Abstract

Narratives of environmental change can overwhelm analysis and interpretation. "Deforestation," for example, usually communicates a unilinear process of agricultural expansion and biomass mining. Using participant observation, aerial photographs, and forest inventories, this article analyzes the social context of forest change in two neighboring indigenous communities in the Lake Patzcuaro Basin, central western Mexico. Deforestation statistics and drastic deficits between fuel wood demand and growth rates suggest severe environmental degradation in this region and in these communities, but case study results show an increase in tree cover and unexpected changes to forest composition. These are related to agricultural abandonment and the ecological effect of selective woodcutting.

Keywords: Mexico, Lake Pátzcuaro, deforestation, fuelwood, common property, remote sensing.

Introduction

Compelling narratives of environmental change can overwhelm analysis, inserting an opaque lens between complex socio-environmental processes and those who observe them. "Deforestation" for example, implies a unilinear and, at least in recent times in developing countries, a unidirectional process of agricultural expansion and biomass mining.

The human imprint on landscape, however, is not unilinear and merely degradational, but rather conflicting and diffuse. Writing on Middle Eastern landscapes that appear used up after millenia of human use, for example, Blumler (1998:218) asserts that "the popular perception of unremitting decline and degradation of natural resources, although not entirely invalid, is overdrawn, and for the most part based on (mis)perception rather than evidence." On the slopes of Mt. Everest in Nepal, Brower and Dennis (1998:186) challenge the assumption that forests there are retreating under pressure from grazing and woodcutting, and warn against the use of "hasty and unsupported intuition as a guide to resource management." They call for a suspension of presumptions in environmental research.

Fairhead and Leach (1998) argue that a presumption of deforestation has shaped landscape interpretation and forest policy throughout West Africa. They describe how colonial foresters drew on theories of climax and equilibrium to interpret forest mosaics and isolated tree-covered areas as remnants of a previously more extensive forest, stubborn survivors of human onslaught. By the 1960s, aerial photographs were available for temporal comparisons, but a deforestation orthodoxy (see Fairhead and Leach 1998: 192) that expected agricultural frontiers to expand and fuelwood demand to outstrip supply was by then so entrenched that these data were not widely consulted. "Historical data were deemed unim- [end p. 47] portant because the underlying assumption of recent, rapid one-way deforestation was so strong and so institutionalized as to render precision unnecessary" (Fairhead and Leach 1998:174). In an analysis of six countries, they argue that the rate and extent of deforestation have been massively exaggerated by governments and forest departments justifying their continued power. The deforestation orthodoxy also obscured the social processes involved in vegetation change.

The observations of these researchers suggest that detailed case studies of recent environmental change and accompanying social processes could uncover results hidden by an overwhelming narrative of unilinear deforestation and merely degradational change. Case studies should help to answer the following questions: In Mexico, does a deforestation orthodoxy misrepresent a more complex process of environmental change and obscure the social processes involved?

Deforestation in Mexico and the Lake Pátzcuaro Basin

Much data support the deforestation orthodoxy in Mexico. Recent national deforestation estimates range from 370,000 to 720,000 ha/year -- an annual rate of 0.8 to 2 percent, with higher rates in the lowlands and slightly less deforestation in the mountains. Immediate causes include agriculture, cattle, logging, woodcutting, and forest fires. Estimates of firewood collection account for nearly five times the authorized cut, contraband logging removes a volume comparable to the authorized cut, and forest fires affect between 90,000 and 500,000 hectares of forests per year (Cairns et al. 1995; World Bank 1995; Masera 1996). These deforestation statistics, however, prevent an image of environmental change that conflates the effects of woodcutting and agricultural expansion. They also imply a narrative of deforestation that excludes the effects of agricultural crisis in marginal areas and precludes the consideration of agricultural abandonment.

A Case Study of Forest Change

One of the highland Mexican regions where deforestation is thought to be particularly intense is the Lake Patzcuaro Basin, an area of intense human impact on vegetation (Caballero et al. 1992; Toledo et al. 1992; Alvarez-Icaza et al. 1993). Available forest inventory data for this micro-region suggest 45 percent deforestation over 30 years (Alvarez-Icaza and Garibay 1992), and comparisons of forest growth and wood demand indicate intense biomass mining (Becerra Luna et al. 1997). But perhaps a deforestation orthodoxy overstates the situation and erases the complexity of forest change. How do woodcutting and agriculture affect forest cover and quality in this region? In particular, does agricultural crisis in this marginal area affect forest cover?

I explore these questions through a paired case study of Santa Fé de la Laguna and San Jerónimo Purenchecuaro, two neighboring indigenous communities on the shores of Lake Patzcuaro. These communities were chosen for further study for several reasons. First, there are much data available. These include forest inventories conducted by professional forester Rafael Sánchez Concha and biomass demand estimates based on kitchen and pottery kiln measurements conducted by Jaime Navia, Omar Masera, and other members of the Grupo Interdisciplinario de Tecnología Rural Apropiada (GIRA), a local nongovernmental organization¹ (Masera et al. 1997a, 1997b; Masera et al. 1998; Sánchez 1998a, 1998b). In addition to these sources, I compiled data on population and economic activities originally gathered by personnel of local health clinics, examined historical data from theses and studies conducted in the region by geographers and anthropologists, and examined aerial photographs of the region from 1960, 1974, and 1990. Most importantly, over the course of two years I enriched these existing data sources with several months of participant observation with woodcutters and interviews with potters, farmers, village authorities, and elders.

Together, Santa Fé and San Jerónimo are home to 7,000 inhabitants, most of whom speak [end p. 48] Purépecha in addition to Spanish. Their common property territories are more than 40 percent covered by pine and oak forests, heavily fragmented by grazing and agricultural lands (Table 1). Most of the forests in the Lake Patzcuaro Basin are found in the common property territories of communities like these, where altitudinal gradients range from just under 2,000 meters above sea level at the level of Lake Patzcuaro to one of the 10 forest-covered peaks of 2,500 or more meters above sea level (Figure 1).

Like other communities in the basin, people in the case-study communities cobble together a rural livelihood based on off-farm labor, some farming, and forest-based activities. However, while the residents of Santa Fé are nearly all potters or merchants of wood-fired pottery, residents of San Jerónimo lack a forest-dependent craft. Instead, they rely on wages earned outside of the community, remittances, and proceeds from weaving straw figures for the tourist trade. This contrasting degree of forest [end p. 49]  dependence provides a natural experiment in the social causes of forest change, and it also influenced the decision to select these communities for further study (Table 2).

Fuelwood Deficits and Forest Impacts

One justification for expectations of intense deforestation in the Lake Patzcuaro region comes from comparisons of the apparent deficit between fuelwood demand and forest growth. In a study sponsored by the Food and Agricultural Organization, Becerra et al. (1997) compared annual demand estimates with annual growth increments to conclude that demand barely met supply for pine and far outstripped the supply of oak fuelwood in the Lake Patzcuaro Basin (Figure 2). Such gaps between supply and demand long have been thought to result in biomass mining and to cause deforestation, desertification, and downstream flooding (Eckholm 1975).

GIRA's work on fuel demand and Sánchez' forest inventory work allow a comparison of demand and supply specific to Santa Fé and San Jerónimo, following the methodology of Becerra et al. (1997).² In Santa Fé, 99 percent of households cook with wood, mainly oak. In San Jerónimo, where households have more access to cash incomes from work performed outside of the community, 80 percent of households use oak for cooking, often supplemented with propane (Masera et al. 1997a, 1997b; Masera et al. in press). Pine firewood is a critical input for pottery production in Santa Fé. There are 600 kilns there, with 450 firing ceramics twice a month. Village-wide inferred demand for pine is conservatively estimated at 2,000 m³ per year (Navia and Ochoa 1998). These communities import very little wood and they do not sell firewood outside of the community. In Santa Fé, current demand for pine appears to be less than growth, but demand for oak outstrips estimates of sustainable supply in both communities (Figure 3).

Available evidence does not correspond to these expectations. Woodcutters report that oak cooking fuel remains readily available in both communities. In Santa Fé, however, woodcutters no longer find sufficient pine in their own territory and increasingly cross into the forests of San Jerónimo Purenchecuaro to find the fuel they prefer for firing their ceramics. Despite demand/growth estimates that indicate mining of oak and sustainable harvests of pine, oak is abundant, but pine is scarce.

Critics of these kinds of demand/growth comparisons provide the tools necessary to explain the discrepancy between expectations and case study findings (Agarwal 1986; Leach and Mearns 1988; Arnold 1992; Foley et al. 1997). Their approach uses inappropriate data for forest fuel stocks and growth rates. Supply estimates, for example, come from timber-focused methods, which look only at forests, not hedgerows, pastures" or agricultural fields where people gather much of their fuel wood. Nor do timber-oriented methods measure the standing volume of fuel wood, because they ignore deadwood, branches, bushes, and small-diameter trees. In highland Mexico, foresters' growth rates for pine come from measuring the spacing of growth rings, but oaks do not produce reliable rings at this latitude. Becerra et al. (1997) used an arbitrary surrogate of 2.5 percent, which follows the practice of professional foresters managing forests for the production of pine. There is little evidence to support this parameter, however.

The approach also ignores the many ways rural people gather fuel without significantly decreasing living forest biomass, such as cutting branches and gathering dead wood. In Santa Fé, roughly half of the wood I observed drying in front of houses consisted of small, dead branches and parts of burned or dead trees. During Masera et al.'s (1997a, 1997b) surveys, 45 percent of woodcutters in Santa Fé and 22 percent in San Jerónimo reported principally gathering dead wood from the ground, including from pastures and abandoned fields outside of the tree-covered forested areas. Furthermore, many local species of oak resprout after cutting, and survive continuous lopping for long periods of time. In the forests of Santa Fé and San Jerónimo, many oaks show evidence of lopping, resprouting, and continued vigorous growth.

Finally, demand/supply comparisons such as Becerra et al.'s (1997) give no consideration to forest ecology, the regeneration requirements of tree species, or how human disturbance affects regeneration. Such comparisons are ignorant of the lessons for conservation coming from current biogeographical research on disturbance regimes (Zimmerer and Young 1998).

Because of the disturbance ecology of pine/oak forests, current patterns of woodcutting shift forest composition towards oak dominance. Woodcutters report a progressive scarcity of pine in the forests of Santa Fé, and cross-sectional data from the forest inventories reinforce their observation. Both San Jerónimo and Santa Fé have similar average stand densities, but there is 250 percent more pine per hectare in San Jerónimo than in Santa Fé (Figure 4).

Forests in both communities are adjacent, cover a similar altitudinal gradient, grow on similar soils, and have similar expositions. Available data cannot preclude different histories of forest clearing, fires, or disease, but the most obvious difference between these forests is human [end p. 51] use; Santa Fé woodcutters more often cut pine. In addition, until the early 1980s, woodcutters from Quiroga, a town immediately to the east, routinely entered Santa Fe's territory to cut pine for local carpentry workshops (Gortaire Iturralde 1971; Dimas Huacuz 1982; Zarate Hernandez 1993).

The effect of this kind of selective cutting of pine in Mexican pine and oak forests is well known. It speeds succession to oak dominance because pine requires more intense disturbances, such as fire. Pine does not regenerate well on the forest floor, so even rates of extraction below growth rates can represent a kind of low-level disturbance that gradually removes the species without creating the conditions for its regeneration (Snook and Negreros 1986; Styles 1993).

Remote Sensing of Forest Change

A second line of support for a presumption of drastic environmental degradation comes from official deforestation statistics. Comparisons of state surveys of commercial forestry resources in the Lake Patzcuaro Basin indicated a 45 percent decrease overall, complete deforestation in the nearby pottery-making county of Tzintzuntzan, and the loss of 77 percent of forests in the county of Quiroga (Alvarez-Icaza and Garibay 1992).

Santa Fé and San Jerónimo together comprise 38 percent of the surface area of Quiroga county. They are well covered by aerial photographs, which enables a comparison of forest cover over time. The Michoacán state forest agency owns black and white photographs at the scale of 1:20,000 from a 1960 forest-mapping flight, and copies were made of these media on a high-resolution laser photocopier. Mexico's national mapping and census agency provided contact prints of 1:50,000 photographs from a 1974 mapping flight and 1:25,000 photographs from a 1990 forest inventory flight. Scanned and rectified, these media show the disappearance not of forests, but of abandoned agricultural fields swallowed up beneath vigorous stands of pine and oak (Figure 5). In a 1,000-hectare area compared between 1960 and 1990, 60 percent of the area remained in forest, 8 percent was deforested, and 20 percent of the area changed from field to forest (Klooster 1999).⁴

Field observations provide further evidence of agricultural abandonment. Crumbling stone fences often intersect eroded areas now used for rainy-season grazing. In areas of large oak and pine trees, stone fences frequently attest to the agricultural history of tree-covered areas. Furthermore, fields that were clear of trees in the 1960 and 1974 photographs are now often covered with vigorous stands of pine and oak trees with diameters of 20 to 30 em at breast height (Figure 5). New clearings are exceedingly rare, and government-assisted tree planting was successful in some areas, especially near the towns. Informants confirm findings from remote sensing: forests are regenerating on abandoned agricultural lands.

The Social Context of Forest Change

Rural depopulation does not explain agricultural abandonment. Santa Fé's population has increased since 1950, while San Jerónimo's resident population has held steady (Table 3). In the Lake Patzcuaro Basin generally, population growth has been above two percent per year for several decades (Castilleja 1992).

Explanations for agricultural abandonment are better sought in changes to the regional and national economy. The railroad reached the Patzcuaro area in 1890 and an all-weather road connecting Mexico City and Guadalajara passed through Santa Fe and San Jerónimo in 1930 (Brand 1951). Increased integration with the rest of Mexico and North America meant a greater availability of maize and beans imported from more productive regions of Mexico and the United States, where large farms benefit from subsidies for irrigation projects, green revolution research into hybrid varieties, and chemical [end p. 53] inputs. Furthermore, programs to alleviate poverty in Mexico impose price ceilings on consumer prices for maize and beans, which discourage subsistence production and dry up markets for local producers.

As regional integration eliminates the economic viability of local agriculture, it also makes urban and international labor markets more accessible and expands the market for local crafts. In Santa Fe, pottery making is the primary activity, while in San Jerónimo temporary emigration, remittances, and wage labor outside the village are the most frequent activities.

As agriculture has become decreasingly viable, pottery and migration have increased in importance. In both communities, agriculture and livestock play decreasingly important roles in household livelihood portfolios. Currently, farming is the principal activity of less than four percent of workers surveyed in these communities (see Table 2). Interviews with village elders indicate a progressive decrease in the importance of agriculture. Middle-aged men frequently work only a fraction of the lands their fathers worked, with the rest abandoned.

Pottery production has long been the principal economic activity in Santa Fé. During field visits in 1946, West (1947) estimated that 80 percent of Santa Fé's workforce engaged in pottery production, which by 1900 had displaced previous woodworking activities in the village. In 1970, 80-90 percent of households included pottery production in their economic activities (Gortaire Iturralde 1971), and this has held steady or increased slightly since then. Currently, the production of wood-fired pottery occupies 89 percent of Santa Fé workers (Durston 1992; Masera et al. 1997a; Navia and Ochoa 1998). While the proportion of households engaged in pottery production has held steady, the importance of pottery production within the household has increased. Santa Fé informants explain that pottery production yields money returns more quickly than agriculture, requires less initial investment, runs less risk of failure due to frost, hail, pest, or drought, and provides the income needed to buy food.

Villagers in San Jerónimo associate the beginning of land abandonment with la norteada, a local name for the United States bracero program that brought temporary laborers to work in the United States agricultural sector. This program, which began during World War II and ended in 1964, forever changed local perceptions of labor opportunities and established patterns of emigration that continue to the present (Foster 1988; ORCA 1997). Housing provides one indication of the cyclical character of the migration process in San Jerónimo; of 866 houses there, only 466 are inhabited. Although some are abandoned, many are in a permanent state of construction by residents planning to return someday.⁵ Villagers from both Santa Fé and San Jerónimo participated in the bracero program, but because of work opportunities in pottery, emigration never took root in Santa Fé as it did in San Jerónimo.

Residents of San Jerónimo also weave figures from wheat straw for the tourist trade, and this occupies 46 percent of San Jerónimo workers. A village priest promoted this activity sometime after 1946, since West (1947) makes no mention of straw figures in San Jerónimo; he identified weaving reed mats called petates as the principal craft activity. San Jerónimo households engaged in craft production also depend on remittances of money, clothing, and goods from relatives working in the United States.

Discussion

Copious statistics on deforestation and forest mining reinforce a presumption of human impact on Lake Patzcauro forests that is unilinear and utterly degradational. However, despite comparisons of wood demand and forest growth that suggest severe mining of oak, forest composition comparisons and woodcutter interviews indicate that pine, not oak, becomes scarcer under uncoordinated woodcutting pressures. Conventional demand growth comparisons reinforce a deforestation orthodoxy while occluding the complexity of forest change under actual cutting pressures.

Similarly, deforestation statistics indicate a 77 percent loss of forest cover in the county [end p. 54] containing these two communities, but aerial photographs reveal agricultural abandonment and an increase in tree cover in both communities in the case study. The apparent contradiction between the official data on deforestation from state forest inventories and the data from aerial photographs and field observations in the case study might stem from differences in the definition of deforestation. Information on methods and sampling intensity used in the state inventories of 1960 and 1990 is lacking, but it is likely that photo interpreters used a very restrictive definition of forest and only interpreted dense stands of commercial-quality trees as "forest."⁶ In this scenario, the case study findings of increased tree cover do not contradict the official "deforestation" statistics; increased tree cover might have accompanied a decreased area of commercial forest. Even in this case, however, this study's findings still provide an important check on the common interpretation of these statistics, which understands "deforestation" as a decrease in tree cover and does not consider the vegetation dynamics accompanying agricultural abandonment and forest mining. Methods of interpretation and forest definition tended to reinforce a deforestation orthodoxy that precluded consideration of other kinds of forest change. The deforestation orthodoxy also precludes consideration of social processes such as agricultural abandonment.

The apparent contradiction between the case study data on agricultural abandonment and forest inventory data indicating drastic deforestation in the Lake Patzcuaro region might also disappear if San Jerónimo and Santa Fé are exceptional. This does not appear to be the case. The social context of forest change observed in Santa Fé and San Jerónimo has three main aspects: agricultural abandonment; emigration with stable or growing rural populations; and intense use of the forest for home and small industries. These factors are all widely present in the Lake Patzcuaro Basin and elsewhere in highland Michoacán.

First, although agriculture maintains a place in livelihood strategies both for its cultural importance and for the strategic stores of food it produces, it is declining regionally. The area planted in maize in the Lake Patzcuaro Basin and the Meseta Purépecha, an adjacent socially and ecologically similar region, decreased by more than half between 1969 and 1993. Although some of these maize fields were certainly converted to avocado orchards in areas where climatic conditions favor the tree, abandoned agricultural areas are common (Carabias et al. 1994; Chapela 1994; Garibay Orozco 1996). Second, despite patterns of emigration, substantial rural populations remain. Inferred emigration rates are near 17 percent in some counties, but population growth regionally approaches three percent per year and only isolated localities actually have declining populations (Castilleja 1992; Garibay Orozco 1996). Third, the region is rich in medium and small forest-based industries. The Lake Patzcuaro Basin and the Meseta Purépecha hold at least 1000 band-saw workshops, sawmills, and wood chippers for cellulose and more than 9,000 small industries reliant on unpaid family labor and wood inputs acquired directly or in an informal, often clandestine, economy. In addition to pottery production, as observed in the case study, these small industries include carpentry workshops, bread and tortilla bakeries, and small brick making factories (Reyes 1992; Masera et al.1998).

Conclusion

This paper explored forest change in two communities on the shores of Lake Patzcuaro in highland central-western Mexico. In one of the communities, cyclical migration to United States and Mexican cities is a dominant household survival strategy. The other community specializes in pottery production requiring large amounts of pine firewood. Regional and community-specific comparisons of fuel demand and forest growth suggest severe over-cutting of oak, with much less of a deficit for pine. In the pottery community, however, forest inventories, woodcutter interviews, and participant observation find no problem with the supply of oak firewood, but suggest instead that woodcutting pressures leads[end p. 55] to oak dominance, forcing woodcutters to seek pine in a neighbor's territory. Similarly, deforestation statistics based on forest inventories indicate severe deforestation, but digitized, rectified aerial photographs from 1960, 1974, and 1990, reveal increased forest cover in both communities. This is the result of a progressive decrease in the importance of agriculture in favor of intensified pottery production and migration.

In the Lake Patzcuaro Basin, a deforestation orthodoxy creates an image of a downward trajectory of deforestation and degradation due to woodcutting and agricultural expansion. The situation in the case study communities, however, is one of complex changes, in which woodcutting has unexpected results on forest ecology and agricultural abandonment leads to increasing tree cover.

While seeming to be straightforward, unbiased, and unambiguous, demand/growth comparisons and a facile interpretation of deforestation statistics misrepresent the relationship between people and the forest. They are methods of assessing human impacts on forests that have favored an exaggerated representation of forest loss; they serve to reinforce the deforestation orthodoxy (Fairhead and Leach 1998). More importantly, they fail to suggest questions about the social factors governing woodcutting and agricultural abandonment.

The research presented here by no means provides the final word on forest change in highland Mexico or in the Lake Patzcuaro Basin. More finely grained studies assessing forest composition changes could use longitudinal data from aerial photographs to compare stands of trees in specific locales. Satellite remote sensing could also provide a valuable smaller scale (regional) view of change. Agricultural and population census data might indicate the role of agricultural abandonment and emigration more generally. In conjunction, such studies would indicate the degree to which findings from San Jerónimo and Santa Fé are representative of the Lake Patzcuaro Basin, the neighboring Meseta Purépecha, and other indigenous highland areas of central and southern Mexico. More importantly, however, they might provide a check on the presumption that an unchecked, unilinear, and homogeneous process of deforestation is sweeping the Mexican countryside. A richer and more subtle narrative of environmental change should lead to a wider array of conceivable and appropriate policy responses.

Notes

1. GIRA's presence in these communities also afforded me the opportunity to connect academic research to on-going action-research and development initiatives. This NGO also was instrumental in helping Santa Fé and San Jerónimo solicit and partly match the federal funds needed to pay for forest inventories.

2. See Masera et al. (in press) for nuanced arguments about the cultural and economic aspects of fuelwood demand in rural Mexico.

3. This estimate is based on GIRA's measurements in 15 pottery workshops (Navia and Ochoa 1998). Their estimates calculated on the basis on interviews about consumption are much higher: 3,200 m³/year, and 5,500 m³/year (Masera et al. 1997a). In 1969, an anthropologist counted mule loads of wood to estimate an annual fuel demand of 3,600 m³ or 6,000 m³ (Gortaire Iturralde 1971:113,124).

4. Qualitative comparisons of other areas in these communities covered by aerial photographs also reveal greater forest cover in 1990 than in 1974 or 1960. In addition, Dr. Helen Pollard kindly provided copies of aerial photographs from a 1942 United States Air Force mapping flight, which partially cover the forests of San Jerónimo. These photographs reveal a landscape with more distinct field margins, less forest cover, and greater forest fragmentation than at present.

5. See Kearney (1996) for a discussion of emigration in the maintenance of Mexican peasant-like villages like this one.

6. Claudio Garibay was kind enough to discuss these data with me and share a copy of the 1960 state-wide inventory in his possession. Unfortunately, the later forest inventories, which he consulted to estimate deforestation rates for the Lake Patzcuaro region (Alvarez-Icaza and Garibay 1992), are now scattered about privatized forestry firms and were not available for review.

Acknowledgements

For financial and institutional support during fieldwork I am grateful to William and Jane Fortune, the Princeton Environmental Institute, and the Science Technology and Environmental Policy program of Princeton University. In the field, personnel of GIRA, the communal authorities of Santa Fé and San Jerónimo, and especially the woodcutters polite enough to allow me to observe their activities, were crucial for data acquisition and analysis. Forester Rafael Sánchez Concha adjusted his forest survey methods to incorporate additional data. Dr. Guillermina Linares and Dr. Soyla Santa Cruz Chavez kindly provided clinic census data.

References

Agarwal, Bin. (1986) Cold Hearths and Barren Slopes: The Woodfuel Crisis in the Third World. New Delhi: Allied Publishers.

Alvarez-Icaza, Pedro, Gabriela Cervera, Claudio Garibay, Pedro Gutiérrez, and Fernando Rosete (1993) Los Umbrales del Deterioro: La Dimensión Ambiental de un Desarrollo Desigual en la Región Purépecha. Mexico City: PAIR-UNAM and Fundación Friedrich Ebert.

Alvarez-Icaza, Pedro and Claudio Garibay. (1992) Producción Agropecuaria y Forestal. In Plan Patzcuaro 2000: Investigación Multidisciplinaria para el Desarrollo Sostenido. V. M. Toledo, P. Alvarez-Icaza, and P. Avila (eds.), pp. 91-134. Mexico City: Fundación Friedrich Ebert.

Arnold, J.E.M. (1992) Community Forestry: Ten Years in Review, rev. ed. Community Forestry Note, 7, Rev. 1. Rome: Food and Agriculture Organization of the United Nations.

Becerra Luna, Francisco, Diego D. Reygadas Prado, and Francisco Moreno Sánchez. (1997) Evaluación de Recursos Forestales para Leña Combustible en la Micro Región Lago de Patzcuaro, Mich. Patzcuaro, México: Grupo Interdisciplinario de Tecnología Rural Apropiada (GIRA) AC. Proyecto FAO.

Blumler, Mark A. (1998) Biogeography of Land-Use Impacts in the Near East. In Nature's Geography: New Lessons for Conservation in Developing Countries, Karl S. Zimmerer, and Kenneth R. Young (eds.), pp. 215-236. Madison: University of Wisconsin Press.

Brand, Donald. (1951) Quiroga: A Mexican Municipio. Washington, DC: Smithsonian Institution, Institute of Social Anthropology Publication No. 11.

Brower, Barbara, and Ann Dennis. (1998) Grazing the Forest, Shaping the Landscape? Continuing the Debate About Forest Dynamics in Sagarmatha National Park, Nepal. In Nature's Geography: New Lessons for Conservation in Developing Countries, Karl S. Zimmerer, and Kenneth R. Young (eds.), pp. 184-208. Madison: University of Wisconsin Press.

Caballero, Javier, Narciso Barrera, and Cristina Mapes. (1992) La Vegetación Terrestre. In Plan Patzcuaro 2000: Investigación Multidisciplinaria Para El Desarrollo Sostenido, V. M. Toledo, P. Alvarez-Icaza, and P. Avila (eds.), pp. 71-90. Mexico City: Fundación Friedrich Ebert.

Cairns, Michael A., Rodolfo Dirzo, and Frank Zadroga. (1995) Forests of Mexico: a Diminishing Resource. Journal of Forestry 93(7):21-24.

Carabias, Julia, Enrique Provencio, Carlos Toledo, and Pedro Alvarez-Icaza. (1994) Manejo de Recursos Naturales y Pobreza Rural. Mexico City: Universidad Nacional Autónoma de México and Fondo de Cultura Económica.

Castilleja, Aida (1992) Población. In Plan Patzcuaro 2000: Investigación Multidisciplinaria para el Desarrollo Sostenido, V. M. Toledo, P. Alvarez-Icaza, and P. Avila (eds.), pp. 239-272. Mexico City: Fundación Friedrich Ebert.

Chapela, Gonzalo. (1994) Aprovechamiento de los Recursos Forestales en la Sierra Purépecha. Mexico City: Universidad Autónoma Metropolitana, Xochimilco.

Dimas Huacúz, Nestor. (1982) Forma y Composición de la Tenencia de la Tierra en Santa Fé de La Laguna. Mexico City: Secretaría de Educación Pública and Instituto Nacional Indigenista.

Dirección General de Estadística (1973) IX Censo General de Población, 28 de Enero de 1970. Localidades Por Entidad Federativa y Municipio con Algunas Características de su Población y Vivienda. Volumen II, Hidalgo a Oaxaca. Mexico City: Dirección General de Estadística.

Dirección General de Estadística (1963) VIII Censo General de Población, 1960. Estado de Michoacán. Mexico City: Dirección General de Estadística.

Dirección General de Estadística (1952) Integración Territorial de Los Estados Unidos Mexicanos. Séptimo Censo General de Población, 1950. Estado de Michoacán. Mexico City: Dirección General de Estadística.

Durston, John. (1992) Organización Social de Los Mercados Campesinos en El Centro de Michoacán. Mexico City: Instituto Nacional Indigenista.

Eckholm, Erik P. (1975) The Other Energy Crisis: Firewood. Worldwatch Paper, 1. Washington DC: Worldwatch Institute.

Fairhead, James, and Melissa Leach. (1998) Reframing Deforestation. Global Analyses and Local Realities: Studies in West Africa. New York: Routledge.

Foley, Gerald, Willem Floor, Gerard Madon, Elhadji Mahamane Lawali, Pierre Montague, and Kiri Tounao. (1997) The Niger Household Energy Project: Promoting Fuelwood Markets and Village Management of Natural Woodlands. World Bank Technical Paper No. 362. Washington DC: World Bank.

Foster, George M. (1988) Tzintzuntzan: Mexican Peasants in a Changing World. Prospect Heights: Waveland Press.

Garibay Orozco, Claudio. (1996) El Dilema de Los Comunes: Un Estudio de Crisis Multiple y La Disputa Política en La Región Purépecha. M.A. Thesis, El Colegio de Michoacán.

Gortaire lturralde, Alfonso. (1971) Santa Fé: Presencia de un Pueblo Hospital. México: Universidad Iberoamericana.

Instituto Nacional de Estadística, Geografía e Informática (1996) Conteo de Población y Vivienda 1995. Michoacán Tomo II. Resultados Definitivos, Tabulados Básicos. Aguascalientes: Instituto Nacional de Estadística, Geografía e Informática.

Instituto Nacional de Estadística, Geografía e Informática (1991) Resultados Definitivos. Datos Por Localidad (Integración Territorial). XI Censo General de Población y Vivienda, 1990. Aguascalientes: Instituto Nacional de Estadística, Geografía e Informática.

Instituto Nacional de Estadística, Geografía e Informática (1987) X Censo General de Población y Vivienda, 1980. Integración Territorial del Estado de Michoacán. Mexico City: Instituto Nacional de Estadística, Geografía, e Informática.

Kearney, Michael. (1996) Reconceptualizing the Peasantry: Anthropology in Global Perspective. Boulder: Westview Press.

Klooster, Daniel. (1999) Environmental Orthodoxy, Environmental History, and Environmental Future: Forest Change Since 1960 in a Pair of Indigenous Communities in Highland Michoacán, Mexico. Honolulu: Paper Presented at the Association of American Geographers 95th Annual Meeting.

Leach, Gerald, and Robin Mearns. (1988) Beyond the Woodfuel Crisis: People, Land and Trees in Africa. London: Earthscan.

Masera, Omar Raúl. (1996) Desforestación y Degradación Forestal en México. Documentos de Trabajo, No. 19. Patzcuaro, Michoacán: Grupo Interdisciplinario de Tecnología Rural Apropiada (GlRA), unpublished manuscript.

Masera, Omar Raúl. (1994) Socioeconomic and Environmental Implications of Fuelwood Use Dynamics and Fuel Switching in Rural Mexico. Ph.D. Thesis. University of California at Berkeley.

Masera, Omar, Jaime Navia, Gerardo Ruiz, Juan Carlos Cedeno, Salvador Ochoa, José, Luis Fabián, Deynira Bautista, and Isabel Alejandre. (1997a) Documento Comunitario: Uso de la Leña en Santa Fé de La Laguna, Michoacán. Patzcuaro, México: Grupo Interdisciplinario de Tecnología Rural Apropiada (GlRA), unpublished manuscript.

Masera, Omar, Jaime Navia, Gerardo Ruiz, Juan Carlos Cedeno, Salvador Ochoa, Sandra Morales, Elida Picho, and Rubén Morales. (1997b) Documento Comunitario: Uso de la Leña en San Jerónimo Purenchécuaro, Michoacán. Patzcuaro, México: Grupo Interdisciplinario de Tecnología Rural Apropiada (GlRA), unpublished manuscript.

Masera, Omar Raúl, Diego Masera, and Jaime Navia. (1998) Dinámica y Uso de los Recursos Forestales en la Región Purépecha: El Papel de las Pequeñas Empresas Artesenales. Patzcuaro, Michoacán: Grupo Interdisciplinario de Tecnología Rural Apropiada (GIRA).

Masera, Omar Raúl, Barbara D. Saatkamp, and Daniel M. Kammen. (In press) From Linear Fuel Switching to Multiple Cooking Strategies: A Critique and Alternative to the Energy Ladder Model for Rural Households. World Development.

Navia, Jaime, and Salvador Ochoa. (1998) Las Pequeñas Empresas Artesanales en la Región Purépecha: El Caso de las Alfarerias, Documento de Trabajo No. 24. Patzcuaro: Grupo Interdisciplinario de Tecnología Rural Apropiada (GIRA), unpublished manuscript.

ORCA. (1997) Estudio Socio-Ambiental de San Jerónimo Purenchécuaro. Organización Ribereña Contra la Contaminación del Lago de Patzcuaro, AC. unpublished manuscript.

Reyes, Javier. (1992) Producción Artesanal. In Plan Patzcuaro 2000: lnvestigación Multidisciplinaria para el Desarrollo Sostenido, V.M Toledo, Pedro Alvarez-Icaza, and Patricia Avila (eds.), pp. 159-180. Mexico City: Fundación Friedrich Ebert.

Sánchez, Rafael. (1998a) Programa de Manejo Forestal de Tipo Especial para la 'Comunidad lndîgena Santa Fé de La Laguna' Ubicado en el Municipio de Quiroga, Mich. Mimeo. Prepared for the Comunidad Indígena Santa Fé de la Laguna and the Grupo Interdisciplinario de Tecnología Rural Apropiada (GIRA) AC.

Sánchez, Rafael. (1998b) Programa de Manejo Forestal de Tipo Especial para la Comunidad lndígena San Jerónimo Purenchecuaro Ubicado en el Municipio de Quiroga, Mich. Mimeo. Prepared for the Comunidad Indígena San Jerónimo Purenchecuaro and the Grupo Interdisciplinario de Tecnología Rural Apropiada (GIRA) AC.

Snook, Laura C. and Patricia C. Negreros. (1986) Effects of Mexico's Selective Cutting System on Pine Regeneration and Growth in a Mixed Pine-Oak (Pinus Quercus) Forest. Current Topics in Forest Research: Emphasis on Contributions by Women Scientists. Proceedings of a National Symposium. General Technical Reports, No. SE-46. Southeastern Forest Experiment Station: US Forest Service.

Styles, Brian T. (1993) Genus Pinus: A Mexican Purview. In Biological Diversity of Mexico: Origins and Distribution, T. P. Ramamoorthy, Robert Bye, Antonio Lot, and John Fa (eds.), pp. 397-420. New York: Oxford University Press.

Toledo, Victor Manuel, Pedro Alvarez-Icaza, and Patricia Avila. (eds.) (1992) Plan Patzcuaro 2000: lnvestigación Multidisciplinaria para el Desarrollo Sostenido. Mexico City: Fundación Friedrich Ebert.

West, Robert C. (1947) Cultural Geography of the Modern Tarascan Area. Washington, DC: Smithsonian Institution, Institute of Social Anthropology Publication No. 7.

World Bank. (1995) Mexico Resource Conservation and Forest Sector Review. Report, 13114-ME. Washington, DC: Natural Resources and Rural Poverty Operations Division Country Department II Latin America and the Caribbean Regional Office.

Zarate Hernandez, Eduardo. (1993) Los Señores de La Utopia. Zamora, Michoacán: El Colegio de Michoacán/CIESAS.

Zimmerer, Karl S. and Kenneth R. Young. (1998) Nature's Geography: New Lessons for Conservation in Developing Countries. Madison: University of Wisconsin Press.

Resumen

Estadísticas sobre la desforestación pueden simplificar el análisis y interpretación de cambio ambiental. La "desforestación", por ejemplo, generalmente se interpreta como un proceso de expansión de la frontera agrícola y la sobre-extracción de madera, sin retrocesos en el proceso. Este artículo utiliza la observación participativa, fotografías aéreas, y estudios dasonómicos para analizar el contexto social de cambio forestal en un par de comunidades indígenas en la Cuenca del Lago de Pátzcuaro, México. Las estadísticas de desforestación y un déficit entre la demanda de biomasa y las tazas de regeneración natural implican un proceso severo de degradación ambiental tanto en la región como en las comunidades estudiadas. Sin embargo, los resultados del estudio de caso demuestran un aumento en la superficie forestal y cambios inesperados en la composición forestal. Estos cambios reflejan procesos simultáneos de abandono de terrenos agrícolas y el impacto ecológico de la extracción selectiva de árboles.