Empirical models of the relationships between early blight (Alternaria solani) and yield of potato (Solanum tuberosum) crop grown under different production environments

Abstract

Four experiments were conducted over two growing seasons during 1995/96 and 1996/97 at Lincoln University and the New Zealand Institute for Crop & Food Research, Lincoln, Canterbury. A large database was used to assess the effects of production environments on disease yield relationships and to compare different models derived from pathogen‐based disease measurement and host‐based disease measurements. All the models were developed using regression analyses. All the single point (SP), multiple point (MP), and integral models derived from host‐based disease measurements explained more variation than models derived from pathogen‐based disease measurements. SP, MP, and integral models were also compared and integral models were selected as the best for comparing disease effects as they combine the effect of disease over the whole growing season in one variable. Further, integral models were tested for the effect of different production environments on disease yield relationships. It was found that the nitrogen‐altered production environments had no significant effect on disease yield relationships, whereas potato cyst nematode‐altered production environments significantly affected the disease yield relationship. A single model was used for the disease yield relationship for nitrogen inputs, whereas separate disease yield relationships were used for each nematode level.