Knowledge, science, and practice in ecological weed management: farmer–extensionist–scientist interactions - Ecological Management of Agricultural Weeds - page 117

ensiformis decreased less desirable vegetation, increased fallow biomass and
nutrient content of the mulch, and also increased bean yields up to 50% (G.
Melendes, unpublished data). Upgraded fallows of mixed vegetation had
higher bean yields with less pest damage than pure improved fallows, sug-
gesting that farmers should introduce certain higher-biomass-producing
species in patches without eliminating selected resident fallow species.
Adaptive management and farmer-extensionist-scientist
interactions
Adaptive weed management following any of the three approaches
can contribute to lower weed losses and costs. How might farmers, extension-
ists, and scientists interact to make this happen?
Precision agriculture is likely to follow the technology transfer diffusion
model that has been employed for the promotion of other purchased inputs in
crop production (Table 3.2). Off-farm scientists develop the sensing devices
for commercial farm machinery, the links for satellite communication, soft-
ware for field data interpretation, and genetically engineered crop varieties.
Select groups of retailers and innovative farmers then pilot-test the products.
Custom applicators, crop consultants, and data processing services, with
support from public extension in regions with large demand, will make the
technology available through contracts. Their focus is likely to be on larger
producers who contract large quantities of inputs (Nowak, 1997). Several
questions are pending (Hewitt & Smith, 1996; National Research Council,
1997).Will the components of precision agriculture be available to the major-
ity of producers? Who will have access to the data banks of crop yield response
by season and soil type? Will the technology be used for yield maximization
for a few farmers or to provide the information base for less risky, more envi-
ronmentally sound agriculture practiced by a majority of farmers who use
mechanization? Hamilton (1995, pp. 146-51) concluded that making compu-
terized data analysis/decision aids more hands-on and transparent would
improve farmer decision-making. Alessi (1996) proposed that a wider and
more diverse group of farmers should participate in pilot-testing of precision
agriculture technology. In this way the technology would serve broader com-
munity interests.
The interaction among farmers, extensionists, and scientists appropriate
for the development of the ecological management of weed patchiness and
uncertainty can be termed participatory learning for action (Table 3.2) (after
Hamiliton, 1995, p. 14). This process draws on field and landscape perspec-
tives of weed control (Firbank, 1993; Cousens & Mortimer, 1995, pp. 217-42,