The effect of different production environments on disease: Yield relationships in potato (Solanum tuberosum L.)

Abstract

The influence of different production environments on disease: yield relationships was investigated to test whether it is feasible to develop robust models of the relationship which are applicable across a range of production inputs. The production environments vary from location to location and from season to season, and it is very important to assess how crops respond to diseases grown under varying production environments.

The experiments were conducted using potato as the model crop and early blight as the disease. Different production environments were developed with a range of nitrogen applications in field experiments and different number of potato cyst nematode per ml of soil in bag experiments. A range of early blight (Alternaria solani) epidemics were generated by inoculating plants with spore suspensions. Disease was measured as per cent disease severity (%DS), per cent reflectance (PR),and the effect on leaf area index (LAI) and radiation interception (RI). Yield was measured as total dry matter (TOM) and tuber dry matter. Green leaf area index (GLAI) was calculated from LAI, corrected for disease with PR ratio or with %DS. The amount of photosynthetically active radiation (PAR) intercepted (RI) was calculated from GLAI corrected for disease. Photosynthetically active radiation corrected with %DS (CPAR2) and with PR ratio (CPAR1) were similar but CPAR2 underestimated the effect of early blight on the host. Therefore photosynthetically active radiation corrected with PR ratio (CPAR1) was considered the most appropriate for calculating photosynthetically active radiation interception in the diseased crop canopies. It was shown for the first time that the RI in the diseased crop can be measured accurately by using PR ratio as a correction factor.

Disease caused a large reduction in RI and RUE which resulted in yield suppressions of 26 and 35% in 1995/96 and 1996/97 respectively. The loss in TDM was associated with a 9% reduction in RI both seasons and 17% and 27% reduction in RUE in 1995/96 and 1996/97 respectively. It has been shown that the calculation of RUE from total dry weight at harvest divided by total intercepted RI provided better estimates of RUE than did cumulative dry weight divided by cumulative intercepted RI. The effect of disease was dependent on the method of calculation of RUE as well as on the type of infection and the pathogen type.

The relationships between disease and yield were compared in relation to the disease measurement method and it was found that the host-based measurements of disease were better predictors of yield than pathogen-based disease measurements. Among the host -based disease measurements, PAR intercepted was found the best because it explained most of the variation in disease: yield relationships and explained yield in a more physiological manner. Accumulated PAR models accounted for the differences in incident radiation, canopy structure and radiation interception by green leaves which are important in determining dry matter production.

Disease: yield relationships were derived from several disease measurements and IDM. The area under curve models of GLAI, PR and RI intercepted were the best to test the effect of different production environments on disease: yield relationships. Nitrogen-altered production environments had no significant effect on the disease: yield relationships and a common relationship between disease and yield was therefore possible. However, a separate relationship between disease and yield was required for nematode affected systems.

The effect of production environments on RI and RUE was further explored in plots without disease. The different production environments affected RUE, and because of this predicting dry matter production using RI alone was not enough and RUE was also needed.