Note: The following units are used in this document:   
1 hectare (ha) = 10,000 m²  =  2.47 acres
1 km² = 100 hectares (ha)
1 metric ton = 2000 kilograms
1 gigaton = 109 metric tons
1 petagram (pg) = 1015 grams (gm)

Unless otherwise stated, all units used in this document are metric.

"The world we want is considerably different from the one we are creating." (Houghton, 1995)

"Our findings show one compelling reason why this [current development practices not providing human benefits as they should] is the case: our relentless conversion and degradation of remaining natural habitats is eroding overall human welfare for short-term private gain." (Balmford, et al., 2002).

What should be the goal of conservation in the 21^st century? One can think of many, but perhaps the most critical are to maintain vital ecosystem "services" (ecological processes which support life on earth), to preserve a considerable degree of genetic diversity, both of variety of species and within species, and to use the resources of nature sustainably. We must plan our conservation efforts with the following points in mind and so choose the areas for maximum conservation effort with care: (i) the less damage, the greater the number of species the forest can sustain (ii) the more species, the greater the value of the forest for conservation purposes (iii) the larger the original area of the forest, the greater the number of species (iv) the more diverse the topography and soil, the greater the number of species (v) each different topographical area will probably contain endemic species (vi) forests with valuable species (forest products) need protection (vii) in conserving species, a viable population of each species is necessary (viii) a sufficient area for each species is essential (ix) all types of forest must be preserved (lowland, montane, peatswamp, dry, floods, etc.) (Jacobs, 1988).

Tropical rainforests, undisturbed, contain as much diversity as they can under present global conditions. We must remember that any change, therefore, will lead to impoverishment of diversity. All of the above criteria must be accompanied, first of all, by a reduction in human population growth rates, and the sustainable use of our global "life-support systems" - food, water, air, energy, "sink" capacities. Our emphases always have been on human life, but without valuing the nonhuman aspects of our environment - including the other living species - it will be impossible to make improvements in, or, indeed, keep stable, the "human condition." We must realize that our current development patterns are impossible to sustain and that our highly resilient life-support systems will eventually reach their limits (and some may already have done so). An improvement in human lives and a reduction in poverty must come from the redistribution of goods and services, qualitative rather than quantitative development, population stabilization, and community action. As Goodland (1995) stated, "The growth debate emphasizes the scale of the growing economic subsystem relative to the finite ecosystem." We must realize that the earth’s ecosystems are, indeed, finite.

The world desires wood; the timber industry is here to supply it. An increasing human population desires food; farmers are here to supply it. How can the demands for farmland, wood and wood products be reconciled with the preservation of forests? Most deforestation, as has been pointed out, results from exploitation of forests to generate foreign exchange or to provide subsistence. This means that rainforests can only be saved by finding alternatives to subsistence goods (food, materials for construction, fuels) taken from rainforests, by providing alternatives to current export goods, and by using forest products sustainably.

Many solutions to the problem of tropical rain forest loss have been proposed, but it will take a combination of many actions to slow the destruction of tropical rainforests. There is disagreement about how forests might best be preserved, and how to reconcile the need for forest products with the need for conservation. Formerly, protection of old-growth forests by keeping them intact was the preferred model. More recently, in the 1980's, the concept of forest dynamism and the potential for regeneration was ascendant. Now that so much forest has disappeared, stress has been laid on the ability of fragments, logged forests and second-growth forests to conserve at least some biodiversity. Since it seems impossible to keep humans away from forests, we must discover ways to preserve them in conjunction with some human activities. What are some means which can be used to save rainforests?

1) Drastic reduction of human population growth: Human population growth drives a great deal of expansionist activity. World population is growing at approximately 2% per year, which will lead to a human population at least 50% higher by 2050, and, quite possibly, more. Many current estimates hover around a global human population of 11 billion by that time, more than 80% higher than the population today. As populations grow, they push farther and farther into forested areas - for agricultural land, for fuel, for game, for timber. Under these circumstances, the global demand for food would at least double. In Asia and Africa, food requirements are projected to increase by a factor of five for the former, and 2.3 for the latter. In some countries with high population growth rates, the demand may be seven times as much by 2030 as at present (Daily, et al., 1998). The demand for fabricated goods, many of which are made from tropical raw materials, will also increase as the world population grows and as people become more affluent, both in developed and less-developed countries. Rapidly-growing human populations may be directly responsible for more than three-quarters of tropical forest loss. Nothing is more vital than the necessity to reduce human population growth if rainforests are to survive for even the next fifty years.

2) Improvement of land use: Land use practices must be improved for timber production, agriculture, housing, and infrastructure. We must improve the productivity of agricultural land, pasture, and tree plantations, and increase recycling efforts to lower the demand for raw materials and products from tropical forests. We also need to reduce urban sprawl, which eats up agricultural land and increases the pressure to convert forest land.

a. Eliminate incursions into pristine rainforests: This is more feasible in countries where population growth is low and where new facilities can be built on the sites of deteriorating ones without using any additional land. This will be considerably more difficult in countries where the population is rapidly growing, and will require considerable funding from richer countries. Even in countries with little population growth, consumerist and expansionist attitudes must change to reduce pressures on rainforests from consumer demand.

b. Utilize second-growth forests for timber production: Increasing interest is being given to using second-growth forests for timber production rather than invading virgin forests. These second-growth forests represent about two-thirds of the global forest cover. Since they are mainly reforested logged or agricultural land, they are frequently fragmented and close to roads, so that access to them is reasonably easy. Many of the tree species found in these forests are not as valuable in the marketplace as currently-utilized species, but are probably just as useful, and they will become more desirable as the supply of prime timber declines.

c. Reforestation: Much previously-forested land - probably hundreds of millions of hectares - can be reforested, although it would take many years before their biodiversity and complex ecological webs are recovered. Many other areas could regenerate naturally if they are left fallow and are not converted to other purposes.

Nevertheless, reforestation is not a simple proposition. Logging and other land use practices damage the soil and vegetation to a greater or lesser extent. In some cases, soil deterioration is so great that the soil turns to laterite (see Part I, Section L2). More often, the soil becomes so degraded through compaction or loss of nutrients and other factors that forest species cannot reestablish themselves, and the land becomes scrub or savannah.

Where logging has occurred and the land has not been converted for agricultural purposes, a forest can sometimes regenerate if the cut area is not too large. The most difficult areas to reforest are logging road tracks (and any other areas where topsoil is removed), which are highly compacted by the movement of heavy machinery and, in addition, are barren of topsoil and seeds. Compaction, still evident more than 10 years after logging, retards root growth and decreases nutrient availability to plants. Old logging roads can be identified since they tend to be dominated by a few light-demanding species, which compete with the normal forest colonizers. The former tend to be long-lived and may be dominant for decades. In many cases shrubby species invade the roads and interfere with colonization by trees. At the least, plant biodiversity is greatly reduced in compacted areas (Guariguata, 1997), which in turn affects the types and diversity of other species. Generalist bird species will generally be found in these areas, rather than the birds of mature forest.

The degree of logging activity is vital in determining whether forest can regenerate. In Kibale National Park, Uganda, uncut trees in heavily-logged areas suffered high mortality because they were often knocked over by falling trees. Twenty-five years later, tree growth rates in these areas were still reduced in comparison with unlogged areas, and the density of all size classes of trees was reduced. There were many gaps which had not been filled by new tree growth. In this forest there are few light-loving species which can colonize large gaps. This feature may partially explain why gaps in heavily-logged areas come to be dominated by herbs and shrubs, which inhibit the germination and development of tree seedlings. In more lightly-logged areas, tree growth rates for small trees were similar to unlogged areas, but larger trees had elevated growth rates, for unknown reasons. Small gaps still remained from logging activities, but there were many fewer changes in the condition of the forest than in heavily-logged plots (Chapman and Chapman, 1997).

Land which has been converted to agriculture, if abandoned, can also sometimes revert to forest. In Puerto Rico, abandoned shade coffee plantations, sugarcane fields and pastureland were able to reforest relatively quickly, with high biomass and biodiversity within 30 to 50 years. Abandoned coffee lands and pastures had approximately equivalent degrees of biodiversity. On former agricultural land, the type of crop, frequency of plowing and other factors influenced the biological heterogeneity of the recovering forest. For example, regenerating forests which grow on former coffee plantations have similar species, despite differences in altitude, soil type, and other features (Zimmerman, 2001). In Puerto Rico, reforestation has occurred because so many people have migrated elsewhere, and much food is imported, conditions not found in most other tropical countries.

3) Improvement of forest management: Obviously, as far as forests are concerned, the best management practice would be to leave them untouched. However, with the current demand for forest products, and with rapidly increasing human populations in tropical regions, this is an unlikely prospect for the majority of large tracts of forest. For most, some sort of management will be required to prevent their complete demolition. There are a wide variety of forest management practices, often with differing and even incompatible goals. Management for ecosystem and genetic resource protection as a primary goal will coexist uneasily with management for sustainable commercial production. As greenhouse gases in the atmosphere increase and carbon sequestration by forests becomes an important issue, and as populations swell in tropical countries, the tension between these goals will increase. Even when the goals are oriented toward conservation, there are different ways of approaching them. Improving the productivity of managed forests is essential, since increasing the yield of timber and other forest products in such forests is far preferable to removing more virgin forest.

a. Sustainable forestry - how it works: By far the most important management practice at present is "sustainable forestry," often allied with "selective logging" practices. Sustainable forestry has been touted as a way to reconcile the demand for tropical woods with preservation of forests. By carefully removing only as many of the trees or other products as can be replaced relatively quickly and leaving others untouched, and, in addition, planting tree seedlings of desirable species, it is thought that forest ecosystems can be maintained so as to provide many crops of timber or other forest products. After an initial harvest, the forests would be left alone for an extended period of time, after which they could again be harvested for valuable products. In order for this system to be effective, there are several imperative preconditions: One, that the populations of organisms in the forest are able to produce a reproductive surplus; two, that there be adequate relatively undisturbed habitat; three, that soil fertility be maintained; and fourth, that erosion, runoff and road construction be kept to a minimum.

In some contexts, the term "sustainable forestry" is equivalent to "sustained-yield logging," that is, a system in which the harvesting of timber occurs at a rate such that timber will continue to be produced at an equitable rate, or, as Vincent (1995) puts it, sustainability is to "...harvest forests to produce an even flow of timber over time." These methods have been used for many years in European temperate forests, where trees are planted and harvested on regular schedules. This gives a sustained yield of timber, but may not be equivalent to "sustainability" of the forest per forest. Nor are these forests in any sense "natural" forests. In general, forestry in temperate countries operates to replace complex natural ecosystems with ecologically simplified forests managed strictly for maximum timber production. They are timber plantations, not true forests. Alternatively, the goal of sustainable forestry can be not just to provide a constant yield of timber, but to maintain the diversity of forests and to ensure that their ecological services remain intact. In this case, sustainable forestry means that timber is extracted in such a way that the forest can regenerate after logging into a complex ecosystem with most of its former components.

Logging in sustainable forestry may be selective, where only certain species or a limited number of individuals may be taken, or clearcut, when all of the trees in a selected area are removed. Then, in either case, the forest is left to regenerate. Management of regeneration ranges from allowing natural processes to occur unimpeded, or cleaning up the site (raking and burning the slash), or preparing the soil for growth of seedlings - either natural or planted. How successful these practices are in promoting forest regeneration depends on many factors such as the intensity of logging, the size of logged areas, the quality of the soil, and the distribution of logged areas within the forest. Proponents of clearcutting believe that it is more efficient, requires fewer roads per hectare, and leaves the logged areas easier to burn, when burning is used to stimulate regrowth. On the other hand, clearcutting of large areas exposes the soil to erosion, causes loss of biodiversity, and makes the cut-over land unreceptive to reforestation. Some of these effects can be minimized by reducing the size of the cut.

Primary forests in many areas are now reduced to patches surrounded by huge disturbed areas - scrub, agricultural land, pasture, and partially-logged forest. Recovery of forests under these circumstances is quite different from recovery from small gaps because of variability in the size of the cut patch, the number and distribution of species remaining (upon which their reproductive success depends), land-use history, and landscape heterogeneity. Forest regeneration may occur if the land has not been too severely degraded. Almost all of the forest in Singapore had been cleared by the early 1880's. Now, more than 100 years later, some areas have regenerated quite well, although there are no dipterocarps or other large-seeded trees in the new forest, mainly because all of their seed dispersers are extinct on the island. Other, more degraded areas have regenerated much more slowly (Chazdon, 1998a). In Sarapiqui, Costa Rica, secondary forests have formed rapidly following deforestation. After 20 years, the species distribution of trees is similar to that of the original forest, although the trees are not yet as large. In this case, there is primary forest near the deforested area, and there are still many seed-dispersing birds and mammals present (Chazdon, 1998a).

b. Problems with sustainable forestry: Many tropical forest experts feel that since primary forests, along with their enormously complex life webs, have taken many hundreds of years to develop, it is probably preposterous to speak of "sustainable" logging. Any logging will damage these biological webs and degrade the forest. Rice, Gullison and Reid (1997), on the basis of research done in Chimanes Permanent Timber Production Forest in Bolivia, suggest that letting loggers take all of the valuable timber rapidly may make more sense, because after the most valuable species (mahogany, in this forest) is gone, the forest will not be of much interest to the logging companies. It may be objected that this will be true only until the market, starved of the most popular species, seeks alternatives in lesser-known species and timber companies return to remove these additional species from the forests. And in other parts of the world, such as Asia and Africa, more than one tree species is sought. Today, as pressures on forests are increasing, "single-species extraction" is becoming rarer as a logging technique. Also, much damage is done during the logging process itself (see below). Thus, it is questionable whether or not sustainable forestry is even an attainable goal.

i) Although only a few trees may be removed per hectare, many other trees may accrue ancillary damage during logging, particularly during mechanized logging. Many smaller trees and non-timber trees sustain damage or are killed when cut trees fall on them, or when bulldozers enter to retrieve cut trees. Selective logging in a diverse, primarily dipterocarp forest in the state of Pahang, Malaysia, resulted in damage to 51% of the trees, although only about 18 trees per hectare (3.3% of total trees) were taken out (Johns, 1991).

ii) Road construction causes erosion and runoff, damages vegetation other than trees, greatly compacts soil, and removes topsoil and seedlings. Logging roads are the most difficult areas to reforest.

iii) Log-loading areas and "skid trails" require clearing of substantial areas of forest.

iv) Climbing plants such as strangler figs and lianas are damaged, or more often, deliberately cut to reduce the probability of pulling down other trees. These plants are essential components of the forest, providing food and shelter for canopy-dwelling species and supporting larger trees.

v). From a practical standpoint, timber quotas are frequently ignored. In Chimanes logging companies basically "mined" the forest for the most valuable species, mahogany, and extracted all specimens of this species which it could locate (Rice, Gullison & Reid, 1997).

vi) The emphasis on certain valuable tree species endangers biodiversity, even if reforestation is permitted. For instance, at Chimanes, the effort to produce mahogany requires thinning of other species, so much so that the forest does not regenerate in a natural fashion and much biodiversity is lost. Too, since mahogany is a climax species and is shade-intolerant, great swaths of other species of trees are cut down to provide gaps where mahogany seedlings can be planted.

vii) Some studies indicate that many animal species are disturbed by even modest logging (Venezuelan antbirds, for instance).

viii) Logging roads provide an ingress for farmers, miners, poachers, and commercial hunters to the forest. They and loggers usually subsist on hunting, which decimates local populations of primates and other large mammals.

ix) Sustainable forestry may not be practiced because it is more profitable to cut down all of the valuable trees quickly (mahogany, for example) than to invest in a slow-growing capital (i.e., mahogany seedlings) and wait for them to mature. There is no economic incentive to practice sustainable management, as unrestricted logging is two to five times more profitable than sustainable forestry.

x) Governments are often unhelpful. They provide little or no money or personnel to monitor timber companies (which are often politically connected and powerful) and, on the contrary, often prefer a short-term benefit in foreign trade to long-term sustainability of their resources.

xi) The effects on plants and animals of sustainable forestry are not well known. On the negative side, many of the pioneer trees which will occupy the gaps created in the forest are not species which can be used for food by animals, and the full diversity of food types will not return until the climax forest is attained (more than 80 years for dipterocarp forests). A study in French Guiana indicated that, for 10 years after selective logging had removed three or fewer trees per hectare from forest plots, biodiversity and population size of plants and animals were reduced by 25% in comparison with unlogged areas (Alper, 1993). On the other hand, there are suggestions that the remaining trees and plants may produce more leaves and fruits, which might offset the loss of other resources. Downed trees themselves are important sources of food and shelter.

There is much we do not know about the potentialities of sustained forestry. (For an excellent discussion of sustainable forestry and an extensive bibliography, see Hartshorn, 1995.)

c. Improving forest management: Less-destructive methods of forest management and logging can certainly be instituted by improving forest management before, during and after resource extraction. What procedures need to be undertaken by forest managers to ensure forest regeneration and preservation?

i) Preservation of forest reproduction: There is little information on the reproductive patterns of valuable nondomesticated tropical species. Therefore, many individual trees may be harvested prior to maturity. If near-adult organisms are taken before they produce seeds, that species may become endangered. The timing of harvesting is also important, as fruiting and setting seed are seasonal for many species. Harvesting should be done after the reproductive season for that species is over. For example, in the Amazon, mahogany is usually cut at the end of the dry season, although it reproduces early in the rainy season which follows. This is a most destructive practice. Also, any species which are keystone species or extremely important in the ecosystem should not be harvested frequently, if at all. For instance, the fruits of figs and palms are essential foods for many animals. Just as critical is the capacity of a species for reproduction. Some species can regenerate by sprouting from stumps and thereby survive cutting. Other species, such as bamboo and some palms, reproduce only after many sterile years and then die, thus cutting them prior to reproduction is fatal for that population.

ii) Allowing natural regeneration: Natural regeneration is required for sustainable forestry. Regeneration in a natural way requires some gaps in the forest, although these cannot be too extensive. This mimics the natural course of events, where the death of trees or wind or fire open gaps in the canopy. It is also necessary to know something of the requirements of seeds for germination and of seedlings for development. In the tropics seedlings of any particular species are often at a low density, and seeds may have only a short period of viability. Many seeds need to pass through an animal gut or to be exposed to heat or moisture in order to germinate. Then, too, many species fruit only sporadically. Thus there are many requirements for adequate reproduction, which vary according to species. The low density of individuals of any one species in tropical forests and their frequently poor capacity for reproduction is a problem in sustainable forestry. Seedlings of the desired species may be planted, but it is often not economically viable to do so, as a great deal of forest may have to be opened for light (as in the case of mahogany), or extensive and labor-intensive weeding may need to be done

. iii) Maintaining ecosystem functions and biodiversity: Forestry is generally thought to be incompatible with the preservation of ecosystem function and natural biodiversity. Natural forest areas need to be maintained for this purpose, a need often ignored by farmers, logging companies, and government planners. Careful management is required to prevent land degradation. Where logging is done so as to minimize erosion and soil compaction, nutrient losses can be replaced during natural regeneration. The use of heavy equipment must be used very sparingly, if at all, and logging roads must be carefully designed to take into account soil type and land topography. Especially, the harvesting cycle must be lengthy enough to allow for complete regeneration.

Immediately after logging, biodiversity drops precipitously. It can rebound to some extent, although the species returning may not be the same as in the original forest. Forest exploitation may be diversified to minimize damage to these areas. For example, sensitive areas should be preserved untouched, and areas which are to be logged can be dispersed and kept as restricted as possible. "Cluster-cutting" from nuclei, or "strip-cutting" (clear-cutting narrow strips) may permit more natural regeneration than traditional cutting. Some areas may be managed intensively for production, others modestly for conservation and ecosystem preservation. If logged areas are converted to agricultural systems, much biodiversity is lost forever, although some may be maintained if many native species are planted. Agroforestry may be useful, although even when local plants are used, biodiversity is much reduced from that in primary forest areas.

iv) Using less-destructive and innovative means to provide timber:

a. Increasing the productivity of popular timber species: The slow growth rate of many large hardwood tree species is an impediment to sustainable forestry. Species which colonize gaps (pioneer species) and which grow rapidly can reach the canopy within 10 to 20 years and could be utilized for sustainable timber production. Plantations of such species could provide a large percentage of needed timber in the foreseeable future. Species which provide high yields can also be planted. In the Philippines, scientists have developed giant species of a timber tree, Leucaena, which produce three times the yield of the natural species (Spears, 1988). If plantations of these giant trees were to be established, cutting of virgin forests could be reduced or eliminated.

b. Utilize less-known tree species for timber: Typically, only a few well-known (and hence valuable) species are utilized by the timber industry. However, many more species could be exploited to increase the yield per hectare of logged forest land. For example, there are 200 potential timber species in Asian rainforests, but only a few species are harvested. If more of these lesser-known and little-utilized species were marketed, rather than being discarded and destroyed, less forest would need to be cut to provide timber resources. In Malaysia, between 1977 and 1981, utilization of lesser-known species was doubled and these species comprised 27% of the log intake of plywood and veneer mills (Spears, 1988).

c. Instituting less-destructive logging activity, using the following methods:

i) Make a forest inventory and map prior to logging. This reduces waste because logging activities can be planned and felled trees located.

ii) Plan the movements of machinery. According to Uhl, et al., (1997), machinery has an impact on approximately 25% of the land area during logging, but careful planning can minimize the area in which machinery moves.

iii) Cut vines two years prior to logging so that when trees are cut, other trees connected to them by vines are not pulled down.

iv) Train loggers in proper logging techniques. One can reduce waste by a factor of three in this way, since improper techniques are a major cause of tree wastage.

v) Reduce machine operating time by planning and preflagging skid trails.

vi) Girdle undesirable trees to provide space for commercial species.

vii) Limit the size of cuts so that natural reforestation can occur and so that effects on biodiversity are minimized.

viii) Exclude clear-cutting of forests and their conversion to pastures and other nonsustainable agricultural uses.

ix) Monitor and assess the condition of the forest in terms of canopy, seedling availability, and biodiversity.

By utilizing these procedures one might reduce harvesting cycles from 70-100 years to 30-40 years. The increase in production from already-logged areas would reduce the need to open up new areas to logging. Healthy populations of commercial species must, however, be maintained in the logged areas.

v) Legal action: Among the actions which can be taken by the forestry sector to conserve rainforests are the following:

a. Adopt and enforce simplified, effective, enforceable forestry codes. Current regulations are conflicting, confusing, and worst of all, seldom adhered to. Limits might be set on the number of trees extracted, cutting cycles specified, and fire zones established to minimize the danger of fires.

b. Regulate permissible areas for logging. For example, one should exclude areas with high degrees of endemism and biodiversity, as well as parks and areas reserved for indigenous peoples. Logging is to be allowed only in designated areas.

c. Reduce levels of timber harvesting to a sustainable level.

d. Establish an "ecology tax," where the costs of forest management are passed on to consumers.

e. Mandate forest management by law. Conservation and biodiversity preservation programs should be incorporated into forest management programs. Logging, where continued, should be low-impact and should preserve forest diversity.

These measures can only be effective where there is strict enforcement of regulations.

vi) Approaches based on traditional agroforestry and shifting agriculture: Traditional agroforestry utilizes few materials external to the forest itself and relies on natural replenishment of soils by decomposition. In agroforestry, crops (food for man and animal, medicines, construction materials) are typically raised under the forest canopy. Small gaps are opened in the forest by cutting some large trees, burning the area, and planting several sequences of crops until the soil loses fertility. Frequently trees such as rubber and various fruits are planted. After this, the area is abandoned and left to regenerate for a number of years. This type of agriculture has been practiced for many centuries, so that probably much tropical forest which appears to be virgin forest is actually secondary forest. Such forests often have higher concentrations of fruit, rubber and other trees than the primary, original forest, and can continue to provide food and income for local inhabitants even during the fallow period. This type of agriculture can be successful only when the human population is small and no other use is being made of the forest.

d. Why is sustainable forestry relatively rarely attempted in tropical areas, even if the timber companies are in agreement with its principles?

i) Unrealistic expectations of yield: The first yields will be at non-sustainable levels (large), as they are taken from pristine forest. Subsequent harvest yields will be lower, as they are taken from a population of growing trees, not mature ones. Unfortunately, the initial (large) harvest leads to unrealistic expectations of yield from the regenerating forest. Timber companies have developed infrastructures to harvest a large yield, and when this does not materialize, they may harvest unsustainably to recoup their investment. This type of exploitation will reduce future yields even more, however.

ii) There is often little comprehension in the harvesting industry that natural populations will fluctuate over time, even without any harvesting. This leads them to attempt a constant harvest rate, regardless of population dynamics.

iii) The tendency of people using rainforest products is to overexploit them; this is particularly true of large companies. Successful sustainable forestry is usually accomplished by small-scale, private or community institutions.

e. What actually happens - Case Study: Sabah, a Malaysian state in north Borneo, has forests with a high density of commercially-useful tree species, and logging has been extremely intensive. In logging concessions, all commercial timber of a useful diameter ("selective logging") is cut, resulting in a very high extraction rate (8 - 14 trees per hectare). What remains after extraction is a mosaic of areas - undisturbed forest (about 20%); disturbed forest where there is some ancillary tree damage and some disruption to the canopy (20%); areas containing debris, where both canopy and the understory are disturbed (30%); skid trails, which are bulldozed through the forest to provide entry for equipment and an exit route for timber; and log landing areas where vegetation, litter and topsoil have been removed (30%). After logging, the soil is left compacted, mainly because of the heavy machinery used. The effects of this type of logging operation are: (Data below are from Nussbaum, Anderson, and Spencer, 1995.)

i) An increase in erosion rates and sedimentation of waterways. Erosion is low in undisturbed forest, but increases considerably after logging. After one year, the increment in erosion in logged areas was approximately 3½ times that of unlogged areas, because most of the exposed topsoil had been washed away. Suspended sediment in waters near the logged areas was eighteen times greater than in unlogged areas.

ii) An increase in soil compaction: Soil compaction is so great where heavy machinery has passed that researchers could not measure any infiltration into soil on a log landing after two hours of rain. In undisturbed areas, the infiltration of rainfall was measured as 154 mm/hour.

iii) Alterations in vegetation: Vegetation was damaged, canopy cover was decreased and, in large gaps, the growth of vines, climbing bamboos and other sun-loving species overwhelmed the growth of dipterocarp (hardwood) seedlings which are light-loving but do not grow rapidly. Where less than 50% of the canopy remained, three-quarters of planted dipterocarp seedlings were overgrown by vines, as compared to about one-tenth in undisturbed areas.

iv) A reduction in biodiversity in comparison with undisturbed areas:

f. "New forestry": Still, some projects, based on "new forestry," give hope for the future. According to the principles of "new forestry," forests should be managed in such a way as to maintain the forest ecosystems, rather than to follow the traditional idea of forests as a source of timber only. Under these principles, forests would no longer be managed to maximize wood production by homogenizing ecosystems. In new forestry, the maintenance of biodiversity is essential, including ground level and underground biodiversity. Even the detritus on the forest floor - dead vegetation and trees - must be maintained. This concept has yet to percolate to tropical forest management, for the most part, although there have been a few attempts.

i) The Programme for Belize, under the auspices of The Nature Conservancy and the Massachusetts Audubon Society, in the Rio Bravo Conservation and Management Area, is an example of new forestry principles. Half of the land is set aside as a nature reserve. In the other half, some cutting is permitted, but all trees more than 12 inches in diameter are mapped. Cutting here is done very carefully to ensure that trees fall and log removal is done in such a way as to do the least damage. PfB then sells the products and the profits are used to maintain the reserve. PfB also works with local inhabitants, hiring them as foresters and rangers, and helping villagers begin small businesses such as honey and fruit production (Clancy, 1998).

ii) Portico S.A. (Brazil) is a company which manages its own forests in an enlightened manner. It has incorporated local farms into its management policy and provided employment for local people. It has a management plan for each of its properties, based on inventories of tree species, their sizes and distribution, and regeneration status. It uses less-destructive logging methods (non-mechanical) and trains its staff how to minimize damage to remaining forest. Moreover, it is involved in research and monitoring and conservation is a component in its policies. No hunting is allowed on company land and its properties are patrolled for compliance. Although these policies increase the raw material price, the additional costs are absorbed by the vertical integration of the company.

iii) In Peru, the Yanesha Forestry Coop (a joint venture of USAID and the Peruvian government) has a community-based project in the central Selva area, which is unsuitable for agriculture and livestock raising. The Cooperative is a group of local people who own and process forest resources and cut timber so as to maintain diversity. Clear-cutting is done in narrow strips to leave large forest areas intact; this attempts to mimic gap formation and to stimulate natural forest regeneration. Trees are to be cut on a 30 to 40-year rotation cycle.

(Information on Portico S.A. and Yanesha from Hartshorn, [1995]; on Yanesha, in Postel and Ryan, [1991]).

However, new forestry - or any kind of forestry - is NOT a substitute for forest preservation. Unfortunately, we know little about how to manage forests in a sustainable manner. Any type of management must realize that forests in the future cannot continue to produce timber at the current rate. Even with the use of "new forestry" and sustainable forestry, reductions in the volume of timber extraction will be required. This, of course, is a substantial economic and political issue.

g. Plantation forestry: Many large international timber companies have established tree plantations in the tropics. Trees in plantations tend to be cut while still young, and are therefore less desirable than trees from primary forests. Only 12% of tropical plantations are stocked with hardwood species which provide wood comparable to wood from natural forests (Postel and Ryan, 1991). Nevertheless, these plantations are often highly productive, and can reduce the need to log old-growth forests. Often the trees planted are hybrids and/or nonnative species. The most popular species are Pinus (pines), Eucalyptus, and Acacia. These species, since they are not in their native areas, are less subject to disease than otherwise, but they are nevertheless attacked by a variety of pathogens, both native and those introduced with them. These disease agents include root, leaf, wood and seed pathogens, which find abundant hosts in these plantations. A rust, Puccinia psidii, native to the Neotropics, has begun to infest stands of Eucalyptus (an Australian tree) on tree plantations in South America (Wingfield, et al., 2001). This and similar infestations reduce the productivity of the plantations and their role in the global timber trade. As yet the infestations have been limited, but new diseases are continually appearing and can be expected to have significant effects on timber production in managed plantations. In addition, species introduced to areas outside their normal range can "escape" and become pests ("weeds").

Many of these problems can be overcome. In the Subri Forest Reserve in Ghana, much of the vegetation and small trees is retained during logging and the establishment of plantations. This method leaves mycorrhizal associations at least partially intact and reduces erosion. Some natural forest is left to minimize the transmission of pathogens (Postel and Ryan, 1991). Unfortunately, most plantations are not so well managed. According to these authors, mixtures of fast- and slow-growing species and fibrous and hardwood species would give a competitive yield of timber and pulp. They estimate that such plantations, on only 5% of the land already cleared of tropical rainforests, could provide twice as much wood as the current harvest.

4) Institution of changes in public policies toward forests: Many public policies encourage and promote the destruction of rainforests for economic or other reasons. These policies (and the great amounts of money which they bring with them) are extremely powerful in comparison to efforts and measures to protect, restore or regenerate natural systems. Many changes are necessary with regard to government land policies in most tropical nations, such as:

a. Government commitment to land reform and redistribution: Much of the land in tropical countries is concentrated in the hands of the few, and must be redistributed to rectify the extreme imbalance in land ownership. Land ownership must not be a privilege simply for the wealthy. The government must provide secure land title for all, without the proviso that the land first be "developed" (that is, deforested).

b. Government commitment to change national agricultural and colonization policies, which offer land free or at low cost to any settlers who "improve" forest land by logging it and establishing farms and pastures: Agricultural subsidies encourage farmers to utilize marginal lands and to clear forests for cropland. Farmers are also encouraged to use pesticides and fertilizers and to irrigate with often scarce water resources. Governments should instead support intensive rather than extensive agriculture to maximize production on already converted land. These subsidies and policies have been discussed extensively in Part III.

c. Government commitment to eliminate or reorient agricultural development plans involving rainforest removal toward better utilization of already-converted land: The government should link agricultural fiscal incentives to soil, water, and timber conservation measures. Along with these efforts, the government should support research on agroforestry and intensive agricultural practices, so as to obtain maximum benefit from land already under agriculture.

d. Government commitment to reform in the granting process for timber concessions, and the elimination of other fiscal incentives to forest destruction: At the moment, governments in most tropical countries have tax and fiscal policies which favor the timber industry (and other activities which impinge on forests, such as agriculture) at the expense of forests and ecological protection. As one sound alternative, new fee systems could be based, not on the area of the logging concession, but on the volume of trees felled. The value of the species of timber taken should also be factored into the concession fees. Concessions should be given by competitive bidding (currently, many, if not most, concessions are given on the basis of political connections, payoffs to officials, and patronage), and the terms of logging concessions should be lengthened to encourage responsible timber extraction and to reduce the incentives to "cut and run." The issuing of timber concessions could additionally be linked to requirements for reforestation after the extraction of valuable timber species.

e. Government commitment to establish fiscal policies to encourage forest retention, not only in the public but also in the private sector: The environment must be considered when policies and economic decisions are made. The separation of economic policymaking from environmental policymaking must be ended, a common occurrence since government agencies are often isolated from one another. Until economic agencies become responsible for the environmental impacts of their policies, little progress can be made. At present, the governments of many tropical countries provide substantial tax benefits for deforestation, a recipe for catastrophe. On the contrary, timber companies must be taxed in line with the real value of the timber, for example, rather than at a "discounted" rate. Another mechanism to tax resource utilization fairly would be to tax resource depletion while lowering general corporate taxes. Governments must also eliminate the fiscal incentives for clearing land for farms and, particularly, ranching.

f. Government commitment to provide the necessary tools for sustainable logging: This should include technical information, legal changes to support this type of operation, and infrastructure for timber management. Timber concessions should be issued for long terms (more than one cutting "cycle"), so that timber companies would have an incentive for reforestation and sustainable logging management.

g. Government commitment to encourage reforestation of deforested land: This might require giving timber companies low-interest loans to fund regeneration and protect biodiversity, for example. Additionally, timber companies and/or individuals could be given tax cuts for reclamation and reforestation efforts. Basically, cheaper capital is required to provide incentives for reforestation. This could be provided in part by international institutions (which instead often support wide-scale destructive "development" projects, such as dams). Timber concessions should be issued with the proviso that reforestation will be undertaken by the timber company.

h. Government commitment to establish and expand reservations for indigenous populations: These areas must be protected from incursions by others, particularly commercial logging, fishing, mining and agricultural interests.

i. Government commitment to set aside large areas as fully protected ecological reserves with no encroachment permitted: Funds must be provided for monitoring, maintaining, and policing these reserves. This must include the provision of technical and scientific assistance in denoting and maintaining reserves, as well as providing trained staff and rangers to assist with and monitor these reserves.

j. Government commitment to assisting local conservation efforts and organizations, rather than opposing their activities with its policies.

k. Government commitment to return a higher proportion of forest tax revenues to the management of forest lands. At present, few of these revenues are returned to benefit the forests or the people living near them.

l. Government commitment to form family planning and population policy organizations, and commitment to establish and support health and family planning clinics.

Of course, such solutions are problematic, as governments are often unwilling to lose the immediate financial benefits of deforestation. They so not wish to think about future losses resulting from their policies. Governments also frequently have relatively little control over the situation on frontiers, and lack political support for conservation. There is enormous pressure to deforest from local politicians and businessmen, as well as multinational corporations, which move into tropical countries under substantial government incentives.

m. Case Studies:

i) China has recently issued some new forest policies, although its previous policies had been mainly exploitative and have degraded and destroyed huge areas of former forest land. Although Chinese forests are mainly temperate, some of the management plans proposed may be instructive. These new policies include technical training and education of forestry officials in new methods of forest management at local and provincial levels; land management planning; mandatory conversion of marginal farmlands to forest; the resettlement and reeducation of forest workers and gatherers; establishing shares in private ownership; and funding research on forestry issues. As an example of novel land management planning, forest lands will be divided into commercial forests and nature preserves, the latter being expanded and protected. Where marginal lands are returned to forest, the loss of farmland will be compensated for by providing 30 years of free food to farmers who have lost their farms under this policy. Government-owned forests will be contracted to communities for management and for use for non-timber products, such as mushrooms and ferns. Under this new policy, forest plantations will provide timber, while natural forests will remain as reserves (Zhang, et al., 2000).

ii) Brazil, about which much has been said in this document, may be realizing the consequences of its current policies. Recently, it has adopted an Amazon forest policy which, if implemented, would conserve large areas of the Amazon rainforest. This policy has provisions which include protecting 80% of rural properties as forest reserves, requirements for licensing logging, burning and deforestation, provision of coordinates for areas to be cleared, prohibition of burning during the dry season, and other regulatory measures. To enforce these provisions is and will be difficult, despite the passage of an "environmental crimes" bill in 1998, due to the vast area of the Amazon forest and the relative independence of Brazilian states. However, the environmental crimes act provides penalties, and there is some attempt to coordinate state and municipal governments in environmental planning, through the G-7 Pilot Program for Conservation of Brazilian Rainforests. The government is also attempting to control access to federal lands (Nepstad, et al., 2002). For the success of these projects, there must be successful integration at various levels of government, adequate funding, and sufficient monitoring capacity, not to mention the political will to resist the forces which have conspired to destroy so much of the Brazilian Amazon.

5) Protection of forest land: Forests must be well protected at various levels, but, in practice, this is rarely accomplished. Although it is essential, the planning of protected areas has been subject to political or economic pressures and has neither been systematic nor effective. Protected areas must be placed in regions where they will contain the full measure of biodiversity of that particular type of ecosystem. One can easily understand that this is not often accomplished, particularly since resource extraction seems economically more attractive in these rich areas. Unfortunately, the areas richest in biodiversity and ecological importance are also those which contain resources most desirable to humans. Thus, reserves are established in places which are relatively barren, or are difficult of access, or appear unviable for profitable development or extraction. There may also be conflicting goals among the advocates of reserves - some may wish remote areas to be preserved for aesthetic purposes, others may wish to preserve areas of high diversity or entire ecosystems. (For an excellent discussion of the issues involved in establishing reserves, see Margules & Pressey, 2000; also, Inamdar, et al., 1999.) Worldwide, only US$6.5 billion is spent annually on established reserves, and half of this is dispensed in the United States. Only a bit more than US$3 billion being spent by the rest of the world for this purpose, a trivial sum.

a. Protect some areas completely (and effectively): This is especially necessary for those areas with poor soils, those which are inherently fragile (such as slopes, which have great potential for erosion and degradation), and those with high degrees of biodiversity and endemism. Even though it is very difficult to preserve large areas of intact rainforest (for economic, political, and social reasons), it is absolutely essential. Huge areas need to be kept roadless, unmanaged, and untrammeled insofar as possible, although there is constant pressure to use forest land (even protected land) for every purpose - agriculture, aquaculture, logging, resource extraction, hunting. Unfortunately, protected areas are often so in name only, and are rarely well-monitored. Furthermore, they are ordinarily not large enough to maintain viable populations of large animals and trees. The case of the Wolong Nature Reserve for giant pandas in China is an instructive case. Since the establishment of this protected area, the situation of the panda has in fact worsened, due to continued forest loss and fragmentation. The number of pandas within the reserve dropped from 145 in 1974 to 72 in 1986. The increase in human population within the reserve has led to an increase in economic activity - agriculture, fuelwood collection, logging, road construction, tourism, and the collection of herbal medicines, with consequent habitat loss for the pandas. Since the people in the reserve are dependent upon wood for fuel, unlike people outside of the reserve who have switched to other forms of energy, such as coal and electricity, the situation within the reserve is actually worse than it is outside of its borders (Liu, et al., 2001).

b. Increase the size of protected forest areas: Rather than conserving numerous fragmented smaller areas, it is essential to protect large intact tracts of forest. Larger areas contain more species and can maintain viable populations of large animals and plants, while smaller areas are more prone to episodes of extinction and invasion. Very large conserved areas are required to ensure that disturbances affect only a small part of the forest area. Then, too, there are many species (both plant and animal) which cannot survive in disturbed areas or near forest edges, and which will be lost if forests are reduced to remnants. Other species, as previously mentioned, require great tracts of intact forest for their foraging, reproductive, and/or shelter requirements. The forest cannot maintain its many and complex levels of organization without large intact areas, although some smaller protected areas are needed to preserve a variety of habitats. The loss of some species may be relatively inconsequential, but loss of others which play key ecological roles is disastrous and results in the disruption of ecological webs. Particularly, forested areas around timber concessions should be preserved to sustain biodiversity and the structure of the original forest and to assist in forest regeneration. It has been estimated that, in Sarawak, protected reserves between 10 million and 100 million hectares would be necessary to preserve full biodiversity and forest functions, but none of the established reserves is larger than 170,000 hectares and most are much smaller, even tiny. Globally, only 3.3% of the natural forest cover lies in reserves (Amelung, Torsten and Diehl, 1992).

c. Protect forest edges against damage: Forest edges are particularly vulnerable, since they are exposed to such adverse occurrences as fire, extraction, hunting, and colonization by exotic species. They are conduits to the inner forest, and as they are damaged, the inner forest becomes more and more threatened. As noted above (Part II, G5f), large trees near edges tend to die at higher than natural rates, among other detrimental effects. It is helpful to establish buffer zones around protected areas to make transitions between natural forest and disrupted areas.

d. Establish "extractive reserves": In such reserves, some limited resource extraction can take place. Rubber tapping, the collection of fruits and medicines, and perhaps even some limited hunting and selective logging might be permitted. These reserves must be managed carefully to prevent unsustainable extraction.

e. Restoration of damaged and disturbed habitats: Although many tropical areas have been converted to pasture or crop land or logged and abandoned, restoration is frequently possible (although not when the soil has become lateritic). Conservation needs to include restoration of damaged lands, which may otherwise become further degraded and useless either as farmland or forest. Often there are some forest organisms or remnants of forest within agricultural areas, and these have the potential for reforesting disturbed areas. For instance, tropical pastures contain many forest organisms (seeds, seedlings, microorganisms, small animals) which aid in restoration. To carry out such programs requires: i) finding areas which are capable of recovery ii) the enlistment of local people to aid in restoration iii) agreement on goals for restoration - whether the reforested area is to act as gene bank, botanical garden, watershed, refuge, or park, and iv) the replacement of private ownership with public, partly because private interests shift, and partly because there will be an economic burden to maintenance which cannot be supplied by the private sector. For such restorations large areas are necessary, as the remnant organisms have usually been scattered, and we have seen the consequences of fragmentation. The forest must be allowed to regenerate over lengthy time periods. Woody plants and many animals will begin to be reestablished within 10 or 20 years, but to restore mature forests takes a century or more. Even if the resulting forest is not identical with the original, it will be more complex than an agricultural system or a tree plantation which might otherwise replace it.

Each type of forest requires a different method of restoration - what works for dry forests may not be effective for rainforests. In areas where much vegetation has been removed, it may be useful to plow the land to loosen the soil, and to provide fertilizer and mulch to restore soil fertility. Elsewhere, simply permitting natural regeneration is justified.

There are many obstacles to renewal, both natural and human. There is the possibility of recurrent fires, of the introduction of foreign organisms, of the loss of many organisms and ecological webs from damaged forests, and there is additionally the long time frame (from a human perspective) required for reforestation to occur. There will be human objections to removing "productive" land from usage, a lack of psychological flexibility about restoration, the low priority most governments hold for restoration, the lack of government interest in rural areas, and a concomitant lack of funds for restoration activities. Many poor tropical countries have few funds available for forest conservation, much less restoration, and international agencies have been reluctant to fund restoration and conservation projects, preferring development schemes such as dams and large agricultural operations. Much money accrues to governments and well-connected individuals from exploiting rainforests, much less from restoration and conservation.

Moreover, we must remember that restoration is uncertain, and will generally result in a forest that is less complex than the original one, at least for a time, and which has a somewhat different species composition. The fact that many previously-forested areas can be restored to some degree should not, however, be a justification for the obliteration of the remaining primary forests.

f. Establish well-protected "zooparks," refuges and botanical gardens: These could act as sanctuaries for a variety of indigenous animal and plant species.

g. Support agroecosystems and agroforestry projects: These are managed systems which can coexist with native species to a greater or lesser extent. Indeed, many agroforestry projects seek to utilize native species rather than to introduce exotics, which can generate many unforeseen difficulties. (See above under reform of agricultural systems; also, follow the link to agroforestry on this website.)

h. Establish effective and permanent management regimes for protected areas, with adequate resources for protection and prevention of illegal incursions.

i. Establish "corridors" to link protected areas, to provide passage for animals between populations and exchange of genetic material among organisms which would otherwise be isolated: Many reserves and parks are too small to provide adequate space to maintain populations of certain species. There is presently a multinational effort (the Mesoamerican Biological Corridor) to establish corridors among conservation areas in Central America. This is based on the sad expectation that most rainforests which are not protected will be lost. Little has so far been accomplished, and corridors are only being established in a few places, partly because there is inadequate funding for the protection of existing parks. The problems, even in setting aside small areas, are great. Politicians have begun to use this project as a means for rural development, rather than conservation, since peasants and indigenous groups feel that they should obtain some benefit from the establishment of corridors. The government of Costa Rica, heretofore a model for conservation in Central America, is currently disinterested, and refuses to make land purchases or to support such ideas as municipal water taxes to pay landowners to reforest watersheds. In some of the proposed corridors, such as one within the Maya Biosphere Reserve (Guatemala/Mexico) illegal logging is uncontrolled, and a route for illegal immigrants to the US and for drug runners lies within it. Mexico also plans to pave a road running through the proposed corridor, although it, as well as Belize, has at least begun to discuss the establishment of corridors (Kaiser, 2001a).

How much land can realistically be preserved? It would be highly desirable to protect 50% of the earth’s terrestrial area, but this would lead to considerable conflicts with human use; even a goal of 1% preservation would do so in some parts of Africa, Asia, and Oceania. But areas of this magnitude would be too small to conserve rainforests in any meaningful (and functional) sense of the word. Several years ago several Central African countries agreed upon the "Yaounde Declaration," which commits them to - among other things - the protection of at least 10% of each country’s forests, as well as to conservation and sustainable forest management. While it is not clear whether or not these goals can be attained, it is remarkable that at last tropical countries are beginning to recognize the problem of deforestation, and, to date, more than 40,000 km² of previously-unprotected forested land has been placed under protection.

As mentioned elsewhere in this document, there are many problems with set-aside areas, from inadequate protection to poaching and other encroachments. A recent study by Bruner, et al.(2001b) assessed the effectiveness of parks in protecting forests. Most parks surveyed had substantially prevented clearing activities, although they did suffer from illegal activities - logging, hunting, grazing and burning. Nevertheless, park areas suffered less impact from these activities than surrounding unprotected areas, particularly with regard to hunting and logging. Those parks which were more effectively guarded and which had other deterrents were in significantly better condition than any non-park areas in the vicinity. Those parks in which local communities were compensated for the demarcation of the park or otherwise benefited from the park were also more successful in preventing encroachment. Since these parks are of considerable benefit to conservation, they need significant increases in support - for personnel, particularly.

6) Improvement of agricultural methods and productivity to reduce the need to cut down additional forest land. Tropical forests cannot be maintained unless agricultural productivity is greatly improved. However, to feed the projected population of the mid-21^st century even at present levels, not to mention a level approaching that of developed countries, agricultural efficiencies would have to be far greater than is currently the case in most countries. We need increases in agricultural productivity of between 1.8% and 3% per year for many years. The average annual growth rate of agricultural productivity in Sub-Saharan Africa between 1991 and 1995 was 2.4%, which must be maintained to prevent a decline in nutritional levels as populations grow (Musters, de Graaf, & ter Keurs, 2000). To provide levels of nutrition equivalent to more developed countries, still higher production growth rates will be necessary. This can occur only if technologies are greatly improved and substantial financial support is provided by wealthier countries. Without such development, conservation strategies are doomed merely by arithmetic.

It is argued that, with proper techniques, including irrigation, sustainable agriculture can be established on former rainforest soils (soybeans in the Brazilian Amazon, for example). Of course, sugar has been raised in the Caribbean and rice has been successfully cropped for thousands of years in Southeast Asia on tropical forest soils. This type of agriculture is possible where soils are fertile, as in the regions of Indonesia which have rich volcanic soils. But elsewhere, as in East Java, only the most marginal agriculture (raising tapioca and other low-nutrient tubers) is possible, and these areas are extremely poor. Some nonvolcanic rainforest areas lie on reasonably fertile soils (especially in deltas) and can sometimes sustain appropriate crops. Recently, archeologists have discovered the remains of ancient agricultural systems in the Beni region of Bolivia. These artefacts lie in seasonally-flooded savannas which have long been thought to be useful only for cattle ranching. However, the many raised agricultural fields, fish ponds and other agricultural constructions indicate that these areas have been productive in the past. Recently, the construction of similar raised fields in savanna areas in Bolivia has permitted the cultivation of tubers, maize, and manioc (Mann, 2000). If more of these areas could be converted to productive agriculture, they could provide a source of land as an alternative to rainforest removal. Often, however, governments encourage and indeed, almost coerce the development of unsuitable agricultural products, although they should not attempt to introduce non-native species and unsuitable crops or domestic animals into forest areas. For example, the Peruvian government has aggressively promoted the introduction of rice cultivation and water buffalo husbandry into many areas of the Amazon rainforest. In most places, rice is a most unsuitable cultivar, and water buffalo have caused serious erosion of riverbanks and destruction of vegetation.

To improve agricultural productivity, a number of things must be accomplished:

a. Reduction of the present rate of degradation and loss of productive farmland due to erosion, salinization, waterlogging, and nutrient depletion: Technologies for these purposes are available, but are little used because of the expense. However, many non-technological methods have been used for years by farmers (contour plowing, abandonment of marginal agricultural lands, planting of wind barriers, fallowing). Erosion can be prevented by the careful selection of appropriate crops, keeping ground cover on the soil, and contour plowing. Irrigation increases crop yields by about 200%, so more land must be irrigated to increase production efficiency, but this uses great quantities of water. More efficient methods must be utilized to prevent water shortages, as only 45% of irrigation water is actually absorbed by plants. Drip irrigation and other efficient delivery systems, better water distribution systems, improved control systems, and raising crops suited to the climate and soil will aid in this endeavor. Excessive or continual irrigation leads to salinization and waterlogging of the soil, which will diminish or destroy its agricultural capacity.

b. Raising the crop yield on current agricultural land, as most land is not producing yields even close to the maximum possible (in part because modern technologies are not used). If the gap between current and potential yield could be bridged, the production of soybeans could be increased by 64%, that of peanuts by 208%, pulses by 472%, and cereals by 170% over a period of several years. The theoretical maximum yield for cereals is 13.4 tons per hectare, but the average cereal yield (1992-1994) was 2.77 tons per hectare - not even close to this figure (Goklany, 1998). There are regional inequities as well. Yields tend to be much lower in the tropical developing countries with large, growing human populations. For instance, in Subsaharan Africa, yields of cereal grains are only one-third of the cereal yield in the northern hemisphere. Increasing yields can raise income for marginal agriculturalists and reduce dependence upon destructive slash-and-burn agricultural methods. This may be done by improving tillage methods to preserve soils and nutrients, which will be more beneficial than removing rain forest to open more agricultural land. Other techniques might include soil testing to determine soil chemistry profiles, crop rotation, nutrient and water management, terracing, instituting appropriate tillage methods for the soil/terrain, crop diversification, and interspersing crops with trees. These methods could reverse the nutrient depletion characteristic of so many cultivated soils in tropical areas. To increase productivity, one must also reduce losses from disease and pests, both during growth periods and after harvest (currently an average of 42% of crop yields is lost to these agents [Goklany, 1998]). Pest and disease-resistant varieties, better storage facilities and improved transportation could help in this, as well as the development of new high-yield crop varieties, suitable to local weather and soil conditions. Except for the "green revolution" with rice, less effort has been put into the development and study of tropical crops than temperate ones.

c. Reforming agricultural practices to be less harmful to forests and forest regeneration: Among these reforms could be reductions in the use of burning, minimizing the use of toxic chemicals, and using swidden land less intensively by increasing fallow times. Zero tillage agriculture should also be utilized. When the soil is left untilled, organic matter is retained, preserving soil fertility and preventing erosion and runoff. Where the soils contain organic matter, forests can often regenerate.

d. Improving the distribution of agricultural products: Distribution systems are extremely unequal in most tropical countries, and often unreliable. Access to food and other agricultural goods must be increased in terms both of availability (delivery) and affordability.

e. Reduction of the environmental impacts of new technologies. To diminish environmental impacts, agricultural management systems must be devised which are suitable for specific areas and crops. This would allow reduction in artificial inputs, so that fertilizer and pesticide use could be considerably reduced.

f. Reformation of policies relating to water management, allocation, and distribution. For instance, governments will frequently subsidize water use for agriculture, reducing incentives for water conservation. That users pay fairly for water is essential (now, frequently, the poor pay more for water than the rich). Many countries have achieved considerable water conservation by this method (Chile, Jordan, India and others), and it could certainly be applied by most tropical countries.

g. Retention of trees as crops to protect water and soil resources. In southern Malaysia, 60% of the forested area has been kept in forest, while the other lands are used for agricultural purposes (Spears, 1988). How much of this land will remain protected with increasing demand for palm oil and other products is questionable, however.

h. Development of agroforestry projects: Cash crops might be raised in small-scale agroforestry plots. Such crops as avocados, papayas, peppers, palm fruits, mangos and many other local crops are being raised in this way in the Amazon. According to Spears (1988), the need for industrial wood could be provided by tree plantations of approximately 25 million hectares, about 10% of the remaining forest area, but as of the date of the article, only 2.6 million hectares of such plantations had been established. They could preclude the need to remove virgin forest, particularly if they consist of rapidly-growing species. Such projects can act as alternatives to the expansion of agricultural areas.

i. Provision of a system of tropical agricultural research stations and, especially, agricultural extension workers and soil experts to assist local farmers. This is absolutely essential for the success of agricultural reform. If farmers don’t know or understand the methods, nothing can be improved.

j. Provision of governmental guidance and regulation: The "green revolution" was successful and widespread only partly because of the dispersal of information to virtually all rice-growers. In addition, some coercive regulation was undertaken by governing bodies - usually local - in some places. In Bali, for instance, water for irrigation is provided only to those farmers who use the new varieties of rice.

Some of these scenarios require that new technologies be developed, others do not. All of these changes require that economic benefits accrue to farmers to provide them with incentives for using different technologies and methods, and for using them effectively. Economic and scientific aid will be required from international agencies as well as national governmental agencies in order to assure that any changes made are sound, adapted to local conditions, and environmentally safe.

7) Modification of economic and legal systems to provide local inhabitants, farmers, and others involved with natural resources with an economic interest in sustainable development. Among these modifications could be permitting long-term land tenure, allowing saleable property rights, improving land productivity, and assuring the right to profits to the provider.

8) Reduction of social and economic imbalances which cause many of the ecological problems in rainforests. In many tropical countries land is very inequitably distributed, with much of the land sequestered in extremely large ranches or estates belonging to a few very wealthy individuals, mainly those with political influence. Very little land is left for the rural and urban poor who increasingly frequently flock to newly-opened lands as migrants and farmers. In Latin America, 40% of the arable land is owned by 1% of landowners and held as huge estates (Ryan, 1992).

9) Reduction of anthropogenic effects on forests from greenhouse gases, pollution, and other climatic alterations. This could be accomplished by instituting conservation measures, reduction of emissions from vehicles, factories and other sources; reformation of agricultural practices; support for public rather than private transportation; and many other actions.

10) Utilization of indigenous species for resources. Many non-indigenous species have been imported into rainforest areas for agricultural and ranching purposes. These decisions have been based on economic demand rather than the ecological suitability of an organism for that particular terrain. Many millions of cattle and water buffalo have been introduced into Amazonia, where they have created environmental havoc. For cattle ranching, many millions of hectares of rainforest have been razed, although the soil rapidly becomes unsuitable for forage grasses. Water buffalo, Southeast Asian creatures, wreak havoc on riparian (river edge) environments, where they wallow. The riverbank vegetation and fauna are devastated and the loss of vegetation causes serious erosion of the banks. Much rainforest land, especially in the delta area of the Amazon River, has been cut for rice cultivation. Rather than introduce such destructive exotic species, one could raise local game to meet demand for meat, rather than cattle or water buffalo, and local crops rather than rice, for instance. These species could be raised with minimal disruption of the natural ecosystem. Butterflies are being raised on farms in Papua New Guinea and Mexico. The paca, a nocturnal rodent of the South American forest, eats fruits, roots, and other plant matter from the forest floor. Thus it could be used to exploit resources not currently being utilized by non-native domesticated animals, which are mostly grazers. The paca does well without damaging the forest; indeed it is part of the natural ecosystem of the forest (Ocana, et al., 1988). Another potential domesticate is the bearded pig of Southeast Asia, which plays a similar role to the paca in the Asian rainforest ecosystem (Robinson, 1988). Iguanas are being raised in Panama, and the "harvest" of meat could produce a sustainable yield of 230 kilograms per hectare per year, while cattle yield after several years drops to 10 or 15 kilograms (Ocana, et al., 1988). The wing bean, a minor crop in Asia, is easily raised, very nutritious - 40% protein and 17% edible oil - and is delicious (Spears, 1988).

11) Establishment of national centers for the conservation of threatened and endangered species. These could include botanical gardens, parks, arboreta, seed banks, zoological parks, wildlife rehabilitation centers, and sanctuaries.

12) Increase in basic research on tropical rainforests: In order to conserve tropical rainforests and wildlife effectively we must have a great deal more information. What we do know about them is extremely inadequate and is geared toward large mammal species and commercially-valuable plants and trees. The global expenditure on tropical biology studies in the 1980's was less than half the cost of a Boeing 747 airplane! (Robinson, 1988)

National institutes could be set up to coordinate research pertaining to conservation and management of resources. But since most tropical countries do not have the funds to carry out significant scientific research projects on their forests, international efforts must be integrated with national ones. Such enterprises as cooperative ventures with international environmental groups, pharmaceutical companies, botanical gardens, and zoos are potentially valuable. For example, Surinam (as well as several other South American countries) is engaging in bioprospecting, training programs, and research in its rainforests in cooperation with the International Cooperative Biodiversity Group, which is supported by Conservation International, the Missouri Botanical Garden, a local pharmaceutical company, and Bristol Myers Squibb (Kaufmann-Zeh, 1999).

In conjunction with this, research is being done to design "bioengineered" trees which would grow quickly, have a high ratio of wood to leaf (not what the tree itself wants) and be resistant to many pests. There are technical, ecological and ethical problems involved with this research and it remains to be seen whether or not bioengineered trees will provide wood to substitute for the wood of virgin forests (Mann and Plummer, 2002).

13) Regarding tropical rainforest preservation as an asset in economic calculations, since rainforests provide essential ecosystem services to everyone. Rainforests are rarely considered as more than short-term assets. There is a need to enact "resource depletion taxes," contractual obligations for logging concessionaires (such as mandating reforestation, the use of careful logging techniques, etc.) and other disincentives to clear-cutting and resource depletion. The revenues from such taxes could be used for reforestation, protection of reserves and other conservation measures.

14) Institution of economic measures favorable to rainforest preservation: Many people have made and are making small-scale decisions to farm, ranch or cut timber on rainforest land. These apparently inconsequential local actions have combined into a conflagration of rainforest destruction, with global consequences. These decisions are economic ones; therefore one of the most important means of preserving rainforests involves providing economic incentives to make conservation more attractive than exploitation. It is essential to make it financially appealing to preserve forest rather than to disturb or convert it, to make regeneration more attractive than rapid harvesting, and to make careful limited extraction more advantageous than clear-cutting. There has so far been little success in promoting sustainable use of resources, primarily since there are so many more financial incentives for rapid resource depletion. There are at present no legal mechanisms to compel those involved in deforestation to make compensation for the losses of rainforest which they engender and which, both immediately and ultimately, cost the public huge financial and amenity losses. Imaginative new measures are sorely needed.

a. International trade in "pollution credits" and the sale of ecosystem services:

i) Sale of pollution credits: There is today a healthy industry involved in selling "emissions" or "pollution" credits on the international market. This business is based on pollution reduction plans established by the Kyoto Climate Accord "Clean Development Mechanism." The Kyoto protocol is an international agreement, signed in 1997 by 174 countries, including the United States, to regulate CO₂ and other emissions to the atmosphere. Each signator country is committed to reducing emission levels of certain pollutants to specified levels by designated dates. However, a country which produces too much of a certain pollutant can purchase "pollution credits" from another country which produces less than its allowed amount of that pollutant. Since trees are thought by some (Schulze, Wirth and Heimann, [2000], for instance) to be net absorbers of carbon (i.e., act as carbon "sinks"), forests might also be used as credits. Under such an accord, countries with large forests could sell "pollution credits" (in reality, "sequestration services"; for more on this topic, see Part I, K1), and thus profit from maintaining their forests. This would compensate for the loss of revenue which is incurred by not logging or otherwise utilizing rainforests. The Conference of the Parties on the Climate Convention has allowed some use of forests in this way, but has not allowed the United States, for example, to obtain as many emission credits as it wished to claim for its forests to compensate for its huge amounts of excess carbon emissions. (The United States, in particular, has been reluctant to reduce its carbon emissions substantially. President Bush has, in fact, rescinded U.S. involvement in the Kyoto accord, citing "lack of scientific proof" for the reality of global warming, although virtually all scientists have agreed that anthropogenic global warming is occurring.) Whether or not forests actually are net carbon absorbers is not clear (see Part I, K1). This is a controversial topic which is being bitterly debated at the Kyoto meetings. The failure of the climate conference at The Hague in November 2000, to reach agreement on CO₂ emissions was largely due to disagreements over land-use change and forestry. Some countries view carbon sequestration by forests as a useful mechanism to mitigate carbon dioxide emissions, while others wish to minimize its importance (a not-unreasonable view, given the uncertainty of whether or not net carbon sequestration by forests occurs). There is also the danger that allowing forests to substitute for a reduction in emissions might endanger or replace forest conservation and careful management of forests. Since young, growing forests are thought to consume more carbon dioxide than older forests, the use of sequestration credits might encourage countries to cut down old-growth forests and replant. This risk could be eliminated by only allowing credits for reforestation on lands which had not been under forest for at least some specified period of time.

Some energy and other companies are investing in forest conservation projects in an effort to reduce carbon levels; by doing so they can earn emission credits. American Electric Power, PacifiCorp and BP-America, in conjunction with The Nature Conservancy, have invested in the Noel Kempff Mercado Climate Action Project in Bolivia. These companies have paid logging companies to retire the logging rights to 1.6 million acres near the Noel Kempff Mercado National Park, an action which doubled the size of the park (Gullison, Rice & Blundell, 2000). Such forest "carbon markets" could provide important incentives for forest protection. In Madagascar, the preservation of 33,000 hectares of forest (Masoala National Park) is preferred by local inhabitants and international interests, which benefit financially from the intact forest, while the government would obtain more revenue by issuing logging concessions. However, the government would benefit more from an intact forest if it were compensated for the emissions reductions provided by the forest (Bonnie, et al., 2000; Kremen, et al., 2000). Kremen and his co-workers calculated that the loss of the forests of the Masoala area would cost the international community between US$68 million and US$645 million from carbon emissions alone (not including such items as the loss for ecotourism companies or the loss of value of biodiversity after deforestation). They estimate that the conservation of carbon (i.e., compensating Madagascar for not deforesting by issuing emission credits) would cost less than US$1 per ton, and that this would represent only one-twentieth of the loss which would be accrued by deforestation. Payment for emissions credits could give Madagascar as much as US$23 million per year, rather than the much lower amount now invested there. Such funds, they estimate, would be sufficient to preserve the park over the long term.

Another proposed mechanism is to pay countries for preserving their forests in order to prevent the release of carbon dioxide which occurs during deforestation. Since the removal of one hectare of tropical forest releases approximately 200 metric tons of carbon, reduction in deforestation levels could represent a significant benefit to the atmospheric carbon load. Such funds are known as "carbon-offset funds" and will probably be important in the future in slowing rainforest destruction. Brazil, the largest deforester, however, opposes allowing carbon-offset funds to be used to combat deforestation. As is generally the case in that country, the Ministry of the Environment has not been consulted in the matter. Laurance (2001a) estimates that Brazil could accrue between US $600 million and $1.96 billion per year from such funds, which would disembowel the economic rationale for rainforest destruction in that country.

In order for forest conservation to be financially beneficial for governments, forest conservation (that is, the retention of forests for carbon sinks) must be incorporated into international agreements as carbon credits.

ii) Sale of ecosystem and other forest services: These services might include watersheds, tourist attractions, genetic resources, and intellectual property. An imaginative way to use market forces to preserve forests is to invest corporations with the right to sell the ecosystem services of the forests. A company might obtain the rights to a watershed, for example, and it would be allowed to provide water which is derived from the preservation of rainforest as a watershed. Since today many vital watersheds are threatened by development, in a world in which trillions of dollars will be needed to provide water in the next 20 years, watershed conservation could greatly reduce the amount of money needed for this purpose. In Costa Rica, conservation areas are credited by the government for services as carbon sinks and as watersheds (at a rate of $10 and $50 per hectare, respectively). And other ecosystem services might also be sold.

In another vein, intellectual property rights can be given to companies in return for funds for forest preservation. Costa Rica and the pharmaceutical company Merck have made an agreement in which Costa Rica conserves forest areas with the financial support of Merck, while Merck in turn receives access to information obtained from forest plants. Costa Rica will also receive a percentage of any profits Merck might obtain from compounds developed from forest organisms (Chichilnisky and Heal, 1998). In order for such schemes to work, it is essential for the resource to provide services which can be commercially valued. As we have seen, rainforests amply fulfill this requirement. In addition, some of the value of this resource must be available for use by private agencies (producers). In these ways private corporations could benefit from the conservation of rainforests.

b."Debt-for-nature" swaps: Tropical countries are liquidating their forests to provide short-term capital and for income to pay debts to developed countries. These debts to other countries and to international agencies such as the IMF and the World Bank are enormous and overwhelm the economies of poorer countries. There appears to be a distinct relationship between the rate of deforestation and the debt level of tropical countries. With "debt-for-nature" swaps, a conservation organization acquires a debt at a discount, and the debtor country redeems the debt by protecting land in reserves, by paying staff at reserves, and so forth. Otherwise, so long as tropical countries have massive debts, they will attempt to resolve them by liquidating their natural resources. For instance, a Bolivian debt of $650,000 was bought by a conservation group for $39,000 and the debt returned to that country with the understanding that Bolivia would set up a forest reserve in return for the debt forgiveness (Katzman and Cale, 1990). Debt-for-nature swaps have some serious disadvantages, however. As in many of these transactions, there is no guarantee that the debtor country will actually carry through with the promise, and if population or political pressures build up, the agreement may be abrogated.

c. "Public-private" partnerships: Non-governmental organizations might avert deforestation by establishing this type of partnership with a government. Dan Jantzen of the University of Pennsylvania has attempted this type of arrangement at Guanacaste National Park in Costa Rica. Private donations bought public grazing land which was adjacent to the park, and reforestation is being carried out on this land (Katzman and Cale, 1990). In the Atlantic forest in Brazil’s Paraná State, a reforestation project is being carried out by the Sociedade de Pesquisa em Vida Selvagem e Educaçao Ambiental (an NGO) and The Nature Conservancy on former pastureland. Funding is being provided by General Motors and American Electric Power. The companies may or may not receive carbon sequestration credits for their efforts (Bright and Mattoon, 2001). This type of arrangement is useful if there is a constituency in the country for conservation; in other places like Amazonia, this is less the case.

d. Private purchases and arrangements by non-governmental organizations: By utilizing their funds to purchase land or logging rights, non-governmental organizations (NGOs) can provide a "safety net" of protected forest. Among such agreements is The Nature Conservancy’s acquisition