Archival Explorations of Climate Variability and Social Vulnerability in Colonial Mexico

Climatic Change (2007) 83:9Ð38 DOI 10.1007/s10584-006-9125-3

Archival explorations of climate variability and social vulnerability in colonial Mexico

Georgina H. Endﬁeld

Received: 27 September 2005 / Accepted: 3 April 2006 / Published online: 8 March 2007 C Springer Science + Business Media B.V. 2007

Abstract In this paper, unpublished archival documentary sources are used to explore the vulnerability toÐand implications ofÐclimatic variability and extreme weather events in colonial Mexico. Attention focuses on three regions covering a variety of environmental, social, economic, and political contexts and histories and located at key points along a north-south rainfall gradient: Chihuahua in the arid north, Oaxaca in the wetter south and Guanajuato located in the central Mexican highlands. A number of themes are considered. First, the significance of successive, prolonged, or combined climate events as triggers of agrarian crisis. Second, a case study demonstrating the national and regional impacts of a particularly devastating climate induced famine, culminating with the so-called ‘Year of Hunger’ between 1785 and 1786, is presented. The way in which social networks and community engagement were rallied as a means of fortifying social resilience to this and other crises will be highlighted. Third, the impacts of selected historical flood events are explored in order to highlight how the degree of impact of a flood was a function of public expectation, preparedness and also the particular socio-economic and environmental context in which the event took place. An overview of the spatial and temporal variations in vulnerability and resilience to climatic variability and extreme weather events in colonial Mexico is then provided, considering those recorded events that could potentially relate to broader scale, possibly global, climate changes.

1 Introduction

There is little doubt that climate change poses one of the greatest threats to humankind world- wide. The precise impacts of predicted changes for different social, economic and ecological systems are less clear (Schneider 2001). Vulnerability is greater among the environmentally, politically, socially or economically marginal groups, whose ability to adapt to and recover from environmental changes and biophysical events can be limited or constrained (Cutter

G. H. Endﬁeld School of Geography, University of Nottingham, University Park, Nottingham, NG 7 2RD, UK e-mail: georgina.endﬁ[email protected] Springer 10 Climatic Change (2007) 83:9Ð38

1996; Ribot et al. 1996; Liverman 1999; Adger et al. 2001; Fraser et al. 2003). For these reasons, it is anticipated that the impacts of future climate changes are likely to be felt most severely by resource dependent societies in the developing world (Liverman 1999). Much more information, however, is needed on the relative vulnerability of different regions, na- tions and socio-economic groups, to climate changes at a range of scales (Parry 2001: 258). Indeed, the assessment of differential spatial and social vulnerability to predicted climate changes has been identified as a priority research area (Adger and Kelly 1999). Investigating such themes poses a number of challenges (Patt et al. 2005). Vulner- ability is a composite and multidimensional concept that can be discussed in environmental and ecological terms, in relation to political economy or as a reflection of social relations and class structure. As a result, there are many definitions of- and approaches to climate change vulnerability (Luers et al. 2003: 255). As Kelly and Adger (2000: 326) have suggested, however, any analysis of human vulnerability to climate change should consider the social, economic and institutional factors that influence differential social vulnerability among societies and that thus promote or constrain opportunities for adaptation. To date, most investigations of climate change vulnerability and adaptive capacity have been based on current and predicted climate change scenarios. As Magaña et al. (1997) have demonstrated, comparisons between current situations and conditions and those that could prevail with predicted climate changes in the future can provide an indication of the most vulnerable regions. Research has also focused on the development of vulnerability indices to identify those societies most at risk from predicted climate changes (Parry 2001; Sullivan and Meigh 2005; Brooks et al. 2005). Deriving insights about the implications of contemporary and future climate changes from the historical record is problematic (McNeill 2005: 178). The fact that past societies differ markedly from those in the modern world makes simple analogies or parallels unrealistic (Ingram et al. 1981: 5; Meyer et al. 1998). However, knowledge of successes and failures in adaptation to past climatic variability can increase the ability to respond to the threats of long-term climate changes (Adger and Tompkins 2004). Yet there have been relatively few historically oriented assessments of vulnerability and adaptation to climate change (Cutter 1996: 533). As a result, we are only partially aware of the ways in which adaptations to climatic variability may lead to socially differential effects. There is considerable debate as to whether modern society is effectively becoming more or less vulnerable as a result of technological innovation and adaptation over time (see Meyer et al. 2000: 240Ð241). The “lessening hypothesis” suggests that societies, who adapt technologically and socially, are able to effectively reduce the degree of impact of recurrent climatic fluctuations of similar magnitude and can thus indirectly lessen the impacts on the entire society. However, according to the “catastrophe hypothesis”, increasingly elaborate technical and social systems, designed to buffer against the impacts of recurrent and expected climatic fluctuations, can actually increase social vulnerability to catastrophe resulting from less frequent natural and social perturbations (Bowden et al. 1981). Time and place specific social, economic, demographic and environmental conditions are thus pivotal to shaping the vulnerability of a particular society or group of people to climate change and particular weather events, but also play a key role in determining the relative success or failure of adaptive strategies. For this reason, it is only by conducting detailed, regionally focused historical studies of social interaction with climatic variability, that changing vulnerabilities, or what Messerli et al. (2000) have referred to as “trajectories of vulnerability” and adaptation can be understood. This paper represents an archival exploration Springer Climatic Change (2007) 83:9Ð38 11 of vulnerability to climatic variability and extreme weather events in colonial Mexico, and provides one such historical study.1

2 Climate change and vulnerability in Mexico

2.1 Climatic characteristics and variability

Mexico comprises a region of climatic variability, sensitivity and diversity. Latitudinally, the central and southern part of the country lies within the Tropics and the northern half, within the Sub-Tropical zone. As a result, there are clear climatic distinctions between northern and the central/southern Mexico. The Trade Winds deliver summer rainfall to the central and southern regions of the country when the Intertropical Convergence Zone (ITCZ) shifts northwards, and the Sub-Tropical High Pressure belt brings stable dry conditions to the country during the northern hemisphere winter when the ITCZ is displaced equator-wards. The northern part of the country tends to be marginal to the summer trade winds, though the north west of the country is affected by the Westerlies during the boreal winter, which can bring rainfall to the region (Metcalfe 1987; O’Hara and Metcalfe 1995). The latter part of the wet season (JulyÐSeptember) is governed by the Mexican monsoon, but there is a relatively drier period between July and August known as the sequia intraestival or mid-season drought that varies in intensity and (exact) timing from year to year. Although, the distribution and total annual rainfall across the country is largely determined by shifts in the strength and location of these dominant atmospheric circulation systems, a number of other climatic features play an influential role. These include hurricanes and tropical cyclones, which typically affect both Caribbean and Pacific coasts between May and November, with September having the greatest frequency. Mid-latitude cyclones can bring rain, hail, sleet and snow. Severe thunderstorms, which form along or ahead of cold fronts, are also common during spring and summer months and are often accompanied by hailstorms (Mosiño Alem«an and Garc«õa 1974). El Niño Southern Oscillation (ENSO) events are thought to be among the most significant causes of inter-annual climate variability across the country. Evidence of ENSO related rainfall anomalies based on instrumental data, for example, suggests that during warm ENSO (El Niño) events, there is an enhancement of the mid-latitude westerlies, bringing above normal rainfall to northern Mexico in the normally dry season (between November and April). In contrast, a reduced flow of air across the country with the ITCZ to the south can result in drought across central Mexico. Cold (La Niña) events have been linked to lower than average winter rainfall in northern Mexico and higher than average summer rainfall in central Mexico (Ropelewski and Halpert 1986, 1989; Hastenrath 1988; Cavazos and Hastenrath 1990; Magaña 1999; Magaña et al. 2003).

1 Documents were consulted in the following archives: Archivo General de la Naci«on, Mexico City (AGN); Archivo General del Estado de Oaxaca (AGEO); Archivo Historico Municipal de la Ciudad de Oaxaca (AHMCO); Archivo Privado de Col Bustamante Vasconcelos, Papeles de San Bartolo; Archivo Historico Municipal de Leon, Guanajuato (AHML), Archivo Casa de Morelos, Morelia, Michoac«an (ACM); Archivo Historico del Estado de Guanajuato (AHEG); Museo Nacional de Antropolog«õa, Archivo Historico (MNA. AH). In all cases, documents will be referenced in the following manner: the abbreviation of the archival repository, the document group (ramo) consulted, the volume or box (caja) number, the expediente (if applicable), and/or the page (foja) number, (denoted by fa (single pages)/fs (multiple pages). (Page numbers may also be accompanied by f (frente) facing page or v (verso) reverse page). Springer 12 Climatic Change (2007) 83:9Ð38

Topographic features also exert an important influence on climatic characteristics and variation across the country. The Mexican plateau or high plain (Altiplano), which has an average elevation of 1500 m above sea- level, is flanked on the western edge by the Sierra Madre Occidental and on the east by the Sierra Madre Oriental. These cordilleras can trap stable air masses, preventing them from entering the central Mexican plateau, and can also deflect winds. Trade winds from the northeast, over the Gulf of Mexico, for example, can be deflected southwards (through the Tehuantepec Pass), where they can become quite violent northerly blowing winds. Topography also plays a critical role in thermal variation and hence also rainfall distribution. Adiabatic cooling of air as it ascends over the cordilleras leads to the release of moisture in the form of heavy rains over the mountain slopes, this being particularly the case with the Sierra Madre Occidental, which is oriented at right angles to the surface trades that blow over the Gulf of Mexico. The extensive Altiplano also exerts an important thermal effect, heating air masses, leading to convective activity and thunderstorms during the summer rainy season (Mosiño Alem«an and Garc«õa 1974). Frosts are also relatively common at high elevations particularly between October and April. Dry atmospheric conditions in the Altiplano, however, can result in radiative cooling and frosts, even during the summer months (Morales and Magaña 1998). Indeed, frosts in June or July, such as those recently recorded in central Mexico (Eakin 2005: 1926), can threaten rain-fed crop yields.

2.2 Climate change and climate history in Mexico

Future climate change scenarios suggest that there could be significant increases in tem- perature and changes in precipitation, together with a possible increase in the frequency and intensity of extreme weather events such as droughts, floods, hurricanes and unusually high or low temperatures (Liverman and O’Brien 1991; Karl and Easterling 1999; McBean 2004; Tompkins and Adger 2004). Such events can have the greatest and most immediate social and economic impacts and repercussions of all climate changes (Berz 1997; Salinger 2005). These trends could bring opportunities to some and increased vulnerability for others (Tompkins and Adger 2004). In Mexico, where a substantial proportion of the population is dependent upon rain-fed agriculture, these predicted changes could threaten livelihoods and result in disaster (Conde et al. 1997; Liverman 1999). Yet climate change is obviously not a new phenomenon. Mexico has experienced climatic change on both long (103) and short (101−2) time-scales (see, for example, J«auregui 1979; Wallen 1955; Kutzbach and Street Perrott 1985; Metcalfe et al. 1991; Leyden et al. 1994; Hodell et al. 2005). Frequent droughts have affected food security and social and economic well being throughout Mexican history and pre-history (Hodell et al. 1995; Conde et al. 1997; Liverman 1999; Florescano 1980). Although widespread instrumental weather data rarely extend back further than the middle of the 19th century (Metcalfe 1987), historical documents can provide high-resolution information on past climate fluctuations (Brown and Isar 1998; Bradley 1999). Moreover, records charting the impacts of extreme events, coping strategies, technological adaptation, narrative, ideology, regulation and recovery, can provide insight into the societal implications of climate during the last few centuries (Hassan 2000). This kind of information could provide a guide to where the most critical sensitivities to future climate changes might lie (McCarthy et al. 2001; Meyer et al. 1998). Recent research has demonstrated the potential of the rich colonial archives of Mexico for reconstructing climatic chronologies and for investigating the impacts of and responses to extreme weather events in different regions of the country over the last six centuries (Metcalfe 1987; O’Hara 1993; O’Hara and Metcalfe 1995; Endfield and O’Hara 1997; End- Springer Climatic Change (2007) 83:9Ð38 13

ﬁeld et al. 2004a,b; Garc«õa-Acosta 1993, 1997; Mendoza et al. 2005). This work has provided a methodological template for the present investigation. Attention focuses on three regions covering a variety of environmental, social, economic and political contexts and histories and located at key points along a north-south rainfall gradient. The Conchos Valley of southern Chihuahua in the arid north (average annual rainfall 350Ð400 mm per annum), the Valley of Oaxaca in the wetter south (1500 mm per annum) and west central Guanajuato located in the central highlands of Mexico, a region of climatic transition (650Ð1000 mm per annum) (Figs. 1Ð3). Each region developed different settlement and land use characteristics during the colonial period. Guanajuato had long history of pre- Hispanic settlement, but by the time Spanish arrived, the area represented something of a frontier zone. The combination of fertile soils and mineral wealth, however, meant that by the second half of the sixteenth century, the area had developed into the country’s main agricultural centre, specialising in the production of wheat, fruits and vegetables. In contrast, Oaxaca retained an indigenous character throughout the colonial period and although livestock and Mediterranean cultivars were introduced into the area, people there continued to focus on the production of maize, beans and chile, employing traditional cultivation methods. Signiﬁ- cantly, there was also a good deal of land retention among indigenous populations in Oaxaca (Taylor 1972). In Chihuahua, the threat of droughts, and hostility from the largely nomadic indigenous groups who occupied the northern part of Mexico, delayed European settlement and colonisation of the region until the seventeenth century. Such attacks continued to cause problems on a number of occasions throughout the eighteenth and early nineteenth century. Despite these obstacles, a lucrative mining and livestock economy developed in the region. Comparative archival analysis of the implications of extreme weather events in these case study areas can provide a unique insight into the way in which vulnerabilities to climate variability may have changed through time and across different contextual spaces.2

3 Social vulnerability to climate change: The preconditions of agrarian crisis

The scale of impact resulting from unusual weather events is very much dependent upon the context in which it takes place. While society can generally cope with or respond effectively to individual years of climate related crisis, successive extreme events, or individual events acting in combination with other natural hazards, or against a backdrop of difﬁcult socio- economic or political circumstances can have ampliﬁed and cumulative impacts, rendering a society much more vulnerable to crisis. A number of recorded climatically driven agricultural crises and famines in Mexican history and prehistory have been associated with successive droughts or drought combined with other unusual or extreme weather events, often frosts (Florescano 1980). Prolonged drought coupled with ‘killing’ frosts in central Mexico between 1452 and 1455, for example, is thought to have contributed to widespread harvest loss, culminating with the famine of ‘1 Rabbit’. This stimulated out-migration, disease and death (Hassig 1981; Therrell et al. 2004). Though undoubtedly one of the most widely reported social calamities in Aztec history, this event was not unique. Indeed, other phases of prolonged drought and associated starvation are thought to have affected the region between 1332 and 1335, and again between 1502 and 1505 (Therrell et al. 2004).

2 Other locations will be referred to throughout the text for comparative purposes. Springer 14 Climatic Change (2007) 83:9Ð38

Fig. 1 Central Guanajuato and locations referred to in the text

The vulnerability of society to such periods of climatic abnormality is very much determined by prevailing demographic, socio-economic conditions, community behaviour and practices (Liverman 1990). There are often differential impacts and losses, depending on the levels of preparedness of different sectors of population, which itself is determined by their socio-economic and political circumstances. The features of the colonial political economy, for example, created a society that was differentially vulnerable to the impacts of climate events (Liverman 1990, 1999). Indigenous depopulation in the 16th century, in the wake of European introduced diseases, prolonged drought (Acu˜na-Soto et al. 2002) and the cruel- ties of the early colonial regime, coupled with the implementation of land granting policies which facilitated the acquisition of usurped or abandoned community territory, allowed the progressive monopolisation of lands by individual landowners, the church and elites and the coincident marginalisation of the indigenous depopulation. This imbalance became particularly problematic in the second half of the seventeenth and throughout the eighteenth century Springer Climatic Change (2007) 83:9Ð38 15

Fig. 2 The Valley of Oaxaca (incorporating the Etla, Tlacolula and Zimatal«an valleys) and locations referred to in the text when indigenous populations recovered numerically and found themselves deprived of land and access to natural resources, including water. Of course, land and water rights were hotly contested by usurped communities throughout the colonial period (Lipsett-Rivera 1992, 1999; Endﬁeld and O’Hara 1997). Yet, possessing little in the way of social storage and reserve seed stocks, and having only limited access to fertile and irrigated lands, it was often the indigenous communitiesÐwho were among the poorest sectors of society Ð that were ﬁrst to suffer during agrarian crises. It was the same groups of people, those least resilient, who disproportionately succumbed to illness and disease during or following drought induced food scarcity and famine. Wealthy landowners, Springer 16 Climatic Change (2007) 83:9Ð38

Fig. 3 The lower Conchos Valley, Chihuahua and locations referred to in the text in contrast, possessing more in the way of cultivable territory, water resources and irrigation, reserve foodstuffs, seed stock and in some cases livestock as a form of ‘social storage’, could buffer themselves against the impacts of drought and would often reap the rewards of climate-induced harvest crises. Indeed, they often capitalised on the desperate and dependent ‘underclass’ by inflating crop and grain prices (Liverman 1999). To counter these problems, grain stores (Reales Alhondigas« ), were established in some towns and cities, and particularly those in the mining districts, which represented the engines of the colonial Mexican economy. Such stores theoretically ensured the availability of some Springer Climatic Change (2007) 83:9Ð38 17 supplies of food and grains for the common good during times of scarcity. Furthermore, where a Real Alhondiga« existed, all grain producers and merchants were obliged to sell their produce there first in order to ensure that a fair price was then offered to the public (Florescano 1976, 1980). While these central stores appear to have generally served communities well during normal years, there is evidence to suggest that they regularly encountered supply problems, specifically during drought years, and were apparently only partially successful in regulating price fluctuations.3 Speculation by individual landowners and merchants would prove problematic throughout the colonial period. Because of the degree of commercial opportunism, any reconstruction of drought history based on an index of grain price records alone could be misleading.

3.1 “The pernicious calamities that occasion hunger”: Drought, famine and disease in colonial Mexico

Investigations of the colonial archives for each of the case study regions have revealed many periods of local as well as more widespread drought-related agrarian crisis throughout the colonial period (Endﬁeld et al. 2004a,b). By comparing multiple archival references,4 it is possible to identify that drought contributed to harvest problems in all three regions of the country on many occasions (see Fig. 4). Drought-related harvest crises are recorded in Guanajuato, and the Baj«õo more generally, in 1590/1, 1601, 1641, 1647, 1651, 1662, 1696, 1751, 1754, 1755, 1768, 1771, 1784Ð1786, 1793, 1803, 1809 and 1816.5 Although relatively few droughts appear to have affected society in Oaxaca, late or non-existent rains and water scarcities are documented in 1550, 1690, 1726, while harvest crisis appears to have descended on the region in 1733, 1746Ð1747 and 1785. 6 Drought was a more frequent phenomenon in the arid north of the country. Records from Chihuahua, the majority of which date from the eighteenth and early nineteenth centuries, indicate that droughts resulted in harvest failures and food scarcities in the region in 1724Ð1727, 1739Ð1741, 1748Ð1752, 1755, 1758, 1760Ð1765, 1770Ð1773, 1785Ð1786, 1804Ð1806, 1809 and 1812Ð1814.7 With the exception of successive droughts in the 1770s and 1780s, which have been recognised elsewhere in the country, archival evidence of prolonged drought (Endﬁeld and Fern«andez-Tejedo, in press)

3 See, for example, AHMCH Justicia, caja 126, exp. 10; AHMCH Demandas de Inconformidad, caja 112, exp. 18; AGN Ayuntamientos, vol. 196, exp. 1. 4 Throughout this study, attempts were made to cross-reference recorded instances of weather events. Duplicate references to the same event from different archival sources was taken as validation of an event. 5 See, for example, AGN Tierras, vol. 187, exp. 2 327 ff.; AGN Tierras, vol. 1110, exp. 18; AGN Indios, vol. 5, exp. 154; AGN Indios, vol. 5, exp. 598; AGN Correspondencia de Virreyes, vol. 12 (series 2), fs. 244; AGN Tierras, vol. 674, exp 1 30ff; Follow the dispute in vol. 675, exp 1; AGN Tierras vol. 192, exp. 1; AGN Tierras, vol. 586, exp 8; AGN Tierras, vol. 1872, exp. 15 10f; AGN Tierras, vol. 671, exp. 3, 18 ff; AGN Tierras, vol. 2901, exp. 36; AGN Tierras, vol. 2959, exp 141; AGN Tierras, vol. 988, exp. 1,2y3,516ff;AGNTierras, vol. 1110, exp. 18; AGN Tierras, vol. 1166 exp. 1 fs. 450; AGN Tierras, vol. 2963, exp.116, f 246Ð308; AGN Tierras, vol. 1352, exp 1; AGN Tierras, vol. 1368. See also Endfield et al. (2004b). 6 AHMCO Actas de Sesiones del Cabildo 1728Ð1733; AHMCO Actas de Sesiones del Cabildo 1746Ð1748; AGEO Real Intendencia, leg 1 exp 46 4 ff; AGN Industria y Comercio, vol, 20, exp. 6, fa 237; AGN Hospital de Jesus, leg 118, exp. 13; See also Endfield et al. (2004a). 7 AHMCH Hacienda, Caja 7, exp. 4; AHMCH Jesuitas, leg II-9, exp. 12; AHMCH Gobierno, caja 20, exp. 5; AGN Mercedes, vol. 76, fs. 95Ð100; AHMCH Guerra, caja 1, exp. 8; caja 1, exp. 7; AHMCH Gobierno, caja 30, exp. 23; AHMCH Notarias, Abasto de Carne, caja 44, exp. 20; AGN Ayuntamientos 173, cuaderno 3; AGN Presidios y Carceles, vol. 4 various fojas; AHMCH caja 40, exp. 5; AGN Historico de Hacienda, vol. 296, exp. 1; AHMCH Justicia, Abasto de Carne, caja 126, exp. 10; AHMCH Hacienda, caja 55, exp. 12; AHMCH Gobierno, caja 49, exp. 13. See also Endfield and Fernandez-Tejedo (in press). Springer 18 Climatic Change (2007) 83:9Ð38

Fig. 4 Recorded droughts and harvest crisis in Oaxaca, Guanajuato and Chihuahua, 1720Ð1820 may support tree ring evidence of two particularly severe phases of prolonged drought in northern Mexico in the middle eighteenth century (1750Ð1765) (Diaz et al. 2002) and between 1801 and 1813 (NOAA International Tree Ring Data Bank, cited in Liverman 1999: 101). The most devastating periods of drought in the three case study areas, were those that compounded the impacts of other weather events, agrarian problems or prevalent social, political or economic difficulties, specifically population pressure and resource stress. The combined impacts of drought and frost, for example, contributed to a loss of the wheat harvests across the Baj«õo in 1641. It is not clear eactly how severe the frost event was per se. This may have been a relatively ‘normal’, seasonal frost, the impacts of which were amplified in the context of prevailing drought. Either way, a document from Guanajuato dated October 15th suggests that “the need and hunger which the local people are ...suffering is very great”.8 In fact, much of central Mexico was affected and the scale of food scarcity was such that appeals for help were made to religious images in Mexico City at this time (Gibson 1964). The droughts that are reported to have affected central Mexico in 1750 and 1751 were similarly damaging. Dramatic loss of harvests and the death of livestock left many inidviduals without any means of survival and with no alternative but to sell off their possessions, property and land. Indigenous residents of Leon (Guanajuato), Juan Santos Ramirez, Martin Fe la Cruz and Luisa and Jos«e Ramirez were among those with no option but to sell off land and farm buildings “due to the terrible dryness of the the (present) year”.9 Even wealthy sectors of society in the region were affected. At the height of the crisis, for example, Christobal Manuel, described as a “rich latino resident of the juristiction of Leon”, recorded his need to sell off part of his property “because of the sterility of the year” which had left him “impeded because of the lack of oxen with which to sow”. His family was apparently “suffering terrible neediness because of the lack of maize which because of the scarcity has such an inflated price”.10 Droughts are also documented elsewhere in the country around the middle of the eighteenth century. Repeated drought induced harvest failure contributed to food scarcity in

8 AHML Communicaciones (benefcio publico) Exp 1, 1641. 9 AHML Notarias 1750Ð1751 fs. 144Ð151. 10 AHML Notarias 1750Ð1757 f.24 recto y verso. Springer Climatic Change (2007) 83:9Ð38 19

Oaxaca between 1746 and 1748.11 Problems started after the limited harvests of 1746. Grain stores were depleted, prices rose and the local administration in Antequera (Oaxaca City) was requested to provide food or financial aid for the poorer sectors of society who faced starvation.12 Drought and food scarcity also affected people in Chihuahua at this time (Endfield and Fern«andez-Tejedo, in press). Indeed, few people had been able to escape “the pernicious calamities” caused by grain scarcity in Durango and Chihuahua in Northern Mexico in 1748.13 Drought was also recorded in the region in 175014 contributing to a lack of grains “especially maize”,15 while shortages of grazing pasture resulted in the death of livestock.16 By 1752 there were “cattle dying in the field...following the rigorous drought,” but problems, both in this region and elsewhere, actually persisted throughout the 1750s.17 By 1755, people in Chihuahua were praying for divine intervention to arrest the “calamity of drought experienced three [years] in a row”,18 while in Leon, Guanajuato, “the late start of the rains and early frosts” between 1754 and 1755 had once again served to reduce the quantity of maize harvested.19 Severe and continuous drought re-appeared in northern and central Mexico in the early 1770s.20 Poor harvests, shortages of flour and a consequent scarcity of bread are recorded in Chihuahua,21 while the want of rainfall coupled with reported earthquake activity in central Mexico in 1771, led to instructions being issued to the whole of the Archbishopric to undertake organised rogation ceremonies.22 Such periods of crisis inevitably affected the health of human populations. Links between famine and disease are far from new (Malvido 1973) and it is well known that some infectious diseases can be aggravated by malnutrition as well as population mobility (Kovats et al. 2003). In this way, prolonged drought may have indirectly stimulated disease outbreaks. Figure 5 summarises possible relationships between reported drought- related harvest problems and archival references to disease in epidemic proportions. While it should be noted that not all outbreaks of epidemic disease necessarily relate to periods of harvest crisis, typhus is recorded in San Felipe del Real (Chihuahua) in 1727,23 following maize shortages and scarcities of wheat flour between 1724 and 1727.24 Similarly, it may be no coincidence that a “general pestilence”, thought to be smallpox, coincided with the prolonged droughts that affected the region in the 1750s and resulted in a third of the population of Chihuahua (City) being buried.25 Epidemics were also recorded in other parts of the country at this time (Cavo 1949),

11 AHMCO Actas de Sessiones del Cabildo 1746Ð1748; fa. 54. 12 AHMCO Actas de Sessiones del Cabildo 1746Ð1748, fa. 54. 13 AHMCH Gobierno, caja 20, exp. 5. 14 AHMCH Guerra caja 1, exp. 8; caja 1, exp. 7. 15 AGN Mercedes, vol. 76, ff. 137r y v. 16 AHMCH Guerra caja 1, exp. 7. 17 Archivo Arzobispado de Chihuahua, Ramo Gobierno y Administraci«on; Cofrad«õas, 1755, caja 3, serie 1.3.3. 18 Ibid. 19 AHML Bandos, caja. 5, exp. 29. 20 AGN Presidios y Carceles, vol. 4, ff. 160Ð163. 21 AHMCH Hacienda, caja 44, exp. 1. 22 AGN Bienes Nacionales, vol. 1182, exp. 13. 23 AHMCH caja 3, exp. 5. 24 AHMCH caja 2, exp. 12. 25 AHMCH Guerra,caja 2, exp. 4. Springer 20 Climatic Change (2007) 83:9Ð38

Fig. 5 Drought, harvest crisis and recorded instances of epidemic disease in Oaxaca, Guanajuato and Chi- huahua, 1720Ð1820 some of which were attributed to the desperate consumption of poor quality foods by the most poverty stricken, starving members of society.26 As Ouweneel (1996: 91) has suggested, however, it is not always clear whether the shortage of food was the “cause or effect” of disease. Both abnormal weather and epidemic disease could lead to harvest shortfalls, in the latter case by effectively reducing the availability of agricultural labour, and it is difficult to identify from archival sources whether harvest problems were a function of weather or disease. Records indicate, for example, that there was “very grave necessity, hunger” experienced in Chihuahua in 1763. Some of the poorest sectors of society offered to work on the haciendas and in silver mines at this time in return for food and shelter.27 Only a year later, however, “the death of diverse people of all classes, especially the poor and devalued people who lack food” was reported. 28 These deaths may have been associated with the well-documented “common and great epidemic”,29 consisting of both typhus and smallpox which swept across Mexico between 1763Ð1764 (Cooper 1965). However, documents indicate that harvests in Chihuahua had also been affected by a “rigorous drought” between 1764 and 176630 and hailstorms in 1763 and 1764.31 In this case, it seems that unusual weather events might have exacerbated the difficulties faced by a society already rendered vulnerable by epidemic disease. Another well-documented period of widespread epidemic disease took place between 1812 and 1814 (Cooper 1965). The first signs of illness in the north of Mexico appear in a document from 1812 when it was suggested that “an epidemic

26 AHMCH Notarias, Abastos de Carne, caja 42, exp. 12. 27 AHMCH Justicia, Demandas de Inconformidad, caja 108, exp. 14. 28 AGN Ayuntamientos, vol. 173, cuaderno 3. 29 AGN Epidemias, exp. 1. 30 AGN Jesuitas, leg II-9, exp. 30. 31 AGN Jesuitas leg II-9, exp. 33. Springer Climatic Change (2007) 83:9Ð38 21 had desolated the towns outside” Chihuahua.32 Another document from 1814 indicates that by that time “a contagious fever...had been introduced to the town [of Chihuahua]” and that “in the 15 days from the 23rd August to the 7th September, 165 people fell ill and 21 died”.33 On 27th July, 1813, however, the local authority in Chihuahua City had recorded “a suspension of the rains” and suggested that a special plea should be made to Nuestra Señora de Guadelupe for help with the “delayed rains and the frosts” which were contributing to harvest failures.34 It seems that the spread of disease might, therefore, have been exacerbated by the drought of 1813.35 Structural and behavioural factors might have contributed to the spread of disease during periods of food scarcity. German polymath scholar, Alexander Baron von Humboldt, reflected on the level of desperation after a visit to Mexico between 1803 and 1804, suggesting that a lack of long-term planning for food crises was one of the key reasons for the prevalence of disease among Mexican communities. “In a country where there are no store houses and the naturales [indigenous people] do not live except for the day” he suggests “the town suffers immensely so the naturales are kept alive with fruits and roots and cacti and this bad food produces illnesses and in general the scarcity observed is accompanied by mortality among children” (Humboldt 1811). Certainly, there is documentary evidence from some parts of the country to indicate that those most poverty-stricken members of society did resort to scavenging to survive the most devastating crises. Portillo (1910), for instance, records a series of poor harvests in 1801 and 1802 brought on in part by a plague of locusts which invaded “the Spanish colonies in south and then central America, affecting the Mexican realm in 1802”. Notices were issued offering “ten pesos for each arroba (11.3 kg/25 pounds) of locusts that were collected and five pesos for the collection of the lar- vae which resulted from their procreation” in an attempt to alleviate the problem. By 1804, there was a great scarcity of grain (particularly maize) recorded in some regions, particularly in Oaxaca, where people were forced to “roam through the fields looking for edible roots and weeds so that they would not die of hunger” (Portillo 1910). Interestingly, there are records of repeated flood events in Oaxaca at this time.36 In Iztlahuaca, “excessive increases” in rainfall and flooding were recorded in 1802.37 Various different forms of anomalous weather and epidemic disasters are also recorded across the country between 1801 and 1804. Flood- ing, followed by “extraordinary drought” was recorded in the north of Mexico in 1804,38 stimulating epidemic disease, while in central Mexico, the River Lerma was recorded as being “higher than the roads and in the houses” at this time.39 This may, therefore, been a period of more widespread climatic anomaly. Whether such conditions may have simultane- ously permitted the “prodigious reproduction” of the locusts (Portillo 1910) remains open to conjecture.

32 AHMCH Gobierno, Actas de Cabildo, caja 2, exp.1. 33 AHMCH Gobierno, caja 49, exp. 13. 34 AHMCH Gobierno, caja 48, exp. 31. 35 AHMCH Gobierno, Actas de Cabildo, caja 2, exp.1. 36 Archivo Privado de Don Luis de Casta˜neda. Maps, Oaxaca. 37 AGN Rios y acequias vol 3. 38 AHMCH Notarais/Gobierno (various). 39 AGN Rios y Acequias, vol. 1: exp. 9 214ff. Springer 22 Climatic Change (2007) 83:9Ð38

3.2 “A time of calamity”: Drought, frosts and widespread famine in late colonial Mexico

Undoubtedly the most devastating and widespread period of weather induced crisis in colonial Mexico took place in the middle of the 1780s. Following a sequence of climatic events thought to be remarkably similar to those leading up to the famine of 1 Rabbit between 1452 and 1455 (Therrell et al. 2004), a combination of droughts in 1780, 1782, 1784 and 1785 and frosts in 1784 and 1785 contributed to the so-called ‘year of hunger’ between 1785 and 1786 (see Fig. 4). This period of crisis affected the majority of the country but had particularly devastating consequences for society in the agrarian heartland of central Mexico. The ﬁrst signs of a problem in this region may have actually been recorded in documents from 1782 when a local councillor from Leon reported that “the population is suffering from ...scarcity and a need for seeds of maize and other crops”. By early 1784, the manager of the grain store recorded “a lack of maize” in Leon “because of the scarcity of the harvest” of that year.40 Scarcity in turn forced prices up and there was concern that this would prevent the poorest sectors of society from accessing the most basic foodstuffs. It seems that the Spanish crown might have recognised the possibility of an imminent crisis around this time. On the 10th May, 1784, a royal order had been issued in Spain requesting that “all the heads of the Indies send each six months notice of the weather experienced in these dominions: noting whether the rains had been scarce or abundant and noting also the nature of the harvests of fruits and other produce.”41 It is not clear whether this order came in response to a fear of more widespread harvest problems or a general growing awareness of the potential for climate induced agrarian problems in the Spanish American colonies as a whole. However, the order preceded one of the most widespread and devastating famines to affect Mexico. By 1785, problems of scarcity were becoming more widespread. A circular sent by the then Viceroy, Conde de Galvez of Leon in 1785, for example, suggests that food shortages were “beginning to be experienced in most of the kingdom because of the lack of rains and in anticipation of the winter”.42 Hunger was rife across the north of the country. Rogation ceremonies were held in Chihuahua and appeals were made once again for divine intervention to alleviate the suffering.43 Although irrigation may not have been successful in hedging against the most severe periods of drought (Yates 1981), the expansion of land under irrigation was a key coping strategy. In 1785, communities across central Mexico were encouraged to grow more crops under irrigation, and in some cases this necessitated ﬁnancial assistance. A letter to the Gazeta de Mexico« , from the Bishop of Michoac«an, dated November 19th 1785, for example, suggested that, “to facilitate more and more sowing of irrigated maize it has been determined that one thousand pesos be distributed in this parish among the poor, well off Spaniards, indios or mulattos who wish to sow”.44 People from all cross-sections of society were being encouraged to grow irrigated maize in stretches of territory across central Mexico that were not normally considered suitable for this particular crop, including the so called hot lowlands (tierra caliente) and cold uplands (tierra fria) in the spring of 1786.45 In part this could be

40 AHML Alh«ondiga exp. 8. 41 AGN Bandos vol. 13, exp. 47, fs. 160. 42 AHML Alh«ondiga exp 10. 43 AGN Ayuntamientos, 173, cuadernos 3. 44 Gazeta de Mexico, 6th December, 1785, Number 52, pp 447Ð449. 45 AGN Alh«ondigas vol. 15. Springer Climatic Change (2007) 83:9Ð38 23 because as one document suggests, there was “more water for irrigation of maize” in many of these districts.46 Community action was central to agrarian crisis management. The local authority in Chihuahua issued a circular on the 11th October 1785 suggesting that “being a time of calamity”, the indio labourers and other workers should be persuaded not to abandon their pueblos and haciendas, but to “provide their personal services to the cultivation of the fields”.47 Elsewhere there were requests for community efforts to alleviate the suffering. A notice from the Gazeta de Mexico« , dated 6th December, 1785 proposed a programme to introduce water to the town of Celaya, Guanajuato, from a source estimated to be two leagues away from the town, with the suggestion that “work begin immediately for the benefits (of the people)...providing occupation for those people who find themselves without resources”.48 Labour was drafted from the stricken population, a strategy that was apparently justified in that it would provide them with paid employment. Despite such efforts, some people had little choice but to sell part or all of their property, usually far below the true sale value, as desperate vendors were keen to point out. Documents from Leon chart the sale of several houses “for more or less half of the just value and price”,49 and highlight the “the extreme need for help because of the hunger”.50 Reports indicate that some benefactors in Chihuahua donated money to aid the poor in 1785.51 Local councils of Guanajuato also began to make charitable donations the following month,52 and circulated the names of wealthy individuals in Leon who could provide food or financial aid.53 Although this crisis was particularly widespread, not all regions of the country were as badly affected. Though far from sufficient to avert the crisis that unfolded, there was trade in foodstuffs from region to region. Maize was regularly purchased from surrounding areas by the local authority of Chihuahua,54 but the scale of crisis in the mid 1780s meant that additional quantities of grains, seed and flour were also transported from much longer distances to supply the Real Alhondiga« in the Villa de Chihuahua55 (Martin 1996: 26). Interestingly, this period of crisis does not appear to have affected society in Oaxaca as dramatically. Indigenous land retention and less reliance on wealthy land owners for food provision in this region gave a large proportion of the population more independence and more subsistence capability compared to other regions. Moreover, in some parts of Oaxaca, the grain harvests in 1786 were particularly good. A letter dated 9th February 1786 from a Señor Narciso Muñoz reports on the “very abundant harvest of maize” reaped in Teutitl«an, Oaxaca. It was felt that “this great abundance might partially help alleviate the scarcity experienced in the surrounding jurisdictions which had not experienced equal luck” (cited in Florescano 1981, Vol. II: 570).

46 Patzcuaro archives, caja 54d. folder 1. fojas 1Ð150. 25th October. 1785. 47 AGN Indios, vol. 91. 48 Gazeta de Mexico, 6th December, 1785, Number 52, pp 449Ð50. 49 AHML Notorias, libro 1785, fs. 53Ð56. 50 AHML Notorias, libro 1786 fs. 66, 88, 98, 101, 124, 129; AHML Notorias, 1785, fs. 53Ð56. (ref is on fs 55). AHML Notorias, 1785, fs. 53Ð56. (ref is on fs 55). There are a number of other references in this document to the selling off of property at “half of the just price”. (See, for example, fs. 90Ð91; fs 162Ð3; fs.171Ð4); AHML Haciendas y Ranchos exp 17; Notarias 1786. 51 AGN Ayuntamientos, vol. 173, cuaderno 1. 52 AGN Alh«ondigas, vol. 10 exp. 5 fs. 250Ð53. 53 AHML Alh«ondiga, exp. 8. 54 AHMCH Justicia, Abasto de Carne, caja 126, exp. 10; AGN Ayuntamientos 173, cuaderno 6. 55 AHMCH Justicia, caja 126, exp.10. Springer 24 Climatic Change (2007) 83:9Ð38

It is thus clear that there were efforts to reduce social vulnerability to the impacts of this period of climatic and agrarian disaster. The extension and consolidation of social networks through local, regional and national trade links, through charity and community engagement in projects intended to benefit entire communities, might well have contributed to improvements in social resilience to later climate events, even if some of these projects were reliant on forced- rather than voluntary labour. Despite these efforts, several years of hardship would follow. There are reports of disease epidemics in various parts of the country during and following the crisis,56 and three years after the first signs of drought, 300 000 people had died and others had been in- capacitated and were thus in no fit state to cultivate lands when the rains did arrive. Tribute demands became impossible to satisfy in the wake of crop shortages,57 while the droughts had served to diminish the amount of seed stock available with which to sow (Florescano 1981). In Chihuahua, the “continuous rains” in August of 1786 actually damaged the seedlings such that hopes of a more successful harvest that year were dashed.58 The magnitude of the crisis and the scale of the losses incurred were a function of both the combination of extreme weather events and the context in which they took place. Successive droughts and frosts compounded already prevalent socio-economic difficulties and emerging popular dissent over resource distribution and access. By the second half of the eighteenth century, population expansion and a growing resource monopolisation by an emerging elite across many parts of the country had effectively created a more marginal Ð and as Tutino (1986) has pointed out for central MexicoÐan increasingly dependent underclass by the second half of the eighteenth century. This sector of the population was far more vulnerable to the impacts of climate changes and faced subsistence crisis when successive and combined weather events served to drastically reduce harvests several years running. This inequality and differential vulnerability is thought to have helped fuel agrarian unrest in the agrarian heartland of the Baj«õo, particularly in the last quarter of the eighteenth century (Tutino 1986). Set against a backdrop of mounting social and political unrest and dissent that would culminate with the independence movement, “scarcity of rains” in 1809 triggered another similarly devastating, widespread agrarian crisis. A “lack of maize, beans and other seeds of basic necessity” had resulted, causing “ want, hunger and calamity” across the country. According to national governmental documents, there was particular concern that “as was common during years of calamity, the poorest sectors of society, and especially the indios” would “abandon their houses, leave the places and towns of resi- dence, causing problems”.59 Such fears over food scarcity, or more particularly, indigenous population mobility and the abandonment of private and rented properties as well as those estates upon which many indigenous workers were employed, possibly reflected a growing awareness at the national governmental level of escalating popular unrest at the time.

56 AGN Historia vol. 72. 57 AGN Tributos vol. 20, exp.15, 2, 5. 58 AGN Historico de Hacienda, vol. 296, exp. 1; AHMCH Justicia, Abasto de Carne, caja 126, exp. 10; AHMCH Guerra, caja 5, exp. 5. 59 AGN Bandos, vol. 25, exp. 45, fs. 62. Springer Climatic Change (2007) 83:9Ð38 25

4 “Most sensitive and saddening events”60: Climate-society interactions and ﬂood risk

4.1 Flood expectation and vulnerability

Most floodplain ecosystems are geared to periodic inundation (Smith and Ward 1998: 19) and social and economic systems have generally evolved to accommodate general, often seasonal, flood pulses. The fertile alluvial soils of the floodplains of the Atoyac River, Oaxaca, for example, were prized for floodwater farming and the controlled use of water for irrigation from expected flash flood events, particularly at the start of the rainy season, was commonly practiced (Kirkby 1973: 32). There was also extensive seasonal use of swamplands adjacent to rivers for grazing purposes in many parts of the country.61 Unexpected flooding, however, had the potential to seriously disrupt agrarian livelihoods, contributing to infrastructural damage in urban settlements and, on some occasions, leading to substantial loss of life. Those events that from documentary records appear to have been the most devastating in terms of economic and human losses, were those that hit communities who were at their most vulnerable and least prepared, or who had already been beleaguered by other environmental or indeed social and economic problems. Agricultural floodplain communities were more or less vulnerable to flooding depending on the degree to which flood risk was anticipated and hence had been inscribed into agrarian practice. Riverside communities in Tehuantepec, Oaxaca for example, were accustomed to storing foodstuffs throughout the wet season, should a flood occur as the water levels rose at the end of the season.62 However, on May 29th, 1721, “a storm of great winds and rains and lasting ten hours from the night of Friday 29th of May [1721] till five o’clock in the afternoon on the Saturday [30th]” was documented.63 The event took place at the beginning of the rainy season that year when “the land was dry and needy”.64 This was a time when the crops had only just been sown, when food stores were depleted after the long dry season and when a flood event was least expected. The impacts were consequently devastating: “wild animals were left in fields, drowned ...the cows and pigs and mules were transported across distances of 22 leagues... dragged along by the flood waters”.65 Recovery was slow, not least because livestock were in short supply but also because land and seedlings had been shrouded in fluvial deposits (Endfield et al. 2004a). A similarly dramatic flood event in Celaya (Guanajuato) on June 28th 1692 hit an equally vulnerable community (see Fig. 6). Occurring and ten o’clock at night, after an afternoon of incessant and torrential thunderstorms, the flood caused a number of fatalities, left an estimated 3000 homeless and many more injured (Endfield et al. 2004b). This event took place a time of “great scarcity of provisions in the majority of the country, the cause being the loss of almost all the harvests the year before.”66 Drought, famine, disease in epidemic

Quote taken from “Descripcion de la Ciudad y Real de Minas de Guanajuato por Jos«e Hernandez Chico, 1788, Archivo de la Marina, Museo Naval de Madrid, Ms 563 in Florescano, E and Gil, I (1976) (eds) Descripciones economicas regionales de la Nueva Espana 1766Ð1827, INAH, Mexico. 61 AGN Hospital de Jesus leg 404 exp 2; Hospital de Jesus 432 exp 5; AGN Mercedes vol. 8, ff 2, 29,40v, 43, 55, 87, 100v, 111, 113, 118, 190. 62 AGN Tierras, vol. 182, exp. 1, fa. 380. 63 AGN Tierras, vol. 182, exp. 1, fa. 380. 64 AGN.R«õos y Acequias 1 exp. 3. 65 AGN Tierras, vol. 182, exp. 1, fa. 380. 66 Lucio Marmolejo (1967 reprint) Efemerides« Guanajuatenses o datos para formar la historia de la ciudad de Guanajuato, Vol. 1, Universidad de Guanajuato, Guanajuato: 181. Springer 26 Climatic Change (2007) 83:9Ð38

Fig. 6 Damaging flood in colonial Guanajuato, 1620Ð1820, as identified from archival documentation. this diagram shows those floods that contributed to documented economic and life losses. A number of these floods may be associated with recorded weather events (unusually heavy rainfall or storms)

proportions and death had already gripped the Baj«õo (Berthe 1970: 247Ð261; Orozco y Berra 1938: 242Ð248). The ﬂood event thus compounded the difﬁculties faced, particularly by the poorer sectors of society in the area. Indeed, given the context in which the event took place, it may well have helped trigger “the emergence of popular uprisings in Guanajuato and Mexico City”.67

4.2 The social construction of ﬂood events

There are numerous floods recorded in Guanajuato in the second half of the eighteenth century and early 1800s, and, as Fig. 6 demonstrates, a cluster of events affected Guanajuato (specifically the towns of Leon, Celeya and Guanajuato City) in 1749, 1750, 1753, 1760, 1770, 1771, 1772, 1788, 1791, 1804 (Endfield et al. 2004b). Though it should be acknowledged that the archival record of flood events is incomplete and partial (see Section 5.2), some of these recorded flood events may have been associated with storms or heavy rains identified in independent reports. The relative impacts of such events, however, were very much dependent upon the context in which they took place. The flood in Guanajuato on July 5th, 1760 which “resulted in turmoil...causing the disaster to survive in lamentable tradition as one of the most memorable that has been experienced” took place at a time when the city’s populace was at their most vulnerable. The event occurred “at twelve at night” when most of the population was asleep, “and finding the river without a bridge or any flood defences...the houses were flooded with water, the roads were covered and there were innumerable people drowned or unspeakably injured” (Ajofr«õn, cited in Moreno 1986). While the cause of the flood was considered to be a “a furious thunder storm which caused a sudden burst of water in the nearby areas” (Ajofr«õn, cited in Moreno 1986), there were other factors which may have exacerbated flood risk. Mining disturbance around the city (Butzer and Butzer 1997: 170), together with more general trends in deforestation to bring more land into to agricultural use as population expanded throughout the eighteenth century, may have also enhanced the risk of siltation and flooding (Endfield and O’Hara 1999; Ouweneel 1996: 38; Metcalfe et al. 1989).

67 Marmolejo, Efemerides« Guanajuatenses, 1967; See Prada 2003; Endﬁeld et al. (2004) presents more detail on the responses to this event (Cope 1994). Springer Climatic Change (2007) 83:9Ð38 27

The scale of water management also increased the vulnerability of society to flooding in this area. A complex myriad of water diversion channels, dams and reservoirs had been constructed by the middle of the eighteenth century. This was largely for irrigation purposes in this, the ‘breadbasket’ of the Mexican economy, to ensure crop growth during the winter dry season and even during short-lived droughts. There was also a need to supply water for the growing urban population in the colonial towns and cities that developed across the region (Murphy 1986). The manipulation of the natural hydrology of the region for these purposes, however, led to a number of problems. Many and varied lawsuits were filed detailing cases of water diversion, storage, monopolisation, deprivation, usurpation and over-abstraction associated with this water management infrastructure.68 More significantly, environmental modification increased flood risk and may account for the clustering of flood events recorded after 1750. Dam breaches appear to have been commonplace events, frequently resulting from poor dam maintenance, though in some cases, flooding may have been the result of deliberate sabotage related to long-standing feuds between neighbouring landowners (Murphy 1986). A document recorded in Celaya on June 26th 1791 by the Mayor of the indigenous residents of the city illustrates such problems. He suggested that “there has been seen twice an overflowing of flood waters which afterwards have caused ruin and damage in the houses and greater damage for the general public.” The blame, he argued, rested with the local hacendados who had changed the natural course of the River Laja with dams “causing damage in times of floods.” The disruption and devastation caused by the flooding had additional repercussions for the community. There was concern among councillors that pools of standing water were “causing notable damage to their health and interests”.69 The development of a complex system of water storage, diversion, abstraction to ensure water for irrigation purposes in the Baj«õo might have thus increased societal vulnerability to flooding and its environmental health implications, particularly in the more densely populated urban areas (Endfield et al. 2004b).

4.3 Flood events, disputes and opportunism

Flooding may have also presented opportunities. The sequence of maps showing part of the River Atoyac and environs in Oaxaca (Plates 1Ð3) demonstrates one case in question. The maps were produced to illustrate the impacts of a series of flood events between 1775 and 1804 on the River Atoyac in Oaxaca and were used as supporting evidence in a land dispute between the community of Xoxocotl«an and the estate owner (hacdendado) of the hacienda of San Miguel. The two parties used the river as a territorial boundary between their respective lands. That this part of the river was prone to flooding is evidenced by the extensive flood defences constructed on the south bank of the river, and a substantial floodwall, 4 varas (just under 4 m) high, erected by the community. Following a flood event in 1775, however, there was an avulsion event70 or a shift of the river channel, as Plate 1 illustrates: “The channel that from time immemorial has been for the Atoyac River up to the year past of 1775...was

68 See, for example, AGN Tierras, vol. 674, exp 1 30ff; Follow the dispute in vol. 675, exp 1; AGN Tierras vol. 192, exp. 1; AGN Tierras, vol. 586, exp 8; AGN Tierras, vol. 1872, exp. 15 10f; AGN Tierras, vol. 671, exp. 3, 18 ff; AGN Tierras, vol. 2901, exp. 36; AGN Tierras, vol. 2959, exp 141; AGN Tierras, vol. 988, exp. 1,2 y 3, 516 ff; AGN Tierras, vol. 1110, exp. 18; AGN Tierras, vol. 1166 exp. 1 fs. 450; AGN Tierras, vol. 2963, exp.116, f 246-308; AGN Tierras, vol. 1352, exp 1; AGN Tierras, vol. 1368. 69 AGN Tierras, vol. 2071, exp. 1, fs. 1Ð110. 70 Avulsion: an abrupt change in the course of a stream or river, generally from one channel to a new one. Often used to describe a shift in a bed of a river which has been used as a boundary by property owners. Springer 28 Climatic Change (2007) 83:9Ð38 Plate illustrating and detailing ﬂooding and avulsion of the Rio Atoyac, 1775 Plate 1.

Springer Climatic Change (2007) 83:9Ð38 29 moved by an tempestuous flood”. The channel in the centre of the map is described as the “...new channel of the river and where the river has flowed since 1775 and does so presently”. The flood event thus resulted in substantial landscape changes. The community seized upon this event, together with continued avulsion of the river channel between 1775 and 1785 (Plate 2), to forward a case for the redefinition of the boundaries between their lands and those of the neighbouring hacienda. According to the hacendado, the community, “against all reason”, attempted to “meddle with the boundaries”. Interestingly, and as Plate 3 illustrates, the migration of this part of the Atoyac, and hence also the degree of contention between the two landowners, would continue into the early years of the 1800s, when, as noted earlier, anomalous weather events would affect the north, south and central regions of the country. A note at the top of the map suggests that “in the year 1804, the river broke free towards the north and east”, not only flooding an extensive part of the property belonging to the hacendado, but effectively prolonging the dispute between the neighbouring landowners.71

5 Contextualising climate variability and vulnerability in colonial Mexico

5.1 Weather events in Mexico and global climatic phenomena

It is to be expected that at least some of the drought, flood and storm events recorded in the colonial archives of Mexico might be explained by historical ENSO activity. Ortlieb (2000) has identified medium El Niño events in both 1784 and 1785, possibly providing one hypothetical explanation of the droughts and frosts recorded across the country at this time. The droughts and harvest failures recorded across the Baj«õo in 1641 and those recorded across the country between 1750 and 1751 also correspond to strong or moderate El Niño years according to Quinn and Neal’s (1992) El Niño chronology. The reliability of the documentary data upon which such classifications were based, however, has been questioned (Hocquenghem and Ortlieb 1992; Ortlieb 2000). Moreover, one needs to consider that there would most probably have been a time lag between the extreme weather event and the emergence, recognition and recording of a crisis situation. Other recorded events appear to have been synchronous in different parts of the country and to have coincided with more widely documented anomalous weather. Unusually heavy rain and flooding recorded in Guanajuato and elsewhere in the country between 1791 and 1792, and also the poor harvests reported from some regions the subsequent year,72 might well be associated with the well-documented El Niño of this period. Elsewhere, as far afield as India, St. Helena and Montserrat, there were simultaneous droughts in 1791 and the period 1789Ð1793 is particularly well-documented as being one of global climatic abnormality generally (Grove 1998). Quinn and Neal’s record suggests there was a strong El Niño in 1783. This was a time when freak weather was reported across the country. The Gazeta de Mexico« of March 10th 1784 reported how, from December 7th the previous year, “evil winds from the south and south east have blown with greater frequency.” This was associated with the re-appearance of “sicknesses”, including “pneumonia, pleurisy, apoplexy, angina, inflammation of the throat; and finally, other ailments, which although customary during this season of the year, began

71 Maps copied with kind permission from the Archivo Privado de Don Luis Casta˜neda Guzman, Oaxaca. 72 AGEO, leg 4, exp. 47 2ff. Springer 30 Climatic Change (2007) 83:9Ð38 Plate illustrating and detailing disputed territorial boundaries in 1785 following avulsion events Plate 2.

Springer Climatic Change (2007) 83:9Ð38 31

Plate 3. Plate illustrating further movement of the river channel and flooding in 1804 to be felt with greater frequency and virulence.”73 In Oaxaca, freak “storm” events, which affected Teotitl«an del Valle on 15th December 1783, followed a series of worrying phenomena. “A great snowfall covered the fields of the area”, while “extremely heavy rainstorms, which were no less damaging to the sown crops than the strong drought of the previous months... the streets were transformed into rivers and the houses were left flooded out”. The events were compounded by “extraordinary heat, rigorous droughts, shattering hurricanes, frightening falls of hail and storms accompanied by prolonged underground noises” (Gay 1950: 427Ð428). Only a few months earlier, volcanic ash clouds were observed and ash falls had covered fields and roads across the region, seriously disrupting agricultural production.74 Although there may well have been a degree of sensationalism in this (second hand) report, catastrophic weather and disastrous events are also recorded further afield. In Mediterranean Europe, the period between 1760 and 1800 saw an increased frequency of both drought and flood episodes thought to be associated with the so called Mald«a climatic anomaly (Barriendos and Llasat 2003). The 1780s in particular are thought to have been

73 Gazeta de Mexico« , March 10th, 1784, cited in Cooper 1965. 74 AGEO, leg 4, exp. 47 2ff. Springer 32 Climatic Change (2007) 83:9Ð38 a period of “very pronounced regional anomalies in the general circulation” across Europe generally (Kington 1988: 2). It is perhaps interesting to note also that the Lakicigar eruption (Iceland), which took place between June 1783 and February 1784, resulted in anomalous weather events across north and west Europe and North America (Strothers 1996; Grattan and Brayshay 1995). The volcanic eruption in Asama, Japan, also in 1783, may have contributed to these atmospheric impacts. Reports from the time indicate the occurrence of dry fogs, unusually warm summers in Europe and North America and a cold winter in 1783 and unusually cold temperatures over the succeeding two to three years, characteristics which Strothers (1996) argues were due to the effect of aerosols from the eruption. The ash falls recorded as affecting Oaxaca in the early summer of 1783 were most probably the result of more local events. But one could perhaps reconsider the anomalous weather conditions across Mexico in 1784 and 1785 in this light. Nevertheless, caution is needed in drawing links between the conditions described in Mexico and events further aﬁeld, especially given the degree of normal climatic variability in Mexico generally.

5.2 Interpreting climatic history from the colonial archival records

The changes, problems and opportunities that future global climate changes might present could be anticipated, or at least better understood, by examining how similar challenges and opportunities have been met in the past (Meyer et al. 1998). Invaluable lessons can be learned from the way in which societies have been affected by and have coped with climate variability and historical extreme weather events in the past (Bradley 1999). The colonial archives of Mexico provide an important resource through which to explore these themes. In particular, it is possible to use these sources to investigate how societal vulnerability and resilience to climatic change and extreme weather events varied over time and in different environmental, social and economic contexts. It is clear that drought has been a regular feature of the Mexican climate (Liverman 1999), stimulating a wide range of adaptive and risk avoidance strategies including food storage, water storage and irrigation (Garc«õa-Acosta 1993). These adaptive strategies effectively less- ened the impacts of recurrent single year droughts for society. In contrast, lower frequency climatic variations with larger amplitudes, tested social resilience and in some cases appear to have exceeded the capacity for human adaptability (Erickson 1999). Archival references indicate that prolonged or successive droughts were particularly devastating, contributing to harvest failure and famine, land abandonment, out-migration, loss of economic livelihood and in some cases, loss of life. There is evidence for organised relief at a range of scales and there appears to have been a degree of community engagement during and following periods of agrarian crisis. Less attention has been paid to the historical record of flooding relative to that of drought in Mexico (J«auregui 1997: 8). Yet documents reveal that in the central and southern regions of the country at least, unexpected flooding presented an equally devastating environmental problem, claiming lives and causing extensive and in some cases irreparable damage. Not all flood events were associated with extreme or unusually rainfall. For instance, there appears to have been an increase in the number of floods in the Baj«õo in the eighteenth century when mining activities and deforestation expanded in the area and, more specifically, when water management and hydrological modification in the region reached their zenith (Murphy 1982). While increasingly elaborate water management systems in the Baj«õo were designed to insulate society from the adverse effects of inter-annual drought and water scarcity, therefore, they may have actually increased vulnerability to catastrophic flooding. Both lessening Springer Climatic Change (2007) 83:9Ð38 33 and catastrophe hypotheses (Bowden et al. 1981) thus appear to pertain when considering “trajectories of vulnerability” (Messerli et al. 2000) in this part of colonial Mexico. On a number of occasions, extreme events may have provided something of a ‘trigger’ to already difficult social, economic, or political circumstances, rendering situations untenable for at least some sectors of society, notably the poorest, and predominantly indigenous communities (Liverman 1990, 1999). Social unrest recorded in Guanajuato and Mexico City in 1692 represents a case in point. There is also evidence from elsewhere in the country of a link between weather events and unrest. Attacks by nomadic groups, referred to in the documentation as Apaches or Conchos, posed a significant threat to colonial investment and settlement in northern Mexico throughout the colonial period. The number of recorded attacks, raids and thefts increased in the 1750s (Griffen 1979: 24; Martin 1996). As discussed in more detail elsewhere (Endfield and Fern«andez-Tejedo, in press), the prolonged period of drought in the 1750s might have contributed to these invasions.75 Despite the presence of a Spanish military defence, violent incursions and attacks on settlements are again documented during periods of anomalous weather and food scarcity in the 1760s, and 1770s, and into the nineteenth century (Endfield and Fern«andez-Tejedo 2005).76 On at least some occasions it seems that climate events may have interacted with prevailing ecological and socio-economic conditions to magnify the impact of infectious disease. It may be no coincidence, for example, that epidemic disease is recorded in the north of the country following prolonged drought in the 1750s and the events that contributed to the ‘Year of Hunger’ between 1784 and 1785. Successive or combined extreme weather events may have also created conditions conducive to so called ‘clusters’ of disease outbreaks on a number of occasions. Sequential extremes, or droughts punctuated by heavy rains and flooding, are thought to destabilize the predator/prey interactions leading to population explosions of opportunistic and disease carrying organisms (Epstein 2001, 2002). A direct link between sequential climate, events and epidemic disease is thought to explain outbreaks of epidemic disease in 1545 and 1576 which contributed to massive indigenous depopulation (Marr and Kiracofe 2000; Acuña-Soto et al. 2000, 2002, 2004). Documentary sources indicate that later phases of epidemic disease re-emergence may similarly relate to sequential weather events. Persistent rains over a period of 11 days and consequent flooding in the northeast of Mexico, for example, (specifically, Coahuila, Nuevo Le«on and along the Lower Rio Grande) in June 1802 combined with a previous drought contributed to economic crisis, but also the re-emergence of epidemic disease across the region (Butzer 2003).77 There are, of course, many problems associated with the recording and interpretation of past events from the colonial archives. The perception of the scale, impact and severity of a specific climatic event will vary between individuals (Duncan and Gregory 1999:4). Accounts of specific climatic events or their impacts must be read with an awareness of the subjectivities applied by individuals. The information contained within historical documents will inevitably be subjective, reflecting the environmental awareness and purpose of the author, but also the nature of the document and its intended audience. It is difficult to establish from historical sources what constituted normal weather conditions and what was perceived to be unusual or extreme (Bryson and Padoch 1980). Whether conditions were classed as normal, unusual or extreme in descriptions and accounts depended on the author’s experience, and the normal

75 AHMCH Guerra, caja 1, exp. 13; AHMCH Gobierno, Caja 27, exp.4; Archivo del Ayuntamiento de Chihuahua, University of Texas at El Paso, Microfilm no. 491, cited in Martin 1996: 25. 76 AHMCH Gobierno, Guerra, caja 2, Exp. 4; AGN Jesuitas, leg II,-9, exp. 30. 77 This year witnessed a ‘year of hunger’ as discussed in Portillo (1910): 188. Springer 34 Climatic Change (2007) 83:9Ð38 range of climatic variation communicated to them through oral history, historical knowledge and social memory (Hassan 2000). Some periods of recorded harvest losses may have little to do with climate phenomena per se. There are, for example, many references to grain scarcity and hunger which were less to do with drought or weather related harvest failure and more a result of profiteering and speculation by wealthy landowners and merchants, particularly in Guanajuato and Chihuahua. Drought speculation by grain merchants artificially inflated prices, and there are also reports of producers and traders actually with-holding and storing grains in order to simulate scarcity and so raise produce prices (Endfield et al. 2004a,b). In addition, there may well have been a degree of sensationalism with at least some of the reports. A letter dated 22nd June 1768, for instance, details the lack of rains in central Mexico at that time and the need to seek divine intervention through rogation ceremonies to the Virgin de los Remedios. The document suggests that the scarcity of rains had “fermented an illness of measles, though there was not much mortality”. There may well have been links between the lack of rain and epidemic disease, as discussed earlier. However, the author of the letter also suggests, somewhat sensationally, that “the lack of rains is the cause of the illness that is suffered and the repeated earthquakes” experienced from the 4th April of that year.78 It was not unusual for contemporary accounts to draw links between some anomalous weather events and other natural hazards. Ortleib (1999) has highlighted how in colonial Peru, earthquakes and abnormal rainfall were closely associated, such that some reports on natural disasters may have led observers to erroneously attribute destruction to unusual weather rather than seismic causes. Though the case reported above deals with the implications of drought rather than heavy rains, it does illustrate the need for caution in interpreting descriptive accounts of weather and weather related events. There are also more temporal recording biases to bear in mind when interpreting the historical documentation. The majority of reported incidences of drought and flood- related crisis are from the eighteenth century and particularly after 1750. Of course this trend might well be significant, especially, as already noted, there is evidence of climatic variability elsewhere at this time. Yet it is felt that in Mexico at least, this trend might reflect not so much an increase in the number of extreme events and natural hazards per se, but the availability of more documentary material available for the eighteenth century relative to the early colonial period. There will have also been an increase in the relative vulnerability of society by this stage. Population expansion coupled with increased economic development effectively meant that there was more disaster potential, while the very hierarchical structure of Mexican society meant that some groups of people were disproportionately affected during such events. Moreover, by the eighteenth century there may have also been more environmental awareness and experience of different types of weather related crisis. This may have helped inform people’s comprehension and conceptualisation of climate risk and perceived vulnerability and hence increased the likelihood of events being recorded (Garc«õa-Acosta 1993). Notwithstanding methodological and interpretative difficulties, it is clear that climate variability and extreme weather events in Mexican history have posed a dynamic set of problems but also opportunities, altering the context of vulnerability for different sectors of society. The rich colonial archives of Mexico represent an invaluable resource for the investigation of historical extreme weather events across the country and provide a unique window through which to view the changing vulnerabilities to- and impacts and implications of these events.

78 AGN Correspondencia de Virreyes, vol. 12 (series 2), fs. 244. Springer Climatic Change (2007) 83:9Ð38 35

Acknowledgements This paper represents research conducted as part of an Arts and Humanities Re- search Board (AHRB, now Arts and Humanities Research Council, AHRC) funded project investigating ‘Agrarian responses to extreme climate events in colonial Mexico: 1521–1821’, reference number B/RG/AN6160/APN10797. Thanks go to Dra. Isabel Fern«andez Tejedo with whom I spent many happy hours in the various archives of Mexico and the many archivists and porteros in the AGN Mexico City and archives and libraries of Oaxaca, Guanajuato and Chihuahua, whose knowledge and assistance is very much appreciated. Thanks also go to Mr. Chris Lewis, School of Geography, University of Nottingham, for car- tographic assistance and to the reviewers for constructive advice on improvements to the ﬁrst draft of this paper.

References

Acuña-Soto R, Caderon Romero L, Maguire JH (2000) Large epidemics of haemorrhagic fevers in Mexico. Am J Trop Med Hyg 62(6):733Ð739 Acuña-Soto R, Stahle DW, Cleaveland MK, Therrell MD (2002) Megadrought and megadeath in 16th century Mexico. Emerg Infect Dis 8(4):360Ð362 Acuña-Soto R, Stahle DW, Therrell MD, Griffin RD, Cleaveland MK (2004) When half the population died: the epidemic of hemorrhagic fevers of 1576 in Mexico. Minireview in FEMS Microbiol Lett 240:1Ð5 Adger N, Kelly PM (1999) Social vulnerability to climate change and the architecture of entitlements. Migration Adaptation Strateg Global Change 4:253Ð256 Adger WN, Kelly PM, Ninh NH (2001) Living with environmental change: social resilience, adaptation and vulnerability in vietnam. Routledge, London Barriendas M, Llasat MC (2003) The case of the Mald«a anomaly in the western Mediterranean Basin (AD 1760Ð1800): an example of a strong climatic variability. Clim Change 61:191Ð216 Berthe, JP (1970) La pese de 1643 en Michoac«an examen cr«õtico de una tradici«on. In Garc«õa Martinez, B (ed) Historia y sociedad en el mundo de habla española. Homenaje a Jos«e Miranda. El Colegio de M«exico, M«exico pp 247Ð262 Berz G (1997) Changing weather extremes: the viewpoint of an international reinsurer. Proceedings of a workshop on indices and indicators for climate extremes. Asheville, NC, June 3Ð6 Bowden MJ, Kates RW, Kay PA, Riebsame WE, Warrick RA, Johnson DL, Gould HA, Wiener D (1981) The effect of climate fluctuations on past human populations: two hypotheses. In Wigley TML, Ingram MI, Farmer G (eds) Climate and history. Studies in past climates and their impact on man. Cambridge University Press, Cambridge, pp 479Ð513 Bradley RS (1999) Paleoclimatology: reconstructing climates of the quaternary. Harcourt Academic Press, San Diego Brooks N, Adger WN, Kelly PM (2005) The determinants of vulnerability and adaptive capacity at the national level and the implications for adaptation. Global Environ Change 15:151Ð163 Brown N, Issar I (eds) (1998) Water environment and society in times of climatic change. Kluwer Press Bryson RA, Padoch C (1980) On the climates of history. J Interdiscip Hist 10(4):583Ð597 Butzer KW, Butzer EK (1997) The natural vegetation of the Mexican Baj«õo: archival documentation of a sixteenth century savannah environment. Quaternary Int 43/44:161Ð172 Butzer EK (2003) Floods, epidemic and human response in Northeastern Mexico, 1802. Paper presented at the Annual Conference of the Association of American geographers, New Orleans, USA, March, 2003 Cavazos T, Hastenrath S (1990) Convection and rainfall over Mexico and their modulation by the Southern Oscillation. International J Clim 10:377Ð386 Cavo AP (1949) Historia de M«exico, anotada por Ernesto Burrus, prologo Mariano Cuevas, M«exico, Editorial Patria, S. A. Mexico Cleaveland MK, Stahle DW, Therrell MD, Villanueva-Diaz J, Burns BT (2003) Tree-ring reconstructed winter precipitation and tropical teleconnections in Durango, Mexico. Clim Change 59:369Ð388 Conde C, Liverman D, Flores M, Ferrer R, Auaujo R, Betancourt E, Villreal G, Gay C (1997) Vulnerability of rainfed maize crops in Mexico to climate change. Clim Res 9:17Ð23 Cooper D (1965) Epidemic disease in Mexico City, 1761Ð1813. Institute of Latin American Studies, University of Texas Press, Austin Cope RD (1994) The limits of racial domination. Plebeian society in colonial Mexico City, 1660Ð1720. The University of Wisconsin Press, Madison, Wisconsin Curtin CG, Sayre NF, Lane BD (2002) Transformations of the Chihuahuan borderlands: grazing, fragmentation and biodiversity conservation in desert grasslands. Environ Sci Policy 5:55Ð68 Cutter SL (1996) Vulnerability to environmental hazards. Prog Hum Geogr 20(4):529Ð539 Springer 36 Climatic Change (2007) 83:9Ð38

Diaz SC, Therrell MD, Stahle DW, Cleaveland MK (2002) Chihuahua (Mexico) winter-spring precipitation reconstructed from treeÐrings, 1647Ð1992. Clim Res 22:237Ð244 Duncan J, Gregory D (1999) Writes of passage: reading travel writing. Routledge, London Eakin H (2005) Institutional change, climate risk, and rural vulnerability: cases from central Mexico. World Dev 33(11):1923Ð1938 Easterling DR, Evans JL, Groisman PY, Karl TR, Kunkel KE, Ambenje P (2001) Observed variability and trends in extreme climate events. A brief review. Bull Am Meteorol Soc 81(3):417Ð425 Endfield GH, O’Hara SL (1997) Conflicts over water in the “Little Drought Age” in central Mexico. Environ Hist 3:255Ð272 Endfield GH, Fern«andez-Tejedo I, O’Hara SL (2004a) Drought, and disputes, deluge and dearth: climatic variability and human response in colonial Oaxaca, Mexico. J Hist Geogr 30:249Ð276 Endfield GH, Fern«andez-Tejedo I, O’Hara SL (2004b) Conflict and co-operation: water, floods and social response in colonial Guanajuato, Mexico. Environ Hist 9(2):221Ð247 Endfield GH, Fern«andez-Tejedo I (in press) Decades of drought, years of hunger: archival investigations of multiple year droughts in late colonial Chihuahua. Clim Change Endfield GH, Davies SJ, Fern«andez-Tejedo I, Metcalfe SE, O’Hara SL (2006) Documenting disaster: climate, crisis and catastrophe in colonial Mexico. In: Pfister C, Mauch C (eds) Natural hazards: cultural responses in global perspective. To be Published by Rowman and Littlefield Epstein PR (2001) Climate change and emerging infectious diseases. Microbes and Infect 3:747Ð754 Epstein PR (2002) Shifting times and paradigms: climate change and public health. Abstract from Leroy S, Steart IS (eds) Conference on environmental catastrophes and recoveries in the holocene. August 29thÐSeptember 2nd, 2002. Brunel University, Uxbridge, UK, pp 30Ð31 Erickson CL (1999) Neo-environmental determinism and agrarian collapse in andean prehistory. Antiquity 74, 281:634Ð642 Florescano E (1976) Origen y desarollo de los problemas agrarios de Mexico (1500Ð1821). Colecci«on Prob- lemas de Mexico. Ediciones Era, Mexico Florescano E (1980) Una historia olvidada: la sequia en Mexico. Nexos 32:9Ð13 Forescano E (1981) Fuentes para la historia de la crisis agr«õcola de 1785Ð1786. Archivo General de la Naci«on, Mexico City Florescano E, Swan S, Menegus M, Galindo I (1995) Breve historia de la sequia en Mexico. Universidad Veracruzana, Veracruz Fraser EDG, Mabee W, Slaymaker O (2003) Mutual vulnerability, mutual dependence. The reflexive relation between human society and the environment. Global Environ Change 13:137Ð144 Garcia SJ (2000) An«alisis de vulnerabilidad agropecuaria por sequia en el estado de Chihuahua. Instituto de Ecologia, Jalapa, Veracruz; available at www.sequia.edu.mx/proyectos/vulnera.html Garc«õa-AcostaV (1993) Las sequias historicas de Mexico. La Red (1):2Ð18 Garc«õa-Acosta V (ed) (1997) Historia y desastres in America Latina. Red de estudios sociales en prevenci«on de desastres en Am«erica Latina. CIESAS, Mexico Gay JA (1950) Historia de Oaxaca. vol 4, 3rd Edn Cap XXI Mexico Gibson C (1964) The Aztecs under Spanish rule. Stanford University Press, California Grattan J, Brayshay M (1995) An amazing and portentious summer: environmental and social responses in Britain to the 1783 eruption of an Iceland volcano. The Geogr J 161(2):125Ð134 Griffen WB (1979) Indian assimilation in the Franciscan area of Nueva Vizcaya. The University of Arizona Press, Tuscon, Arizona Grove R (1998) Global impact of the 1789Ð1793 El Niño. Nature 393:318Ð319 Hassan F (2000) Environmental perception and human responses in history and prehistory. In: McIntosh RJ, Tainter JA, McIntosh SK (eds) The way the wind blows: climate, history and human action. Columbia University Press, New York, pp 121Ð140 Hassig R (1981) The famine of one rabbit: ecological causes and social consequences of a pre-Columbian calamity. J Anthropol Res 37:172Ð182 Hastenrath S (1988) Climate and circulation of the tropics. D Reidel Publishing, Boston Hodell DA, Brenner M, Curtis JH, Medina-Gonz«alez R, Idelfonso Chan Can E, Abornaz-Pat A, Guilderson TP (2005) Climate change on the Yucatan Peninsula during the little ice age. Quaternary Res 63:109Ð 121 Hocquenghem A, Ortlieb L (1992) Historical records of El Nino events in Peru (XVI-XVIIIth centuries: The Quinn et al. (1987) chronology revisited. In Ortlieb L, Machar«e J (eds) Paleo ENSO records International Symposium Extended Abstracts. OSTROM/CONCYTEC, Lima, pp 143Ð149 Humboldt A (1811) (1973 reprint). Ensayo politico sobre el reino de la Nueva Espana, estudio preliminar y notas de Juan A. Ortega y Medina, Mexico: Editorial Porrua S.A, Mexico

Springer Climatic Change (2007) 83:9Ð38 37

Ingram MJ, Farmer G, Wigley TML (1981) Past climates and their impact on man: a review. In Wigley TML, Ingram MJ, Farmer G (eds) Climate and history. Studies in past climates and their impact on man. Cambridge University Press, Cambridge, pp 3Ð49 J«auregui E (1979) Algunos aspectos de las fluctuaciones pluviom«etricas en Mexico en losultimos « cien años. Bolet«õn del Instituto de Geograf«õa 9:39Ð64 J«auregui E (1997) Climatic changes in Mexico during the historical and instrumented periods. Quaternary International 43/44:7Ð17 J«auregui E, Kraus D (1976) Some aspects of climate fluctuations in Mexico in relation to drought. Geofisica Internacional 16:45Ð61 Karl TR, Easterling DR (1999) Climate extremes: selected review and future research directions. Climatic Change 42:309Ð323 Kelly PM, Adger WN (2000) Theory and practice in assessing vulnerability to climate change and facilitating adaptation. Climatic Change 47:325Ð352 Kington J (1988) The weather of the 1780s over Europe. Cambridge University Press, Cambridge Kirkby AVT (1973) The use of land and water resources in the past and present Valley of Oaxaca, Mexico. Memoirs of the Museum of Anthropology, No. 5. University of Michogan, Ann Arbor Kovats RS, Bouma MJ, Hajat S, Worrall E, Haines A (2003) El Niño and health. The Lancet 362(9394):1481Ð 1489 Kutzbach JE, Street-Perrott E (1985) Milankovitch forcing of fluctuations in the level of tropical lakes from 18 to 0 kyr BP. Nature 317:130Ð134 Leyden BW, Brenner M, Hodell DA, Curtis JH (1994) Orbital and internal forcing of climate on the Yu- catan Peninsula for the past ca. 36 ka. Palaegeography, Palaeclimatology, Palaeoecology 109:193Ð 210 Lipsett Rivera S (1992) Indigenous communities and water rights in colonial puebla: patterns of resistance. The Americas 48(4):463Ð483 Lipsett Rivera S (1999) To defend our water with the blood of our veins: the struggle for resources in colonial Puebla. University of New Mexico Press, Albuquerque Liverman D (1990) Drought impacts in Mexico: climate, agriculture, technology, and land tenure in Sonora and Puebla. Annals of the Association of American Geographers 80(1):49Ð72 Liverman DM (1999) Vulnerability and adaptation to drought in Mexico. Nat Res J 39:99Ð115 Liverman DM, O’Brien KL (1991) Global warming and climate change in Mexico. Global Environ Change 1(5):351Ð364 Luers AL, Lobell DB, Sklar L, Addams CL, Matson PA (2003) A method for quantifying vulnerability, applied to the agricultural system of the Yaqui Valley, Mexico. Global Environ Change 13:255Ð267 Magaña VO (1999) Los impactos de El Niño en Mexico. Dirrecci«on General de Protecci«on Civil, Secretaria de Gobernacion, D.F. Mexico Magaña V, Conde C, Sanchez O, Gay C (1997) Assessment of current and future regional climate scenarios for Mexico. Clim Res 9(1Ð2):107Ð114 Magaña VC, V«azquez JL, P«erez JL, P«erez JB (2003). Impact of El Niño on precipitation in M«exico. Geof«õsica Internacaional 42:313Ð330 Malvido E (1973) Cronolog«õa de las epidemias y crisis agricolas de la epoca colonial. Historia Mexicana 89:96Ð101 Marr JS, Kiracofe JB (2000) Was the Huey Cocoliztli a Haemorrhagic Fever? Med Hist 44:341Ð362 Martin CE (1996) Governance and society in colonial Mexico. Chihuahua in the eighteenth century. Stanford University Press. Stanford, California McBean G (2004) Climate change and extreme weather: a basis for action. Nat Hazards 31(1):177Ð190 McCarthy JJ, Canziani OF, Leary NA, Dokken DJ, Kasey S (2001) Climatic Change 2001: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Third Assessment Report of the Intergovern- mental Panel on Climate Change. Intergovernmental Panel on Climate Change Mendoza B, J«auregui E, Diaz-Sandoval R, Garc«õa-Acosta V, Eelasco V, Cordero G (2005). Historical droughts in central Mexico and their relation with El Niño. J Appl Meteorol 44(5):7009Ð716 Messerli B, Grosjean M, Hofer T, Nuñez L, Pfister C (2000) From nature dominated to human dominated environmental changes. Quaternary Sci Rev 19:459Ð479 Metcalfe SE (1987) Historical data and climatic change in Mexico Ð a review. The Geogr J 153:211Ð222 Metcalfe SE, Street-Perrott FA, Perrott FA, Harkness DD (1991) Palaeolimnology of the Upper Lerma Basin, Central M«exico: a record of climatic change and anthropogenic distrurbance since 11,600 years BP. J Palaeolimnol 5:197Ð218 Metcalfe SE, Bimpson A, Courtice AJ, O’Hara SL, Taylor DM (1997) Climate change at the monsoon/Westerly boundary in Northern Mexico. J Palaeolimnol 17(2):155Ð171

Springer 38 Climatic Change (2007) 83:9Ð38

Meyer WB, Butzer KW, Downing TE, Turner BL II, Wenzel G, Wescoat J (1998). Reasoning by Analogy. In: Raynor S, Malone EL (eds) Human choice and climate change, no. 3, Tools for Policy Analysis, Batelle Press, Columbus, Ohio, pp 218Ð289 Morales T, Magaña V (1999) Unexpected frosts in central Mexico during summer. Proceedings of the 10th Symposium on Global Change Studies, 10Ð15 January, 1999, Dallas, TX, Preprint Volume. American Meteorological Society (AMS), Boston, MA, USA, pp 262Ð263 Moreno H (1986) Francisco de Ajofr«õn, Diario del Viaje a la Nueva España Secretaria de Educaci«on Publica, Mexico Mosiño Alem«an P, Garc«õa E (1974) The climate of Mexico. In: Bryson RA, Hare KF (eds) Climates of North America. Elsevier, New York, pp 345Ð404 Murphy ME (1982) Irrigation in the Baj«õo region of colonial Mexico. Westview Press, Dellplain Latin American Studies, No. 19 NOAA, International Tree Ring Data Bank. ftp://ftp.ngdc.noaa.gov/paleo/treering/ O’Hara SL (1993) Historical evidence of fluctuations in the level of Lake P«atzcuaro, Michoac«an, Mexico over the last 600 years. The Geogr J 159:51Ð62 O’Hara SL, Metcalfe SE (1995) Reconstructing the climate of Mexico from historical records. The Holocene 5(4):485Ð490 Orozco and Berra M (1938) Historia de la dominaci«on española en M«exico, con una advertencia de Genaro estrada, 2 vols, Mexico, Biblioteca Historica Mexicana de obras ineditas N«um 10:242Ð248 Ortlieb L (2000) The documented historical record of El Niño events in Peru: an update of the Quinn record (sixteenth through nineteenth centuries). In: Diaz HF, Markgraf V (eds) El Niño and the southern oscillation: multiscale variability and global and regional impacts. Cambridge University Press, Cambridge, pp 207Ð295 Ouweneel A (1996) Shadows over An«ahuac. An ecological interpretation of crisis and development in central M«exico 1730Ð1800, University of New Mexico Press, Albuquerque. Parry M (2001) Viewpoint: climate change: where should our priorities be? Global Environ Change 11:257Ð 260 Patt A, Kelin RJT, de la Leinart Vega-Leinart A (2005) Taking the uncertainty in climate change vulnerability assessment seriously. C.R Geoscience 337:411Ð424 Portillo A (1910) Oaxaca en el centenario de la Independencia nacional. Oaxaca Prada SN (2003) Cultural strategies in the 1692 uprising in Mexico City: contributions for reconstructing the history of ancient political culture. Historia Mexicana 53(1):5Ð63 Quinn WH, Neal VT (1992) The historical record of El Niño events. In: Bradley RS, Jones PD (eds) Climate since AD 1500. Routledge, London, pp 623Ð648 Ribot JC, Magalhães AR, Panagides SS (1996) Climate variability, climate change and social vulnerability in the semi-arid tropics. Cambridge University Press, Cambridge Ropelewski CF, Halpert MS (1987) Global and regional scale precipitation patterns associated with the El Niño Southern Oscillation. Monthly Weather Review 115:1606Ð1626 Ropelewski CF, Halpert MS (1989) Precipitation patterns associated with the high index phase of the Southern Oscillation. J Clim 1:268Ð284 Salinger MJ (2005) Increasing climate variability and change: reducing the vulnerability. Guest Editorial. Clim Change 70:1Ð3 Schneider SH (2001) What is dangerous climate change? Nature 411:17Ð19 Smith K, Ward R (1998) Floods: physical processes and human impacts Wiley & sons Chichester Stahle DW, Villanueva-Diaz J, Cleaveland MK, Therrell M, Paul GJ, Burns BT, Salinas W, Suzan H, Fule PZ (2000) Recent tree-ring research in Mexico. In Roig FA (compiler) Dendrochronologia en America Latina: Dendrochronology in Latin America. Editorial de La Universadad nacional de Cuyo, Mendoza, Argentina, pp 285Ð305 Strothers R (1996) The great dry fog of 1783. Clim Change 32:79Ð89 Sullivan C, Meigh J (2005) Targeting attention on local vulnerabilities using an integrated index approach: the example of the climate vulnerability index. Water Sci Technol 51(5):69Ð87 Taylor WB (1972) Landlord and peasant in Colonial Oaxaca. Stanford University Press, California Therrell MD, Stahle DW, Acuña-Soto R (2004) Aztec drought and ‘the curse of one rabbit’. Bull Am Meteorol Soc 85(9):1263Ð1272 Tompkins EL, Adger WN (2004) Does adaptive management of natural resources enhance resilience to climate change? Ecol Soc 9(2) 10 (online url: http://www.ecologyandsociety.org/vol9/iss2/art10) Tutino J (1986) From insurrection to revolution in Mexico. Princeton University Press. New Jersey Wallen C (1955) Some characteristics of precipitation in Mexico. Geofisika Annaler 37:51Ð85 Wigley TML (1985) Impact of extreme events. Nature 316:106Ð107 Yates PL (1981) Mexico’s agricultural dilemma. University of Arizona Press, Tuscon

Springer