(introduction...)

Introduction
TRF and its conversion
Soils of the TRF ecosystem
Forest conversion and soil productivity
Deforestation and the emission of radiatively active gases
Deforestation and hydrological balance
Sustainable use of the TRF ecosystem
Research needs
References

 

Rattan Lal

Introduction

The humid tropics comprise about 31 per cent of all tropical biomes, cover 11 per cent of the earth's total surface, occupy about 1.5 billion ha of land area, and are home to about 2 billion people (WRI 199091). Of the 1.5 billion ha of the humid tropics, 45 per cent lie in the Americas, 30 per cent in Africa, and 25 per cent in Asia and Oceania. Within the generic term "tropical rain forest" (TRF), there are three principal types of forest vegetation including: lowland rain forest (80 per cent of the humid tropical vegetation), premontane forest (10 per cent), and lower montane and montane forests (10 per cent). The TRF ecosystems are characterized by constantly high temperatures and relative humidity, high annual precipitation, highly weathered and leached soils of low chemical fertility, and high total biomass. High total biomass production, despite low soil fertility, is due to the effect of high temperatures and relative humidity, abundant rainfall, and low moisture deficit. The natural vegetation of the TRF is characterized by a high degree of biodiversity. The TRF ecosystem has global importance in terms of soil and climatic interactions and its impact on several processes. For example, local and global climatic patterns are influenced by the interaction of the TRF with the atmosphere (Salati et al. 1983; Myers 1989; Houghton 1990). An important aspect with global influences involves the impact of the TRF on biogeophysical cycles, e.g. C, N, S, and H2O. Conversion of the TRF to other land use disrupts these cycles, which are critical in regulating several global processes; e.g. emission of radiatively active gases into the atmosphere, change in the total water vapour present in the atmosphere. It is because of these local, regional, and global interactions that the TRF and its conversion are a major concern.

TRF and its conversion

In prehistoric times, the geographical area of undisturbed TRF was about 1.5 billion ha. It is estimated that 45 per cent of the original TRF has been converted to other land uses, with a regional loss of about 52 per cent in Africa, 42 per cent in Asia, and 37 per cent in Latin America (Richards 1991). Because of the wide diversity in vegetation type and in the mode and degree of conversion, however, there is a large variation in the estimates of the extent of the remaining TRF and rates of its conversion. The areal extent of TRF from 1700 to 1990 for three regions is depicted in table 9.1. Over the 290 years, the TRF decreased by 36 per cent in tropical Africa, 26 per cent in Latin America, and 30 per cent in Asia. The most drastic conversion happened between 1920 and 1950. The data in table 9.2 are an estimate of the total deforestation that occurred in different regions over a 328-year period. Low and high estimates of total forest conversion range from 484 million ha (32 per cent of the total) to 538 million ha (36 per cent of the total).

Present estimates of the remaining area of tropical rain forest and annual rates of deforestation are also highly variable and erratic (Myers 1991). Estimates of the total area of TRF for the year 1990 range from 1,282 million ha (FAO) to 1,715 million ha (WRI) (table 9.3). The principal discrepancy in the data in table 9.3 lies in the estimate of TRF for Africa. The WRI estimate of 600 million ha includes both closed forest and wooded areas. There are several categories of vegetation called TRF. These include closed forest, forest land, woodland, shrub land, forest land under shifting cultivation, and miscellaneous land (FAO 1981). The closed forest is the true TRF. The distinction between these categories is difficult to make, and estimates vary widely. Estimates of the area of closed forest and wooded land and the rate of conversion are shown in table 9.4.

Table 9.1 Global change in tropical rain forest and woodland, 1700-1990 (million ha)

Region	1700	1850	1920	1950	1980	1990	Total change
Tropical Africa	1,358	1,336	1,275	1,188	1,074	869	- 489
Latin America	1,445	1,420	1,369	1,273	1,151	1,067	- 378
South & South-East Asia	558	569	536	493	415	410	- 178

Sources: Richards (1991); WRI (1990).

Table 9.2 Esffmated area of deforestation, 1650-1978 ('000 km²)

Region	Levela	Pre-1650	1650-1749	1750-1849	1850-1978	Total deforestation 1650-1978
						High	Low
Central America	H	18	30	40	200	288
	L	12					282
Latin America	H	18		170	637	925
	L	12	100				919
Oceania	H	6	6	6	362	380
	L	2	4	-			368
Asia	H	974	216	606	1,220	3,016
	L	640	176	596			2,632
Africa	H	226	80	-16	469	759
	L	96	24	42			631
Total	H	1,242	432	806	2,888	5,368
	L	762	334	848			4,832

Source: Williams (1991). a. H = high estimate; L = low estimate.

Table 9.3 Present estimates of TRF and the annual rate of deforestation

Region	WRI (1992 93)		FAO (1991)
	Total area	Annual rate	Total area	Annual rate
	(ha m.)	(%)	(ha m.)	(%)
Africa	600.1a	5.0	241.8	4.8
Latin America	839.9	8.3	753.0	7.3
Asia	274.9	3.6	287.5	4.7
Total	1,714.9	16.9	1,282.3	16.8

a. Includes wooded land area and closed forest.

Table 9.4 Different categones of TRF and their conversion rate

Region	Total area (ha m.)		Conversion rate (ha m./yr)
	Closed forest	Wooded land	Closed forest	Wooded land
Africa	217	652	1.33	2.34
Latin America	679	388	4.12	1.27
Asia	306	104	1.82	0.19
Total	1,202	1,144	7.27	3.81

Source: WRI (1988 89).

There are several problems with the available data. Original data based on recent and direct surveys are not available. Most estimates are 10-20 years old and obsolete. Furthermore, there are differences in the criteria used, and the accuracy of most estimates is questionable. As with the area, the rate of deforestation is also hard to estimate. However, reliable estimates of the areas of TRF and rate of conversion are needed for: (i) land-use planning, and (ii) predicting the impact of forest conversion on soil and environment.

Soils of the TRF ecosystem

The predominant soils of the humid tropics are oxisols, ultisols, and alfisols (table 9.5). Oxisols and ultisols comprise 63 per cent of soils of the TRF (table 9.6). These soils are highly weathered, leached, devoid of basic cations, and relatively infertile. Young soils of moderate to high fertility (mollisols, inceptisols, and entisols) occupy about 15 per cent of the total land area. There are several soil-related constraints on intensive food crop production in the humid tropics. The principal constraints are listed in table 9.7. Oxisols and ultisols have low nutrient reserves and are prone to toxicity owing to high concentrations of Al and Mn. In general, these soils have high P-fixation capacity. Alfisols are relatively more fertile than oxisols and ultisols. However, alfisols have weakly developed structure and are highly prone to accelerated soil erosion. The effective rooting depth for food crops and annuals is generally 20-30 cm owing to either physical (compacted, concretionary, or gravelly subsoil) or chemical (Al or Mn toxicity, low P) limitations. Coupled with low plant-available water reserves, water deficiency can be a problem for shallow-rooted annuals. In contrast, upland crops can be subjected to periodic inundation and anaerobiosis. With proper management, however, the agricultural productivity of these soils can be greatly improved while minimizing risks of soil and environmental degradation. An impor tent strategy in enhancing the productive potential of these soils is to reduce the adverse effects of forest conversion.

Table 9.5 Geographical extent and distribution of major soils of the humid tropics (ha million)

Soil type	Region
	America	Africa	Asia	Total
Oxisols	332	179	14	525
Ultisols	213	69	131	413
Inceptisolsa	61	75	90	226
Entisolsb	31	91	90	212
Alfisols	18	20	15	53
Histosols		4	23	27
Spodosols	10	3	6	19
Mollisols	-	-	7	7
Vertisols	1	2	2	5
Aridisols	-	1	1	2
Total	666	444	379	1,489

Source: N RC (1993).
a. Inceptisols include Aquepts, Tropepts, Andepts, and Entisols.
b. Entisols include Fluvents, Psamments, and Lithic Entisols.

Table 9.6 Soils of the humid tropics (% of the total area)

Principal feature	Soil type	Region
		America	Africa	Asia	Total
· Acid, infertile	Oxisols and ultisols	82	56	38	63
· Moderately fertile, & well-drained	Alfisols, vertisols, mollisols,inceptisols, andisols, fluvents	7	12	33	15
· Poorly drained	Aquepts	6	12	6	8
· Very infertile, sandy	Psamments, spodosols	2	16	6	7
· Shallow	Lithic entisols	3	3	10	5
· Organic	Histosols	-	1	6	2
		100	100	100	100

Source: NRC (1982).

Table 9.7 Soil-related constraints on intensive land use for food crop production in the TRF ecosystem

Constraint	Oxisols	Ultisols	Alfisols	Inceptisols	Mollisols	Andisols
Physical
Accelerated erosion	2	2	3	2	1	1
Soil compaction & crusting	2	2	3	2	1	1
Root impedance	3	3	3	1	1	1
Moisture imbalance	2	2	3	1	1	1
Shallow depth	2	2	3	1	1	1
Nutritional
N deficiency	3	3	2	2	1	1
P deficiency	3	3	2	1	1	1
Al & Mn toxicity	3	3	1	1	1	1
Micro-nutrient deficiency	3	3	2	1	1	1
Biological
Soil fauna	2	2	2	1	1	1
Biomass carbon	2	2	2	1	1	1

3 = severe; 2 = moderate; 1 = slight.

Forest conversion and soil productivity

Deforestation and conversion to arable land use have drastic impacts on soil properties, water and energy balance, and soil erosion hazard (Lal 1987). The worst-case-scenario local effects are outlined in table 9.8. The principal soil degradation effects include adverse effects on soil structure leading to crusting, compaction, and hardsetting. Reduction in infiltration, increase in surface runoff, and soil exposure to raindrop impact and to the shearing effect of overland flow accentuate soil erosion risks. Alterations in pore size distribution and reduction in the colloid content of the surface soil, owing to eluviation, and preferential removal of clay and organic carbon by erosion drastically reduce plant-available water reserves. High soil temperatures, often reaching 40-45°C at 0-5 cm depth for 4 to 6 hours a day, further aggravate the frequency and intensity of drought stress experienced by shallow-rooted crops.

The principal impact of deforestation on chemical and nutritional properties is related to a decrease in the organic matter content of the soil and to disruption in nutrient-recycling mechanisms owing to the removal of deep-rooted trees. The decrease in soil organic matter content is mostly due to the high rate of mineralization caused by high temperatures. The absence of actively growing roots in the subsoil horizon leads to leaching of bases (e.g. Ca, Mg, K, Na) and increase in soil acidity. In addition to leaching, loss of N and S also occurs owing to volatilization.

Table 9.8 Worst-case scenario regarding the adverse effects of deforestation on soil productivity

Physical effects	Chemical and nutritional effects
· Compaction, crusting, increased strength	· Loss of soil organic matter, nitrogen, and sulphur
· Accelerated erosion	· Leaching of bases
· Loss of clay and soil colloids	· Acidification
· Drought stress	· Reduction in soil biological activity
· High soil temperatures	· Disruption of nutrient recycling

Table 9.9 Technological options to minimize the adverse effects of deforestation

Activity	Recommended practice
Time of land clearing	· During dry season when soil moisture is low
Method of land-clearing	· Manual felling with chainsaw
Mechanized clearing	· Preferably with shearblade
Management of fell biomass	· In situ burning, no windrows
Stumping and root removal	· Remove manually from the top 30 cm, or leave intact
Protective cover	· Plant an aggressive cover crop, e.g. Mucuna, Desmodium spp. Puereria, etc.
Seedbed preparation	· No-till or conservation tillage
Erosion management	· Vegetative hedges, e.g. Vetiver, Leucaena, etc.

The magnitude of these adverse effects depends on the method of deforestation and on the soil and crop management practices. In addition, there exists a strong interaction with soil type, rainfall regime, nature of the existing vegetation, ambient soil moisture content, and microclimate. Technological options for minimizing the adverse effects of deforestation are outlined in table 9.9. It is well known that the adverse effects of deforestation are more severe for mechanical than for manual land-clearing. The adverse effects of mechanical clearing (soil compaction and accelerated erosion) are generally less severe for shearblade than for treepusher or bulldozer blade methods of tree felling. Compaction and structural degradation are more severe when soil wetness is high at the time of landclearing (Ghuman and Lal 1992). Stumping and removal of roots, which is necessary only to facilitate mechanized farm operations, should preferably be done manually. Root ploughing is disruptive and causes considerable soil disturbance. Windrowing also scrapes topsoil and concentrates nutrient-rich ash in narrow strips. Management of the soil structure and erosion control can be achieved by sowing a quick-growing cover crop. The cover crop should preferably be a legume, e.g. Mucuna utilis, Pueraria phaseoloides, Centrosema spp., or Desmodium spp. Erosion control on sloping lands can be achieved by establishing vegetative hedges (e.g. Vetiver, Leucaena, Gliricidia) and other multi-purpose trees and woody shrubs. Adoption of agro-forestry practices also enhances nutrient recycling and minimizes leaching losses of bases.

Enhancing the nutrient capital of the soil is critical to increasing the agricultural productivity of these soils of low inherent fertility. Soil fertility is further depleted by deforestation and biomass removal and/or burning. Therefore, judicious application of fertilizer needs careful consideration. To some extent, nitrogen can be supplied through biological fixation. However, other nutrients, including Ca, Mg, and P, must be made available from off-farm sources. Admittedly, resource-poor farmers cannot afford capital-intensive inputs. None the less, essential nutrients must be supplied, through application of either organic manure or mineral fertilizer, if high yields are expected on a sustained basis.

Deforestation and the emission of radiatively active gases

Deforestation strongly affects the dynamics of soil organic matter. Experiments conducted in Africa (Greenland and Nye 1959; Nye and Greenland 1960; Lal 1976; Juo and Lal 1977; Aina 1979; Lal et al. 1980; Ghuman and Lal 1991,1992) show rapid decline in soil organic matter content following deforestation and cultivation. The magnitude of carbon decline in the top 5 cm depth can be as much as 50 per cent in 12 months and 60 per cent in 18 months. The organic carbon (C) content of the top 30 cm depth declines by about 50 per cent within 10 years of deforestation and intensive cultivation. Examples of the carbon loss from soils of the humid tropics within 10 years of deforestation and intensive cultivation are shown in tables 9.10-9.12. The rate of C loss may be as much as 1.13 mg/ha/yr from soil managed by conservation tillage and agro-forestry to 5.60 mg/ha/yr for soils managed with a plough-based conventional tillage system. That being the case, newly cleared land in the humid tropics may release between 98.7 billion kg C/yr and to 218.8 billion kg C/yr, with a mean emission rate of about 154.3 billion kg C/yr.

Table 9.10 Loss of organic curbon with continuous and intensive cultivation with no-till and agro-forestry in 10 years following deforestation

Depth (cm)	Organic C (%)		Bulk density (mg/m3)		Total soil carbon (mg/ha)		Carbon emission in 10 years (mg/ha)
	Initial	Final	Initial	Final	Initial	Final
0-10	2.50	1.50	1.10	1.40	27.5	21.0	6.5
10-25	1.40	1.00	1.25	1.45	26.3	21.8	4.5
25-50	0.90	0.80	1.30	1.45	29.3	29.0	0.3
Total					83.1	71.8	11.3

Source: Lal (1991).

Table 9.11 Loss of organic carbon with continuous and intensive cultivation using plough-based mechanized systems in 10 years following deforestation

Depth (cm)	Organic C (%)		Bulk density (mg/m3)		Total soil carbon (mg/ha)		Carbon emission in 10 years (mg/ha)
	Initial	Final	Initial	Final	Initial	Final
0-10	2.5	0.5	1.10	1.5	27.5	7.5	20.0
10-25	1.40	0.4	1.25	1.45	26.3	8.7	17.6
25-50	0.9	0.3	1.30	1.45	29.3	10.9	18.4
Total					83.1	27.1	56.0

Source: Lal (1 991).

The loss of organic C from soils under shifting cultivation is less than that from soils under intensive cultivation. Nye and Greenland (1960) observed that the loss of organic carbon in 100 years may be 20 per cent for a soil with 12-year fallow cycle to 45 per cent for a soil with 4-year fallow cycle. The annual loss of C due to shifting cultivation may be as much as 0.27 mg/ha. If shifting cultivation is prac tised on about 25 million ha, the total loss of C due to shifting cultivation is estimated at 6.25 billion kg C/yr. In addition to C, biomass burning also causes release of several other greenhouse gases, e.g. CO₂, CO, CH₄, and NO_x.

Table 9.12 Changes in soil organic carbon (SOC) content of the surface 0-5 an layer of two soils in southern Nigeria

Alfisol at Ibadana				Ultisol at Okomub
		DC				DC
Year	Organic carbon (%)	%/yr	Average (%/yr)c	Year	Organic carbon (%)	%/yr	Average (%/yr)c
1978	2.17	1984	1.8
1979	1.61	- 25.8	- 25.8	1985	1.4	- 22.2	- 22.2
1982	1.54	-1.5	-7.3	1986	1.45	+3.6	-9.7
1984	1.14	-13.0	-7.9	1987	1.05	-27.6	-13.9
1985	1.24	+8.8	-6.1	1988	1.15	+9.5	-9.0
1986	1.30	+4.8	-5.0
1987	1.09	-16.2	-5.5

a. The data from Ibadan are from Watershed 1.
b. The data from Okomu are from the manually cleared plots; data recalculated from Ghuman and Lal (1991).
c. The average (%/yr) is calculated for each year on the basis of the original SOC content.

Deforestation and hydrological balance

Deforestation of TRF can drastically alter the components of the hydrological cycle:

P = I + R + DS + D + > Ed_t,

where P is precipitation, I is infiltration, R is surface runoff, DS is soilwater storage, D is deep drainage, E is evapotranspiration, and t is time. Deforestation decreases I and DS and increases R and D components. In general, deforestation may also increase E. The change in E, however, may also depend on the land use.

Several experiments have demonstrated the effects of clear-cutting on the increase in total water yield. The impact of deforestation on the hydrological balance of a 44 ha watershed was studied at the International Institute of Tropical Agriculture (IITA), Ibadan, Nigeria. Prior to partial deforestation in 1978 and complete defor estation in 1979, measurements of surface and subsurface flow were made under the forest cover from 1974 to 1977. Under the forest cover, the interflow was 0.4 per cent to 1.4 per cent and total flow 0.8 per cent to 2.7 per cent of the total rainfall. Partial clearing in 1978 increased interflow to 1.2 per cent and increased total flow to 6.6 per cent of the total rainfall (table 9.13).

Table 9.13 Effects of partial clearing in 1978 on total water discharge from Watershed 1

Parameters
Rainfall (mm)	785.8
Surface flow (mm)	42.7
Surface flow (% of rain)	5.4
Subsurface flow (mm)	9.4
Subsurface flow (% of rain)	1.2
Total yield (mm)	52.1
Total yield (% of rain)	6.6

Note: The partial clearing was of 3.1 ha out of 44.3 ha.

Table 9.14 Hydrological components on an annual basis for Watershed 1, 1979-1986

Year	Annual rainfall (mm)	Subsurface flowb (mm)	Surface flow (mm)	Total water yield		Apparent evapo transpirationa
				mm	% of rainfall	mm	% of rainfall
1979	1,435.5	28.0	73.4	101.4	7.1	1,334.1	92.9
1980	1,449.7	73.1	90.0	163.1	11.3	1,286.6	88.8
1981	1,074.5	58.9	28.9	87.8	8.2	986.7	91.8
1982	851.5	50.9	25.9	76.8	9.0	774.7	91.0
1983	897.6	45.8	21.3	67.1	7.5	830.5	92.5
1984	1,162.2	58.9	27.1	86.0	7.4	1,076.2	92.6
1985	1,675.7	18.5	93.2	111.7	6.7	1,563.9	93.3
1986	1,164.1	1.9	51.7	53.8	4.6	1,110.3	95.3

a. Evapotranspiration includes soil water storage and groundwater recharge.
b. Subsurface flow is underestimated during wet years because it is computed as a part of surface flow during the storm runoff.

The entire watershed was cleared in 1979 and cultivated to food crops. The data in table 9.14 show that the total water yield ranged from 4.6% to 11.3% of the rainfall received. Because of the bimodal distribution of the rainfall, the hydrologic balance was computed separately for each growing season. The hydrologic balance showed that total water yield ranged from 1.4% to 11.8% for the first season (table 9.15) and from 0.8% to 18.1% for the second season (table 9.16). The intermittent stream, with a trace of flow after heavy rain and no flow during the dry season, became a perennial stream that recorded a measurable flow throughout the dry season (table 9.17).

Table 9.15 Hydrological components for the first growing season (March-July), 1979-1987

Year	Annual rainfall (mm)	Subsurface flowb (mm)	Surface flow (mm)	Total water yield		Apparent evapo transpirationa
				mm	% of rainfall	mm	% of rainfall
1979	846.1	7.0	89.8	96.8	11.4	749.3	88.6
1980	604.3	1.2	7.0	8.2	1.4	596.1	98.6
1981	636.8	20.2	17.3	37.5	5.9	599.3	94.1
1982	615.2	28.4	17.6	46.0	7.5	569.2	92.5
1983	580.9	22.3	15.2	37.5	6.5	543.4	93.5
1984	681.6	23.6	13.9	37.5	5.5	644.1	94.5
1985	935.7	10.8	52.1	62.9	6.7	872.8	93.3
1986	714.2	1.8	36.9	38.7	5.4	677.3	94.8
1987	723.5	36.4	49.2	85.6	11.8	637.9	88.2

a. See notes to table 9.14.
b. See notes to table 9.14.

Table 9.16 Hydrological components for the second growing season (AugustNovember), 1979-1986

Year	Annual rainfall (mm)	Subsurface flowb (mm)	Surface flow (mm)	Total water yield		Apparent evapo transpirationa
				mm	% of rainfall	mm	% of rainfall
1979	585.8	0.03	4.6	4.6	0.8	581.2	99.2
1980	845.4	71.90	81.1	153.0	18.1	692.4	81.9
1981	432.4	33.50	11.6	45.1	10.4	387.3	89.6
1982	223.6	19.20	8.1	27.3	12.2	196.3	87.8
1983	230.6	19.20	6.1	25.3	11.0	205.3	89.0
1984	480.6	30.50	13.2	43.7	9.1	436.9	90.9
1985	735.5	6.90	41.1	48.0	6.5	687.5	93.5
1986	379.2	0.10	13.9	14.0	3.7	365.2	96.3

a. See notes to table 9.14.
b. See notes to table 9.14.

Table 9.17 Hydrological components for the dry season (December-February) for Watershed 1, 1979-1987

Year	Seasonal rainfall (mm)	Subsurface flow (mm)	Surface flow (mm)	Total water yield (mm)
1979	3.6	0.0	0.0	0.0
1980	23.6	0.0	0.0	0.0
1981	5.3	2.0	0.08	2.1
1982	12.7	5.0	0.10	5.1
1983	0.0	3.6	0.03	3.6
1984	86.1	4.1	1.1	5.2
1985	0.0	3.6	0.0	3.6
1986	7.6	0.0	0.0	0.0
1987	18.8	3.9	0.0	3.9

Note: The data for December are taken from the previous year.

An increase in the magnitude of interflow and its continuous discharge throughout the dry season may be attributed to the replacement of deep-rooted perennials with high water requirements with shallow-rooted annuals with relatively fewer water requirements. Further, annuals were not grown during the dry season.

Sustainable use of the TRF ecosystem

Criteria for sustainable land use

The tropical rain-forest ecosystem must be used, improved, and restored. Continuous depletion of these resources has economic and ecologic ramifications at local, regional, and global scales. Sustainable use of soil and water resources in the TRF ecosystem should take the following into consideration:

· the nutrient capital of the soil resources should be enhanced by applications of chemical and organic fertilizers;

· the management systems adopted must optimize energy flux as well as energy use efficiency - energy efficiency alone is not adequate in view of increasing population pressure;

· Iosses of nutrients and water out of the ecosystem should be minimized;

· nutrient recycling mechanisms from subsoil to surface horizons must be an integral aspect of the land-use system;

· land degraded by past mismanagement must be restored by afforestation with ecologically adapted and quick-growing species.

Land capability assessment

Land capability assessment is necessary for the rational utilization of forest resources. Sustainable use of TRF resources necessitates a detailed and accurate inventory of the soil, water, vegetation, and climatic characteristics of the region. These inventories/surveys should be conducted at reconnaissance scales (1: 50,000 to 1 :1,000,000) and detailed scales (1 :10,000 to 1:50,000). The land resources should then be classified according to their potential capability as follows (FAO 1982):

Forest land

There are several types of forest land:

(a) Natural forested land should be preserved as natural forest and left alone. It has limitations of topography, shallow/stony soils, poor water regime, etc. Some examples are marginal steep lands, forests in the vicinity of regions with short supplies of firewood, inaccessible areas, small islands, and regions with other sociopolitical connotations.

(b) Production forests are suitable for managed logging of timber and other forest products.

(c) Planted or man-made forests are fertile, prime lands and are suitable for tree plantations, e.g. Gmelina, teak, Cassia.

(d) Protected forests are forest reserves protected in order to preserve the natural biodiversity.

Arable land

Arable land is prime agricultural land and is suitable for supporting continuous and intensive agriculture for food-crop and livestock production. Such land should be developed and managed according to ecologically compatible methods of deforestation and land development. When deforestation for arable land use is inevitable, land development should be carefully planned and implemented according to scientific guidelines.

Guidelines for land use in the TRF ecosystem

The development of TRF for alternative land uses has become a global issue. For some countries, the question is no longer whether to remove tropical forest for alternative land uses; the important consideration is how much to remove and by what method so that ecological concerns are adequately addressed. It is the ill-planned and improper management of TRF that has created severe ecological, economic, and socio-political problems. The sequence of steps needed to achieve a rational use of the TRF ecosystem is outlined below:

1. Iand capability assessment;
2. choice of proper land use (e.g. arable land, protected forest, manmade forest);
3. use of proper methods and time of deforestation (e.g. manual, chainsaw, shearblade);
4. adoption of soil conservation measures (e.g. cover crop, mulch farming, vegetative hedges);
5. use of science-based agronomic techniques of soil and crop management (e.g. balanced fertilizer use, proper crop rotation and cropping sequences, appropriate tillage methods, and integrated pest management).

Best management practices for sustainable agriculture

Some soils supporting the TRF ecosystem can be converted to intensive arable land use with sustained production provided that:

(a) expectations of agronomic yields are not too high,

(b) the soil and crop management systems adopted ensure the replenishment of plant nutrients harvested in crops and the maintenance of biophysical resources,

(c) the soils are taken out of production and put to restorative land use long before the degradative processes are set in motion.

Some research-proven agronomic practices based on these guidelines are listed in table 9.18. Just as use of prime agricultural land is essential not only for food-crop production but also for establishing pasture and forest plantations, so is the use of chemical and organic fertilizers for enhancing soil fertility. Most soils of the TRF ecosystem are of low inherent fertility. Enhancing soil fertility, therefore, is crucial to sustained agricultural productivity.

Imperata control

Land misuse and severe soil degradation encourage encroachment by Imperata cylindrica and other noxious weeds. It is important to maintain soil fertility at a high level to curtail encroachment by lmperata and to reclaim already infested lands. Reclamation of Imperata-infested land requires a combination of mechanical, chemical, and biological measures. Soil inversion, to uproot rhizomes and expose them to high temperatures during the dry season, followed by the use of systemic herbicides and sowing an aggressively growing cover crop, is essential to eradicate the noxious weed. Biological methods of Imperata control, slow as they may be, are often effective on a long-term basis. Preventing encroachment by adoption of the best management practices (BMPs) outlined in table 9.18 should be the best overall strategy.

Table 9.18 Best management practices for sustainable land use in TRF ecosystems

Arable land use	Pasture development	Agro-forestry	Forest plantations
· Use prime land, and avoid marginal, steep, or shallow soils	· Use prime land, and avoid steep and shallow soils	· Use prime land of high inherent fertility	· Use prime land with no serious limitations
· Remove forest by manual methods, or by shearblade	· Use proper clearing methods, e.g. manual, slash and burn, etc. Tree defoliants can also be used in regions with low tree density. Dead trees can be left standing	· Clear land by manual methods or shearblade techniques	· Clear existing vegetation by manual methods of slash and burn or by shearblade. Some roots and stumps can be left intact
· Use cover crop and mulch farming techniques for soil and water conservation	· Seed with suitable and ecologically adapted mixture of grass and legumes	· Choose native tree spe- cies that do not aggressively compete with annuals	· Seed a leguminous cover crop immediately
· Make frequent use of planted fallows	· Maintain soil fertility as per soil test values. Balanced fertilization is important	· Proper tree management is crucial.	· Establish tree seedlings through the leguminous cover by suppressing it through chemical or mechanical means. Cover crop management is crucial to tree establishment
· Wherever feasible, integrate woody perennials and livestock with food-crop annuals		· Choose appropriate crops and cropping sequences	· Use balanced fertilizer based on soil test values and tree requirements
· Use chemical and organic fertilizers judiciously		· Manage soil fertility in relation to cropping intensity and soil test values	· Use effective soil and water conser vation techniques
· Choose appropriate crops and cropping sequences

Restoring degraded forest lands

Restoration of degraded lands in TRF ecosystems is a high priority if the rate of new deforestation is to be reduced. The choice of land restorative measures to be adopted depends on the type of degradation, the processes involved, and antecedent soil properties and vegetation. Knowing the critical/threshold levels of soil properties, beyond which the soil's life support processes are severely jeopardized, is crucial in this endeavour. Land restorative techniques for soils degraded by different processes are outlined in table 9.19.

Research needs

In view of the ever-increasing demand on limited and fragile resources, the question most often asked is whether soil productivity in TRF ecosystems can be sustained with intensive and continuous farming. The available research data indicate that most tropical soils can be intensively cultivated and produce high and sustained yields by adopting BMPs based on an ecological approach to agriculture. In this connection, land-clearing techniques play an important role. The effects of improper land-clearing methods are observed even 8-10 years after the land has been cleared, and especially when the overall soil fertility has drastically declined. Adopting a land-use system that may produce, say, 60-80 per cent of maximum returns and that avoids causing environmental degradation is a better choice than land-use systems that bring high short-term returns but severely degrade the resource base.

An optimum resource utilization should be based on scientific data obtained through well-designed and adequately equipped long-term experiments. To start meeting this objective, additional research information is needed on evaluating the following:

· land capability and the development of criteria for the choice of rational land use and for appropriate methods of removing vegetation,

· the economic and environmental consequences of different land-use systems,

· methods of restoring forest vegetation and soil quality degraded by land misuse,

· ecologically compatible methods of Imperata control,

· adaptability of those methods of soil and crop management that enhance production from existing land, thereby reducing the need to clear new land.

Considering the limited resources available and the urgent need to use forest resources efficiently, it is important that priorities are defined and research goals are sharply focused. A coordinated effort is needed to achieve these objectives.

Table 9.19 Land restorative techniques

Soil degradative process	Strategies	Land restorative techniques
Soil compaction	Enhance soil structure	Grow planted fallows and deep rooted perennials
	Improve aggregation	Use mulch farming techniques
	Enhance activity of soil fauna, e.g. earthworms	Avoid excessive vehicular traffic
		Use subsoiling discriminatingly and judiciously
Soil erosion	Divert run-on	Isolate the area
	Prevent runoff	Construct diversion channels
	Minimize raindrop impact	Establish permanent ground cover
	Enhance soil structure	Use fertilizets and manures
		Establish vegetative hedges on the contour
		Establish micro-catchments and water-spreading devices to enhance water infiltration
Nutrient depletion	Stop fertility mining practices	Take land out of production and establish planted fallows
	Use balanced fertilizer	Augment nutrient capital by the addition of chemical and organic fertilizers
	Develop nutrient recycling mechanisms	Establish native trees and deep rooted shrubs to facilitate nutrient recycling

References

Aina, P. O. 1979. Soil changes resulting from longterm management practices. Soil Science Society of America Journal 43: 173-177.

FAO (Food and Agriculture Organization of the United Nations). 1981. Forest Resources of Tropical Africa, Asia and Americas. Rome: FAO.

---- 1982. Tropical Forest Resources. FAO Forestry Paper no. 30. Rome: FAO.
---- 1991. Production Year Book. Rome: FAO.

Ghuman, B. S. and R. Lal. 1991. Land clearing and use in the humid Nigerian tropics. II. Soil chemical properties. Soil Science Society of America Journal 55: 184 188.

---- 1992. Effects of soil wetness at the time of land clearing on physical properties and crop response on an Ultisol in southern Nigeria. Soil & Tillage Research 22: 1 11.

Greenland, D. J. and P. H. Nye. 1959. Increase in the carbon and nitrogen contents of tropical soils under natural fallows. Journal of Soil Science 9: 284 299.

Houghton, R. A. 1990. The global effects of tropical deforestation. Environment Science Technology 24: 414 422.

Juo, A. S. R. and R. Lal. 1977. The effect of fallow and continuous cultivation on the chemical and physical properties of an Alfisol in the tropics. Plant Soil 47: 567 584.

Lal, R. 1976. No-tillage effects on soil properties under different crops in western Nigeria. Proceedings Soil Science Society of America 40: 762 768.

---- 1987. Tropical Ecology and Physical Edaphology. Chichester: Wiley.
---- 1991. Agricultural activities and carbon emission from soils of the tropics. Washington D.C.: US Environmental Protection Agency, mimeo.

Lal, R., D. DeYleeschauwer, and R. M. Nganje. 1980. Changes in properties of newly cleared Alfisol as affected by mulching. Soil Science Society of America Journal 44:827 833.

Myers, N. 1989. Deforestation Rates in Tropical Forests and Their Climatic Imp/ications. London: Friends of the Earth.

---- 1991. Tropical forests: Present status and future outlook. In: N. Myers (ed.), Tropical Forests and Climate. Boston: Kluwer Academic Publishers, pp. 3-32.

NRC (National Research Council). 1982. Ecological Aspects of Development in the Humid Tropics. Washington D.C.: National Academy of Sciences.

---- 1993. Sustainable Agriculture and the Environment in the Humid Tropics. Washington D.C.: National Research Council.

Nye, P. H. and D. J. Greenland. 1960. The Soil Under Shifting Cultivation. Technical Comm. 51. Harpenden, England: Commonwealth Bureau of Soils.

Richards, J. F. 1991. Land transformation. In: B. L. Turner, W. C. Clark, R. W. Kates, J. F. Richards, J. T. Mathews, and W. B. Meyer (eds.), The Earth as Transformed by Human Action: Global and Regional Changes in Biosphere over the Past300 Years. New York: Cambridge University Press, pp. 163 178.

Salati, E., T. E. Lovejoy, and P. B. Vose. 1983. Precipitation and water recycling in tropical forests. Environmentalists 3: 67-72.

Williams, M. 1991. Forests. In: B. L. Turner, W. C. Clark, R. W. Kates, J. F. Richards, J. T. Mathews, and W. B. Meyer (eds.), The Earth as Transformed by H uman Action: Global and Regional Changes in Biosphere over the Past 300 Years. New York: Cambridge University Press, pp. 179-201.

WRI (World Resources Institute). 1988 89. Forests and rangelands. In: An Assessment of the Resource Base That Supports Global Economy. Washington D.C.: WRI, pp. 285 295.

---- 1990 91. Forests and rangelands. In: A Guide to the Clobal Environment. Washington D.C.: WRI, pp. 101-120.

---- 1992 93. Towards Sustainable Development. Washington D.C.: WRI.