Agroecological Intensification of Smallholder Farming - The Oxford Topic of Food, Politics, and Society - page 114

reduced pest pressure, but no effect was seen in 40 percent of studies, and complexity
increased pest pressure in 15 percent of studies.
Host genetic diversity can influence pathogen population structure, which can in
turn affect disease epidemics. In an experimental study of the effect of landscape hetero-
geneity on the spread of wheat stripe rust, host frequency and the size of the initial epi-
demic focus were found to have significant effects on disease spread (Mundt et al. 2011).
In a study involving joint analysis of three large, country-scale data sets on the wheat leaf
rust epidemics in France, it was found that the extent to which specific varieties were
cultivated affected the frequencies of the corresponding pathogen races, which in turn
influenced the performance of varietal resistance. The results of both of these studies
imply that greater varietal diversity will reduce epidemic pressure on a given host geno-
type, as expected.
Models can contribute to an understanding of plant disease epidemics and the roles of
host resistance and diversity, pathogen population characteristics, and the environment
(including farmers' management options as they affect any of these). Modeling can pro-
vide insights on trade-offs in pest management. For instance, simulation has been used
to explore the utility of various innovations and how they interact with farm types. The
costs and benefits were found to vary with the type of farm (Blazy et al. 2009). In the
Collaborative Crop Research Program's (CCRP) Andean region, Rebaudo et al. (2011)
are using agent-based, cellular automaton models to understand how decision making
influences pest dynamics over time and space.
Plant breeding can make use of resistance at the gene, genotype, and population lev-
els. Through the use of natural allelic diversity, conventional breeding can be effective
at solving most pest and disease problems when well-resourced breeding programs
apply well-designed strategies. Effective resistance breeding requires an understand-
ing of the diversity of types of resistance genes and phenotypes available in crop biodi-
versity. Although breeding for forms of resistance governed by single dominant genes
is relatively straightforward, it has often led to boom-and-bust cycles because insects
and pathogens can rapidly evolve to overcome the resistance. For pest-disease systems
with high evolutionary potential (McDonald and Linde 2002), breeding programs thus
aim to accumulate multiple genes of small effect, which can be achieved by recurrent
selection.
For some pests and diseases for which natural allelic variation for resistance is lim-
ited, it may be worthwhile and possible to seek genetically engineered forms of resis-
tance. Many transgenic schemes have been designed for pest resistance, such as insect
resistance through genes obtained from a bacterium ( Bacillus thuringiensis , or Bt) and
virus resistance (Marra et al. 2002). In sweetpotato, for example, weevils are a signifi-
cant problem, and solutions have been sought in vain though conventional breeding,
integrated management, and even pesticides. Given that sweetpotato is vegetatively
propagated on small plots, it would be possible, in principle, to manage the pest through
transgenic resistance provided by Bt genes without pollen contamination (through the
use of nonflowering varieties) or excessive selection pressure for resistance build-up
on pest populations (since sweetpotato is not grown in uniform monocultures).