A quantitative approach for assessing bark beetle infestations: A study of Pityokteines spinidens Reitt. egg gallery densities in windthrown Abies alba Mill

摘要

The presented statistical method enables monitoring of bark beetles that colonise different parts of the wind-thrown tree stem at different densities. The bark beetle, Pityokteines spinidens Reitt. occurring in forests with the fir Abies alba Mill. was used as an example. We propose a three-step statistical procedure. In the first step, a hierarchical cluster analysis (HCA), the single gamma function, and the gamma mixture distributions are used to model the distribution of egg galleries of P. spinidens along an A. alba windthrown stem. In the second step, a generalised linear model (GLM) and a generalised additive model (GAM) are employed to analyse the relationships between the number of P. spinidens egg galleries in selected stem sections and the total P. spinidens egg gallery density on the entire A. alba windthrown stem. In the third step, the mean total P. spinidens egg gallery density in the investigated area is estimated using survey sampling with simple random sampling without replacement (SRSWOR). The first and second steps allow us to identify which sections of the A. alba windthrown stem should be evaluated to adequately estimate the total egg gallery density (to save time and money and to reduce the amount of debarking performed per tree). The third step enables the calculation of the estimated relative standard error (to analyse the precision of the proposed method).Data were collected between 1994 and 2003 (data set D1) and in 2014 (data set D2) in forests including A. alba in the Swietokrzyski National Park, Poland. D1 was used to build the GLM and GAM, while D2 was employed to validate the models and to illustrate the proposed method.Six main types of egg gallery distributions along the stem were distinguished. P. spinidens inhabited all parts of the stem of A. alba windthrowns. Validation of the models suggested that the GAM performed better than the GLM. The predicted and observed data were closely consistent. This statistical evaluation of P. spinidens egg gallery densities will aid in the assessment of the ecological roles played by this species in A. alba ecosystems. Similar statistical approaches can be developed for other bark beetles colonising tree stems.