Overestimation of yield loss of tobacco caused by the aggregated spatial pattern of Meloidogyne incognita

Overestimation of yield loss caused by Meloidogyne incognita on tobacco was calculated as a function of the statistical frequency distribution of sample counts. Sampling frequency distributions were described by a negative binomial model, with parameter k, and the resulting probability generating function was used to calculate discrete damage probabilities. Negative binomial damage predictions were compared to mean-density estimates of damage. Predictions based on mean density alone overestimate yield loss by values ranging from 300% at a k of 0.1 to less than 10% at a k of 1.0. Damage overestimation was described as an exponential function of k and mean density. Preplant sampling data for M. incognita were used to derive a linear model for the estimation of k from mean density, allowing the calculation of yield-loss overestimation based on one parameter, the field mean density. Overestimation of damage ranged from 288% at a density of 50 juveniles/500 cm(3) soil, to 5% at a density of 1,000 juvelfiles/500 cm(3) soil.