Fallow Area and Duration

We looked at three main factors for this analysis: (1) whether or not farmers had land in fallow at the time of the interview, (2) the amount of land in fallow, and (3) the length of time the land has been in fallow. For all three of these analyses, there was no statistical difference between SA Users and SA Non-Users. For our first factor, actual land in fallow, there were, in fact, almost exactly equal numbers for SA Users and SA Non-Users.

Number of SA Users and SA Non-Users Who Had Land in Fallow in 1998* ­ Guatemala and Mexico

* At the time of the survey.

With respect to the amount of land in fallow, farmers in the two groups in Guatemala were almost identical. In Mexico, however, while there is similarly no statistical difference between the two groups, it appears that SA Users tend to have slightly less land in fallow.

Amount of Land in Fallow (ha) for SA Users and SA Non-Users in 1998 ­ Guatemala

Amount of Land in Fallow (ha) for SA Users and SA Non-Users in 1998 ­ Mexico

Concerning the amount of time farmers have left land in fallow, both groups of farmers are nearly identical in both Guatemala and Mexico.

Amount of Time (Years) Land Has Been in Fallow for SA Users and SA Non-Users in 1998 ­ Guatemala

Amount of Time (Years) Land Has Been in Fallow for SA Users and SA Non-Users in 1998 ­ Mexico

Contamination of the Environment

It turns out that relatively few farmers in either the SA User or SA Non-User groups use chemical fertilizer or pesticide. At both sites, farmers reported that they have problems with agricultural pests, but the relationship is the opposite between Guatemala and Mexico. In Guatemala, SA Plots were more likely to have problems with pests than were Non-SA Plots. In Mexico, SA Plots were less likely to have problems with pests than were Non-SA Plots. Both relationships are statistically significant.

Reported Problems With Agricultural Pests in SA Plots and Non-SA Plots, for 1997 ­ Guatemala and Mexico

According to the project teams, some SA Users in Guatemala have problems with insect infestations in plots they do not burn because burning acts to control insect populations. In Mexico, project team members often found that initial use of sustainable agriculture techniques caused an increase in rodent infestations. They also found, however, that SA Users sometimes used a variety of integrated pest management techniques, primarily natural insect repellents made from garlic and onion, which proved to be very effective.

In Guatemala, only 15 farmers reported using pesticides in their plots and virtually none reported using chemical fertilizer. In Mexico, farmers used pesticides more frequently. In both sites, however, pesticide use was lower in SA Plots than in Non-SA Plots. According to the project managers in Mexico, SA Users are less inclined to use pesticides because velvetbean decreases insect infestations. Nevertheless, the number of farmers who use pesticides is so small that this result is not programmatically significant.

Use of Pesticides in SA Plots and Non-SA Plots, for 1997 ­ Mexico

As we saw in a previous analysis, there is a significant difference between SA Users and SA Non-Users in the use of fire in agriculture. Smoke is another form of contamination, and it is clear that sustainable agriculture is effective at reducing the amount of smoke released into the atmosphere from preparation of agricultural lands. In addition, because agricultural fires sometimes escape into nearby forests — especially during particularly dry periods — reduced burning by farmers most likely reduces contamination of the atmosphere from forest fires.

Indirect Impact of Sustainable Agriculture on Biodiversity

The second section of our framework looks at the indirect impacts of sustainable agriculture projects on biodiversity, including attitudes related to biodiversity and conservation and involvement in community-level organizations.

Attitudes Concerning Conservation

According to our quantitative data, significantly more SA Users than SA Non-Users perceive that sustainable agriculture has positive effects. But it is less clear the extent to which SA Users have more positive attitudes about the environment. We asked farmers if sustainable agriculture had any positive effects on water, soil, air, and the forest. The only significant difference between SA Users and SA Non-Users was in Guatemala, for water.

Number of SA Users and SA Non-Users Who Perceive Positive Effects of Sustainable Agriculture ­ Guatemala and Mexico

According to the results from Mexico, 65 (44.8%) of farmers who use sustainable agriculture say that the reason they do not use fire to prepare their agricultural plots is to protect the forest. In Guatemala, 143 (93%) SA Users and 103 (66.7%) SA Non-Users report that they do not use fire in their fields in order to protect the forest (P = 0.035).

During some of the focus group interviews we conducted in both Guatemala and Mexico, SA Users demonstrated a clear awareness of the connection between their actions and the environment. Representatives of both Defensores de la Naturaleza and Línea Biósfera attribute this awareness to the environmental education activities that accompanied their sustainable agriculture extension projects. The following quotes are a sample of perceptions expressed during the focus group sessions:

"At times, it is nice to burn your fields, but the truth is we haven't realized that we, in fact, are the ones responsible for polluting the environment — we're destroying nature. We're the ones lighting the fires, but still we ask ‘Where is all this smoke coming from?'"
     — José, SA User, Mexico

"We could reduce the amount of pollution if people would just understand the causes — if all farmers would quit burning we could then stop the forest fires."
     — Antonio, SA User, Mexico

"Fire is a problem because when we burn our fields, fertility drops and the rains wash away the soil..."
     — Anonymous SA User, Guatemala

"[Sustainable agriculture] is good, because we know that if we were to cut down the entire forest, after awhile, it would stop raining."
     — Juan, SA User, Mexico

Referring to the use of herbicides to prepare plots:

"To unite the community, we must teach everyone to prepare their plots with hoe and machete alone because that way, we will not contaminate the water."
     — Miguel, SA User, Mexico

Participation in Community Organizations

We looked at various ways that farmers might participate in community activities or be exposed to other opportunities that might foretell their involvement in future conservation activities. In addition to looking at membership in community organizations, we also looked at the extent to which farmers are visited by an extensionist. Finally, we looked at the extent to which farmers have participated in cross-community exchanges to learn new agricultural techniques or other technological innovations.

In Guatemala, SA Users were significantly more likely to report belonging to a community organization than SA Non-Users. In Mexico, however, there was no significant difference. According to the project teams, these results reflect the relatively low level of official, community organization in Guatemala and the high level of community mobilization in Mexico. In Mexico, most farmers belong to ejido organizations that are part of greater regional organizations, so the need for community organizations per se is minimal. In Guatemala, on the other hand, communities are often required to self-mobilize in order to take any collective action.

Number of SA Users and SA Non-Users Who Participate in a Community Organization ­ Guatemala and Mexico

In both Guatemala and Mexico, SA Users are significantly more likely to have been visited by an extensionist than SA Non-Users. This makes sense because the programs of both Defensores de la Naturaleza and Línea Biósfera rely heavily on extension programs to train farmers. In addition, these results are clearly biased by the fact that these organizations conducted the data collection and necessarily sampled SA Users with whom they had interacted to include in the study. We also looked at the extent to which community members participated in cross-community exchanges or visits, but very few farmers were involved in these.

Number of SA Users and SA Non-Users Visited by an Extensionist ­ Guatemala and Mexico

Putting the Findings in Perspective

We divide our discussion in this section into three parts that parallel our three goals: (1) the conditions under which sustainable agriculture contributes to conservation, (2) principles for using sustainable agriculture as a conservation tool, and (3) process lessons — learning to learn better. In the first two sections, we try to make sense of our findings by looking across factors and our two study sites. In the last section, we take a critical, internal look at how we planned and implemented this study.

Many of the insights and interpretations we include in this section came about from the final working meeting we held in August 2000. During that meeting we discussed each site individually and then compared and contrasted the two sites in an effort to determine how different characteristics might have contributed to our findings. The last day of the meeting was spent evaluating the research process itself.

The Conditions Under Which Sustainable Agriculture Contributes to Conservation

The first goal of this study was to better understand the conditions under which sustainable agriculture can be used as an effective strategy to reach conservation goals. For this portion of our discussion, we will examine our findings using the conventional wisdom as a framework. As you go through this section, keep in mind that we necessarily narrowed our sample to a fairly limited set of social, cultural, and environmental characteristics to come up with a relatively precise understanding of what makes for a successful sustainable agriculture program. While we believe that many of these insights and observations are generalizable, we also realize that they are derived from some fairly specific situations.

Area planted to subsistence crops

We found that area planted was associated with sustainable agriculture use in completely opposite ways in Guatemala and Mexico. In Guatemala, SA Users plant significantly more area to maize than SA Non-Users. In Mexico, SA Users plant significantly less area to maize than SA Non-Users. We conclude from these findings that, in Guatemala, use of sustainable agriculture does not reduce expansion of agriculture and, therefore, reduce rates of deforestation. In fact, it appears to have just the opposite effect: It appears that sustainable agriculture as practiced in the Sierra de las Minas increases agricultural expansion and, thus, deforestation. In Mexico, on the other hand, we conclude that sustainable agriculture use does reduce expansion of agriculture, and, therefore, reduce rates of deforestation.

During our analysis, we discovered that SA Users in both countries produced more maize than their SA Non-User counterparts. But this increase in production was a function of two different factors in our two sites. In Guatemala, it was a function of increased area planted because yield was equal between SA Users and SA Non-Users. In Mexico, however, SA Users planted less than SA-Non-Users, but demonstrated yields that were, on average, 1.5 times greater than SA Non-Users. In the case of Mexico, the difference in production is a function of increased yields achieved by SA Users. So, according to our user-level data, area planted is a function of the efficiency with which farmers manage their fields. Our plot-level data confirm this: in Guatemala, yield between SA Plots and Non-SA Plots was virtually the same but in Mexico the yield from SA Plots was significantly higher than that from Non-SA Plots.

But why then are SA Users in Mexico so much more efficient in their use of land than SA Users in Guatemala? Why was the conventional wisdom supported in Mexico, but not in Guatemala? What difference between these two sites would cause this divergence?

As we mentioned in our findings, area planted and yield are also a function of many other factors. But we controlled for many of the potentially determining factors in our sample selection within and between each site. In addition, there was virtually no difference between the SA User and SA Non-User samples within sites and between Guatemala and Mexico in the use of fertilizer and pesticide and access to credit. In fact, almost no farmers had access to these inputs because of their high costs.

When we looked at labor inputs, however, we found a different story. It appears that SA Users in Mexico invest proportionately more labor per hectare than SA Users in Guatemala. In fact, SA Users in Mexico use almost 10 days of labor per hectare more than their Guatemalan counterparts. One of the distinctions between the two study sites that probably accounts for some of this difference is that SA Users in Mexico use integrated pest management in addition to the two techniques also used in Guatemala — velvetbean and minimum tillage. But it is unlikely that this difference accounts for all of the dissimilarity in labor between the two sites.

So, SA Users in Mexico invest more labor in their maize fields, which leads them to achieve higher yields and require less land to feed their families than SA farmers in Guatemala. This relationship in Mexico is strongly supported by our multivariate analysis: Agricultural fields in Mexico with high yields tend to be smaller plots farmed by SA Users. So, SA Users in Mexico and Guatemala use sustainable agriculture to increase production of maize in different ways: In Mexico, SA Users use sustainable agriculture as part of a strategy of intensive agricultural production whereas in Guatemala, SA Users use sustainable agriculture as a strategy of extensive agriculture.

The differences in labor investment between Guatemala and Mexico explain much of the differences in yields and area planted between the two sites. SA Users in Guatemala ultimately invest less labor per hectare than Guatemalan SA Non-Users and Mexican SA Users and SA Non-Users. But we also saw that SA Users in Guatemala plant significantly more area to maize than SA Non-Users in Guatemala. It is clear that these SA Users in Guatemala are using their surplus labor — saved through the use of sustainable agriculture — to invest in planting more area to maize. These results support earlier work on the relationship between technological innovation and deforestation mentioned by some researchers in the literature (Kaimowitz and Angelsen 1998; Angelsen and Kaimowitz 1999).

Project managers in Guatemala conjecture that farmers with extra time on their hands are also investing it in cash crops. These project managers report that the main driver of deforestation on the north side of the Sierra de las Minas is indeed the planting of cash crops such as coffee and cardamom rather than the expansion of maize fields. In fact, we found that SA Users in Guatemala were significantly more likely to plant fruit trees and programmatically more likely to plant coffee (102 SA Users and 89 SA Non-Users plant coffee).

We thought that perhaps the desire or need to sell surplus maize might be driving farmers to increase production. Indeed, we found that SA Users were more likely to sell surplus maize than SA Non-Users. We also found that SA Users in Mexico sold about 1.5 times more maize than SA Non-Users and sold three times more maize than SA Users in Guatemala. It appears that SA Users in Guatemala and Mexico use sustainable agriculture as different economic strategies as well. SA Users in Guatemala seem to use sustainable agriculture to save labor that then can be used on greater agricultural expansion and investment in cash crops to generate income. SA Users in Mexico, however, use their labor to focus on maize production for household consumption and as a way to earn additional income.

What else accounts for the differences in labor input, yield level, and area planted that we found between SA Users and SA Non-Users in Guatemala and Mexico? Because the differences are so marked between our two study sites, it seems that there must be some other factor influencing farmer behavior. And, indeed, there is. It turns out that the biggest influence on increases in farming efficiency — reflected in yield level, area planted, and labor investment — is access to land for agriculture.

In the study site in Mexico, farmers in the ejido are given about 20 hectares of land; they do not have access to any other land. The incentive to be efficient is very high. In Guatemala, on the other hand, farmers on the north side of the Sierra de las Minas live in an area in which few community residents actually have title to their land (although they say they "own" it, ownership is more a function of claim than right). At this site, farmers actually have incentive to plant more area—the opposite situation as in El Ocote, Mexico. In Guatemala, government policy up to the 1980s actually encouraged farmers to deforest land to create larger agricultural plots. Farmers could claim land, including primary forest, that did not appear to be owned by anyone. According to the government, as long as farmers improved "the productive capacity" of the land — that is, used it for crops or livestock — they could maintain indefinite usufruct rights to it. Although this policy is no longer formally in effect, it still influences farmers' perceptions and behavior related to the acquisition and use of land. In the absence of owning title to their land, Guatemalan farmers have less land security. They have every incentive, therefore, to plant more area, not less, and to make as little investment in the land as possible.

We conclude that, in tropical forest sites similar to our sample, unless access to land is limited, sustainable agriculture techniques such as those used at our two study sites will not successfully reduce the amount of area planted and, therefore, rates of deforestation. Conversely, in areas where farmers have a greater incentive to be as efficient as possible because of limited access to land, sustainable agriculture can reduce area requirements and, thus, rates of deforestation. Farmers must be secure in their rights to land and government policy must support — rather than discourage —­ agricultural practices that are compatible with conservation. Adoption rates for sustainable agriculture techniques support these relationships: from 1994 to 1997, adoption in Mexico increased fivefold to 500 SA Users while in Guatemala it increased only 1.4-fold to 613 SA Users. In Mexico, farmers were more eager to try new agricultural techniques that held the promise of increasing yield. In addition, it appears that SA Users in Mexico are more committed than their counterparts in Guatemala because SA Users in Mexico tended to participate in the project longer.

Our analysis clearly and consistently demonstrates that the use of sustainable agriculture techniques is associated with reduced use of fire. As we mentioned, forest fires — many of which start as a result of poor fire management during the burning of agricultural plots — are perhaps the greatest threat to biodiversity conservation in both the Sierra de las Minas and El Ocote. In fact, 20,000 hectares and 19,000 hectares of primary forest in the Sierra de las Minas and El Ocote reserves, respectively, were lost to fire in 1998. At both sites, SA Users were overwhelmingly less likely to use fire to prepare their lands than SA Non-Users. After controlling for all other variables in our multivariate analysis, use of fire was the most important determinant of user status in Guatemala and Mexico. Similarly, our multivariate analysis also showed that older SA Non-Users were more likely to use fire to prepare their lands.

Fallow Area and Duration

We found no differences in fallow amount and duration between SA Users and SA Non-Users in either Guatemala or Mexico. Sustainable agriculture does not appear to have an impact on fallow land under the conditions found in our two study sites. We did find, however, that plots with reduced fallow times produced higher yields if farmers had used sustainable agriculture.

Contamination of the Environment

Use of chemical fertilizers and pesticides by farmers was limited in both Guatemala and Mexico. Pollution from these threats, therefore, was not a major concern in our study sites. Nevertheless, we infer that contamination of the environment from smoke was greatly reduced by use of sustainable agriculture in both Guatemala and Mexico.

Attitudes Concerning Conservation

As expected, SA Users in both Guatemala and Mexico tend to perceive more positive effects of sustainable agriculture than SA Non-Users. In Guatemala, SA Users were more likely than SA Non-Users to report that they do not use fire in order to protect the forest. According to the results of the focus groups in Guatemala and Mexico, SA Users generally perceive the importance of biodiversity conservation.

Perhaps most important is the conclusion by both Defensores de la Naturaleza and Línea Biósfera that their sustainable agriculture programs are crucial for building trust and confidence in the communities in which they work. This result is supported by field experience and the results of the focus group analysis. Both organizations believe that their sustainable agriculture programs have served as an effective bridge approach to reach conservation goals. In both cases, farmers originally perceived little connection between their agricultural practices and conservation and the importance of conservation alone. By working with farmers on their self-perceived priorities and building relationships with community members, both organizations were able to demonstrate to farmers the links between agriculture and biodiversity conservation. This link was supported by outreach and education programs to clarify and bolster perceptions of this relationship. At the same time, establishing these relationships enabled each organization to work on other conservation issues that were not originally perceived by community residents as top priority.

Participation in Community Organizations

In Guatemala, SA Users were more likely to belong to a community organization than SA Non-Users. In Mexico, there was no difference because everyone belonged to the same ejido and regional organizations.

Organizations and the organization of communities played different roles in Guatemala and Mexico. In Guatemala, communities on the north side of the Sierra de las Minas are dispersed, with little communication and interaction between communities. Similarly, within communities, organization is fairly decentralized — there are few formal community structures to pull residents together. In addition, these communities are subject to a high level of conflict, perhaps due in part to their lack of organization.

At the beginning of the program in the Sierra de las Minas, Defensores de la Naturaleza found it very difficult to gain access to individual farmers because of the lack of formal community mechanisms. In this case, the sustainable agriculture program served as a catalyst for community mobilization. It increased formal mechanisms for communication within communities and decreased the amount of conflict between communities and individuals. The increased level of organization provided Defensores de la Naturaleza with the means to more efficiently work with communities on important conservation issues.

In Mexico, the level of community organization was quite high and influenced adoption of sustainable agriculture in a different way. While Línea Biósfera reports that it was difficult initially to gain access to communities because ejido organizations acted as a buffer to outsiders, once accepted and trusted, official endorsement by the ejido gave the organization virtually unlimited access to all community members. Adoption rates of sustainable agriculture were very high in Mexico compared with Guatemala.

Principles for Using Sustainable Agriculture as a Conservation Tool

Our second goal was to determine key principles that can help project managers use sustainable agriculture projects more effectively to reach conservation goals. The principles in this section are the result of our analysis of the data and discussions we had concerning conditions at each site. Although they are not meant to be a recipe that guarantees success, we offer these principles as guidelines to help you implement effective sustainable agriculture programs. These principles are derived from two sites that are very similar in many ways. We believe, therefore, that these principles can be generalized to other similar sites. The more dissimilar the site, the more unlikely it is that the principle will hold. We divide the principles as they relate to two phases of project management: design and implementation.

Design

Be clear about the threats to conservation that sustainable agriculture is designed to address.

Sustainable agriculture is effective as a conservation tool only if it is appropriately directed at addressing a particular threat. We have seen that sustainable agriculture, as defined in this study, does not necessarily reduce deforestation resulting from expansion of subsistence crops. In Guatemala, we saw that, although sustainable agriculture activities focused on subsistence crops, the main threat was expansion of cash crops. In our sample, sustainable agriculture proved to be an exceptional tool for reducing forest fires — although this was not an explicitly intended objective of the sustainable agriculture program at its inception. In our sample, focusing sustainable agriculture on the reduction of pollution caused by use of chemical pesticide and fertilizer turned out to be a worthless endeavor. Had proper assessments occurred before the project began, it is likely the program would not have been designed to influence environmental contamination. Similarly, the assumption that sustainable agriculture would increase recovery of forested lands proved unfounded.

Be clear about the mechanism through which sustainable agriculture impacts conservation.

The mechanism through which sustainable agriculture influences conservation outcome is relatively complex. There is a long series of assumed intermediate steps to get from intervention to outcome: Adoption of sustainable agriculture leads to increased yield, which leads to the need for less land, which leads to lower area planted and reduced labor needs, which leads to a reduced likelihood of cutting down forested areas to plant new fields…. To make a causal link between sustainable agriculture and conservation, project managers must understand each intermediate step. We saw, for example, that adoption of sustainable agriculture does not necessarily mean increased yield or decreased area planted.

Do not use sustainable agriculture to reduce rates of deforestation where there is relatively open access to land.

Access to land may be one of the biggest predictors of sustainable agriculture's utility as a conservation tool. In our study, where land was relatively available, the sustainable agriculture techniques adopted by farmers did not work to reduce area planted. Where access to land was limited, farmers had greater incentive to be efficient in their use of land, and sustainable agriculture was associated with reductions in area planted.

Use sustainable agriculture where farmers have greater land security.

Land security — either in the form of tenure or usufruct rights — provides the opportunity for farmers to make investments for future production in their agricultural plots and is, therefore, another important factor related to the efficacy of sustainable agriculture as a conservation tool. In areas where farmers have little land security and perceive that the land they work could be taken away from them at any time, they have little incentive to make investments in their plots that might increase yield and reduce area needs. In Mexico, where there is relatively high land security, farmers were willing to make greater investments in their agricultural plots.

Consider the use of sustainable agriculture in areas where local, state, and national government policy is supportive of greater agricultural efficiency.

Government policy can influence farmers to plant either more or less area. In countries where the government's policy is to exploit natural resources and promote development without consideration for conservation, it is likely that sustainable agriculture will have minimal impact on conservation. In countries where government policy encourages greater efficiency in land use and management and where conservation is valued, chances are sustainable agriculture will be a more effective conservation tool. In Guatemala, government policy is a major driver in the expansion of agricultural fields. In Mexico, the ejido system actually encourages farmers to be more efficient in their use of land.

Do not assume that labor saved using sustainable agriculture techniques will be used on activities that are supportive of conservation.

Over time and in some situations, the sustainable agriculture techniques adopted by farmers included in this study may reduce the amount of labor required to work a particular field, but farmers may put this saved labor to use in destructive ways. In Guatemala, for example, we saw that farmers who use sustainable agriculture techniques actually plant more total area and invest their time in other activities — such as extensive cash crop expansion without intensification — that can work against conservation.

Implementation

Begin your sustainable agriculture project by testing only a few techniques.

Inundating farmers with too many tools at the beginning of a project may discourage them from participating in sustainable agriculture activities. Farmers are more likely to use a few, very effective techniques rather than many moderately useful ones. This proved to be the case in both Guatemala and Mexico. At the beginning of the program, both organizations promoted up to 15 different techniques at the insistence of the organization that trained and supervised them in sustainable agriculture. Farmers, however, only wanted to use two or three techniques. In addition, rather than assuming that each technique has high rates of return, it is important to fully test each technique on a small scale before promoting it widely. This supports earlier findings on what makes for successful sustainable agriculture extension (Bunch 1982).

Select specific sustainable agriculture techniques carefully on the basis of returns to labor.

Farmers look for ways to increase productivity while reducing labor demands. They will not adopt practices that require high amounts of labor, especially if the return on labor is not favorable. In our study, farmers were more likely to use those techniques that required extremely low labor investments or that were sure to save them time over the long run. Planting velvetbean, perhaps the easiest technique to use, was the most popular in both sites. Furthermore, investing less labor in smaller areas adds to the efficiency sought by farmers. This result was clear in our regression analysis on yield.

Be prepared not to see immediate results.

The effects of sustainable agriculture take time to become apparent. Investments are often incremental over multiple years, so results might be slow in coming or difficult to discern. Reduction of area planted will occur only after improvements in yields are attained, which requires significant amounts of time. In our study sites, project managers reported that the effects of sustainable agriculture were not observable for three to five years. As a project develops, it is important to keep this in mind to address potential concerns farmers may have during the initial phases. With agricultural production and yield improvements slow in coming, it will undoubtedly be longer before conservation benefits are apparent.

Establish a flexible system of sustainable agriculture extension that will adapt to local conditions.

Sustainable agriculture projects must be based on the needs of local farmers to have any chance of conservation success. In addition, the way an organization works with farmers is extremely important. In some cases, as in Mexico, it may be best to work through local, volunteer promoters. In other cases, as in Guatemala, paid employees may need to be contracted to carry out project activities. When opportunities arise to promote new avenues of extension, such as cross-community exchanges, the implementing organization must be ready make the most of them.

Integrate sustainable agriculture activities with other interventions that create the conditions for sustainable agriculture to contribute to conservation success.

Sustainable agriculture, like most other interventions, will not achieve conservation on its own. Other project activities, such as environmental education or community mobilization, create the conditions necessary for sustainable agriculture to take hold, flourish, and positively affect conservation outcomes. In both Guatemala and Mexico, supplemental project activities led farmers to more readily support sustainable agriculture activities.

Use sustainable agriculture as a bridge approach to conservation — to win the trust and confidence of communities.

The bridge approach provides the opportunity for an organization to gain the confidence of a community while promoting the natural links that occur between a particular activity and conservation. In effect, it builds a bridge between an intervention and conservation. In the case of sustainable agriculture, reduction in erosion through the use of cover crops, for example, may also contribute to improved drinking water quality in surrounding rivers and streams. The bridge approach may also create the conditions for future conservation actions. In Guatemala, sustainable agriculture participants formed the nucleus of natural resources management and environmental committees that were established several years into the program.

Use sustainable agriculture as a mechanism to organize communities and help reduce conflict.

Sustainable agriculture can serve as a mechanism to organize communities that are highly decentralized or unorganized. In this way, sustainable agriculture contributes to the social and political conditions that are required to interact with and mobilize communities to take conservation action. It can also serve as a neutral opportunity for farmers who do not normally interact to work together to solve problems of mutual concern. In Guatemala, in particular, sustainable agriculture served this purpose.

When implementing sustainable agriculture as a conservation tool, stay focused on conservation!

Sustainable agriculture interventions are necessarily highly social endeavors that have many intermediate production, economic, and social outcomes. Because so many of these intermediate social outcomes are required, there is a higher risk that project managers may lose sight of the ultimate conservation goals determined at the beginning of the project. To prevent project managers from being satisfied merely with increased yields, for example, they should constantly have conservation goals at the forefront of their minds.

Process Lessons — Learning to Learn Better

The third goal of this research ­ primarily BSP's in its role as organizer and facilitator of this project ­ was to learn how to determine the conditions under which a specific conservation tool works across multiple projects and to determine how to build capacity in local project partners to facilitate their own applied research and learning.

This project proved to be extremely rewarding to those of us who worked on it together. Working as equal partners, BSP, Línea Biósfera, and Defensores de la Naturaleza, with the support of CIFOR, constantly shared ideas, creating a strong, mutual learning environment. During our many meetings to discuss results of the study, honest and constructive exchanges allowed us to determine what was and wasn't working, and why, in the Sierra de las Minas and El Ocote sustainable agriculture programs. We were able to openly discuss success and failures, and this level of objectivity allowed us to gain a better understanding of the conditions under which sustainable agriculture works as a conservation tool. (For candid self-assessments of the programs managed by Línea Biósfera and Defensores de la Naturaleza, see their respective case studies at www.BSPonline.org.)

This study provided Defensores de la Naturaleza and Línea Biósfera an opportunity to ask specific operational questions about the efficacy of one of their cornerstone programmatic interventions. It also allowed them to figure out the best way of addressing those questions in order to learn how to improve project success. It improved their capacity in applied research design and implementation and in data analysis and communications. It also sparked interest in future research to analyze other pressing issues at each site. This interest was demonstrated by Línea Biófera's declaration that it would use what it learned from this study to research the effects on conservation of cattle and the opening of new pastures in El Ocote — a threat identified as being perhaps greater than agricultural encroachment. In the Sierra de las Minas, Defensores de la Naturaleza determined it would further investigate the role that expansion of cash crops plays in deforestation in the Reserve.

Based on our final meeting and discussions related to the process of conducting this research, we can recommend the following process principles for conducting similar inquiries.

Focus research on a specific tool.

By focusing on a specific tool, we can learn about the conditions under which it is most effective. We can also come up with concrete, operational recommendations for other practitioners around the world to use the tool more effectively in the future.

Design research around the interventions your partners are most interested in learning about.

Experience in implementing a specific conservation tool is extremely important. This experience allows project partners to ask the right questions, determine the best way to answer them, interpret the results of the analysis, and put the results to immediate use. Working across multiple sites provides the opportunity to look across many different conditions to determine which are most associated with successful implementation of the tool.

Work with project partners that demonstrate a high level of individual and institutional curiosity.

Field-based practitioners and organizations that spend most of their time implementing and managing projects often have little time to sit back and analyze what they are doing. But the desire to improve, willingness to question the efficacy of interventions, and drive to learn are all-important ingredients in a successful joint-learning effort.

Involve project partners in all phases of the research.

Include all project partners from the beginning of the research project, especially during the conceptualization and design phases. Their questions should drive the research. Constant involvement in the research and responsibility for its successful completion will help keep partners engaged throughout the process. By being involved in all phases of the research, project partners will more likely see the benefit and utility of the results and put them to good use.

Identify potential gaps in capacity and plan to fill them early on.

In any joint-learning endeavor, there will be gaps in capacity to carry the work through to the end. Be prepared to fill these gaps with additional training or technical assistance. By building capacity throughout the research cycle, partner organizations will be able to conduct the research on their own in the future. If possible, work with partners that have similar research-related skill levels to ensure that partners move through the learning process in unison. We discovered that the aspect of research most likely to be deficient in partner organizations is data analysis. If you intend to contract someone from outside to assist in data handling and analysis, hire this person at the beginning of the project so he or she can participate in the design, planning, and implementation of the research.

Select sites to be included in the study in a systematic and precise fashion to obtain specific principles.

Developing the sampling framework for the study is key. To obtain relatively precise principles, you must limit your sample so you can control for potentially confounding factors that may affect the outcome of interest. To obtain principles that are also generalizable, however, your sample must be large enough to be able to look across a range of important conditions. By including multiple sites in your sample, you can vary the circumstances you are studying in order to determine the conditions under which the conservation tool of interest is most effective.

Standardize data-collection instruments and methods and analytical approaches in order to learn across sites.

To learn across sites effectively, data must be collected and analyzed in a standardized fashion. To standardize approaches and instruments it is crucial to have as much time to interact as possible. To increase the power of the analysis you wish to conduct, it is essential to collect and analyze the same variables using the same analytical and statistical tools.

Develop an agreed-upon learning framework that maps out the questions you will ask, the way you will research and analyze the questions, and the results you hope to communicate.

Developing a learning framework at the beginning serves as a guide throughout the life of the research project. When in doubt, project partners can always refer back to the learning framework to orient their work. This provides partners the opportunity to avoid miscommunication and misunderstandings about the goals or approach of the research. The framework must, however, be flexible. You must be prepared to modify it based on the reality of the situation as you collect and analyze data.

Develop and stick to a mutually agreed-upon workplan and timeline.

Coordinating this type of research is extremely challenging. Multiple partners, sites, and instruments all add to the need to develop and adhere to a common workplan. As research moves through various phases — design, data collection, analysis, communication — it is important to keep the process going at a steady rate. We found that periodic field visits and periodic review and analysis meetings helped keep us on track throughout the learning process. We also found, however, that competing job responsibilities meant that project partners were not always available to work on this project, which caused costly delays. Managing and sticking to the workplan is especially critical for the organization that takes on the responsibility of facilitating and coordinating the learning project.

Try to keep it simple.

Plan to ask what you believe to be relatively easy operational questions. We guarantee that, as you move through the learning process with multiple partners and sites, the challenge to effectively address your questions will become increasingly difficult and complex!

Conclusions and Next Steps

This project has clearly demonstrated that sustainable agriculture does not always have its intended impact. But no tool or strategy is ever completely and singularly effective in meeting conservation goals. The purpose of this study was not to prove that sustainable agriculture works or does not work. Rather, we undertook this research to determine the conditions under which sustainable agriculture serves as an effective conservation tool. The results included in this publication shed some light on these conditions.

This research project also was designed to develop concrete guidance for practitioners around the world who are working with sustainable agriculture. We have addressed this goal by distilling relevant principles from our analysis. We hope they prove useful.

Finally, we had hoped to learn the best way to go about asking and answering important operational questions in conservation. During the research process, we learned many things about how to work effectively across a portfolio of projects to learn most effectively. Our learning on this theme is summarized in the process principles we present in the preceding section.

This research has generated a number of questions that conservation practitioners must address to advance our collective understanding of how to best integrate agriculture interventions into conservation projects. These questions include the following:

• What role does sustainable agriculture play in reducing deforestation and other threats when cash crops such as coffee and cardamom are involved?

• What role does household ownership of cattle play in deforestation in tropical forest areas, and how can sustainable agriculture be used as a tool to address this issue?

• Since one of the major contributions of sustainable agriculture to conservation is fire prevention, are there approaches to this goal that will engage farmers more quickly than aiming to reduce rates of deforestation through increased yields? Can we promote techniques that require less labor and that allow farmers to see more immediate results so that the effects on fire prevention are even more extensive and immediate?

• To what extent does sustainable agriculture affect other conservation-related issues such as emigration from communities to fragile forested lands? Do the investments that sustainable agriculture requires serve to encourage farmers to remain on their land rather than leaving their community in search of more productive areas?

• What are the conditions under which sustainable agriculture will work as a conservation tool in areas of the world that are dissimilar to our sample?

• To what extent would labor-intensive sustainable agriculture techniques, unlike those used by the farmers in our sample, contribute to reducing rates of deforestation? What labor-intensive techniques are most useful, appropriate, and equitable in socioeconomic and conservation terms?

• What is the role of diversified, agroforestry programs that can be sustainably managed to produce cash crops and increase family livelihoods while reducing or substituting for subsistence grain production?

As we conducted this learning inquiry, the study generated a number of questions. To continue to learn how to learn better, we provide the following questions as potential guides:

• What is the optimum number of projects and sites that make up a learning portfolio?

• What are the basic skills required of project partners for their participation to contribute effectively to the learning process?

• What is the best way to deal with gaps in knowledge and capacity — for example, in data handling and analysis — to complete the learning process?

• What is the most effective role for a facilitating organization to encourage and support local partners to conduct sound and precise learning?

• What is the role of outside technical support and assistance from third-party individuals and institutions that are not directly involved in the day-to-day implementation of the learning project?

• What is the best way to communicate the results of individual project partners and the work across the entire learning portfolio?

This study has provided a wealth of learning on both the conditions under which sustainable agriculture succeeds as a conservation tool and the process by which partner organizations can develop precise, operational, and useful management principles across sites. We encourage others to continue to question, investigate, and improve our understanding of the use of sustainable agriculture and other conservation tools around the world and to share what they learn with the rest of the conservation community.

To Help You on Your Way

If you or the organization with which you work is considering using sustainable agriculture as a conservation tool, you may want to ask yourself the following questions.

First, ask about your organization's capacity to do sustainable agriculture

• To what extent are we willing to get involved in an agriculture project when we are a conservation organization?

• Do we have the necessary staff, capacity, and funding to design, manage, and monitor a sustainable agriculture project?

• Do we adequately understand the pros and cons of promoting sustainable agriculture at the project site?

• Can we team up with other local organizations to promote sustainable agriculture at the project site?

Then, ask about the underlying assumptions and design of your project

• What are our conservation goals and objectives? How can we measure them? How will we be able to tell if our sustainable agriculture project affects our goals and objectives?

• What are the major threats to conservation at the site? To the best of our knowledge, is sustainable agriculture the best possible tool to address these threats? What is the mechanism through which we think sustainable agriculture will affect conservation success?

• How will we determine which specific sustainable agricultural techniques to promote?

• What other project activities would best complement and support sustainable agriculture?

• What is our long-range strategy to reduce threats? Once sustainable agriculture is established, what follow-up activities would be best suited to long-term conservation?

• What is the best way to assess the needs of the communities in which we plan to work and what are the best mechanisms for interacting with community members?

Then, ask about the environmental conditions at your site

• What environmental conditions at the site in which we work will affect rates of adoption of sustainable agriculture techniques? What will affect the extent to which sustainable agriculture is effective?

• Have other sustainable agriculture or related programs been attempted at the site before? How successful were they? How did environmental factors influence their success?

• Is the site at the agricultural frontier? Where is agriculture expansion most likely to occur?

Then, ask about the social conditions at your site

• To what extent are community members organized? How unified or dispersed are the communities? What existing communications channels can be used?

• Is there preexisting community infrastructure that can be used to promote sustainable agriculture activities?

• To what extent are the communities open to involvement of outside organizations and individuals?

• Do community members legally own their land or have formal usufruct rights? How secure are they in their access to the land they farm? To what extent do government policies encourage or discourage the efficient use of land resources? Is government policy supportive or unsupportive of biodiversity conservation?

• What other projects have been promoted in the area in the past? To what extent did residents participate?

Finally, ask about the costs and benefits of sustainable agriculture

• How much will it cost to implement the program over the next five years? Over the next 10 years? How many farmers do we intend to reach?

• Are there other projects that may have higher returns than sustainable agriculture? What are the short-, medium-, and long-term benefits of sustainable agriculture? How do these compare with other potential interventions?

• What are the opportunity costs of implementing this project? Are there other activities we will not be able to undertake because of our focus on sustainable agriculture?

To Learn More

For inquiries related to the results of this study, please contact Richard Margoluis at Richard@FOSonline.org or www.FOSonline.org.

We encourage others to continue learning more about the conditions under which sustainable agriculture programs can be used as an effective strategy for achieving conservation success. These resources can help support that learning.

Results From the Two Study Sites

The final reports from Guatemala and Mexico are available in Spanish only in the Publications section of the BSP Web site at www.BSPonline.org.

Defensores de La Naturaleza. 2001. Impacto de la Agricultura Sostenible sobre la Conservación de la Biodiversidad, Reserva de Biósfera Sierra de las Minas, Guatemala. Washington, D.C.: Biodiversity Support Program.

Línea Biósfera. 2001. Impacto de la Agricultura Sostenible en la Conservación de la Biodiversidad, Reserva de la Biósfera Selva el Ocote, Chiapas, Mexico. Washington, D.C.: Biodiversity Support Program.

Summary Literature Review

A brief literature review for this study is available online in the Publications section of the BSP Web site at www.BSPonline.org.

References

References marked with "±" are available at www.BSPonline.org.

Altieri, M. 1999. Multifunctional dimensions of ecologically-based agriculture in Latin America. Paper prepared for the FAO-Netherlands Conference on the Multifunctional Character of Agriculture and Land, Cultivating Our Futures, at Maastricht, The Netherlands, 12-17 Sept..

Angelsen, A., and D. Kaimowitz. 1999. Rethinking the causes of deforestation: Lessons from Economic Models. The World Bank Research Observer 14(1):73-98.

Angelsen, A., D. Kaimowitz, S. Holden, J. Smith, and S. Vosti. 1999. Technological change in agriculture and tropical deforestation: Definitions, theories, and hypotheses. Paper presented at the CIFOR Workshop on Technological Change in Agriculture and Deforestation. CATIE: Centro Agronómico Tropical de Investigación y Ensenza, Turrialba, Costa Rica, 11-13 March.

Borrini-Feyerabend, G. 1996. Collaborative management of protected areas: Tailoring the approach to the context. Gland, Switzerland: IUCN: World Conservation Union.

Brandon, K., K. Redford, and S. Sanderson, eds. 1998. Parks in peril: People, politics and protected areas. Washington, D.C.: Island Press.

Brown, M., and B. Wyckoff-Baird. 1992. Designing integrated conservation and development projects. Washington, D.C.: Biodiversity Support Program.

Buckles, D., B. Triomphe, and G. Sain. 1998. Cover crops in hillside agriculture: Farmer innovation with mucuna. Ottawa, Canada: International Development Research Center (IDRC)/Centro Internacional de Mejoramiento de Maiz and Trigo (CIMMYT).

Bunch, R. 1982. Two ears of corn: A guide to people-centered agricultural improvement. Oklahoma City, OK: World Neighbors.

Bunch, R. 1988. Güinope integrated development program, Honduras. In The greening of aid, eds. C. Conroy and M. Litvinoff. London: Earthscan Publications

CARE. 1993. Diagnóstico y estrategia agroforestal para la zona de amortiguamiento de la Reserva de Biósfera Maya, Petén. Guatemala.

Castellon, M. 1997. Dynamics of Q’eqchí-Maya colonists in Guatemala Sierra de las Minas. University of Wisconsin-Madison, USA.

Centro Internacional De Informacion Sobre Cultivos De Cobertura. 1997. Experiencias sobre cultivos de cobertura y abonos verdes. Honduras.

Current, D., E. Lutz, and S. Scherr, eds. 1995. Costs, benefits, and farmer adoption of agroforestry: Project experience in Central America and the Caribbean. Washington, D.C.: World Bank.

Davenport, R., and A. Kaus. 1995. Programa de ecodesarrollo de México: Evaluación de medio plazo. Washington, D.C.: Biodiversity Support Program.

Diaz Plaza, R., R. Flores Escamilla, and T. Vera Pren. 1992. Evaluación agronómica y social de la transferencia de un paquete tecnológico para el control de la mosquita blanca con agricultores Mayas de Yucatán, México.

Donovan, R. 1994. BOSCOSA: Forest conservation and management through local institutions. In Natural connections: Perspectives in community-based conservation, eds. D. Western, M. Wright, and S. Strum. Washington, D.C.: Island Press.

Edwards, M. 1997. Estudio antropológico de las políticas de desarrollo de la Fundación Defensores de la Naturaleza en dos comunidades indígenas de la Reserva de Biósfera Sierra de las Minas.

Faris, R. 1999. Deforestation and land use on the evolving frontier: An empirical assessment. Development Discussion Paper No. 678 (February). Harvard Institute for International Development.

Ferraro, P., and R. Kramer. 1995. A framework for affecting household behavior to promote biodiversity conservation. Washington, D.C.: EPAT/Winrock International Environmental Alliance.

Flores Rodas, M. 1993. Memoria: I Congreso Forestal Centroamericano. Guatemala.

Fundacion Defensores de la Naturaleza. 1992. Primer plan maestro para la Reserva de la Biósfera Sierra de las Minas.

Fundacion Defensores de la Naturaleza. 1996. Plan de trabajo de ecodesarrollo. Sierra de las Minas.

Fundacion Defensores de la Naturaleza. 1993. Diagnóstico de integración humana en la Reserva de Biósfera Sierra de las Minas.

Fundacion Defensores de la Naturaleza. 1996. Evaluación del primer semestre de actividades de los programas desarrolladas en el distrito Polochic, Sierra de las Minas.

Gliessman, S.R. 1998. Agroecology: Ecological process in sustainable agriculture. Ann Arbor, Michigan: Ann Arbor Press.

Gómez-Pompa, A. 1982. La etnobotánica en México. Biótica 7(22).

Guatemala, Comision Nacional del Medio Ambiente de. 1986. Ley de protección y mejoramiento del medio ambiente. Guatemala.

Guatemala, Congreso de La Republica de. 1989. Ley de areas protegidas y su reglamento. 5th ed. Guatemala: Consejo Nacional de Areas Protegidas.

Guatemala, Congreso de La Republica de. 1990. Reserva de Biósfera Sierra de las Minas. Decreto 49-90. Guatemala: Consejo Nacional de Areas Protegidas.

Guatemala, Instituto Geografico Nacional de. 1977. Mapa geológico de la República de Guatemala, escala 1:500,000.

Guatemala, Instituto Geografico Nacional de. 1959. Mapa de clasificación de reconocimiento de los suelos República de Guatemala. Guatemala: Servicio Interamericano de Cooperación para la Agricultura. Esc. 1:250,000.

Guatemala, Instituto Geografico Nacional de. 1968. Mapa Forestal de la República de Guatemala. Esc.1:250,000. Color.

Gutierrez Castellanos, H. 1997. Análisis del impacto de las actividades agropecuarias sobre el suelo, agua y bosque, en la comunidad del Paxte Poptún, Petén. Tesis del Ing. Agr. Guatemala, Universidad de San Carlos de Guatemala, Facultad de Agronomía.

Hjarsen, T. 1997. Effects of rural agriculture and plantation forestry on high Andean biodiversity. Denmark: Centre for Research on the Cultural and Biological Diversity of Andean Rainforests (DIVA).

Humphries, S. 1998. Milk cows, migrants, and land markets: Unraveling the complexities of forest-to-pasture conversion in Northern Honduras. Economic Development and Cultural Change 47(1), October.

INE SEMARNAP. (1999). Programa de restauración ecológica del polígono "El Ocote," Ocozocoautla Chiapas, México. INE SEMARNAP.

INE. 1998. Estudio para la propuesta de declaratoria para el establecimiento de la zona de restauración "Selva El Ocote." INE: Chiapas, México.

Instituto de Historia Natural. 1993. Plan operativo 1993: zona de protección forestal y faunica selva El Ocote. Instituto de Historia Natural, Gobierno del estado de Chiapas: Chiapas, México.

International Center for Research in Agroforestry (ICRAF). 2001. http://www.icraf.cgiar.org/ Accessed 04-15-01.

Kaimowitz, D. 1996. Livestock and deforestation, Central America in the 1980s and 1990s: A policy perspective. Jakarta, Indonesia: Center for International Forestry Research.

Kaimowitz, D. 1999. Review of R. Faris (1999) article. POLEX Listserv.

Kaimowitz, D., and A. Angelsen. 1998. Economic Models of Tropical Deforestation: A Review. Bogor, Indonesia: Center for International Forestry and Research.

Langholz, J. 1999. Exploring the effects of alternative income opportunities on rainforest use: Insights from Guatemala's Maya Biosphere Reserve. Society and Natural Resources 12:139-149.

Larson, P., M. Freudenberger, and B. Wyckoff-Baird. 1998. WWF integrated conservation and development project: Ten lessons from the field, 1985-1996. Washington, D.C.: World Wildlife Fund.

Lehnhoff, A., and O. Nuñez. 1995. Sierra de las Minas Biosphere Reserve, Guatemala. Unpublished report produced by The Nature Conservancy.

Leiva, J. 1984. La agricultura migratoria y sus efectos sobre el suelo; un enfoque para su restauración. Guatemala, Universidad de San Carlos de Guatemala, Facultad de Agronomía.

Little, P. 1994. The link between participation and improved conservation: A review of issues and experiences. In Natural Connections: Perspectives in Community-based Conservation, eds. D. Western, M. Wright, and S. Strum. Washington, D.C.: Island Press.

Lutz, E., S. Pagiola, and C. Reiche, eds. 1994. Economic and institutional analyses of soil conservation projects in Central America and the Caribbean. Washington, D.C.: World Bank.

Magdaleno, J. 1995. Diagnostico de la Asociación Rural de Interés Colectivo "Tzobolutic." Unpublished document from ARIC Tzobolutic; Chiapas, México.

Margoluis, R. 1994. Conservation for health: Small-scale commercial utilization of non-timber forest resources and human health in the Sierra de las Minas Biosphere Reserve. Final Report. Guatemala: Fundación Defensores de la Naturaleza.

Margoluis, R., and N. Salafsky. 1998. Measures of success: Designing, managing, and monitoring conservation and development projects. Washington, D.C.: Island Press.

±  Margoluis, R., S. Myers, J. Allen, J. Roca, M. Melnyk, and J. Swanson. 2001. An ounce of prevention: Making the link between health and conservation. Washington, D.C.: Biodiversity Support Program.

McNeely, J. 1994. Protected areas for the twenty-first century: Working to provide benefits for society. Unasylva 45:3-7.

Mulleried, Fk. G. 1957. La geología de Chiapas. Gobierno del estado de Chiapas: Tuxtla Gutiérrez, Chiapas.

Neill, S., and D. Lee. Forthcoming. Explaining the adoption and disadoption of sustainable agriculture: The case of cover crops in Northern Honduras. Submitted to Economic Development and Cultural Change.

Oliva, F. 1990. Diagnostico Comunitario de la Unión de Ejidos Triunfo de los Pobres. Unpublished document from la Unión de Ejidos Triunfo de los Pobres; Chiapas, México.

Patton, M. 1990. Qualitative Evaluation and Reserarch Methods. London: Sage Publications.

Puentes R., T. Carranza, and M. Gonzalez. 1997. Abonos verdes : evaluación del impacto de los abonos verdes y cultivos de cobertura en la sustentabilidad del manejo de los recursos naturales, en las comunidades del sureste de México: Proyecto Pachuca, Rockefeller. Paper presented 6-12 April 1997, Chapecó, Brasil.

Robinson, A. 1991. Sustainable agriculture: The wildlife connection. American Journal of Alternative Agriculture 6(4):161-167.

Robinson, J. 1993. The limits to caring: Sustainable living and the loss of biodiversity. Conservation Biology 7(1) (March):20-28.

Russell, V. 1996. The Chimalapas Ecological Campesino Reserve: The golden gourd of conflict and its role in protected area management. Unpublished MS thesis. Ithaca, NY: Cornell University.

Salafsky, N., and L. Wollenberg. 2000. Linking livelihoods and conservation: A conceptual framework and scale for assessing the integration of human needs and biodiversity. World Development. August V28(8):1421-1438.

±  Salafsky, N., and R. Margoluis. 1999. Greater than the sum of their parts: esigning conservation programs to maximize results and learning. Washington, D.C.: Biodiversity Support Program.

±  Salafsky, N., R. Margoluis, and K. Redford. 2001. Adaptive management: A tool for conservation practitioners. Washington, D.C.: Biodiversity Support Program.

Schelhas, J., and R. Greenberg, eds. 1996. Forest patches in tropical landscapes. Washington, D.C.: Island Press.

Scrimshaw, S., and E. Hurtado. 1988. Procedimientos de asesoría rápida para programas de nutrición y atención primaria en salud. UCLA.

Smith, N., P. Alvim, A. Homma, I. Falesi, and A. Serrão. 1991. Environmental impacts of resource exploitation in Amazonia. Global Environmental Change 1(4):313-320.

Thrupp, L. 1998. Cultivating diversity: Agrobiodiversity and food security. Washington, D.C.: World Resources Institute.

Thrupp, L., S. Hecht, and J. Browder. 1997. The diversity and dynamics of shifting cultivation: Myths, realities, and policy implications. Washington, D.C.: World Resources Institute.

Valenzuela de Pisano, I. 1996. Agricultura y bosque en Guatemala, estudio de caso en el Petén y Sierra de las Minas. Guatemala, Instituto de Investigaciones de las Naciones Unidas para el Desarrollo Social/Fondo Mundial para la Naturaleza/Universidad Rafael Landirvar.

Vásquez Sánchez, M. A., and I. March Mifsut. 1996. Conservación y Desarrollo Sustentable en la Reserva El Ocote, Chiapas. Chiapas: ECOSUR – CONABIO.

Wells, M., and K. Brandon. 1992. People and parks: Linking protected area management with local communities. Washington, D.C.: The World Bank, World Wildlife Fund, USAID.

West, P., and S. Brechin, eds. 1990. Resident peoples and national parks: Social dilemmas and strategies in international conservation. Tucson: University of Arizona Press.

Western, D., M. Wright, and S. Strum, eds. 1994. Natural connection: Perspectives in community-based conservation. Washington, D.C.: Island Press.

Yanggen, D., T. Reardon, and D. Bandy. 1999. Kudzu improved fallows in the Peruvian Amazon: A case study of technological change’s impact on deforestation. Paper presented at the CIFOR workshop on Technological Change in Agriculture and Deforestation. CAITE, Turrialba, Costa Rica, 11-13 March.

About the Biodiversity Support Program

The Biodiversity Support Program (BSP) is a consortium of World Wildlife Fund, The Nature Conservancy, and World Resources Institute, funded by the United States Agency for International Development (USAID). BSP's mission is to promote conservation of the world's biological diversity. We believe that a healthy and secure living resource base is essential to meet the needs and aspirations of present and future generations. BSP began in 1988 and will close down in December 2001.

A Commitment to Learning

Our communications activities are designed to share what we are learning through our field and research activities. To accomplish this, we try to analyze both our successes and our failures. We hope our work will serve conservation practitioners as a catalyst for further discussion, learning, and action so that more biodiversity is conserved. Our communications programs include print publications, Web sites, presentations, and workshops.

BSP Web Sites

We invite you to visit our Web sites.

*Until the end of 2006, these two sites will be available at the addresses above. WWF-US will be hosting these sites on the WWF site at www.worldwildlife.org. BSP thanks WWF for providing this service.

BSP Publications

Many of our publications are available online at www.BSPonline.org. On our home page, click on Publications. You can view publications online until the end of 2006. You may contact us by mail, phone, or fax until December 2001.

Biodiversity Support Program c/o World Wildlife Fund 1250 24th St. NW Washington, DC 20037 USA

Phone: 202-861-8347 Fax: 202-861-8324 E-mail: BSP@wwfus.org Web Site: www.BSPonline.org

Publication Credits

Please cite this publication as: Margoluis, R., V. Russell, M. González, O. Rojas, J. Magdaleno, G. Madrid, and D. Kaimowitz. 2001. Maximum Yield?: Sustainable Agriculture as a Tool for Conservation. Washington, D.C.: Biodiversity Support Program.

Acknowledgments

This study would never have been initiated had it not been sparked by discussions with Andreas Lehnhoff, Oscar Nuñez, and Estuardo Secaira in Guatemala. We are extremely grateful for their ideas and guidance. We also would like to thank Mary Ann Seday for her contribution as our lead statistician. We are convinced that, without her, we would not have been able to complete this project. She put up with endless requests and revisions — always with good nature and a smile. We also would like to thank Carlos Rivera for his enormous contribution to the analysis, and we thank Luis Mejía for assistance in data entry and management. In Guatemala, we thank César Tot for his contribution to the management of the project and input on the analysis; Carlos Velásquez for his assistance in analysis; César Castañeda for input into the design; Fernando Cha May, Oscar Cac, and the Polochic field team; and Jenny Vides for her assistance in editing. In Mexico, we would like to thank the entire field team that collected all field data and Dr. Rubén Puentes for his input on the Mexico case study.

We thank Larry Fisher, David Lee, Steve Sherwood, Carlos Perez, Lori Anne Thrupp, and Doug Mason for their participation during the design phase of this study. We also thank Oscar Brenes for his involvement in the initial stages of this project and for reviewing a draft of the results. For their extensive review, insightful evaluation, and constructive suggestions related to various drafts of this publication, we thank Jim Adriance, Jethro Pettit, Nick Salafsky, Carlos Perez, and Judy Oglethorpe. For assistance in management, editing, and production, we thank Janice Davis, Sheila Donoghue, Stacy Springer, and the good folks at Ellipse Design and Balmar Solutions in Print.

Finally, we thank the 608 farmers who live in and around the Sierra de las Minas and El Ocote Biosphere Reserves who graciously answered our many, many questions and took time to share their experience and knowledge with us. Without them, the learning that resulted from this project would not have been possible.

The Biodiversity Support Program (BSP) is a consortium of World Wildlife Fund, The Nature Conservancy, and World Resources Institute, funded by the United States Agency for International Development (USAID). This publication was made possible through support provided to BSP by the Global Bureau of USAID, under the terms of Cooperative Agreement Number DHR-A-00-88-00044-00. The opinions expressed herein are those of the author and do not necessarily reflect the views of USAID.

© 2001 by World Wildlife Fund, Inc., Washington, D.C. All rights reserved. Reproduction of this publication for educational and other noncommercial purposes is authorized without prior permission of the copyright holder. However, WWF, Inc. does request advance written notification and appropriate acknowledgment. WWF, Inc. does not require payment for the noncommercial use of its published works and in no way intends to diminish use of WWF research and findings by means of copyright.