Spatial and temporal patterns of balsam fir mortality in spaced and unspaced stands caused by spruce budworm defoliation

Spatial and temporal patterns of balsam fir (Abies balsamea (L.) Mill.) mortality were studied during a spruce budworm (Choristoneura fumiferana Clem.) outbreak from 1976 to 1984 on Cape Breton Island, Nova Scotia. Natural mortality in four insecticide-protected plots was 0% in spaced and 9-15% in unspaced stands, with only the smallest trees dying. Budworm-caused mortality (i.e., total minus natural) was 31-49% and 11-32% in spaced and unspaced young fir plots, respectively, but reached 94-100% in severely defoliated spaced plots, unprecedented in the literature for young fir mortality caused by the spruce budworm. Mortality began in the fourth to sixth year of defoliation, being earliest in the severely defoliated plots. From 80 to 90% of trees that died had >75% cumulative defoliation, and most (64-100%) of the smallest (2 cm DBH) trees died. There was no significant difference in percent mortality between 25 spaced and 13 unspaced plots (p = 0.434), although, on average, mortality was 10-22% higher in the spaced plots. About 20-30% more of the intermediate-sized and largest trees were killed in the spaced plots. High spatial plot to plot variability in mortality occurred, which was apparently related to observed differences in the amount of defoliation and especially the incidence of "back-feeding" (damage to noncurrent foliage), as well as to plot size. Because budworm-caused mortality exhibit a distribution that tends to form large "holes" in stands, the degree of between-plot variability is related to plot size, and it is recommended that small plots that may miss these patches of mortality be avoided.