ObregÃ³n, 1988; de Ortega MontaÃ±ez, 1955; de Torquemada, 1969; Farfan, 1592; Keber, 1995; Lopez de Hinojoso, 1578, 1595; Mendieta, 1997; SahagÃºn, 1997; Somolinos, 1956.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 185 TABLE 3-1â Famines in the Valley of Mexico Years Causes Remarks â1 1003-1011 Drought, war Famine, epidemic, high mortality â2 1029-1035 Drought Famine, high mortality â3 1253 Drought Famine, epidemic â4 1332-1335 Drought Famine â5 1382-1385 Floods Famine â6 1403 Locust Famine â7 1430 Drought Famine â8 1434 Drought Famine â9 1446-1456 Locust, floods, frost, snow, drought Famine, probably associated with epidemic of hemorrhagic fever, high mortality 10 1464 Heat waves, strong winds Famine 11 1488 Locust Famine 12 1492 Floods Famine 13 1498-1499 Floods Famine 14 1503-1505 Drought Famine, high mortality 15 1514 Drought Famine 16 1520-1521 War, first smallpox epidemic Famine 17 1538-1539 Second smallpox epidemic Famine 18 1541-1548 Drought, frost, strong winds, crop Famine, epidemic of hemorrhagic disease. fevers, high mortality 19 1550-1555 Drought, mumps epidemic Famine 20 1558-1559 Frost, locust Famine, epidemic of hemorrhagic fevers 21 1562-1564 Drought, epidemics Famine, epidemics of measles, smallpox, and hemorrhagic fevers; high mortality 22 1571-1573 Drought, epidemic Famine, epidemic of unknown origin 23 1576-1579 Drought Famine, epidemic of hemorrhagic fever, high mortality 24 1584-1588 Drought Famine, epidemic of hemorrhagic fever, high mortality 25 1594 Drought, frost, heat waves Famine 26 1610-1613 Drought, frost, strong winds Famine 27 1615-1517 Drought, frost Famine, epidemic of measles 28 1621-1623 Drought Famine 29 1629 Floods Famine 30 1634-1635 Drought Famine, epidemic of whooping cough 31 1639-1642 Drought Famine, epidemics of measles, whooping cough, and hemorrhagic fever 32 1658-1663 Drought Famine, epidemic of measles 33 1691-1697 Drought, frost, floods, speculation Famine, riots, epidemics of measles and other unknown diseases continued
186 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS TABLE 3-1â Continued Years Causes Remarks 34 1713-1714 Drought, frost Famine, epidemic of unknown disease, high mortality 35 1718 Drought Famine 36 1736-1739 Drought Famine, epidemic of hemorrhagic fever, high mortality 37 1746-1749 Drought Famine, epidemics of smallpox and typhus 38 1784-1789 Drought Famine, epidemic of hemorrhagic fever, high mortality SOURCE: Adapted from Chimalpahin (2001); Cooper (1965); de Alva IxtlixÃ³chitl (1975); Flores and Malvido (1985); Garcia et al. (2003); Gerhard (1986); Gibson (1964); Therrell et al. (2004). TABLE 3-2â Major Epidemics in the Valley of Mexico Number of Epidemics That Caused High Diseases (n = 15) Number Duration (years) Mortalitya â1 Hemorrhagic fevers: 24 1.78 Â± .62 4 cocoliztli, matlazahuatl â2 Smallpox 20 1.22 Â± .42 5 â3 Typhus 18 1.25 Â± 1.94 3 â4 Unknown 19 1.30 Â± 1.41 3 â5 Measles 13 1.21 Â± .42 3 â6 Influenza 7 1.0 Â± 0 1 â7 Typhoid fever 4 1.25 Â± .50 0 â8 Poliomyelitis 3 1.4 Â± .54 0 â9 Cholera 3 3.0 Â± 2 2 10 Whooping cough 2 1.0 Â± 0 0 11 Scarlet fever 2 1.5 Â± .70 0 12 Mumps 1 1.0 Â± 0 1 13 Chickenpox 1 1.0 Â± 0 1 14 Meningitis 1 1.0 Â± 0 0 15 Croup 1 1.0 Â± 0 0 a >1% of the total population. SOURCE: Adapted from Chimalpahin (1998); Cooper (1965); de Torquemada (1969); Flores (1888); Flores and Malvido (1985); Garcia et al. (2003). involved in 31 (26.05 percent) of the cases. For the remaining 33 (27.73 percent) epidemics, including 24 described as hemorrhagic fevers, the cause remains to be identified. All 13 known diseases that led to epidemics were caused by exclusive human pathogens with known human-to-human transmission (smallpox, measles, typhus, etc.). Aerosols were the most common mechanism for person-to-person
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 187 contagion, accounting for 48 (40 percent) of all the epidemics. This was followed by insect vectors, with 18 (8.4 percent) epidemics, and water-borne diseases accounting for 10 (8.4 percent). For the remaining 43 epidemics the mechanism of transmission remains unknown. As in many parts of the world, five diseases were emerging infections of their respective times. All of them were imported and became permanently established in the country (AIDS, chickenpox, measles, mumps, and smallpox). For reasons that remain unclear, cholera disappeared after each introduction. Influenza behaved with the same periodic outbreaks as it does in the rest of the world, and hemorrhagic fevers reemerged locally from a distant past. The four most destructive epidemics (see Table 3-3), with mortality rates ranging from 20 to 90 percent of the entire population, were associated with a sequence of climate extremes, with drought in the years preceding the epidemic followed by wetness during the year of the epidemic (AcuÃ±a-Soto et al., 2000, 2002). Drought and the Collapse of TeotihuacÃ¡n The city of TeotihuacÃ¡n, located about 40 km north of Mexico City, was one of the largest and most sophisticated human conglomerates of the preindustrial world. With a complex urban design, the city was the cultural, religious, and military center of a vast area in Mesoamerica. Following its splendor between the years 300 A.D. and 600 A.D, the city went into decline between 650 A.D. and 750 A.D. (Millon, 1970). Undeniably severe and sustained drought occurred in the eighth and ninth centuries in North America (AcuÃ±a-Soto et al., 2005). TABLE 3-3â Deadliest Epidemics in Central Mexico Mortality of the Diseases Year Total Population (%) â1 Unknown 1003-1011 90 (?) â2 Hemorrhagic fevers (cocoliztli) 1545-1548 70-80 â3 Hemorrhagic fevers (cocoliztli) 1576-1578 50 â4 Hemorrhagic fevers (matlazahuatl) 1736-1737 ~20-30 â5 Smallpox 1520 ~15 â6 Hemorrhagic fevers (mysterious fevers) 1813-1815 ~10 â7 Influenza 1918 ~5 â8 Cholera 1833 ~5 â9 Smallpox-typhus 1763-1764 ~5 10 Smallpox 1779 ~4 11 Smallpox 1797-1798 ~4 12 Cholera 1850-1854 4 13 Typhus 1915 1 14 Smallpox 1840 1 SOURCE: Adapted from Chimalpahin (1998); Cooper (1965); de Torquemada (1969); Flores (1888); Flores and Malvido (1985); Garcia et al. (2003).
188 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Based on the similarities of the climate (drought) and demographic (large popu- lation loss) events of the sixteenth century in the same area, it has been pro- posed recently that drought-associated epidemics of hemorrhagic fever may have contributed to the massive population loss during the collapse of TeotihuacÃ¡n (AcuÃ±a-Soto et al., 2005). The specific co-occurrence of drought and abandon- ment of TeotihuacÃ¡n has not been proved, but it is an attractive hypothesis given the well-documented occurrence of megadrought in adjacent areas (Hodell et al., 1995; Metcalfe and Hales, 1994). The Fall of Tula As a result of the meticulous labor of certain educated Indian nobility and Spanish friars during the sixteenth century, pre-Hispanic Mexicoâs written historyâlost as a consequence of the Spanish order to set fire to the library of the city of Texcocoâwas partially recovered. Indian authors such as Fernando Alvarado TezozÃ³moc (TezozÃ³moc, 1975), Domingo Chimalpahin (Chimalpahin, 1998, 2001), and Fernando de Alva IxtlixÃ³chitl (de Alva IxtlixÃ³chitl, 1975) inter- viewed many elders and studied some of the texts that survived the fire but have since disappeared. Europeans, such as Friar Bernardino de SahagÃºn (SahagÃºn, 1997), worked with informants. Fernando de Alva IxtlixÃ³chitl related the fall of Tula (de Alva IxtlixÃ³chitl, 1975). He described a series of climatic disasters that plagued the Toltecs for 20 years before a huge epidemic that resulted in a dramatic population loss. Beginning in the year 984 A.D., heavy rains âthat destroyed most buildings and lasted for 100 daysâ were followed by a year of intense heat that âdried all plants and trees.â The next year came with frost âthat took all the land without leaving anything.â The year after, heavy rains came again âwith great hailstorms and lightning, so abundant that all the surviving trees were destroyed.â This period was followed by an intermission of 12 years of normal weather, but 4 years before the epidemic, a plague of worms infested the grain. Fernando de Alva IxtlixÃ³chitl (1975) describes what appears to have been an epidemic in the year 1003, in the style of pre-Hispanic legends: In the year 1003, when in the first days, a little boy that was white, blond and beautiful, that had to have been a demon, was on a hill. They took him to the City to show him to the king. When the king saw him, he demanded that they bring him back from where they had taken him, because it did not seem to be a good sign. And then the little demon boyâs head began to rot, and many people died from the horrible smell. The Toltecs decided to kill him when one of them was able to reach him, because every one who arrived near the boy died soon after. With this horrible smell, disease spread all over the land and out of the 1,000 Toltecs, 900 died. . . . From this time forward, there was a law that on its fifth birthday, any blond creature would be sacrificed, and this lasted up until the arrival of the Spanish.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 189 The collapse of Tula ended with an extremely violent war, as evidenced by archeological data. During the sixteenth century, several Indian historians wrote about the history of the fall of Tula. Other authors, using independent sources of information, narrate the same events (Chimalpahin, 1998, 2001; TezozÃ³moc, 1975). Yet the climatic history of the collapse of the Tula Empire is waiting for high-resolution proxy evidence of rainfall. Drought-Associated Epidemics of Hemorrhagic Fevers of the Sixteenth Century The post-Conquest collapse of the Mexican population occurred predomi- nantly during the sixteenth century megadrought (AcuÃ±a-Soto et al., 2000, 2002, 2004; Stahle et al., 2000). According to all witnesses (Farfan, 1592; Lopez de Hinojoso, 1578, 1595; Somolinos, 1956), the events that caused the highest mortality were a series of epidemics of hemorrhagic fevers referred to as cocoliztli (Nahuatl word for lethal âpestilenceâ) that probably began in 1536 (see Table 3-4). The epidemics of 1545 and 1576-1580 were particularly lethal. Together, they were responsible for approximately 12 million to 15 million deaths in the highlands of Mexico. During the epidemics, a large proportion of the population was incapacitated. Some witnesses described whole families dying of starvation rather than disease, even when not severely ill. Cocoliztli epidemics evolved as an expanding wave originating in central Mexico that radiated outward over the highlands of central Mexico and caused severe social and economic disintegration. The cause of cocoliztli remains elusive. A brief consensus description, based on contemporaneous attending physicians, is the following: The disease had a very short course, started abruptly with high fever, vertigo, severe headache, insatiable thirst, red eyes, and weak pulse. Shortly after, patients became intensely jaundiced, demented, and restless. Then hard and painful nodules appeared behind one or both ears, sometimes so large that they occupied the entire neck and half of the face. This process was accompanied by intense chest and abdominal pain, as well as dysentery. Toward the end, blood flowed from the ears, anus, vagina, mouth, and nose. The disease was almost inevitably fatal for the native population. The Spaniards were minimally affected, and when they occasionally acquired the disease, it had a benign course (Farfan, 1592; Lopez de Hinojoso, 1578, 1595; Somolinos, 1956). Drought was particularly important for the epidemics of hemorrhagic fevers in Mexico. Using tree-ring reconstructions of rainfall over central Mexico, cocoliztli epidemics were found to occur in years of abundant rain embedded in the midst of the sixteenth-century megadrought (AcuÃ±a-Soto et al., 2002, 2004; Stahle et al., 2000). The 1736 and 1813 epidemics were also highly lethal and were asso- ciated with a similar sequence of drought and dryness. The hemorrhagic fevers mysteriously disappeared after 1815 (AcuÃ±a-Soto et al., 2000; Cooper, 1965). At
190 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS TABLE 3-4â Epidemics of Hemorrhagic Fevers in the Valley of Mexico Years Drought Mortality (%) Other â1 1545-1548 Yes 70-80% Famine â2 1559 Yes Unknown Famine â3 1563-1564 Yes Unknown Measles, smallpox, famine â4 1566 Yes Unknown â5 1570 Yes Unknown â6 1576-1578 Yes 50% Famine â7 1587-1588 Yes Unknown Famine â8 1602-1603 Yes Unknown â9 1606-1607 Yes Unknown 10 1613-1614 Yes Unknown Famine 11 1616 Yes Unknown Famine and âotherâ diseases 12 1629-1631 Yes High Famine 13 1633 Yes Unknown 14 1634 Yes Unknown Whooping cough, famine 15 1641-1642 Yes Unknown Whooping cough, famine 16 1651 Yes Unknown 17 1731 Yes Unknown 18 1736-1737 Yes 20-30% Famine 19 1742 Yes Unknown 20 1761-1762 Yes 25,000 in Mexico City 21 1768-1769 Yes High infant mortality Measles 22 1772-1773 Yes Unknown Measles and whooping cough 23 1784-1786 Yes 300,000 in the Typhus, famine country 24 1813-1814 Yes 10%, 25,000 in Mexico City SOURCE: Adapted from Chimalpahin (1998); Cooper (1965); de Torquemada (1969); Flores (1888); Flores and Malvido (1985); Garcia et al. (2003). this time, the specific factors that relate hemorrhagic fevers with drought, as well as the cause of cocoliztli, remain unknown. Communication Networks Promote Epidemics Data presented here indicate that the frequency of famines decreased and eventually disappeared in the centuries following the Conquest. Obviously, droughts, the main cause of famines, did not become less severe or less frequent after the Conquest. A possible explanation is that the effects of droughts were
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 191 reduced by the system of roads developed by the Spanish. In pre-Hispanic times, the native population was fragmented over a vast country; ethnic groups lived in relative isolation, surrounded by enemies and fighting for space and resources. The network of roads built under Spanish rule unified the territory for com- mercial (collection of silver) and military (control of Indian revolts) purposes. As the road system expanded, the traffic of food, silver, and people was set in motion. Development of the colonial road network had an unintended impact on the introduction and movement of infectious diseases. Cocoliztli, smallpox, and measles are the diseases that contributed the most to the population collapse of the sixteenth century (AcuÃ±a-Soto et al., 2000, 2004; Flores, 1888; Flores and Malvido, 1985; Marr and Kiracofe, 2000). Since the cause of cocoliztli remains unknown, this discussion is centered on the impact of improving communication networks on smallpox and measles, the then-emerg- ing infectious diseases in Mexico. Both diseases are transmitted exclusively by human pathogens and require between 500,000 and 600,000 individuals to remain indefinitely in circulation. Because of the lack of immunity, the first epidemic events of either disease were invariably catastrophic for the native population. After the first wave of infection, the viruses behaved differently depending on the size of the population and the proportion of immune individuals. In large popu- lations, both viruses probably circulated at low frequency during the interepi- demic periods. Waiting until a new generation of unprotected children reached a critical number, they then reappeared as epidemics that preferentially affected children. In small and isolated human groups, the situation was different. After a fast outbreak, the infection disappeared quickly because very soon, nonimmune individuals were unavailable; the diseases returned only when they were reintro- duced years or decades later. The fragmentation and isolation of Indian groups offered protection from smallpox and measles only until their settlements were connected by the advancing network of roads and missions. Once these settle- ments were linked by a system of roads, exposure to the repeated introductions of smallpox and measles had devastating effects and correlates with the order in which local Indian populations collapsed. The population loss in Mexico due to smallpox and measles epidemics was not uniform in space and time. In the Valley of Mexico, it occurred during the sixteenth century, with high mortality registered in the north until the second half of the seventeenth century. In the Baja California peninsula, it occurred until the eighteenth century, shortly after the first mission was founded in 1697â178 years after the arrival of the Spaniards in Mexico (Gerhard, 1996). The Past and the Present Compared Droughts have been a central factor for at least three of the major population collapses that occurred in central Mexico during the last 1,000 years. Droughts have been associated with both famines and epidemics of hemorrhagic fevers that
192 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS caused the death of millions of people. The emergence of new pathogens and the increased connectivity of native populations also appear to explain much of the temporal and spatial patterns of famine and epidemic disease in colonial Mexico. To some degree, the globalization of the modern world resembles the colonization of Mexico in the sixteenth to eighteenth centuries. The increased connectivity of populations and the emergence of aerosol-borne pathogens have proven to be a dangerous combination. New roads and trade leveraged the impact of measles and smallpox in colonial Mexico, as globalization has leveraged SARS and influenza in recent times. Acknowledgments The authors wish to thank Yvonne Rosenstein for her helpful comments on the manuscript. WILDLIFE HEALTH AS AN INDICATOR OF CLIMATE CHANGE Pablo M. Beldomenico, M.V., M.P.V.M., Ph.D. Wildlife Conservation Society Damien O. Joly, Ph.D.7 Wildlife Conservation Society Marcela M. Uhart, M.V. 7 Wildlife Conservation Society William B. Karesh, D.V.M. 7 Wildlife Conservation Society Introduction The changes in climate we are experiencing as global warming and distur- bance in precipitation regimes (IPCC, 2001) are having an impact on the health of wild animals, with resulting deleterious impacts on major human interests. In this paper, we review the relationship between climate change and wildlife health and argue that monitoring wildlife health provides an effective and sensitive indicator and predictor of climate-related emerging infectious diseases. â Global Health Programs.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 193 Effects of Climate Change on Wildlife Health After a long period of neglect, pathogens have recently been suggested as important drivers of host population dynamics (Hudson et al., 1998; Tompkins and Begon, 1999). At the same time, disease has been implicated in major wild- life population declines (Pounds et al., 2006; Roelke-Parker et al., 1996). Wildlife health, therefore, is an important factor for population sustainability and system resilience and, hence, is drawing conservation attention. Because most human emerging infectious diseases originate from a wildlife reservoir (Jones et al., 2008), wildlife health is critically linked to public health. Wildlife species may serve as an early warning for some diseases, as is the case for howler monkeys and yellow fever in South America (Rawlins et al., 1990) or great apes and Ebola hemorrhagic fever in central Africa (Karesh and Reed, 2005). In addition, due to the flux of pathogens occurring at the wildlife-livestock interface, wildlife health is also important for domestic animal health. Suggested Mechanisms Climate change may affect wildlife health in several ways because the deter- minants of disease incidence are numerous and specific to the disease in question and climate change may influence each of these factors. Moreover, the direct and indirect impacts of climate change on host-pathogen interactions might favor some host species because they could release hosts from the population control exerted by pathogens by interfering with the precise conditions required for pathogen viability (either directly or indirectly; i.e., changes in vector abun- dance) and shift host population regulation to other factors, such as food or other resource availability. It is important to recognize that the components of a dynamically functioning ecosystem are interconnected; thus, the influence of climate change on animal distribution, abundance, or demography via infectious disease may be indirect due to shifts in relationships such as competitive advan- tages among conspecifics, predator-prey dynamics, and so forth. In the interests of simplification, we propose that climate change may directly modify the patterns of infectious disease basically in two ways: 1. Favoring pathogens (increasing pathogen and/or vector proliferation, or pathogen and/or vector survivability) 2. Increasing the hostâs susceptibility to infection Climate Change Favoring Pathogens and Their Vectors Changes in climate shift the relationship among pathogen, host, and the envi- ronment. Focusing first on the pathogen, we know that climatic conditions play a significant role in the geographic and temporal distribution of pathogens and, as
194 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS appropriate, their vectors. Environmental conditions also affect the viability and reproductive success of both pathogens and vectors, which can be thought of as the nonhost components of severity of infection. Changes in distribution and seasonalityâ It is known that, within limits, arthro- pod populations are favored by heat and moisture. Therefore, we anticipate that climate change will influence vector-borne diseases. In fact, a number of vector-borne human and domestic animal diseases have increased in incidence or geographic range in recent decades (e.g., malaria, African trypanosomiasis, tick-borne encephalitis, yellow fever, plague, dengue, African horse sickness, bluetongue) (Harvell et al., 2002). These changes have been or can be identified because the diseases are important for public health or domestic animals, and hence records to detect their occurrence currently exist in many cases. It is close to impossible to know if vector-borne diseases are changing in wildlife unless more widespread, systematic monitoring is put into place. A large body of research has been conducted to attempt to predict the expan- sion of vector-borne diseases due to climate change. Largely, these efforts involve modeling. For example, a study addressed how climatic variables determined the abundance of Ixodes scapularis (the tick that transmits Lyme disease) and ehrlichiosis and babesiosis in eastern North America (Ogden et al., 2005). The authors used current knowledge of tick biology, originating from empirical and experimental data, to construct a theoretical model that predicted the abundance of ticks based on the climatic and seasonal variables they had measured. In a sub- sequent study, they used this model to simulate the expansion of tick distribution under two projected climate change scenarios, predicting a substantial northward movement of tick range (Ogden et al., 2006). These examples support the basic concept that climate change is expected to influence the geographic range of vec- tors and their transmitted diseases. A similar situation may be observed for pathogens affecting ectothermic (cold-blooded) hosts or those that proliferate outside affected individuals, because the pathogens are more exposed to ambient temperature as opposed to those hav- ing a life cycle that is completed almost entirely inside a host that preserves a constant temperature (endothermic). A rise in average temperature may not only affect the proliferation of the pathogen, but also have the potential to modify the seasonality of the diseaseâwhich could occur earlier every yearâand remain infective or active for a longer period of time (Harvell et al., 2002). Harmful algal blooms (HABs), also known as âred tides,â are events in which single-celled protists (dinoflagellates) proliferate rapidly and accumulate in the water column. These events are associated with wildlife mortalities because under certain circumstances these organisms can produce potent toxins. A systematic increase in seawater temperature may contribute to the occurrence of HABs (Juhl, 2005). For example, the Wildlife Conservation Society investigated an episode of mass bird die-off in the Malvinas-Falkland Islands in which high levels of
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 195 toxins produced by these dinoflagellates were detected in sick or dead gentoo penguins as well as in the marine prey species found in the digestive tracts of affected animals (Uhart et al., 2004). This was the first report of paralytic shellfish poisoning affecting seabirds in the southwest Atlantic, which might suggest that climate change is aiding the expansion of this type of disease to more extreme latitudes. Increased severity of diseaseâ An increased intensity of parasitism or severity of infection may also result from favorable conditions for pathogens. In the St. Kilda archipelago of Scotland, for example, feral populations of Soay sheep experience periodic mass mortalities (Coulson et al., 2001). Although the proximate cause of death has been determined to be protein-energy malnutrition, parasites have been implicated as a contributory factor (Grenfell et al., 1995). At first observation, the depth of the population crashes was critically dependent on the weather, and large numbers of trichostrongylid nematodes were found in dead animals. An experi- mental study showed that the administration of antihelminthic therapy reduced mortality considerably (Gulland et al., 1993), which supported the link between parasites and death. Trychostrongylids have a life cycle that involves several stages outside their hosts, making them highly vulnerable to environmental con- ditions. In particular, larvae are very susceptible to desiccation, so humidity and precipitation regimes are crucial for their survival outside their hosts (Wharton, 1982). The emerging scenario would be that increased precipitation would allow for increased larval survival, which in turn would lead to higher parasite burdens, contributing to elevated mortality. Climate Change Increasing Hostâs Susceptibility to Infection As mentioned, changes in climate shift the relationship among pathogen, host, and the environment. Focusing on the hosts, we know that climatic condi- tions can affect their behavior and hence susceptibility to infectious organisms due to change in exposure or contact rates. While genetics provides a framework, host immunity or disease resistance is also dependent on physiologic mechanisms affected by environmental conditions. The impact of changes in climate can occur at a rate more rapid than a hostâs ability to adapt. Increased exposure to pathogensâ One way in which climate change can result in an increased susceptibility to infection is by inducing changes in host behav- ior, which may determine increased exposure to pathogens. While some parts of the world are projected to become more moist, others are projected to become drier (IPCC, 2001). On the Patagonian Steppe, for example, water supplies are threatened (Barros et al., 2000). As a result, the concentration of individuals â A common parasitic gastrointestinal nematode found in many ungulate species.
196 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS around water resources may grow, thereby increasing intraspecific interaction and indirect contact rates, and possibly shifting density-dependent infectious disease relationships. Since water supplies are frequently shared among wildlife and domestic animals, these alterations may increase the risk of pathogen exchange at the wildlife-livestock interface. Another example of the way in which climate change may result in increased susceptibility to infection is its impact on feeding behavior. For instance, the reduction in sea ice is causing a change in the behavior and diet of walruses, which are becoming more pelagic and are preying more on ringed seals (a carnivore) and less on invertebrates. This, in turn, may increase the prevalence of trichinellosis in walruses (Rausch et al., 2007). Finally, climate change may determine that some vertebrate hosts expand their distribution, and thus expose immunologically naÃ¯ve species to their pathogens and nonpathogenic commensal organisms, as well as expose themselves to new pathogens. For example, tropi- cal deglaciation is causing an increase in the elevational limit of some anurans, which have taken with them the agent of chytridiomycosis (Batrachochytrium dendrobatidis) to unprecedented altitudes (Seimon et al., 2007). Decreased host resistanceâ The susceptibility of hosts may also increase if their intrinsic vulnerability to disease is affected. For many species, climate change will serve as an additional form of stress. The effect of stress on vertebrates is well known: a cascade of neuroendocrine mechanisms triggered by stress result- ing in a reduction in immune function (Lochmiller and Dabbert, 1993). Since wild species usually live on tight energy budgets (Beldomenico et al., in press), a number of physiological functions compete for these limited resources. An increased demand by one system results in fewer resources for the rest. If climate change causes resources to become more scarce, or of poor quality, or if other physiological systems increase their demands (reproduction, molting, migration, etc.), then the share left for immunological investment will be reduced. A recent study on rodents demonstrates that poor body condition predisposes individuals to a variety of infections, and these infections further decrease the condition of individuals, triggering a âvicious cycleâ that eventually ends up in death and, therefore, population declines (Beldomenico et al., 2008). Thus, maintaining good body condition is important to reducing infection, and avoiding infection is essential to maintaining good condition. If climate change-induced food resource limitations or stress impoverish the condition of many individuals in a population, this type of infection-declining condition cycle may be triggered and the population will fail. Amphibians are particularly sensitive to climate disturbance. In the last three decades, thousands of species have experienced population declines worldwide and more than 100 have disappeared (Stuart et al., 2004), many of them in seem- ingly undisturbed environments. The chytrid fungus B. dendrobatidis has been implicated in many of these amphibian population crashes (Berger et al., 1998;
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 197 Weldon et al., 2004); however, amphibians have also been declining in regions where the fungus is absent, and the fungus has been found in places with no affected frogs (Di Rosa et al., 2007). In declining populations where the chytrid fungus was not present, other pathogens were found at high prevalences, namely, Saprolegnia ferax (Kiesecker et al., 2001), Amphibiocystidium ranae (Di Rosa et al., 2007), ranaviral disease (Bollinger et al., 1999), and metazoan parasites (GarcÃa et al., 2007). Amphibian declines have been correlated with climatic change, and a hypoth- esis for climate-driven epidemics arising from climate favoring pathogens dif- ferentially over hosts has been proposed to explain amphibian declines (Pounds et al., 2006). However, the vicious cycle hypothesis (Beldomenico et al., 2008) may also be a satisfactory explanation. Reading (2007) presents evidence that environmental warming negatively affects the body condition of toads. Alford et al. (2007) observed that frog population declines are preceded by an increase in indicators of stress. The emerging vicious cycle hypothesis proposes that climate disturbance is affecting the condition of amphibians, which predisposes them to more frequent infections and/or infections of increased severity, which triggers a vicious cycle with the potential to cause amphibian population declines. Thus, while the synergy between poor condition and infection may be a proximate cause of these declines, the ultimate cause would be a condition impoverished by climatic changes. In summary, there are multiple mechanisms by which climate change could affect wildlife health including, but not limited to, the following: â¢ Expansion in the geographic distribution of pathogens, vectors, or hosts â¢ Changes in the seasonality of some diseases â¢ Increased severity of disease â¢ Increased exposure to pathogens â¢ Decreased host immunity All may result in a disruption of population and system health dynamics. Thus, independent of mechanism, monitoring the health of wildlife populations provides a sensitive and quantitative method to detect changes and serve as an early warning system. One subject in need of further investigation is the relationship between evolution and an organismâs ability to adapt to these rapid changes. It is possible that genetic shifts will modulate local effects of climate change. However, there is little evidence that evolution will mitigate negative effects of climate change at the species level (Parmesan, 2006). Moreover, it should be considered that the speed of evolution is different among different taxa. Bacteria and viruses, for example, have the capacity to evolve rapidly, adapting to environmental changes before their hosts. This might result in a differential adaptation favoring patho- gens and inadvertently causing host population declines or extinctions.
198 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Summary Many studies have shown that climate change can influence the dynamics of wildlife diseases. The question is: Can this relationship be utilized to monitor the ecological effects of climate change and to predict and prevent the emergence of new diseases threatening wildlife, livestock, and human health? We argue that in the context of health and climate change, monitoring wildlife health is of direct relevance for several reasons. First, the emergence of human and livestock diseases is closely tied to wildlife health (Jones et al., 2008; Karesh and Cook, 2005). As a result, detection of climate-related emergence of disease in wildlife populations provides an early warning of system disturbance and, thus, potential human and domestic animal health concerns (early-warning capability being the useful feature of any indicator of ecological change [Carignan and Villard, 2002]). Second, the range of population turnover times in hosts and wildlife pathogens, from short generation times in bacterial and viral pathogens to rela- tively longer generation times in helminths and other parasites, and even decades in some hosts, provides an opportunity to evaluate change at a variety of temporal scales. Third, for some wildlife populationsâparticularly hunted or managed populationsâgood long-term baseline health data exist. Consequently, we know the range of what is normal and can more easily determine what is abnormal or different. Finally, and most relevant to this discussion, as âsystem integrators,â the health of wild animals is tuned to a set range of natural variation and, therefore, provides a sensitive indicator of change. USE OF CLIMATE VARIATION IN VECTOR-BORNE DISEASE DECISION SUPPORT SYSTEMS William K. Reisen, Ph.D. University of California, Davis Christopher M. Barker, M.S.9 University of California, Davis Introduction 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 â Center for Vectorborne Diseases, School of Veterinary Medicine, University of California, Old Davis Road, Davis, CA 95616, Reisen E-mail: email@example.com; Barker E-mail: cmbarker@ ucdavis.edu.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 199 FIGURE 3-2â West Nile virus transmission cycle. SOURCE: CDC (2005). some extent domestic animals causes disease but is a âdead endâ for the virus. Figure 3-2 Surveillance data on arthropod vectors or infection in reservoir hosts typically are skeletal, often leaving the passive detectionfrom powerpoint pdf made 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 interven- tion 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 Californiaâs mosquito-borne encephalitis virus surveillance and control program as an example. The California program
200 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS provides an excellent model because (1) the state encompasses multiple biomes that vary markedly across north-south latitudinal and east-west elevational gradi- ents; (2) an intensive surveillance program has been consistently monitoring mos- quito 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 tem- perature and wetness, which normally are averaged over a defined time period to delineate âaverageâ conditions for a specific geographic region. Climate variation describes deviations about these long-term means that may be measured at a vari- ety of scales from days to years, whereas climate change is directional and con- sists 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 Dif- ference 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 NASAâs Terrestrial Observation and Prediction System (TOPS; see Figure 3-3), a model- ing 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 Ã 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 anoma- lies can be tracked at varying scales. Short-term changes (weather) can be fore- cast days to weeks in advance and predict events such as rainstorms or heat waves that may immediately affect vector-borne pathogen transmission. Interannual variationâdriven by global cycles such as El NiÃ±oâmay be used to forecast ecosystem change seasons in advance and therefore forecast changes in vector
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 201 Weather networks Orbiting satellites Terra/Aqua/Landsat/Ikonos Landcover/ change, leaf area index, surface temperature, snow cover and Temperature/rainfall/ cloud cover radiation/humidity/wind Ancillary data Topography, river Ecosystem Weather & climate networks, soils simulation models forecasts Monitoring & forecasting Stream flow, soil moisture, phenology, fire risk, forest/range/crop production FIGURE 3-3â TOPS system brings ground and remote measures of climate into ecological models to monitor and forecast risk. SOURCE: NASA (2007). Figure 3-3 abundance (Reisen et al., 2008a) and outbreaks of arbovirusesand isas Rift Valley graphics are bitmapped, but type has been replaced such real type fever virus (Anyamba et al., 2002; Linthicum et al., 1991). Longer, interdecadal trends may indicate shifts in baselines (i.e., climate change), and these gradual changes may elongate transmission seasons and extend vector and pathogen dis- tributions. Change has been most clearly detected at northern latitudes (Githeko et al., 2000) and in urban landscapes that present their own microcosms for climate change and variation (Kalnay and Cai, 2003). Certainly the consistently
202 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS high incidence of WNV in the U.S. central plains and central Canada (data not shown) was unexpected and generally has tracked above-normal summer tem- peratures (see Figure 3-4), even though these relatively low mean temperatures were considered suboptimal for virus amplification within the primary mosquito vector Culex tarsalis (Reisen et al., 2006b). Impact of Climate Variation on Vectors and Pathogens Vectors and the pathogens they transmit are especially subject to climate variation because climate markedly affects arthropod population size and dynam- ics, and because pathogen replication rates are influenced directly by ambient temperatures during infection of the poikilothermic10 arthropod vector. This is especially true for the mosquito-borne encephalitides at temperate latitudes where temperature delineates amplification and transmission season duration. Climate variation also indirectly determines the size and age structure of avian maintenance and amplifying host populations by impacting primary productivity and therefore the abundance and distribution of food sources. In contrast, the impact of climate on mosquito populations is rapid and direct. Many vector mosquitoes utilize surface water accumulations for larval develop- ment that depend directly or indirectly on precipitation. For rural species such as Culex tarsalis, the timing and size of the adult population peaks depend directly on winter and spring rainfall and snowmelt runoff, and have been related to El NiÃ±o conditions and the associated wet winter and spring seasons (Reisen et al., 2008a). Conversely, urban species, such as those in the Culex pipiens complex, utilize peridomestic and underground drainage systems and may be favored by La NiÃ±a conditions of high temperature and low rainfall. In urban centers such as Los Angeles, high rainfall and associated runoff scour the underground larval habitats and actually reduce vector abundance (Su et al., 2003). Warm temperatures increase the rate of mosquito population growth, reduce adult survival, and increase the frequency of blood feeding as well as the chances of virus acquisition and transmission (Reeves et al., 1994; Reisen, 1995). Tem- perature also is positively associated with encephalitis virus replication within the mosquito vector. The time from virus infection to transmission (the extrinsic incubation period [EIP]) is directly related to ambient conditions and can be described by degree-day models (Reisen et al., 1993, 2006b). These regimens frequently delineate episodic waves of transmission during outbreaks (Nielsen et al., 2008) as well as transmission seasons and the geographic distribution of virus amplification. 10â Any organism whose body temperature varies with the temperature of its surroundings.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 203 FIGURE 3-4â Incidence of human West Nile virus cases per million population and tem- perature anomalies for the United States, 2003-2007. Figure 3-4 SOURCES: CDC (2007); NOAA (2008). bitmapped, though each map is separate
204 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Components of a Vector-Borne Pathogen Surveillance Program Comprehensive vector-borne disease surveillance programs include measures of early-season meteorological conditions that may be used to forecast risk for the coming virus transmission season. As temperatures increase and vectors become active, these programs begin monitoring vector abundance and virus activity within the primary enzootic cycle as a measure of the risk of pathogen transmis- sion to humans and to validate earlier forecasts. These data can be combined in models that forecast or measure risk. For WNV in California, climate measures focus on temperature because the effects of precipitation vary among regions (Reisen et al., 2008a) and vector species (e.g., urban Cx. pipiens Âmosquitoes do well under hot, dry conditions, whereas rural Cx. tarsalis do well under wet con- ditions in many areas). Culex abundance is monitored by New Jersey or American light traps (Mulhern, 1985), the Centers for Disease Control and Preventionâs (CDCâs) dry ice-baited traps (Newhouse et al., 1966), and gravid female traps (Reiter, 1987) in urban situations. Mosquitoes and dead birds are tested Ârapidly for infection by detecting viral RNA using robotic extraction and real-time reverse transcriptase-polymerase chain reaction (RT-PCR) testing procedures (Shi et al., 2001). Avian infection indicating active transmission is measured by sequentially bleeding sentinel birds, such as chickens, for seroÂconversion11 and by recording and testing dead birds reported by the public. Equine and human cases are diagnosed by healthcare providers and confirmed serologically at local laboratories. The temporal cascade of events and surveillance data along a typi- cal WNV amplification curve is shown in Figure 3-5. The horizontal line shows the hypothetical level of amplification required before tangential transmission to humans is frequent. The key to effective surveillance is the rapid dissemination of results and analysis to individuals responsible for intervention decisions. To expedite data exchange in California, a web-based management system called the Surveillance GatewayÂ© has been implemented in which data are entered, stored, analyzed, and displayed (see Figure 3-6). When combined with rapid laboratory diagnostics, mosquitoes collected on Monday or Tuesday and immediately shipped to the laboratory are routinely tested, recorded, reported, and mapped online as early as Thursday or Friday of the same week. These data are then compared to his- torical records, and a risk score is assigned for each parameter (see Table 3-5). The scores for individual surveillance parameters are then averaged to obtain an overall risk score ranging from 1 to 5, where 1.0-2.5 denotes a ânormal season,â 2.6-4.0 represents increasing risk requiring âemergency planning,â and 4.1-5.0 indicates âepidemic conditions.â Risk levels for WNV can be forecast and then measured based on average daily temperatures (see Figure 3-7). These risk levels, based on temperature, 11â Conversion of a sentinel host from antibody negative to positive after infection.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 205 Human cases Equine cases Amplification Avian infection Mosquito infection Climate Mosquito abundance Time [months] FIGURE 3-5â Sequence of surveillance data collected during seasonal virus amplification. Figure 3-1 also Field data retrieved and laboratory test results entered Figure 3-5 MVCAC agencies enter data Central server Results automatically sent to CDC Interactive maps automatically updated Historical Results reported to database client automatically updated after entry Arbovirus bulletins FIGURE 3-6â Data flow through the Surveillance GatewayÂ© system. Figure 3-6
206 TABLE 3-5â California Mosquito-Borne Virus Surveillance and Response Plan Model Scores for Each Surveillance Parameter Risk Average Daily Adult Mosquito Mosquito Chicken Dead Bird Level Temperature Abundance MIR/1,000 Seroconversions Infections Equine Cases Human Cases 1 <13Â°C <50% 0 0 in regiona 0 in region 0 in region 0 in region 5 year avg. 2 13-18Â°C 50-90% 0.1-1.0 â¥1 in region,a â¥1 in region, 5 year avg. 0 in agency 0 in agency 3 18-22Â°C 91-150% 1.1-2.0 1 flock in agencyb 1 in agency â¥1 in region, â¥1 in region, 5 year avg. 0 in agency 0 in agency 4 22-26Â°C 151-300% 2.1-5.0 2 flocks in agencyb 2-5 in agency 1-2 in agency 1 in agency 5 year avg. 5 >26Â°C >300% >5.0 >2 flocks in agencyb >5 in agency >2 in agency >1 in agency 5 year avg. aRegion refers to areas encompassed by the formalized regions of the California Mosquito Control Association. bAgency refers to the boundaries of the various mosquito districts. SOURCE: Adapted from California Department of Public Health (2007).
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 207 relate directly to the impact of temperature on the duration of EIP estimates from the degree-day model (Reisen et al., 2006b). Cool temperatures, which require elongated periods before transmission is possible, are assigned low levels of risk; hot temperatures, which result in rapid and efficient transmission, are assigned elevated risk scores. Figure 3-7A shows risk scores based on tempera- ture assigned to mosquito control districts in California during each month in 2007, with elevated risk shown in red. Above-normal temperatures in the southern San Joaquin Valley produced earlier-than-normal increases in urban Cx. pipiens complex mosquito abundance and virus infection rates, and epidemic risk levels by June were higher and earlier than in other areas of California (Figure 3-7B). Coincidentally, the 2007 WNV epicenter for California was Bakersfield in Kern County, and the number of human cases increased with risk, peaking in late July (Figure 3-7C). Although temperature was predictive of risk and receptivity for amplification in the Central Valley, other epidemiological factors are clearly needed because there was little enzootic activity in the hot southeastern deserts during 2007. This area has had limited annual enzootic amplification and few human cases, even though temperatures are very high for most of the year (Reisen et al., 2008b). Vector mosquitoes are plentiful here, but virus amplification rarely has reached outbreak levels, because American crows and other corvids are uncommon in the desert biome of California (Reisen et al., 2006a). Lack of abundant corvid populations may thereby limit infection in peridomestic Cx. pipiens complex mosquitoes that are less competent hosts than Cx. tarsalis (Reisen et al., 2005). Response to Surveillance Data Integrated vector management programs are driven by decision support sys- tems based on surveillance data (see Figure 3-8). The California Mosquito-borne Encephalitis Virus Surveillance and Response Plan describes the attributes of a comprehensive surveillance program, provides methods for calculating risk, and provides a series of response guidelines for each level of increased risk. 12 Climate variation and forecasts provide the only early-season surveillance infor- mation upon which to gauge the response program to pending risk. Early-sea- son response activities include preventive methods, such as equine vaccination, source reduction, larval mosquito control, and public education. Surveillance activities, such as mosquito collection and bird testing, typically begin during spring, and the timing and sensitivity of these programs determine the limit of virus detection and therefore the time available for control operations (control window) to arrest amplification at levels where human risk is minimal (see Figure 3-8A, dashed line). If routine control methods fail to arrest virus amplification and risk increases to emergency planning or epidemic levels, emergency adult 12â See http://westnile.ca.gov/resources.php.
208 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS A Kern Co. 35 Duration of EIP50 Risk level 2 30 25 20 3 Days 15 5 4 10 5 0 30 26 22 18 Temp[C] B 5 4 COAV GRLA Figure 3-7A KERN Risk level MERC bitmapped, except gray box (days/temp) lower right PLCR each map is separate 3 SAYO Kern Co. SHAS SJCM SUYA 2 TEHA TRLK WEST WVAL 1 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 JAN FEB MAR APR MAY JUN JUL AUG SEP OCT Half-month FIGURE 3-7 Figure 3-7B
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 209 C Number of cases Kern Co. Half-month FIGURE 3-7 California mosquito district risk levels 1-5 for WNV transmission: esti- mated from (A) temperatures downloaded from the TOPS system (inset shows the increas- ing duration of extrinsic incubation period with decreasing temperature and associated risk), (B) the entire risk assessment model calculated bimonthly for selected mosquito control agencies, and (C) the number of human cases within each mosquito control region of California. mosquito control and recommendations for human personal protection provide the only methods to interrupt transmission and protect the public. Use of forecasts and nowcasts13 for making control decisions have inherent problems, especially in California and other areas where the public (especially anti-insecticide advocates) frequently believes that the risk of applying insec- ticides, even in ultralow volumes (ULVs), exceeds the risk of illness or death from viral infection. Using forecasts and surveillance measures, it is possible to accurately determine that the risk of an epidemic is imminent and to apply large-scale aerial ULV adulticide treatments to immediately reduce vector abun- dance and transmission. In this scenario, cases are prevented, but the rationale for a large monetary expenditure and exposure of the human population appears unjustiï¬ed to some because we cannot know whether the prevented cases would have occurred in the absence of vector control measures. The actions of vector control and public health ofï¬cials are then questioned in the press, despite the fact that risks from pyrethrin insecticides and synergists are minimal (Peterson et al., 2005). If concurrent measures of risk are used instead of forecasts, the virus may have ampliï¬ed to higher levels, some humans will have been infected before 13 Forecasts or real-time measures of events in the immediate future.
210 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS A Reactive Public education: emergency Personal Protection EPIDEMIC intervention Amplification AMPLIFICATION Proactive CONTROL population MAINTENANCE WINDOW management Time [months] Preventive methods B Figure 3-8A Integrated vector management Personal protection: - avoidance - repellents Reservoir removal or vaccination Vaccination FIGURE 3-8â Intervention options for WNV shown in relation to (A) the amplification curve and (B) the transmission cycle. Figure 3-8B SOURCE: Figure 3-8B is modified from CDC (2005).
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 211 adulticides are applied, and the public health consequences can be considerable. In this scenario, cases of severe illness and some deaths occur, the epidemic peak frequently is realized and recognized by the public, and the population is exposed to the same level of adulticides as during preventive sprays. In the latter case, both the antispray advocates and the families of affected cases are perplexed and the cost-benefit ratio is increased. Surveillance in a Changing World West Nile virus will not be the last invading zoonosis to reach North Amer- ica. Rapid globalization of commerce and travel increases the probability of additional vector or pathogen introductions. For example, Aedes albopictus mos- quitoes, which are vectors of dengue, chikungunya, and other viruses, were transported to California and distributed throughout Los Angeles with imported âLucky Bambooâ (Dracaena spp.) from China (Linthicum et al., 2003; Madon et al., 2002). Fortunately, surveillance detected this introduction and the invading mosquitoes were eradicated. Vector-borne pathogens also frequently enter the United States. Travelers from India with chikungunya viremia levels sufficient to infect mosquitoes have been detected in the United States during the current epidemic (Lanciotti et al., 2007), but the risk for local transmission is unknown because the vector competence of the local mosquito populations is unknown. Climate change exacerbates these situations in time and space by elongating transmission seasons and increasing the geographic area receptive to pathogen introduction (Epstein, 1998; Patz et al., 2000). The WNV epidemic has provided a wake-up call for public health agencies and clearly delineated the inability of current control programs to contain an invading, mosquito-borne zoonosis (Holloway, 2000). Unfortunately, the waning WNV epidemic has resulted in the diversion of both research and surveillance funding to other health problems, and many of those programs still in place focus surveillance diagnostics only on WNV and will not detect the next invading pathogen. Acknowledgments This research was funded by the Climate Variability and Human Health and the California Applications Programs, Office of Global Programs, NOAA; Deci- sion Support through Earth Science Results, NASA; Research Grant AI55607 from the National Institute of Allergy and Infectious Diseases, NIH. We are especially indebted to the corporate members of the Mosquito and Vector Control Association of California and the California Department of Public Health who granted permission for us to utilize their surveillance data for this project. Forrest Melton, NASA, and Bborie Park, UC Davis, assisted with data manipulations.
212 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS REFERENCES Overview References Leroy, E. M., B. Kumulungui, X. Pourrut, P. Rouquet, A. Hassanin, P. Yaba, A. Delicat, J. T. Paweska, J. P. Gonzalez, and R. Swanepoel. 2005. Fruit bats as reservoirs of Ebola virus. Â Nature 438(7068):575-576. Linthicum, K. J., A. Anyamba, C. J. Tucker, P. W. Kelley, M. F. Myers, and C. J. Peters. 1999. Climate and satellite indicators to forecast Rift Valley fever epidemics in Kenya. Science 285(5426):397-400. Pinzon, J. E., J. M. Wilson, C. J. Tucker, R. Arthur, P. B. Jahrling, and P. Formenty. 2004. Trigger events: enviroclimatic coupling of Ebola hemorrhagic fever outbreaks. American Journal of Tropical Medicine and Hygiene 71(5):664-674. Reisen, W. K., and C. M. Barker. 2008. Use of climate variation in vector-borne disease decision support systems. In Institute of Medicine. Global climate change and extreme weather events: understanding the contributions to the emergence, reemergence, and spread of infectious dis- ease. Washington, DC: The National Academies Press. Reisen, W. K., D. Cayan, M. Tyree, C. M. Barker, B. F. Eldridge, and M. Dettinger. In press. Impact of climate variation on mosquito abundance in California. Journal of the Society for Vector Ecology. Tucker, C. J., A. Anyamba, K. Linthicum, and J. Small. In press. The 2000 Rift Valley hemorrhagic fever outbreak in southwestern Arabia: bio-terrorism or natural occurrence? Paper submitted for publication. AcuÃ±a-Soto et al. References AcuÃ±a, R., ed. 1982. Relaciones GeogrÃ¡ficas. Mexico City: Instituto de Investigaciones AntropolÃ³gi- cas, Universidad Nacional AutÃ³noma de MÃ©xico. AcuÃ±a-Soto, R., L. CalderÃ³n-Romero, and J. H. Maguire. 2000. Large epidemics of hemorrhagic fevers in Mexico, 1545-1815. American Journal of Tropical Medicine and Hygiene 62(6):733-739. AcuÃ±a-Soto, R., D. W. Stahle, M. K. Cleaveland, and M. D. Therrell. 2002. Megadrought and mega- death in 16th century Mexico. Emerging Infectious Diseases 8(4):360-362. AcuÃ±a-Soto, R., D. W. Stahle, M. Therrell, D. Griffin, and M. K. Cleaveland. 2004. When half of the population died: the epidemic of hemorrhagic fevers of 1576 in Mexico. FEMS Microbiology Letters 240(1):1-5. AcuÃ±a-Soto, R., D. W. Stahle, M. D. Therrell, S. Gomez Chavez, and M. Cleaveland. 2005. Drought, epidemic disease, and the fall of classic period cultures in Mesoamerica (AD 750-950). Hemor- rhagic fevers as a cause of massive population loss. Medical Hypotheses 65(2):405-409. Anonymous. 1978. Tira de Tepechpan. In Biblioteca EnciclopÃ©dica del Estado de MÃ©xico, edited and commentary by X. Noguez. Mexico City: Estado de MÃ©xico. Anonymous. 1980a. Anales de Tlaltelolco, unos anales histÃ³ricos de la NaciÃ³n Mexicana, edited by Rafael PorrÃºa. Mexico City. Anonymous. 1980b. CÃ³dice Aubin (CÃ³dice de 1576), edited by A. PeÃ±afiel (1902). Mexico City: Editorial InnovaciÃ³n. Anonymous. 1981. Anales de Tecamachalco, crÃ³nica local y colonial en idioma Nahuatl 1398-1590. Edited by A. PeÃ±afiel. Mexico City: Editorial InnovaciÃ³n. Anonymous. 1995. Anales de Tula, Hidalgo. In Fuentes y estudios sobre el MÃ©xico indÃgena, Vol. 6. Mexico City: Instituto Nacional de AntropologÃa e Historia, Universidad de las AmÃ©ricas.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 213 Anonymous. 1998. Histoire mexicaine depuis 1221 jusquâen 1594: Manuscrito num. 40 del Fondo de Manuscritos Mexicanos, Biblioteca Nacional de Francia (Coleccion cientifica). Translation by X. M. GonzÃ¡lez. Mexico City: Instituto Nacional de AntropologÃa e Historia. Camargo, D. M. 1999. Relaciones geogrÃ¡ficas de Tlaxcala. Tlaxcala/San Luis PotosÃ, Mexico: Gobi- erno del Estado de Tlaxcala/El Colegio de San Luis. Chimalpahin, D. 1998. Las Ocho Relaciones y el Memorial de Colhuacan, 2 vols., edited and trans- lated by R. Tena. Mexico: Consejo Nacional para la Cultura y las Artes. Chimalpahin, D. 2001. Diario. Mexico: Consejo Nacional para la Cultura y las Artes. Clavijero, F. J. 1945. Historia antigua de Mexico, Vol. 4. Mexico City: Editorial PorrÃºa. Cook, S. F., and F. Sherburne. 1985. El Hospital del Hambre de Guadalajara: un experimento de asistencia mÃ©dica. In Ensayos sobre la Historia de las Epidemias en MÃ©xico, edited by C. E. Flores and E. Malvido. Mexico City: Instituto Mexicano del Seguro Social. Cook, S. F., and L. B. Simpson. 1948. The population of central Mexico in the sixteenth century. ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ Ibero Americana (31):43. Cooper, D. B. 1965. Epidemic disease in Mexico City, 1761-1813: an administrative, social, and medical study. Austin, TX: University of Texas Press. de Alva IxtlixÃ³chitl, F. 1975. Obras HistÃ³ricas, 2 vols. Mexico City: Instituto de Investigaciones HistÃ³ricas, Universidad Nacional AutÃ³noma de MÃ©xico. de CÃ¡rdenas, J. 1945. Problemas y secretos maravillosos de las Indias. In ColecciÃ³n Incunables Americanos, Vol. 9. Madrid, Spain: Ediciones de Cultura Hispanica. de Grijalva, J. 1924. De las seÃ±ales que en esta tierra se vieron por este tiempo, y de la gran peste que llamaron cocoliztli, CrÃ³nica de la Orden de San AgustÃn en las provincias de la Nueva EspaÃ±a. In Cuatro edades desde el aÃ±o de 1533 hasta el de 1592. Mexico City: Imprenta Victoria. de Lorenzana, F. A. 1992. Historia de la Nueva EspaÃ±a. Mexico City: Universidad de Castilla-La Mancha and Miguel Angel PorrÃºa. de ObregÃ³n, B. 1988. Historia de los descubrimientos antiguos y modernos de la Nueva EspaÃ±a escrita por el Conquistador en el aÃ±o de 1584, edited by M. Cuevas. Mexico City: Editorial PorrÃºa. de Ortega MontaÃ±ez, J. 1955. InstrucciÃ³n reservada al conde de Moctezuma. Mexico City: Editorial Jus. de Torquemada, J. 1969. Monarquia Indiana, Vol. 5. Mexico City: Editorial PorrÃºa. Farfan, A. 1592. Third book, Of fever and corrupt blood, Treatise of Medicine. Libro tercero. De la Calentura de Sangre corrompida y su Cura. Tractado Breve de Medicina y de Todas las E Â nfermedades que a Cada Paso se Ofrecen. Hecho por el Padre fray Agustin Farfan, Doctor en Medicina y Religioso de la Orden de San AgustÃn en esta Nueva EspaÃ±a. En la casa de Pedro Ocharte, Mexico City. Flores, C. E., and E. Malvido, eds. 1985. Ensayos Sobre la Historia de las Epidemias en MÃ©xico, 2 vols. Mexico City: Instituto Mexicano del Seguro Social. Flores, T. A. 1888. Historia de la medicina en MÃ©xico. Mexico City: Oficina TipogrÃ¡fica de la Oficina de Fomento. Garcia, A. V., C. J. M. PÃ©rez, and V. A. Molina. 2003. Desastres agrÃcolas en MÃ©xico. CatÃ¡logo histÃ³rico. Centro de InvestigaciÃ³n y Estudios Superiores en AntropologÃa Social. Fondo de Cultura EconÃ³mica. Gerhard, P. 1986. GeografÃa histÃ³rica de la Nueva EspaÃ±a 1519-1821. Mexico City: Universidad Nacional AutÃ³noma de MÃ©xico. Gerhard, P. 1996. La frontera norte de la Nueva EspaÃ±a. Mexico City: Universidad Nacional AutÃ³noma de MÃ©xico. Gibson, C. 1964. The Aztecs under Spanish rule. A history of the Indians of the Valley of Mexico. Stanford, CA: Stanford University Press. Hodell, D. A., J. H. Curtis, and M. Brenner. ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ 1995. Possible role of climate in the collapse of classic Maya civilization. Nature 375(6030):391-394.
214 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Hugh, T. 1993. Conquest, Montezuma, CortÃ©s and the fall of old Mexico. New York: Simon and Schuster. Keber, E. Q. 1995. Codex telleriano remensis. Austin: University of Texas Press. LeÃ³n, N. 1982. What were the Matlazahuatl and Cocoliztli in pre-Columbian and in Colonial Period. In Ensayos sobre la historia de las epidemias en MÃ©xico. Edited by C. E. Flores and E. Malvido. Mexico City: Instituto Mexicano del Seguro Social. Lopez de Hinojoso, A. 1578. Tratado VII. De Pestilencia. Summa y RecopilaciÃ³n de Cirugia con un Arte para Sangrar muy Util y Provechoso. Mexico City: Imprenta de Antonio Ricardo. Lopez de Hinojoso, A. 1595. Libro Octavo. De tabardete y Cocoliztle. Summary RecopilaciÃ³n de Cirugia con un Arte para Sangrar Muy Util y Provechoso y Examen de Barberos. Mexico City: Imprenta de Pedro Balli. Marr, J. S., and J. B. Kiracofe. 2000. Was the huey cocoliztli a haemorrhagic fever? Medical History 44(3):341-362. Mendieta, G. 1997. Historia EclesiÃ¡stica Indiana, 2 vols. Mexico City: Consejo Nacional para la Cultura y las Artes. Metcalfe, S. E., and P. Hales. 1994. Holocene diatoms from a Mexican crater. Memoirs of the Cali- fornia Academy of Sciences. Pp. 501-515. Millon, R. 1970. Teotihuacan: completion of map of giant ancient city in the Valley of Mexico. Sci- ence 270(3962):1070-1082. Ocaranza, F. 1933. Las grandes epidemias del siglo XVI en la Nueva EspaÃ±a. Medicina (Mex) 13:176-178. SahagÃºn, B. 1997. Primeros Memoriales: Paleography of Nahuatl text and English translation. Translated by T. D. Sulllivan. Norman: University of Oklahoma Press. Somolinos, A. G. 1956. Hallazgo del manuscrito sobre el cocoliztli (original del Dr. Francisco HernÃ¡ndez). La Prensa mÃ©dica mexicana 21:115-123. Stahle, D. W., E. R. Cook, M. K. Cleaveland, M. D. Therrell, D. M. Meko, H. D. Grissino-Mayer, E. Watson, and B. H. Luckman. 2000. Tree-ring data document 16th century megadrought over North America. Eos, Transactions American Geophysical Union 81(12):121, 125. TezozÃ³moc, F. A. 1975. CrÃ³nica Mexicayotl. Mexico City: Instituto de Investigaciones HistÃ³ricas, Universidad Nacional AutÃ³noma de MÃ©xico. Therrell, M. D., D. W. Stahle, and R. AcuÃ±a-Soto. 2004. Aztec drought and the curse of one rabbit. Bulletin of the American Metereological Association 85:1263-1272. Yu-ping, C. N., and L. Heligman. 1994. Growth of the worldâs megalopolises. In Mega-city growth and the future, edited by R. J. Fuchs, E. Brennan, J. Chamie, L. Fu-Chen, and J. I. Uitto. Tokyo: United Nations University Press. Beldomenico et al. References Alford, R. A., K. S. Bradfield, and S. J. Richards. 2007. Ecology: global warming and amphibian losses. Nature 447(7144):E3-E4. Barros, V., M. E. CastaÃ±eda, and M. Doyle. 2000. Recent precipitation trends in Southern America to the East of the Andes: an indication of a mode of climatic variability. In Southern hemi- sphere paleo and neoclimates: key sites, methods, data and models, edited by P. P. Smolka and W. Volkheimer. ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ Berlin: Springer-Verlag. Beldomenico, P. M., S. Telfer, S. Gebert, L. Lukomski, M. Bennett, and M. Begon. 2008. Poor condi- tion and infection: a vicious circle in natural populations. Proceedings of the Royal Society B: Biological Sciences, epub ahead of print. Beldomenico, P. M., S. Telfer, S. Gebert, L. Lukomski, M. Bennett, and M. Begon. In press. The ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ dynamics of health of wild field vole (Microtus agrestis) populations: a haematological perspec- tive. Journal of Animal Ecology.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 215 Berger, L., R. Speare, P. Daszak, D. E. Green, A. A. Cunningham, C. L. Goggin, R. Slocombe, M. A. Ragan, A. D. Hyatt, K. R. McDonald, H. B. Hines, K. R. Lips, G. Marantelli, and H. Parkes. 1998. Chytridiomycosis causes amphibian mortality associated with population declines in the rain forests of Australia and Central America. Proceedings National Academies of Science 95(15):9031-9036. Bollinger, T. K., J. Mao, D. Schock, R. M. Brigham, V. G. Chinchar. 1999. Pathology, isolation and preliminary molecular characterization of a novel iridovirus from tiger salamanders in Saskatch- ewan. Journal of Wildlife Diseases 35:413-429. Carignan, V., and M.-A. Villard. 2002. Selecting indicator species to monitor ecological integrity: a review. Environmental Monitoring and Assessment 78(1):45-61. Coulson, T., E. A. Catchpole, S. D. Albon, B. J. Morgan, J. M. Pemberton, T. H. Clutton-Brock, M. J. Crawley, and B. T. Grenfell. 2001. Age, sex, density, winter weather, and population crashes in Soay sheep. Science 292(5521):1528-1531. Di Rosa, I., F. Simoncelli, A. Fagotti, and R. Pascolini. 2007. Ecology: the proximate cause of frog declines? Nature 447(7144):E4-E5. GarcÃa, I., R. Albornoz, and E. La Marca. 2007. Perturbaciones climÃ¡ticas y disminuciÃ³n ï¿½ï¿½ï¿½ Atelopus de oxyrhynchus (Amphibia: Anura) en los Andes de Venezuela. Herpetotropicos 2:64-72. Grenfell, B. T., K. Wilson, V. S. Isham, H. E. Boyd, and K. Dietz. 1995. Modelling patterns of parasite aggregation in natural populations: trichostrongylid nematode-ruminant interactions as a case study. Parasitology 111(Suppl):S135-S151. Gulland, F. M., S. D. Albon, J. M. Pemberton, P. R. Moorcroft, and T. H. Clutton-Brock. 1993. Parasite-associated polymorphism in a cyclic ungulate population. Proceedings Biological Sci- ences 254(1339):7-13. 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. Sci- ence 296(5576):2158-2162. Hudson, P. J., A. P. Dobson, and D. Newborn. 1998. Prevention of population cycles by parasite removal. Science 282(5397):2256-2258. IPCC (Intergovernmental Panel on Climate Change). 2001. Climate change 2001: the scientific basis. Contribution of Working Group I to the third assessment report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press. Jones, K. E., N. G. Patel, M. A. Levy, A. Storeygard, D. Balk, J. L. Gittleman, and P. Daszak. 2008. Global trends in emerging infectious diseases. Nature 451(7181):990-993. Juhl, A. 2005. Growth rates and elemental composition of Alexandrium monilatum, a red-tide dino- flagellate. Harmful Algae 4(2):287-295. Karesh, W. B., and R. A. Cook. 2005. The human-animal link. Foreign Affairs 84(4):38-50. Karesh, W. B., and P. Reed. 2005. Ebola and great apes in central Africa: current status and future ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ needs. Bulletin de la SociÃ©tÃ© de Pathologie Exotique 98(3):237-238. Kiesecker, J. M., A. R. Blaustein, and L. K. Belden. 2001. Complex causes of amphibian population declines. Nature 410(6829):681-684. Lochmiller, R. L., and C. Dabbert. 1993. Immunocompetence, environmental stress, and the regula- tion of animal populations. Trends Comparative Biochemical Physiology 1:823-855. Ogden, N. H., M. Bigras-Poulin, C. J. OâCallaghan, I. K. Barker, L. R. Lindsay, A. Maarouf, K. E. Smoyer-Tomic, D. Waltner-Toews, and D. Charron. 2005. A dynamic population model to investigate effects of climate on geographic range and seasonality of the tick Ixodes scapularis. International Journal of Parasitology 35(4):375-389. Ogden, N. H., A. Maarouf, I. K. Barker, M. Bigras-Poulin, L. R. Lindsay, M. G. Morshed, C. J. OâCallaghan, F. Ramay, D. Waltner-Toews, and D. F. Charron. 2006. Climate change and the potential for range expansion of the Lyme disease vector Ixodes scapularis in Canada. International Journal of Parasitology 36(1):63-70.
216 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Parmesan, C. 2006. Ecological and evolutionary responses to recent climate change. Annual Review of Ecology, Evolution, and Systematics 37:637-639. Pounds, J. A., M. R. Bustamante, L. A. Coloma, J. A. Consuegra, M. P. Fogden, P. N. Foster, E. La Marca, K. L. Masters, A. Merino-Viteri, R. Puschendorf, S. R. Ron, G. A. Sanchez-Azofeifa, C. J. Still, and B. E. Young. 2006. Widespread amphibian extinctions from epidemic disease driven by global warming. Nature 439(7073):161-167. Rausch, R. L., J. C. George, and H. K. Brower. 2007. Effect of climatic warming on the Pacific walrus, and potential modification of its helminth fauna. Journal of Parasitology 93(5):1247-1251. Rawlins, S. C., B. Hull, D. D. Chadee, R. Martinez, A. LeMaitre, F. James, and L. Webb. 1990. Syl- vatic yellow fever activity in Trinidad, 1988-1989. Transactions of the Royal Society of Tropical Medicine and Hygiene 84(1):142-143. Reading, C. J. 2007. Linking global warming to amphibian declines through its effects on female body condition and survivorship. Oecologia 151(1):125-131. Roelke-Parker, M. E., L. Munson, C. Parker, R. Cock, S. Cleaveland, M. Carpenter, S. J. OâBrien, A. Pospochil, R. Hofmann-Lehmann, H. Lutz, G. L. M. Mwamengele, M. N. Mgasa, G. A. Machange, B. A. Summers, and M. J. G. Appel. 1996. A canine distemper virus epidemic in Serengeti lions (Panthera leo). Nature 379(6564):441-445. Seimon, T. A., A. Seimon, P. Daszak, S. R. P. Halloy, L. M. Schloegel, C. A. Aguilar, P. Sowell, A. D. Hyatt, B. Konecky, and J. E. Simmons. 2007. Upward range extension of Andean anurans and chytridiomycosis to extreme elevations in response to tropical deglaciation. Global Change Biology 13:288-299. Stuart, S. N., J. S. Chanson, N. A. Cox, B. E. Young, A. S. Rodrigues, D. L. Fischman, and R. W. Waller. 2004. Status and trends of amphibian declines and extinctions worldwide. Sci- ence 306(5702):1783-1786. Tompkins, D. M., and M. Begon. 1999. Parasites can regulate wildlife populations. Parasitology Today 15(8):311-313. Uhart, M., W. Karesh, R. Cook, N. Huin, K. Lawrence, L. Guzman, H. Pacheco, G. Pizarro, R. ÂMattsson, and T. MÃ¶rner. 2004. Paralytic shellfish poisoning in gentoo penguins (Pygoscelis papua) from the Falkland (Malvinas) Islands. Proceedings AAZV/AAWV/WDA Joint Confer- ence, San Diego, California, August 28-September 3, 2004. Pp. 481-486. Weldon, C., L. H. du Preez, A. D. Hyatt, R. Muller, and R. Spears. 2004. Origin of the amphibian chytrid fungus. Emerging Infectious Diseases 10(12):2100-2105. Wharton, D. A. 1982. The survival of desiccation by the free-living stages of Trichostrongylus colu- briformis (Nematoda: Trichostrongylidae). Parasitology 84(Pt 3):455-462. Reisen and Barker References Anyamba, A., K. J. Linthicum, R. Mahoney, C. J. Tucker, and P. W. Kelley. 2002. Mapping potential risk of Rift Valley fever outbreaks in African savannas using vegetative index time series data. Photogrammetric Engineering and Remote Sensing 68:137-145. California Department of Public Health. 2007. California mosquito-borne virus surveillance and response plan, http://westnile.ca.gov/downloads.php?download_id=324&filename=2007%20 CA%20Mosq%20Surv%20and%20Resp%20Plan%20JULY.pdf (accessed February 25, 2008). CDC (Centers for Disease Control and Prevention). 2005. West Nile virus transmission cycle, http:// www.cdc.gov/ncidod/dvbid/westnile/cycle.htm (accessed February 25, 2008). CDC. 2007. Human incidence maps, http://www.cdc.gov/ncidod/dvbid/westnile/surv&control.htm (accessed February 6, 2008). Epstein, P. R. 1998. Global warming and vector-borne disease. Lancet 351(9117):1737. Githeko, A. K., S. W. Lindsay, U. E. Confalonieri, and J. A. Patz. 2000. Climate change and vector-borne diseases: a regional analysis. Bulletin of the World Health Organization 78(9):1136-1147.
APPROACHES TO STUDYING THE CLIMATE-DISEASE CONNECTION 217 Holloway, M. 2000. Outbreak not contained. West Nile virus triggers a reevaluation of public health surveillance. Scientific American 282(4):20-22. Kalnay, E., and M. Cai. 2003. Impact of urbanization and land-use change on climate. Nature 423(6939):528-531. Lanciotti, R. S., O. L. Kosoy, J. J. Laven, A. J. Panella, J. O. Velez, A. J. Lambert, and G. L. Campbell. 2007. Chikungunya virus in U.S. travelers returning from India, 2006. Emerging Infectious Diseases 13(5):764-767. Linthicum, K. J., C. L. Bailey, C. J. Tucker, D. R. Angleberger, T. Cannon, T. M. Logan, P. H. Gibbs, and J. Nickeson. 1991. Towards real-time prediction of Rift Valley fever epidemics in Africa. Preventive Veterinary Medicine 11:325-334. Linthicum, K. J., V. L. Kramer, M. B. Madon, K. Fujioka, and the Surveillance-Control Team. 2003. Introduction and potential establishment of Aedes albopictus in California in 2001. Journal of the American Mosquito Control Association 19(4):301-308. Madon, M. B., M. S. Mulla, M. W. Shaw, S. Kluh, and J. E. Hazelrigg. 2002. Introduction of Aedes albopictus (Skuse) in southern California and potential for its establishment. Journal of Vector Ecology 27(1):149-154. Mulhern, T. D. 1985. The New Jersey mechanical trap for mosquito surveys. 1942. Journal of the American Mosquito Control Association 1(4):411-418. NASA (National Aeronautics and Space Administration). 2007. Terrestrial Observation and Predic- tion System, http://ecocast.arc.nasa.gov/content/view/76/133/ (accessed February 6, 2008). Nemani, R. R., C. D. Keeling, H. Hashimoto, W. M. Jolly, S. C. Piper, C. J. Tucker, R. B. Myneni, and S. W. Running. 2003. Climate-driven increases in global terrestrial net primary production from 1982 to 1999. Science 300(5625):1560-1563. Newhouse, V. F., R. W. Chamberlain, J. G. Johnston, Jr., and W. D. Sudia. 1966. Use of dry ice to increase mosquito catches of the CDC miniature light trap. Mosquito News 26:30-35. Nielsen, C. F., V. Armijos, S. S. Wheeler, T. E. Carpenter, K. Kelley, W. M. Boyce, D. A. Brown, and W. K. Reisen. 2008. Risk factors associated with the 2006 West Nile virus outbreak in Davis, a residential community in Northern California. American Journal of Tropical Medicine and Hygiene 78(1):53-62. NOAA (National Oceanic and Atmospheric Administration). 2008. U.S. climate at a glance, http:// www.ncdc.noaa.gov/oa/climate/research/cag3/cag3.html (accessed February 25, 2008). Patz, J. A., D. Engelberg, and J. Last. 2000. The effects of changing weather on public health. Annual Review of Public Health 21:271-307. Peterson, R. K. D., P. A. Macedo, and R. S. Davis. 2005. A human-health risk assessment for West Nile virus and insecticides used in mosquito management. Environmental Health Perspectives 114(3):366-372. Reeves, W. C., J. L. Hardy, W. K. Reisen, and M. M. Milby. 1994. Potential effect of global warming ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ on mosquito-borne arboviruses. Journal of Medical Entomology 31(3):323-332. Reisen, W. K. 1995. Effect of temperature on Culex tarsalis (Diptera: Culicidae) from the Coachella and San Joaquin Valleys of California. Journal of Medical Entomology 32(5):636-645. Reisen, W. K., R. P. Meyer, S. B. Presser, and J. L. Hardy. 1993. Effect of temperature on the transmis- ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ sion of western equine encephalomyelitis and St. Louis encephalitis viruses by Culex tarsalis (Diptera: Culicidae). Journal of Medical Entomology 30(1):151-160. Reisen, W. K., Y. Fang, and V. M. Martinez. ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ Culicidae) 2005. Avian host and mosquito (Diptera: vector competence determine the efficiency of West Nile and St. Louis encephalitis virus trans- mission. Journal of Medical Entomology 42(3):367-375. Reisen, W. K., C. M. Barker, R. Carney, H. D. Lothrop, S. S. Wheeler, J. L. Wilson, M. B. Madon, R. Takahashi, B. Carroll, S. Garcia, Y. Fang, M. Shafii, N. Kahl, S. Ashtari, V. Kramer, C. Glaser, and C. Jean. 2006a. Role of corvids in epidemiology of West Nile virus in southern California. Journal of Medical Entomology 43(2):356-367.
218 GLOBAL CLIMATE CHANGE AND EXTREME WEATHER EVENTS Reisen, W. K., Y. Fang, and V. M. Martinez. 2006b. Effects of temperature on the transmission ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ï¿½ of West Nile virus by Culex tarsalis (Diptera: Culicidae). Journal of Medical Entomology 43(2):309-317. Reisen, W. K., D. Cayan, M. Tyree, C. M. Barker, B. F. Eldridge, and M. Dettinger. 2008a. Im- pact of climate variation on mosquito abundance in California. Journal of Vector Ecology 33(2):89-98. Reisen, W. K., H. D. Lothrop, S. S. Wheeler, M. Kensington, A. Gutierrez, Y. Fang, S. Garcia, and B. Lothrop. 2008b. Persistent West Nile virus transmission and the displacement St. Louis encephalitis virus in southeastern California, 2003-2006. Journal of Medical Entomology 45(3):494-508. Reiter, P. 1987. A revised version of the CDC gravid mosquito trap. Journal of the American Mosquito Control Association 3(3):325-327. 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 Diseases 8(1):6-13. Shaman, J., J. F. Day, M. Stieglitz, S. Zebiak, and M. Cane. 2004. Seasonal forecast of St. Louis encephalitis virus transmission, Florida. Emerging Infectious Diseases 10(5):802-809. Shi, P. Y., E. B. Kauffman, P. Ren, A. Felton, J. H. Tai, A. P. DuPuis, S. A. Jones, K. A. Ngo, D. C. Nicholas, J. Maffei, G. D. Ebel, K. A. Bernard, and L. D. Kramer. 2001. High-throughput detection of West Nile virus RNA. Journal of Clinical Microbiology 39(4):1264-1271. Su, T., J. P. Webb, R. P. Meyer, and M. S. Mulla. 2003. Spatial and temporal distribution of mosquitoes in underground storm drain systems in Orange County, California. Journal of Vector Ecology 28(1):79-89. Tucker, C. J. 1979. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sensing of Environment 8(2):127-150.