Macroecological patterns in tree defenses affect forest pest dynamics and patterns of tree mortality. Forecasting forest vulnerability to insect epidemics is rendered more difficult in the absence of knowledge regarding drivers of tree defenses across environmental and temporal (age/ontogeny) gradients. The elevational gradient in plant defense (EGPD) hypothesis posits that natural enemy pressures are positively linked to temperature along climatic gradients, thereby selecting for enhanced defenses at lower elevations while leaving plants less defended at higher elevations. Extensive tree mortality in high-elevation forests during recent bark beetle epidemics has led to presumed support for the EGPD even in the absence of direct tests. Tree properties such as age, growth rate, and size can also affect defenses and, in isolation or via interactions, could be key factors underlying observed patterns of tree mortality. Thus, in a phylogenetically constrained test of the EGPD, I sought to disentangle the interacting influences of tree properties from possible elevational climatic effects on defenses of three widespread pines across a 1500-m elevational transect in the Rocky Mountains, Colorado, USA.
Results/Conclusions
I assessed defenses by measuring the total area occupied by resin ducts in the most recent 5 xylem rings (i.e., 5 years of retrospective assessment) in 304 individuals of Pinus ponderosa (PIPO), 254 of Pinus contorta (PICO), and 242 of Pinus flexilis (PIFL). In addition to resin duct traits, I measured tree age, size, and growth rate (based on xylem ring widths). I assessed the importance of these tree properties and position along the climatic (elevational) gradient on pine defenses via machine learning techniques. Growth rate consistently had the greatest influence on resin duct area in all three species, accounting for 73% of variation in PIPO, 53% in PICO, and 61% in PIFL. Tree age accounted for 17% of variation in resin duct area in both PIPO and PIFL, and 13% in PICO. Position along the elevational gradient had greater influence on resin duct area in PICO and PIFL (23 and 20 % of variation, respectively) than in PIPO (7% of variation) suggesting a smaller influence of climate on defenses in ponderosa pine compared to its congeners. While I found support for the EGPD hypothesis, influences of tree growth rate and age on resin duct area highlight the existence of a hierarchy of drivers of pine defenses and point to the necessity of including these factors in studies of conifer defenses across environmental gradients.