Abstract Fuelwood is the main energy source for
households in rural Africa, but its supply is rapidly
declining especially in the densely populated areas.
Short duration planted tree fallows, an agroforestry
technology widely promoted in sub-Sahara Africa for
soil fertility improvement may offer some remedy.
Our objective was to determine the fuelwood production
potential of 6, 12 and 18 months (the
common fallow rotation periods) old Crotalaria
grahamiana, Crotalaria paulina, Tephrosia vogelli
and Tephrosia candida fallows under farmer-managed
conditions in western Kenya. Based on plotlevel
yields, we estimated the extent to which these
tree fallows would meet household and sub-national
fuelwood needs if farmers planted at least 0.25 hectares,
the proportion of land that is typically left under
natural fallows by farmers in the region. Fuelwood
yield was affected significantly (P\0.05) by the
interaction between species and fallow duration.
Among the 6-month-old fallows, T. candida produced
the highest fuelwood (8.9 t ha-1
), compared
with the rest that produced between 5.6 and
6.2 t ha-1
. Twelve months old T. candida and C.
paulina also produced significantly higher fuelwood
yield (average, 9.6 t ha-1
) than T. vogelli and C.
grahamiana of the same age. Between the fallow
durations, the 18-month fallows produced the most
fuelwood among the species evaluated, averaging
14.7 t ha-1
. This was 2–3 times higher than the
average yields of 6 and 12-month-old fallows whose
yields were not significantly different. The actual
fuelwood harvested from the plots that were planted
to improved fallows (which ranged from 0.01 to
0.08 ha) would last a typical household between 11.8
and 124.8 days depending on the species and fallow
duration. This would increase to 268.5 (0.7 years)
and 1173.7 days (0.7–3.2 years) if farmers were to
increase area planted to 0.25 ha. Farmers typically
planted the fallows at high stand densities (over
100,000 plants ha-1 on average) in order to maximize
their benefits of improving soil fertility and
providing fuelwood at the same time. This potential
could be increased if more land (which fortunately
exists) was planted to the fallows within the farms in
the region. The research and development needs for
this to happen at the desired scale are highlighted in
the paper.