Factors associated with bark beetle infestations of Colorado Plateau ponderosa pine using repeatedly-measured field plots

Abstract

Decades of research have brought increasing focus to the biotic and abiotic factors associated with bark beetle infestation of North American conifers. Information is lacking, however, regarding beetle-caused mortality for ponderosa pine of the Colorado Plateau. From 1995 to 2019, we installed and periodically remeasured permanent plots in this forest type with a goal of identifying factors related to infestation probability including tree and stand characters, and abiotic factors such as temperature and precipitation. Forty-eight sites were distributed throughout the Colorado Plateau, each with two transects of 10 contiguous ∼ 0.04 ha circular plots. Stand and tree conditions were measured or interpolated for each year from plot establishment in 1995 until the final measurements in 2019. Temperature, precipitation, and other abiotic conditions were estimated from Daymet interpolations. We used a generalized linear mixed model to identify significant explanatory variables resulting in the best accuracy of predicting stem-level bark beetle-caused pine mortality. Tree and stand conditions were found related to the probability of infestation. Pines >15.2 cm dbh were more likely to become infested compared to pines below that threshold; no relationship was found above that breakpoint. Similarly, stems 80–180 years old had elevated probability of infestation compared to stems < 80 years; stems > 180 years were intermediate. Pines with poor crown vigor (low crown volume, suppressed or intermediate canopy position) were more likely to become infested compared to pines with good crown vigor (high crown volume, dominant or open grown canopy position). Pine basal area (BA), a surrogate for intra-species competition, was positively related to probability of infestation. Pine BA infested the previous year, a surrogate for beetle population size, was also positively related to probability of infestation. Several drought- and precipitation-related variables resulted in similar model accuracy. Best model fit among these, by a slight margin, was contemporaneous climate water deficit (CWD), a measure of plant water stress, summed from April-August. Increasing CWD was associated with increased probability of infestation. Finally, counts of days above 16 °C the previous year were negatively related to infestation probability. That is, relatively warm weather during the previous year was associated with reduced beetle success, possibly because it can result in fractional beetle life cycles that disrupt synchronized attacks on new hosts or leave overwintering brood in less cold hardy life stages.