Fact Sheet 4.11. Replenishing Soil Fertility through Agroforestry
in Subhumid Tropical Africa
Leguminous fallows of Sesbania sesban, Tephrosia vogelii, Gliricidia sepium,
Crotalaria grahamiana, and Cajanus cajan accumulate 0.1-0.2 t N ha-1
in their leaves and roots in 1-2 years. These large amounts of nitrogen are
uncommon in the organic farming literature; they are equivalent to mineral fertilizer
input levels in modern agriculture. Upon incorporation of leguminous biomass
into the soil and subsequent mineralization, these improved fallows provide
sufficient nitrogen for one to three subsequent maize crops-doubling to quadrupling
maize yields at the farm scale (Rao et al., 1998; Kwesiga et al.,
1999). There are no transport costs involved because all of the nitrogen is
fixed in the same fields where crops are grown in rotation.
In phosphorus-deficient soils, farmers are beginning to use phosphate rock
applications of 0.125-0.25 t P ha-1 as a capital investment, with an expected
residual effect of 5 years. In addition, biome transfers from hedges of the
wild sunflower tithonia (Tithonia diversifolia) have shown large yield
increases of maize and high-value crops such as vegetables in western Kenya
(Jama et al., 1999a,b). Tithonia leaves contain high nutrient concentrations
(3 percent N, 0.3 percent P, 3 percent K) and decompose rapidly in the soil,
providing a source of soluble carbon that enhances nutrient cycling (Gachengo
et al., 1999). Combinations of tithonia biomass with phosphorus fertilizers
have been particularly effective (Palm et al., 1997; Nziguheba et
al., 1998; Rao et al.,1998). Farmers incorporate leguminous fallows,
tithonia, and phosphate rock into their farming systems in a variety of ways.
Food security has been effectively achieved with these practices. Economic analysis
shows high net present values for these technologies (Sanchez et al.,
The next step envisioned is planting vegetables that produce high-value products
as a way to increase small farmer income and reduce poverty. Some farmers have
reported increases in their net profits from US$91-1665 yr-1 when they have
shifted from maize to vegetables in their now-fertile soils (Nyasimi et al.,
1997). A further step will be the switch to newly domesticated tree crops that
produce high-value products. These "Cinderella" species-so called because their
value has been largely overlooked by science although they are appreciated by
local people-include indigenous fruit trees and other plants that provide medicinal
products, ornamentals, or high-grade timber (Leakey et al., 1996). One
example is Prunus africana, a timber tree that is indigenous to montane
regions of Africa. A substance extracted from its bark to treat prostate gland-related
diseases has an annual market value of US$220 million (Cunningham and Mbenkum,
1993; Simons et al., 1998). Because these trees are cut and killed in
indigenous forests and the bark shipped to Europe, Prunus africana is
now in the CITES Appendix II list of endangered species. With domestication,
this tree is now being turned into a crop, as researchers select superior ecotypes,
ways to harvest the bark sustainably, and eventually the development of extraction
industries located in nearby rural areas (Simons et al., 1998).
Use and Potential
These practices are quite new, having been tested in research in the 1990s;
about 20,000 smallholder farmers currently practice them on roughly 20,000 ha,
primarily in western Kenya and eastern Zambia. The total potential area could
become very large, assuming enabling policies in 10 percent of smallholder farms
in subhumid Africa (8.1 Mha) and 25 Mha in the subhumid tropics of Latin America
and non-paddy rice areas of Asia, all during the next 20 years.
Changes in time-averaged aboveground and soil carbon can be measured via methods
described elsewhere in this Special Report.
Current Knowledge and Scientific Uncertainties
Estimates cited are very preliminary. Hard data are now being developed at ICRAF.
Starting from soils that are 40-60 percent depleted in carbon and have very
little aboveground biomass, measurement of differences can be made and modeled.
The largest uncertainty is the area that will benefit from this technology.
A 10-year period is recommended to assess impact on soil carbon stocks.
Monitoring, Verifiability, and Transparency
Direct measurement (time-averaged) of aboveground and soil stocks should be
used for monitoring. Combining present algorithms for estimating biomass in
shrubs and small trees, standard soil carbon sampling, and GIS techniques appears
feasible. The level of additional sequestered carbon can be readily estimated
by the techniques described above. Assumptions and methodologies associated
with this practice can be explained clearly to facilitate replication and assessment.
Scientific methods are open to review and are replicable over time.
About half of the carbon stored in the soil is likely to have a turnover rate
of >50 years. Cessation of the activity would lead to a loss of soil carbon,
which has been estimated to be 40-60 percent in about 20 years.
Major increases in food security and poverty reduction seem assured. Spillover
effects could occur in developing rural industries and employment. There would
be less dependence on nitrogen fertilizer, the manufacture of which entails
major consumption of fossil fuels (Schlesinger, 1999). Using biologically fixed
nitrogen would lead to potential savings in N2O emissions. These practices also
would reduce dependence on superphosphates, whose manufacture is also highly
fossil fuel-intensive with high risk of pollution. Increases in soil conservation
and below-ground biodiversity are likely. Extinction of the endangered tree
species Prunus africana would be less likely.
Relationship to IPCC Guidelines
See Fact Sheet 4.10.