Innovating in a dynamic technical context - Water Scarcity, Livelihoods and Food Security - page 107

failure. New ideas are tested, accepted or discarded in a dynamic and iterative
process of learning selection (Douthwaite, 2002; Perrin, 2002). This learning
process takes more time than is available in a short-duration project . A history
of the successful introduction of conservation agriculture in rice-wheat systems
in the Indo-Gangetic Plains describes the setbacks and unanticipated directions
(Harrington and Hobbs, 2009). The process took more than 20 years.
Learning selection can be applied to complex systems such as agroecosystems
(Douthwaite and Gummert, 2010). Learning strategies can be structured into
theories of change and formalized into outcome logic models (Alvarez et al.,
2010, see Chapter 3). This implies that researchers will be well served if they
can foresee the trajectory of innovation using time frames longer than that of
their current project. This allows them to build on past learning and to link up
with concurrent projects on similar issues being managed by other partners.
The experience of the CPWF is that using systems of innovation in a longer
time frame is often the key to progress and ultimate success. We describe three
examples from CPWF projects. In all three examples, engagement and the
social processes of innovation were as important as the technical research.
Institutions and policies were as important as technologies and getting to
outcomes covered multiple projects only some of which were managed by the
CPWF. Moreover, the innovation trajectory covered 8-10 or more years.
Slash and mulch
Project PN15 in Central America worked on the replacement of slash and
burn on hillsides with slash and mulch (locally named Quesungual slash and
mulch agroforestry system, QSMAS). Both feature long-term, multi-year
rotations, with grain crops alternating with tree regrowth. Slash and burn
involves slashing a plot, letting plant residues dry, and then burning them,
leaving nutrient-rich ash but little soil cover. QSMAS also involves slashing
(and removing valuable wood for other purposes while preserving key tree
saplings), letting plant residues dry, and then planting crops directly into the
soil cover. The project showed the benefit of slash and mulch on soil fertility,
soil microorganisms, soil water-holding capacity, improved drought tolerance,
reduced erosion, reduced risk of landslides during heavy rain, more rapid tree
re-growth, reduced deforestation, higher crop yields, increased productivity
and higher family incomes (Pavon et al., 2006; Castro et al., 2009).
PN15 did not discover or invent QSMAS, which comes from work in the
early 1990s when the Food and Agriculture Organization of the United
Nations (FAO) and other partners developed it with farmers in the village
named Quesungual in Honduras. QSMAS proved itself during the El Niño
drought of 1997 and Hurricane Mitch in 1998. There was reduced drought
damage and little erosion and no landslides during the hurricane.
Project PN15 first reviewed and synthesized existing information collected
by the FAO project 1995-2005. Based on this synthesis, the project carried
out research to accomplish three tasks: (1) measure and quantify the ecological