Tropical Permaculture Agroforestry in Central America

Christopher Nesbitt of Maya Mountain Research Farm in Belize shares some thoughts about woody agriculture, stacked polycultures and agroforestry systems.

My experience is limited to the tropics, where soils tend to be poor, and easily damaged, and informed by my time working where I live. The land I am working was formerly citrus and cattle, severely degraded, compacted and a monoculture at the end of its short productive life cycle. This is similar to a lot of land in the lowland humid tropics, damaged land left behind by poor management practices, and is an excellent avenue for young people to get into owning land. I bought my land in a remote area, no road, no electricity in 1988. I was 22. It was cheap.

I took my  PDC in 1991 with Michael Pilarski, Chuck Marsh, Rick Valley, Jose Caballero, and Mark Cohen. I got excited by the concept of what would now be called “biomimicry”, this idea of creating habitat through tree planting, and decided to replicate the form and function of the primary rainforest. Much of what I have observed contradicts the idea that annuals are more productive than perennials, at least in terms of energy invested to energy returned. Perhaps an acre of grain will produce more than an acre of tree crops, but not in terms of calorie based accounting. Annual systems take a lot of energy to plant, maintain and harvest. Agroecologies built on trees are much more human scaled. The two tools I use every day are a machete and a 5 gallon bucket, and my own two feet.

As a farmer of over 25 years of work with annual crops and perennial crops I can state with some conviction that the energy returned on energy invested of a stacked polyculture that is well designed and well established, with staple crops like breadfruit (Artocarpus ailtilis), breadnut (Artocaprus camansi ),  jackfruit (Artocarpus heterophylla), coconut, avocado, mango, the various spondias, herbaceous perennials like papaya and banana/plantain, plants such as chaya (Cnidoscolus aconitifolius), and moringa can provide greens, with things like banana, pineapple, papaya among a small percentage of potential components, and the eventual sub-canopy crops such as coffee, cacao, ginger, cardamom, tumeric and vanilla, some of which are extremely high value, is much better than the return on energy invested of any perennial system. It takes much less energy to maintain a well designed agroforestry system once established than a set of annual crops, be it sorghum, corn, rice or beans or any vegetable crops. Annuals, here, when compared to a stacked polyculture that is mature, are like spinning your wheels on a car.

Vegetable gardens and corn, bean and rice crops, including the “three sisters” or corn, beans and pumpkin, the energy to establish and maintain those annual cropping systems is significant, more prone to pests, and subject to problems related to weather for growth and harvesting.

Lets think of our classic abandoned “emerald wasteland”, what I had, the green desert left behind after the green revolution has spoiled all the habitat for soil biota and reduced the landscape to an inherently biologically and financially unstable monoculture, and that high energy input to high energy output system has run up into the limitations of the productive life span of the target crops, banana, pineapple, citrus, sugarcane or intensive papaya, for some examples and the increased damage to the soil. That land has some specific needs:

1. Soil covers to retain soil so it does not lose more soil

2. Money earner/calorie exporting crop

3. Soil microorganism habitat creation


The key to a successful establishment of an agroforestry system here is to take advantage of temporal microclimates. When you have sun, you can grow sun loving plants. We grow annuals when the perennialr are getting established, corn, rice, beans, vegetables, interplanted with pioneer species like pineapple planted on contour, to retain soil and provide a yield within two years, and bananas for cover and shade, and to provide a yield within a year, with massive amounts of biomass for mulching and long pseudo-stems for use to make mechanical barriers to soil loss, especially around target species. Additional crops like papaya, sugar cane, cocoyam, cassava and sweet potato will give a yield within one year. Pioneer species make the rest of what is coming possible.

A gallery of pioneers, cassava, chaya and banana. Planted in year one.
A gallery of pioneers, cassava, chaya and banana. Planted in year one.



One species that is especially noteworthy is pigeon pea or Cajanus cajan. This is a semi-perennial bean plant that grows to a height of 8 feet, spreading a crown. it lives for up to three years and has a large and deep tap root. The canopy of this species, when planted densely, will shade out the ground, completely, and reduce the impact of torrential rains. It gives a lot of food, up above where the wild pigs, agoutis and other animals can reach it, and mines nutrients from the soil. At the end of its life, it  dries out, and the whole “tree” becomes fuel wood, very important to farmers in rural areas who do not use propane. The roots, left behind, from the deep and large tap root to the many lateral and hair roots, all die. The nitrogen associated with the roots is released, and the resulting woody material becomes food for soil microorganisms from termites down to bacteria and fungi. The space taken up by the roots becomes habitat for those organisms and improves the ability for both water and oxygen to infiltrate the soil.

Taking care of these plants for the intervening years between initial planting and production of your long term target species gives the farmer a good return in calories and marketables while the woody crops are getting established. That is key as few farmers have the surplus energy to maintain a future “food forest” without a yield in the interim. Some of those species are improving soil structure, like banana, or retaining soil, like pineapple, vetiver or lemon grass.

This mimics the biological pattern of succession here with plant species like Cecropia, the Heliconias, Bay cedar (Guazuma ulimifolia), hog plum (Spondias mombin), balsa (Ochroma sp), Cuam wood (Schizelobium parahybum) which emerge after a catastrophe like a fire, hurricane or monoculture, only the transition we are making from damaged land to ecosystem replication is much, much faster, and our species selection is largely human centric.

By year five, the microclimates have changed, and the sections of my land that have been expanded into have some shade from fruit and timber trees. This allows us to take advantage of those microclimates.

In the initial years we have selected a mixture of trees for the canopy based on need, and the desired goods and services. Timber trees like samwood (Cordia alliadore), mahogany, cedar and santa maria (all Meliaceaes), and teak all have different wood qualities, canopy behaviours and time frame between planting and harvesting. Samwood, for example, is an excellent canopy specie for cacao and coffee, keeping its shade in the dry season, harvestable within 20 years, while the melliaceaes all kick their leaves off when you need a canopy the most, and take 40-70 years to harvest. Samwood also has the wonderful habit of shedding lateral branches as it grows skyward, dropping lots of fuel wood to the forest floor, a desirable attribute for us as we use fuel wood, exclusively, to cook. Teak, on the other hand, makes shade that is much too dense for those species, and can be taken out after 16 years or a bit longer.

In the agroforestry system we have made, food trees, all of the Artocarpus species, mango, avocado, peach palm, coconut, the spondias, caimito, guava, soursop, annona, amongst a small number of the species we manage, take up a large percentage of the canopy, and provide much of the calories after years five to seven when the “pioneer species” have reached senescence or been shaded out. Some species, like mamey, may take fifteen years to give fruit. Planting early and late bearing cultivars of various species expands the seasons of those species, giving better food security and access to markets.

Leguminous trees are a part of the agroforestry system too, and provide some products and services. Guanacaste (Enterolobium cyclocarpum) has a huge spreading crown, makes great wood… but takes over 50 years to get to the point where is should be harvested. It is also a tree legume, so fixes a lot of nitrogen, drops hundreds of lbs of small leaflets, which seem to be all surface and no core mass, so excellent for making soil, and attracts birds when in flower. The pods are loved by our pigs. Leucaena is used for fire wood, the leaves are edible by animals (with some limitations because of the alkaloid mimosine), and attracts parrots when in fruit. The erythrinas attract pollinators, carpet the floor with mulch and retain soil. Tamarind fixes nitrogen and gives a fruit very high in vitamin C. The bauhinias can be coppiced, repeatedly, and make excellent fuel wood. Inga edulis and the other ingas gives pods high in nutrients, excellent fuel wood and a spreading crown. Mayflower, (Tabebua sp.) can make excellent canopy, are leguminous and make great rot resistent wood.

The coconut trees provide us with marketable oil, and the pressed partially de-fatted flesh is excellent animal fodder. We have a lot of them planted out, in what I call “induced patchiness”, which replicates natural patterns of patchyness, increasing rates of pollination and eases harvesting.

Banana, coffee, cacao, erythrina, peach palm, chi'kai/  coconut, sugar cane (left over from the pioneer phase), Spondias dulcis, Artocarpus camansi, cacao. 12 years into the transition.
Banana, coffee, cacao, erythrina, peach palm, chi’kai/ coconut, sugar cane (left over from the pioneer phase), Spondias dulcis, Artocarpus camansi, cacao. 12 years into the transition.

Under that newly established canopy at year five is where it gets interesting. Cacao and coffee are often called “shade loving”, a misnomer, in my opinion. They are really shade requiring. With inadequate shade, those trees will live short miserable lives. With enough inputs, they may be productive. Once the canopy gets to a certain point, coffee and cacao can be planted. The cacao will yield within another five years. the coffee only take two or three years, depending on soil conditions and species of coffea. The energy to maintain this area of land is decreased as less sunlight hits the ground and the grasses and other sun loving species die back. The harvest increases of longer term crops as the pioneer species become less important.

By year 10, this system is now producing lots of food, literally food dropping from the canopy to the forest floor. If not previously established, ginger, tumeric and cardamom can be propagated, complexing the agroforestry system by adding additional harvest/revenue streams into the matrix of plant assemblies of the farm without cannibalizing or otherwise detracting from what is there, already. Vanilla can be planted amongst those crops, adding a very high value crop into the system, as can black pepper. These crops diversify the production and take advantage of the shade and structure provided by previous work. The presence of medicinal crops, too many to mention here, is another important aspect of stacked polycultures. They fit easily into a diversified system.

Soil profiles in many locations on the farm are changing, with accumulation and aeration of the soil from root activities and mechanical or botanical barriers to erosion. We swale here, but not so much for water infiltration than to create long settling ponds for the run off to deposit soil nutrients. The water washes down biomass and soil, which settles. The soil and leaves, sticks, seeds, flowers, manure all become rich zones oWe lay banana stems in V’s open to the hill slope around the base of target species like cacao and coffee or young fruit or timber species, and fill in those V’s with mulch, banana leaves, compost and biochar. The water washing down the hill s carry nutrients in the form of leaves, stems, soil particles, animal and bird manure, and the web of biomas slows the water so that the nutrients settle. The larger trees are also mining nutrients from the subsoil and dropping them to the forest floor in the form of leaves, branches, flowers and fruit, where they are broken down by mocroorganisms to create more soil.

I spend half an hour to an hour walking around collecting food for my pigs, which really compliment the production model we have made, and food for my students and family, and to export from the farm to town. I don’t sell much food, not for any ethical reason, but because the time involved in getting to market and selling it is not worth the money I might make. We sell cacao and ginger, and vanilla, and pigs, all things that have a good value. We donate food weekly to a program that feeds 22 elderly people in nearby Punta Gorda town. At this point, except where I am expanding, I spend most of my working time, two to three hours a day, collecting food and fuel wood from the “forest”, literally foraging in the forest I created. Right now I have about 25 acres under active management.

Produce stall at Mata Mountain permaculture, Belize
Bountiful production from Maya Mountain

Because of the level of diversity we are very -protected from both market vagaries and plant specific pathogens. We have a lot of species. In addition to the high level of species diversity, we also have high diversity within species, making species specific pathogens less threatening.

One thing that is important for me is the timber species I have planted are increasing in value every day. I have thousands of timber trees planted out, all of which are growing at a steady rate. Some of them are quite valuable, already. My children will get many thousands of board feet of cedar, mahogany, santa maria, teak and samwood. Where does that fit into the argument that annuals produce more than perennials?

As for closed nutrient cycles, I import few things directly for the farm except for the occasional bag of distressed, weevil infested corn I get from my campadre in the village, or rice bran from the rice mill, for my pigs, and their manure is cycled back to the land. We live up a river with no road, and can only come or go in a small motorless dug out canoe. We cannot haul in much inputs!

My pigs eat the food from the farm, mostly fallen food from trees, and return it to the farm in the form of composted pig manure. We bring in foods we do not grow, and of course defecate and urinate and that ends up as part of our soils, too.. I have animals passing through my land from the Columbia Forest Reserve to the Columbia river, my land comprising a biological corridor between intact habitat and water, since the karst landscape to the north of us has no surface water in our dry season. We get large cats, jaguars, ocelots, margays, puma and janguarundi passing through, as well as collared and white lipped peccaries (who ate my bean crop this year), and brocket deer and tapir, as well as many species of birds. Some of these bring in seed for the farm, most of the Chaemadora tepejilote and Chaemadore ernesti-augusti initially was brought in by birds, and the important food, much prized by the Kekchi Maya, “chi’kai” (Calathea lutea) was also originally introduced by birds, but we now encourage it and it is a wonderful forest floor food that produces several pounds of food a day for us in its season. Because of intensive management and rotational practices, the chi’kai is largely absent from areas close to expanding village populatoins. Birds also propagate some of the chili peppers we eat, passing the seeds to the soil while roosting. Some of the predators, take things from us, especially poultry, and, over the years, too many dogs for me to think about without some degree of melancholy, and the wild pigs raid our corn, beans, cocoyam and cassava.

So, at this point we are producing significant amounts of food, fiber, timber, medicinals and marketable crops. Because we have so many species, over 500, the harvest cycle is spread out through the year, and with our timber species, some of them are spread out into the future in a time span of decades. Large parts of the farm can be managed by one person working alone with a LOT of free time left over. Those parts also produce a lot of food in relation to the amount of energy invested to maintain them.

The ecosystem services this farm provides, habitat for animals, a biological corridor betwen the forest and the river, carbon sequestration, soil and soil moisture rentention, they have value, as well, but those values, important, yes, critical in the face of rapidly expandinf population densities, yes, are very diffocult to quantify when the dominant paradigm in agricultural economics is the limited kilograms per hectare x dollar value per kilogram versus the cost of labour and inputs and capitalization model that most people assess value with.

Where we are, now, as a species, the old economic model is woefully inadequate. Comparing annual crops to perennial crops is, in my opinion, also inadequate. In terms of straight calorie expenditure to calories returned, and the possibility of high value crops like vanilla, cardamom, turmeric, cacao and other crops, the increasing value of timber species, annual crops cannot come close to the value of a stacked polyculture. In terms of ecosystem service replication in human centered habitat that meets food, fiber, medicinal, timber and marketable needs, there is no valid comparison to annual production.


Christopher NesbittMaya Mountain Research Farm sign

Maya Mountain Research Farm

San Pedro Columbia, Toledo

PO 153 Punta Gorda Town, Toledo


Central America

Country code 501-630-4386


Leave a Reply