2.8.9
INFLUENCE OF PRODUCTION ENVIRONMENT ON CROP YIELD RESPONSES TO DISEASE

RE GAUNT, NA LUCAS, SFA SHAH and HG WHELAN

Plant Science Department, PO Box 84, Lincoln University, Canterbury, New Zealand

Background and objectives
The derivation of descriptive models of the amount of yield loss caused by single disease constraints is straightforward if confined to a narrow range of production inputs, and especially if the duration of the epidemic is short. When applied outside this range, the models are often inaccurate and therefore of limited value [1]. The use of physiological variables such as healthy area duration [2] in the models improves the range of application. Several pathosystems were tested to determine how well this approach can cope with different production environments.

Materials and methods
Field plots of barley exposed to brown rust, potatoes exposed to early blight and peas exposed to ascochyta diseases were grown under normal farmer practice. The barley plots had yields adjusted by nitrogen and water inputs. The potato plots were manipulated either by nitrogen inputs or by different intensities of cyst nematode. Two cultivars of peas with different growth habits (determinate/indeterminate) were sown at different densities. There were at least five subplots in each plot with different disease intensities throughout the season. Yield, dry matter accumulation, healthy area duration, reflectance and total intercepted photosynthetically active radiation (PAR) were measured in all subplots. Models were developed for each pathosystem to test the hypothesis that it was possible to model yield response in a wide range of production environment using relatively simple predictors

Results and conclusions
Single descriptive models for predicting yield, using green area duration or intercepted radiation (PAR) as the predictor, were valid over a wide range of agronomic inputs. However, separate models were required to accommodate the presence of nematodes in potatoes and with different cultivars in peas. The differences in these models are discussed in relation to the separate effects of disease on PAR and radiation use efficiency (RUE) in the pathosystems tested.

References
1. Gaunt RE, 1995. Annual Review of Phytopathology 33, 119-44.
2. Waggoner PE, Berger RD, 1987. Phytopathology 77, 393-98