Crop diversification for weed management - Ecological Management of Agricultural Weeds - page 359

shoots, where their roots are concentrated (Atkinson, 1980). Thus, resource
use in open-canopy tree systems is spatially variable and incomplete: the tree
crops fail to fully use light, water, and nutrients that are available in inter-row
areas.
Patterns of resource use in open-canopy tree crops have led to the differen-
tiation of close-to-tree and inter-row zones for weed and cover crop manage-
ment (Glenn & Welker, 1989). Weeds or cover crops growing close to tree
trunks can compete with trees for resources and reduce tree performance. In
an experiment conducted in West Virginia, Welker & Glenn (1989) grew
young peach trees within a tall fescue ( Festuca arundinacea ) sod and found that
leaf nitrogen, canopy width, trunk diameter, and fruit yield of the trees
increased as the size of vegetation-free areas surrounding the trees increased
up to 9 m 2 . To prevent this type of competitive stress, farmers can suppress
vegetation close to trees by mowing, surface cultivating, grazing, mulching,
or applying herbicides (Glenn & Welker, 1989). In contrast, weeds and cover
crops growing farther away from trees pose little or no competitive threat to
tree growth and production, and can be used to improve and conserve soil,
and aid pest management.In California, Hendricks (1995) found that almond
orchards with cover crops grown between tree rows had more soil organic
matter, higher earthworm and predatory insect densities, reduced insect pest
damage, and similar yields compared to conventionally managed almond
orchards with bare soil. The major additional cost associated with cover crop-
ping was for more irrigation water.
Perennial cover crops have been tested for weed control in established
coffee plantations. In a region of Nicaragua with a five-month dry season, a
three-year-old stand of sun-grown coffee was used to compare the use of two
perennial legume cover crops ( Arachis pintoi and Desmodium ovalifolium ) against
local grower weed management practices that combined mowing and herbi-
cides (paraquat and simazine with spot applications of glyphosate) (Bradshaw,
1993; Bradshaw & Lanini, 1995). Both cover crops and weeds were removed
manually from coffee rows in all treatments during the course of the experi-
ment, and cover crops were hand-weeded for three months after they were
sown. Weed biomass in cover cropped plots was lower than in plots managed
with mowing and herbicides at three of four sampling dates; no difference
was detected among treatments at a fourth sampling date, after growers had
mowed twice and applied herbicides twice (Bradshaw & Lanini, 1995). No dif-
ferences were detected in coffee growth and yield among the legume cover
crop treatments, a weedy control, and the mowing plus herbicide treatment,
though cover crops increased moisture stress in coffee leaves during the last
month of the dry season (Bradshaw, 1993).