Crop Diseases - Stripe rust in wheat and Sderotinia rot in canola
Stripe rust in wheat is caused by Puccinia striiformis f.sp. tritici (Figure 16.1). Infection and growth is favored at temperatures between 12 to 15°C, with longer time required at lower and higher temperatures. At ideal temperatures, the cycle from infection to new spore production takes about 12 to 14 days, given susceptible plants and sufficient humidity. In the warmer months up to two cycles of the disease can occur per month (Murray et al. 2005).
Sclerotinia rot is caused by Sclerotinia sclerotiorum (Lib.) de Bary (Figure 16.2). Sclerotia (asexual resting propagules) remain viable for many years in the soil. When weather conditions are favorable, the sclerotia germinate to produce apo-thecia (sexual fruiting bodies) (Le Tourneau 1979; Morrall and Uomson 1991). Apothecia produce thousands of air-borne ascospores that can be carried several kilometres by the wind (Brown and Butler 1936; Schwartz and Steadman 1978). Spores that land on canola petals may lodge in the lower canopy of the crop during senescence at the end of flowering. Germinating spores use the petal as a source of nutrient, producing a fungal mycelium that grows and invades the canola plant. Germination and infection are enhanced by wet weather (McLean 1958; Rimmer and Buchwaldt 1995).
Sclerotinia is a monocyclic disease in Australia in keeping with the flowering periodicity of canola, although in some parts of New South Wales (NSW) where a
summer-irrigated crop is grown, two cycles may occur in one year (Hind-Lanoise-let, unpublished data, 2006).
A potentially sustainable way of controlling crop diseases is the breeding of resistant plants, although to optimise control other management strategies such as fungicide use and good land management practices have to also be used, tte latter can include crop rotation and general crop hygiene (Murray and Brown 1987). tte incorporation of multiple disease management strategies reduces the chance and hence severity of attack and limits fungicide use which in turn reduces the risk of the pathogen acquiring resistance to the fungicide. Conventional breeding has, however, not always been successful at producing resistant plant varieties, as can be seen until recently with attempts to breed resistance for S. sclerotiorum into canola (Buchwalt et al. 2003). Even when resistance for a pathogen such as stripe rust has been successfully bred into a variety for several decades this can be overcome by the introduction of a new stripe rust race as occurred in Western Australia in 2002, with spread of rust to the eastern Australian states by 2003 (Murray et al. 2005).
When plant resistance cannot be relied on for the level of pathogen management required, other management strategies such as fungicide application are generally used (Sansford et al. 1995). Routine seasonal application of fungicides is, however, not profitable as procurement and application costs are high, and disease incidence varies greatly with year, region and locality (Sansford et al. 1995; Twengstrom et al. 1998). A major consideration is the potential removal or dilution of fungicides by early or unexpected rains, and in this regard short-term climatic modeling is highly desirable, tte result of climate-pest modelling only take on meaning when interpreted in terms ofbroader risk management considerations.
Models can range in complexity from a simple set of anecdotal rules applied by the subsistence farmer to complex, computer-based models such as those constructed by researchers in collaboration with state departments, tte simplest models are likely to be based on relatively simple causal or "push-pull" relationships (deterministic) whereas complex models are likely to be based on webs of such relationships involving a large number of agrometeorological factors and confounders, with ability to take into consideration the chance of each causal factor potentiating with time, or in terms of some spatial distribution (probabilistic).
All models offer a predictive dimension, giving opportunities for anticipating and hence limiting crop damage. Some use early warning systems, such as changing weather patterns, to allow for early, corrective action (proactive), whereas others rely on the onset of disease as an action trigger (reactive), tte latter models are likely to be of limited effectiveness in controlling damage, hence the need to explore new data and methodologies which can be effectively used in modelling for proactive disease management (Gugel and Morrall 1986; Zadoks 1984). A cornerstone of epidemiological modelling is the collection of relevant local data for disease occurrence and related risk factors, from which risk management models can be developed to allow for a range of actions based on the excedence oflimit values within a predetermined data range (Abawi and Grogan 1975; Last 2001).
Was this article helpful?