2.6 Forests, sediment and water quality

Contents - Previous - Next

Sustainable food production depends on water quality as well as quantity. High levels of sediment and dissolved minerals in rivers and streams can have a variety of negative effects on downstream agriculture and fisheries as well as on peoples' nutritional well-being. By helping to preserve the quality of water supplies, trees and forests play an important role in maintaining food security. Water quality is directly linked to the prevalence of human diseases, especially gastro-intestinal disorders which directly affect people's ability to absorb foods (and thus their nutritional status). It is important to note that the issue of food security includes problems relating to people's ability to use food which is available to them.

From the point of view of nutrient cycling, undisturbed forests are the most efficient of all land use systems (Bormann and Likens, 1981). These systems are even capable of removing and immobilising some potentially harmful pollutants arriving from rain deposition (Sicamma and Smith, 1978). Removal of the forest, partially or completely, breaks this tight chemical cycling and releases minerals and nutrients into the drainage water. This has been shown in studies in Nigeria (Kang and Lal, 1981), Indonesia (Bruijnzeel, 1983) and a number of other countries. Besides causing a loss of nutrients from the site this may also result in an unwelcome addition of nutrients to irrigation water, which can contribute to the eutrophication of water bodies.

The effects of increased sediment levels are usually serious. While modest amounts of sediment may benefit food production in some circumstances - floodplain farmers in Bangladesh, for example, depend on periodic inundation and deposition of nutrient-rich sediment to maintain soil fertility - much more often the effects of sediment are harmful and costly. It can bury crops in floodplains, clog the gills of fish, damage marine fisheries by destabilising mangroves and blanketing seagrass beds and coral reefs, impair the quality of drinking water leading to increased prevalence of diseases, reduce irrigation reservoir capacity, block irrigation canals, and aggravate flooding by filling up flood control dams.

Vegetation cover is not the only factor that influences sediment yield from a given watershed or basin. It is also affected by climate (particularly rainfall), geology, and soils as well as forest fires (Pearce, 1986). If soils are unstable, and rainstorms severe, heavy sedimentation can occur even from watersheds that are entirely forested.

Under given conditions, however, forests do have an important impact in reducing sediment yield from watersheds. In studies in Indonesia, it was found that the sediment yield from areas that had been reforested was only a third of that from an agricultural watershed (Hardjono, 1980). Introducing trees into grazing or cropping land in a well-managed agroforestry system can also have valuable impact on reducing sediment yield (Hamilton, 1983).

The benefits of tree planting for reducing sediment levels may take years to materialize depending on the transport and storage mechanisms at work. Because sediment can be trapped and stored by vegetation and other physical barriers, eroded soil does not always appear in rivers immediately. There is usually a time lag, and the larger the watershed and the greater the opportunities for sediment storage, the longer this lag tends to be. Changes in erosion rates resulting from altered land-use practices may therefore take a considerable time before they are reflected in sediment loads in rivers.

In the case of large watersheds, trapped sediment will continue to flush out for decades. Reforestation of upland areas may therefore have little effect in the short term. Reservoirs will continue to fill with sediment even if the entire above-reservoir watersheds have been planted in forest. This time lag means that actions to prevent sedimentation need to be incorporated into the startup of development projects which are likely to cause (e.g. road building and logging) or be affected by increased sedimentation (e.g. dams).

Where sedimentation is a problem, it is important to identify the precise sources. In a particular watershed, ninety percent of the problem may be coming from five percent of the land. On steep terrain, a major source of sediment is often roads and forest logging activities. Where roads are constructed across or along stream channels, they put substantial amounts of soil directly into the waters during the construction phase. If they are badly located or designed, or improperly maintained, they may continue to be a serious sediment source for years to come.


2.7 Forests and the global climate

2.7.1 The Albedo effect
2.7.2 Carbon dioxide

In the long term, one of the most important ways in which forests may influence food production is through their effect on the global climate: altering rainfall patterns, world temperatures, and seasonal climate variations. Cutting of tropical forests has been implicated as being one of the contributory factors behind the gradual increase in the levels of carbon dioxide and certain other trace gases in the atmosphere. The impact this has on the global heat balance, the 'greenhouse effect', has become a cause of widespread concern (Swaminathan, 1986). The two most significant ways in which forests are thought to influence the global climate are through reflected heat from forest areas, and on the level of carbon dioxide in the atmosphere.

While the fact that the carbon dioxide levels are increasing is now widely accepted, the impact of this on the global climate is extremely difficult to estimate, and remains the subject of considerable controversy. Short-term effects may be different from those in the long term, and the effects will almost certainly vary between regions (Henderson-Sellers and Gornitz, 1984).

Agriculture will also be seriously affected if the observed increase in sea level continues, especially in low lying coastal areas. Bangladesh, for example, could lose 10 percent of its land area. Many coastal wetlands and mangrove areas would also be destroyed, with serious consequences for the world's fisheries.

2.7.1 The Albedo effect

Closed-canopy forests absorb more solar radiation than any other vegetation type and reflect less heat back into the atmosphere. The fraction of reflected radiation is known as "albedo". In recent years there have been many warnings that large scale forest removal may result in an increase in albedo (Hamilton, 1976; Chambers, 1980).

The overall effect of massive deforestation is not easy to predict because, along with increasing the albedo, deforestation may alter other variables that may produce countervailing effects. Two of the most comprehensive ' Global Circulation Model' studies have produced almost exactly opposite predictions about the effects of tropical forest removal. One suggests a slight warming and an increase in rainfall (Lettau et al, 1979). The other predicts a slight cooling in the equatorial region, and an 11 percent reduction in rainfall in the tropics (Potter et al, 1975). A more recent study examining the impact of deforestation of the Amazon rainforest asserts that although radical alteration in forest cover would increase local albedo, there would be no major impact on regional or global climate Henderson-Sellers and Gornitz, 1984).

With the inherent complexities of climatic systems, all of these models suffer from problems. Until there are more reliable input data and better models, there is unlikely to be a clear verdict on the impact of increased albedo from forest removal on the global climate.

2.7.2 Carbon dioxide

Many doubts also exist with regard to the effect of forest clearance on atmospheric carbon dioxide levels (Woodwell et al, 1978; Hampicke, 1979). Although the cutting and burning of trees does release carbon dioxide, it is not the only factor involved; the burning of fossil fuels and cement manufacture are reckoned to be more important contributors to the increased carbon dioxide burden.

The global carbon cycle is still only partly understood and there is considerable scientific disagreement about how much contribution forest disappearance is really making to increases in carbon dioxide levels. The fact, for example, that the forest area in the northern temperate zone has been increasing over the past five decades may partially offset the large forest loss occurring in the tropics (Sedjo and Clawson, 1984).

While allowing for factors such as forest burning, the carbon fixation of forest regrowth, and the effect of carbon levels on plant photosynthesis levels most models agree that there will be a considerable net transfer of carbon to the atmosphere from forest clearance and burning in the tropics. One estimate puts the total contribution at between 1 and 4.5 billion tons per year, plus an additional 2 billion tons from oxidation of exposed organic matter in the soil. This is clearly a substantial amount, given that fossil fuel burning is currently estimated to supply around 5 billion tons (Myers, 1980). However, as was noted above the whole carbon cycle is not well understood. The annual increase in carbon in the atmosphere is estimated to be only about 2.3 billion tons. Thus the increased carbon levels being emitted both from forest clearance and fossil fuel burning are being absorbed: either by the oceans or by unknown terrestrial sinks.

Without doubt, a great deal of further research will be required before the effects of forests on the global climate are properly understood.


2.8 Forests, rainmakers?

2.8.1 The Amazon basin
2.8.2 Cloud forests

Equally controversial is the effect of forests on local rainfall. There is a widespread belief that deforestation causes a decrease in local precipitation, and that conversely, restoring forest cover will lead to an increase (Goodland and Irwin, 1975; World Water, 1981). If it was indeed true, such an effect would have a great impact on agriculture.

The scientific literature on this subject is far from conclusive. In India, the influence of forests on precipitation has been debated for nearly a hundred years (Singh, 1988). Some studies have reported a decrease in rainfall in certain districts after forest clearance (Warren, 1974), while others have noted an increase following reforestation (Eardley-Wilmot, 1906). A beneficial effect on the number of rainy days per year has also been recorded in some studies (Ranganathan, 1949). However, no clear overall pattern has emerged and it is generally concluded that although there may be some relationship between forest cover and rainfall, the effects on total precipitation are relatively small (Hill, 1916).

A study in the Central Congo Basin found no evidence of any influence of forests on rainfall. It was suggested, however, that forest clearing, by increasing the heat reflectance, might introduce some instability into weather patterns which can be equally important as total rainfall to production systems (Bernard, 1953).

Throughout much of the tropics, most local precipitation is the result of monsoons or major storms generated by large weather systems, or else is caused by moisture-laden air being forced upwards as it passes over hills and mountains. In neither case is tree cover likely to have any dramatic influence on total rainfall.

There are two special cases which merit attention, however, the Amazon Basin and the "cloud" forests.

2.8.1 The Amazon basin

The Amazon Basin is a horseshoe-shaped plain, open on the east to the moisture-bearing trade winds from the ocean and ringed by mountains and high plateaux on the other sides. Recent studies have shown that in the Amazon Basin the recycling of water vapour by forest vegetation might indeed constitute an important source of atmospheric moisture for rainfall within the basin (Salati and Vose, 1984).

It has been estimated that conversions of 10, 20 and 40 percent of the forest to shrubs and crops would result in annual rainfall reductions of around 2,4, and 6 percent, respectively (Brooks, 1985). These reductions may not seem large, given the fact that the average rainfall in the region is in excess of 2000 mm. Nonetheless, since the dry period in the Amazon already has the forest ecosystem under stress, even a change of this magnitude might produce irreversible changes in the natural forest (Salati and Vase, 1984). Even if these changes do not effect the global climate, the effect on agricultural production in the region could be disastrous.

2.8.2 Cloud forests

The second case occurs where persistent moisture laden clouds or fog are driven through forests or belts of trees by wind. "Cloud" forests occur at higher elevation on many mountains and are often made up of unusual plant and animal communities. Worldwide, cloud forests cover about 500,000 square kilometres, or almost 5 percent of the closed moist tropical forest (Persson, 1974). "Fog" forests which occur in some coastal areas (sometimes in areas with normally very low precipitation, such as the coast of Peru and Chile) act in similar ways. These forest areas can have a significant impact on a region's hydrological system, and thus on agricultural production.

These tree barriers strip the moisture from the clouds or fog. Single trees or narrow belts of trees are most effective. In closed forests there is a mutual sheltering effect. A study in Hawaii found that a single Araucaria heterophylla tree added 760 mm of 'horizontal' precipitation in a year to the normal vertical precipitation of 2600 mm (Ekern, 1964).

This extra moisture is added to the hydrological system and can add to groundwater and streamflow levels. Because of their height and the large amount of moisture-stripping surface, trees are much more effective in this function than other vegetation types. Maintaining forests in these areas is therefore critical to local hydrological regimes. Conversely, where wind-driven fog or persistent cloud phenomena occur in areas that have been cleared, planting trees can re-establish a water capture system.


2.9 Forests and genetic resources

A final, and important, connection between forests and food security is their role as reservoirs of genetic diversity. Though not strictly an environmental link, the fact that forest environments provide a habitat for a great diversity of plant and animal species makes them important biological resources.

Forest areas represent the largest single store of genetic diversity. From the point of view of future agricultural production, the species they contain - both known and yet to be discovered - may have a critical role to play in providing the genetic variation needed to combat the ever-adapting pests and diseases that prey on food crops. They may also provide a range of entirely new foods and medicines - of both plant and animal origin - that could have a major impact on human health and nutrition.

Conserving these genetic resources for future generations is being increasingly recognised as both a moral and practical imperative. The problem is in devising ways of achieving this.

Ex-situ conservation of genetic resources using gene and seed banks undoubtedly have important roles to play. But there are limitations to these methods: especially their high costs and associated technical problems such as genetic drift within breeding populations. In the foreseeable future, in-situ approaches in which species are conserved within their natural habitat will have to shoulder the major burden of genetic resource conservation.

This means preserving certain areas of forests in an intact, or near intact state. This poses a set of difficult challenges, given the many human and economic pressures on forests. Simply fencing off areas of forest is not possible in many cases, as huge numbers of people depend on those forests for their livelihoods. Compromises will have to be made, and ways will have to be found of combining conservation with sustainable utilisation of forest resources by local people. For unless local people have a stake in the survival of forests, conservation efforts are doomed to fail.

Forests and genetic resources

Contents - Previous - Next