(introductory text...)

Nutrient cycling and re-use is an important strategy for sustainable crop production in the humid tropics. It involves returning nutrients removed by crops and animals to the soil for future use. In addition to crops and animals, soil fauna (e.g., earthworms, termites) also play an important role in cycling of several elements, including C, N. P. S. B. Cu. Zn, and Mo. Growing deep-rooted plants is important in order to cycle nutrients from the sub-soil by returning them through crop residue to the surface where the following shallow-rooted crops can use them. Used effectively, recycling can substantially reduce chemical fertilizer requirements. Some important recycling strategies outlined in Fig. 17 include crop residue mulch, deep-rooted perennials, and animal wastes.

Fig. 17 Strategies for recycling nutrients

1. Crop residue mulch

Crop residues contain substantial quantities of plant nutrients. The data in Table 31 show the nutrient composition of the crop residues of some crops grown in the humid tropics. The concentration in oven-dry tissue ranges from 0.58% to 4.0% for N. 0.1% to 1.1% for P. and 0.2% to 3.4% for K. Nitrogen and phosphorus concentrations are generally higher in legumes than in cereals. On a weight basis, the major plant nutrients contained in 1 Mg of crop residue may range from 15 to 60 kg of N. P. and K (Table 32).

The beneficial effects of returning crop residue as mulch on crop yield are well known. These benefits are due not only to the recycling of plant nutrients but also to improvements in soil moisture and temperature regimes, enhancement of soil structure, and erosion control. However, the use of crop residues as fertilizers is especially important to resource-poor farmers. Some examples of the beneficial effects of crop residue mulch on crop yields are shown by the data in Tables 33 through 37. The data in Table 33 show that compared with an unmatched control, crop yields were improved with any mulch material. Rice husks increased maize yield by 0.7 Mg/ha and cassava yield by 12 Mg/ha. The data in Table 34 on yam production on an acid soil in eastern Nigeria show that mulching significantly increased the yam tuber yield. Mulching increased tuber yield by 20% on both ridge till and flat seedbed.

Table 31 Nutrient composition of crop residues of some crops grown in the humid tropics

(kg/ha/yr)
Crop/species	N	P	K	C/N ratio
Cowpea stem	1.07	1.14	2.54
Cowpea leaves	1.99	0.19	2.20
Rice	0.58	0.10	1.38	105.0
Maize	0.59	0.31	1.31	55.0
Oil palm (processed fibre)	1.24	0.10	0.36
Sesbania leaves	4.0	0.19	2.0
Crotolaria spp.	2.89	0.29	0.72
Tephrosia spp.	3.73	0.28	1.78
Water hyacinth	2.04	0.37	3.40	18.0
Azolla spp.	3.68	0.20	0. 15
Typha spp.	1.37	0.21	2.38

(Modified from FAO, 1990)

Returning crop residue as mulch may also have synergistic effects with fertilizer use. The data from the eastern Amazon by Schoningh and Alkamper (1984) showed that crop residue mulches with low C:N ratios had more beneficial effects than those with high C:N ratios (Table 35). On an Ultisol in eastern Nigeria (Tables 36 and 37), the yield of plantain and bananas was drastically improved by residue mulch.

Table 32 Plant nutrients contained in 1 Mg of dry straw

(kg/Mg)
Crop/species	N	P	K	Total
Cowpea stem	10.7	11.4	25.4	47.5
Cowpea leaves	19.9	1.9	22.0	43.8
Rice	5.8	1.0	1 3.8	20.6
Maize	5.9	3.1	13.1	22.1
Oil palm (fibre)	12.4	1.0	3.6	17.0
Sesbania leaves	40.0	1.9	20.0	61.9
Crotolaria spp.	28.9	2.9	7.2	39.0
Tephrosia spp.	37.3	2.8	17.8	57.9
Water hyacinth	20.4	3.7	34.0	58.1
Azolla spp.	36.8	2.0	1.5	40.3
Typha.spp	13.7	2.1	23.8	39 6

(Recalculated from the data in Table 31)

Plantain yield was five times more with mulch than with chemical fertilizers alone. The data in Fig. 18 show that returning crop residue mulch enhanced the beneficial effects of fertilizer application in maize yield (Kang, 1993). Without fertilizer application, residue retention had little effect on maize grain yield.

Table 33 Crop yield response to 22 different mulch materials applied on Alfisols in Nigeria

(Mg/ha)
Mulch	Cassava (fresh roots)	Maize	Cowpea	Soybean
Bare soil (control)	16.4 def	3.0 e	0.6 a	0.6 de
Maize stover	16.4 def	3.3 cd	1.1 a	1.5 abc
Maize cobs	17.8; cdef	3.3 cd	1.1 a	1.4 abed
Oil palm leaves	17.1 def	3.2 cd	1.2 a	0.9 bcde
Rice straw	17.9 cdef	3.5 bed	1.0 a	1.5 abc
Rice husks	28.3 a	3.7 abc	1.1 a	0.8 de
Kikuyu grass straw	14.2 ef	3.3 cd	1.2 a	1.4 abed
Elephant/napier grass (Pennisetum)	16.6 def	3.3 ed	0.9 a	1.3 bed
Guinea grass	15.5 f	3.6 bed	2.1 b	1.5 ab
Andropogon straw	18.5 cdef	3.5 bed	1.0 a	1.2 bcde
Cattail straw (Typha)	16.7 def	3.1 cd	1.0 a	1.1 bcde
Cassava stem (chipped)	20.9 cd	3.8 abc	0.9 a	1.4 abcd
Pigeon pea tops	22.9 be	3.7 abc	1.1 a	0.9 cde
Pigeon pea stem (chipped)	19.9 café	3.5 bed	1.0 a	1.3 bcd
Legume husks	26.4 ab	4.4 a	1.0 a	1.5 abc
Soybean tops	22.9 be	4.2 ab	1.0 a	1.2 bcde
Hemp (Eupatroium)	18.8 cdef	3.6 abc	1.0 a	1.2 bcde
Mixed twigs (chipped)	18.5 cdef	3.4 bed	1.0 a	1.2 bcde
Sawdust	20.5 cde	3.7 abc	0.9 a	1.9 a
Black plastic	30.5 ab	3.0 cd	0.9a	1.1 bcde
Transluscent plastic	27.7 ab	2.7 d	1.0 a	1.1 bcde
Fine gravel	22.9 be	3.1 cd	1.0 a	1.0 bcde

Figures followed by similar Ietters are stastically similar within vertical roust (Okigbo and Lal. 1980)

Table 34 Effects of tillage methods and mulching on yield and yield components of yam tubers in eastern Nigeria

Treatment	Diameter (cm)	Length (cm)	Number	Mean tuber yield (Mg/ha)
Ridge, mulch	16.7 a	23.5 b	9936 b	15.4 a
Flat, mulch	17.1 a	21.7a	12916a	16.1 a
Ridge, no mulch	13.3 b	20.7 a	10385 b	12.8 b
Flat, no mulch	13.8 b	20.6 a	10128 b	13.4 b

Figures followed by similar letters are statistically similar within vertical rows.

(Maduakor et al., 1984}

Table 35 Yield response of maize and cowpea to different mulch materials and mineral fertilizer in an eastern Amazon Oxisol, Capitao Poco (CPATU) Para, Brazil, 1983

(kg/Mg)
Mulches used (10 Mg/ha of dm)	First crop (maize)*		Second crop (cowpea)**
	NPK (kg/ha) 20-80-60	NPK 0-0-0	NPK 30-80-60	NPK 0-0-0
Elephant grass	4646	2144	1227	80
Pueraria	5697	3342	1187	114
Weeds	4911	2215	1394	105
Sec. Forest (2 3 years)	4462	1560	1191	35
Sec. Forest (4 5 years)	4479	1807	1397	95
Rice husks	4398	1146	1487	163
Maize cobs + husks	4863	2101	1302	41
Bare soil	3539	78	1169	7
LSD. 5%	987		281
LSD. 1%	1321		376
LSD. 0.1%	1729		493

* Grain moisture content 14. 5%
** Grain moisture content 13, %

(Schoningh and Alkamper 1984)

2. Agroforestry systems

Agroforestry systems involve growing woody herbaceous species and perennials in association with food crops and livestock on the same piece of land. Agroforestry systems have been described extensively in several reports (i.e., Kang et al., 1981, 1989, 1990; Harwood, 1987;
Nair. 1989; Szott et al., 1991). They are known to increase ecological diversity within a landscape unit and optimize the use of limited resources through the integration of complementary components. There are three principal types of agroforestry systems (Fig. 19).

Table 36 Effect of mulching and fertilizer on yield of plantain and banana on an acid soil in eastern Nigeria

(Mg/ha)
Treatment	Plantain	Banana
	Giant	Medium
No mulch, no fertilizer*
No mulch, fertilizer	18.0 a	16.7 a 7.5 a
Mulch, no fertilizer	17.2 a	15.8 a 9.5 a
Mulch and fertilizer	31.3 b	19.8 a 13.3 b

* Most plants broke.

Numbers in the same column followed by the same letter are not significantly different at 5%.

(IITA. 1981)

Table 37 Comparative effects of fertilizer and mulch on plantain yield on an acid soil in eastern Nigeria

Parameter	Mulch	Fertilizer
Yield (Mg/ha)	22.8	4.8
Bunch weight (kg)	11.8	8.1
Plants harvested (% of planted)	116.0	36.0
Harvest duration (months)	10.0	6.0

(IITA. 1981)

(i) Agrisilvicultural: This system involves simultaneously growing crops and trees on the same piece of land. Some commonly used agrisilviculture systems include alley cropping (Plate 38) and hedgerow cropping.

Fig. 18 Synergistic effects of using fertilizers in combination with returning crop residue

* N₀P₀K₀, = control with no fertilizer. (a) years 1 4: N₁ = 80 kg N/ha, N₂ = 160 kg N/ha; (b) years 5 8: Ni = 100 kg N/ha, N₁ = 200 kg N/ha; (c) years 9 10: Ni = 75 kg N/ha, N₂ = 150 kg N/ha. P₁ = 30 kg P/ha. P₂ = 60 kg P/ha, K₁ = 40 kg K/ha. K₂ = 80 kg K/ha.

(Kang, 1993)*

Fig. 19 Types of agroforestry systems

(ii) Silvopastoral: This system involves raising livestock on improved pastures grown in association with trees. Some commonly used systems are alley farming and live fences (Plate 39).

Table 38 Commonly recommended species for agroforestry systems in the humid tropics

Species	Growth characteristics	Uses
Acioa bateri	Fast-growing shrub	Alley cropping, nitrogen fixation
Albizia falcate	Tree grows to 30 m	Erosion control, nitrogen fixation
Albizia lebbeck	Tree grows to 25 m	Erosion control, nitrogen fixation
Anthonotha	Fast-growing shrub	Alley cropping, nitrogen fixation macrophylla
Calliandra calothyrsus	Fast-growing shrub to 8 m, on acid soils	Alley cropping, nitrogen fixation
Cassia siamea	Shrub grows to 8 m. vigorous coppicing	Fuelwood, nitrogen fixation, lumber
Erythrina spp.	Tree grows to 20 m, often thorny, coppices shell	Live fences, nitrogen fixation, fuelwood, fodder
Flemingia macrophylla	Shrub grows to 3 m	Alley cropping, nitrogen fixation
Gliridia sepium	Fast-growing tree to 20 m. vigorous coppicing	Alley cropping, nitrogen fixation. forage, fodder. staking material
Inga spp.	Nitrogen-fixing shrub, acid- tolerant	Alley cropping. nitrogen fixation
Leucaena leucocephala	Tree grows to 20 m, fast-growing on non-acid soils. vigorous coppicinig	Fodder. fuelwood. erosion control, nitrogen fixation, alley cropping. staking material
Panagomia pinneta	Small tree. grows to 8m	Erosion control live hedges
Sesbania spp.	Fast-growing loss: tree	Erosion control, nitrogen fixations

(NRC, 1993a)

Table 39 Merits and limitations of agroforestry systems

Merits	Limitations
Reduction in fallow period and high cropping intensity over longer time period	High labor input
Erosion control and runoff management	Highly skilled management
Strengthening of nutrient cycling mechanisms leading to savings in fertilizer use	Low yields due to allelopathic effect and competition among trees and food crops for light, water, and nutrients
Alternate products (e.g., fodder, fuel, staking water mulch and food crops)	Limited application on soils of moderate to low soil fertility
Saving, land and decrease in need for clearing new land	Potentially high risks of pests and diseases
Improved traditional system, therefore, ecologically compatible	Difficulties in adoption due to traditional land tenure system

(iii) Agrisilvopastoral: This system involves a three-way mixture based on a combination of crops, trees. and animals. Such a system requires skillful management, and can be sustainable even in harsh environments and fragile soils.

A wide selection of tree species and woody shrubs can be used for agroforestry systems (Table 38). Some of these trees are suited for acid soil conditions and others for erosion control, some are more appropriate as forage trees, and still others are useful for pruning to be used as mulch. The choice of appropriate species is critical to the success of agroforestry systems. In addition to the intended use, the choice of tree and associated crop species also depends on cultural and ethnic factors of social importance.

The merits and limitations of agroforestry systems are shown in Table 39. A principal advantage of these systems is the reduction in the length of the fallow period and the potential for continuous and intensive cropping. Agroforestry may facilitate intensive land use for multiple uses on relatively fertile soils. It may also enable relatively more intensive use on steep lands and marginal soils, which cannot be used otherwise. A major advantage of agroforestry systems on sloping lands is erosion control. Closely spaced contour hedgerows of suitable woody perennials and shrubs can drastically reduce the risks of runoff and accelerated soil erosion. The data in Table 40 indicate large reductions in runoff and soil erosion with hedgerows of Gliricidia and Leucaena established at 2 and 4 m intervals.

Table 40 Alley-cropping effects on runoff and soil erosion from maize-cowpea rotation measured in 1984

Treatment	Runoff		Erosion (Mg/ha/yr)
	Total (mm)	Fraction of rainfall (%)
Plow-till	232	17.1	14.9
No-till	6	0.4	0.03
Leucaena 4 m	10	0.7	0.9
Leucaena 2 m	13	1.0	0.1
Gliricidia, 4 m	50	1.5	1.7
Gliricidia. 2 m	38	2.8	3.3

(Lal, 1989a)

Table 41 Net primary production of biomass for commonly recommended multi-purpose tree species in the humid tropics

Species	Net primary production of biomass (ka/ha/yr)
Acacia auriculiformis	3000-4000
Acacia mangium	2500 3500
Albizia falcata	4000-5000
Alchornea cordifolia	2000-3000
Calliandra calothyrsus	2500-3500
Cordia alliodora	2500-3500
Dalbergia latifolia	4000-5000
Erythrina poeppigiana	4000-6000
Gmelina arborea	1500-5000
Leucaena leucocephala	3000-5000

(NRC, 1993a)

Table 42 Nutrient composition of foliage of some trees and woody perennials grown with agroforestry systems in the humid tropics

(%)
Species	N	P	K
Cassia siamea (leaves)	1.91	0.18	1.03
Tephrosia sp.	3.73	0.28	1.78
gliricidia sp.	4 15	0.27	300
Leucaena leucocephala	3.85	0.17	1.46
Erythrina sp.	400	0.29	3.05

(FAO. 1190)

Another benefit of growing woody perennials and trees in association with crops is the large quantity of biomass produced. The net primary production of biomass for perennials ranges from 1.5 to 6 Mg/ha/yr (Table 41). This biomass is a valuable resource for small land holders of the humid tropics. In addition to its use as forage, fuel, and staking material, the biomass can also be returned to the soil as mulch for soil protection and nutrient recycling. The nutrient content of the foliage of some of these species ranges from 2% to 4% for N. 0.2% to 0.3% for P. and 1% to 3% for K (Table 42). Consequently, a substantial amount of plant-available nutrients can be added to the soil by returning 3 to 7 Mg/ha of biomass. The data in Table 43 show that nutrients contributed by the biomass returned to the soil from Leucaenu can be 275 to 440 kg/ha of N. 15 to 49 kg/ha of P. 133 to 264 kg/ha of K, 74 to 195 kg/ha of Ca, and 17 to 52 kg/ha of Mg. Nutrients contributed vary widely among different species (Table 44). Rather than prunings, even the litter fall from some trees can add substantial amounts of nutrients (Table 45). Not all the nutrients recycled in the biomass, however, are available to crops. The nutrient use efficiency may be only 20% to 30% or less.

Table 43 Amount of nutrients in Leucaena leucocephala prunings

(kg/ha)
Year	Nutrients
	N	P	K	Ca	Mg	Total
1985	440	49	264	195	52	1489
1988	408	37	244	155	29	873
1989	275	15	133	74	17	514
1990	281	16	188	106	26	617

(Hauser and Kang, 1993)

Table 44 Estimated nutrient addition through prunings of four woody shrubs grown at 4 m intervals on an Alfisol in western Nigeria

(%)
Species	Biomass yield (Mg/ha/yr)	Nutrient yield
		N	P	K	Ca	Mg	Total
Acioa barteri	30	41	4	20	15	5	85
Alchornea cordifolia	4.0	85	6	48	42	8	189
Gliricidia sepium	5.5	169	11	149	104	18	451
Leucaena leucocephala	74	247	70	184	98	16	565

(Kang and Wilson. 1987)

Table 45 Biomass and nutrient addition by litter fall of a three-year-old plantation of Cassia siamea

(kg/ha)
Month	Leaf fall (Mg/ha)	Nutrient addition
		N	Ca	Mg	K
January	5.9	110.3	88.3	13.0	39.1
February	5.7	106.4	85.2	12.6	37.7
March	2.4	43.7	35.0	5.2	15.5
April	1.5	28.1	22.5	3.3	10.0
May	1.9	35.1	28.1	4.1	12.5
June	1.4	25.8	20.7	3.0	9.2
July	2.1	40.0	32.0	4.7	14.2
August	3.4	62.5	50.1	7.3	22.2
September	9.0	167.8	134.4	19.8	59.5
October	9.6	179.3	143.6	21.2	63.6
November	12.7	236.5	189.7	28.0	84.0
December	16.4	305.0	244.3	36.1	108.2
Total	172.0	1340.8	1073.9	158.4	475.7

(Ghuman and Lal. 1990)

Table 46 Grain yield of maize alley cropped with Leucaena leucocephala hedgerows and in plots with no hedgerows (control) on an Alfisol in southwestern Nigeria, and absolute and relative yield advantages of alley cropping

(kg ha)
Grain yield
	1985	1986	1988	1989	1990	Mean
Alley cropping	4890 a	3470 a	5657 a	4141	5376	4707
Control	3990 b	2240 a	5084 b	3353	4032	3740
Difference	900	1230	573	788	1344	967
Relative %	2.6	54.9	11.3	23.5	33.3	25.8

(Kang., 1993)

Table 47 Effects of agroforestry systems on relative grain yields of cowpea a tropical Alfisol

Treatment	Relative grain yield for different years *
	1982	1983	1984	1985	1986	1987	Mean
Plow-till	128	78	79	77	176	65	100
No-till	270	147	212	139	177	38	164
Leucuena, 4 m	177	91	103	73	51	39	89
Leucaena, 2 m	129	57	89	28	26	42	62
Gliricidia 4 m	168	106	119	105	80	37	103
Gliricidia, 2 m	124	95	120	72	41	40	82

* Relative yield is calculated as a ratio of actual yield to average yield of all seasons and all treatments. expressed in percent. No fertilizer was applied.

(Lal 1989a)

Fig. 20 Grain yield as affected by position of the crop row from the perennial hedgerow

Table 48 Grain yield of rice from an experiment conducted at Yurimaguas, Peru, with three species of shrubs

Species	Row from the shrub	Grain yield (kg/ha)
Inga	1	723
	2	1942
	3	2162
	4	1986
	5	1975
	6	2052
	8	1923
	10	1875
Erythrina	1	1582
	2	2014
	3	1888
	4	2059
	5	2056
	6	2044
	8	2107
	10	1997
Leucaena	1	1718
	2	2084
	3	1948
	4	2128
	5	2253
	6	2321
	8	2106
	10	2419

(TROPSOIL.1987)

Because of added nutrients and other favorable soil physical factors, total production can be more with agroforestry than with simple crop-based, tree-based, or animal-based systems. However, the yield of individual components may be decreased. The data in Table 46 show that the yield of maize was the same or better with alley cropping than without. However, the data in Table 47 on cowpea yield show a significant yield reduction with alley cropping, probably due to allelopathic effects. The reduction in average cowpea yield was as much as 11% by Leucaena at 4 m intervals and 38% by Leucaena at 2 m intervals. Yield reduction may also happen due to competition for nutrients between perennials and annuals. An example of the competition is shown by the data in Table 48 on an acid soil at Yurimaquas, Peru. Rice crop yields increased with wider spacing between hedgerows, and with increase in distance from the hedgerow (Fig. 20). The grain yield of the row next to the hedge of Inga or Erythrina was about 1300 kg/ha. The yield was about 1500 kg/ha or less for the fourth spacing The grain yield increased to more than 2000 kg/ha for the eighth row from the hedgerow. The grain yield of the row next to the hedgerow decreased by about 40% regardless of the hedgerow species. In addition, the high labor requirement is another limitation of agroforestry systems. The system is labor-intensive and complete mechanization is often difficult and not feasible to achieve.