The stochastic backward shifts model better corresponds to the fine-scale structural heterogeneity of old-growth Abies-Fagus-Picea forests than the ontogenic life cycle model

Abstract

The ontogenic forest life cycle of regeneration, maturation, senescence and death is a common starting point for the conceptualization of developmental pathways in forest ecosystems. Nonetheless, the usefulness of this framework under natural stand dynamics in small tree collectives formed by individuals with overlapping and weakly synchronised life cycles is not evident. This research investigated empirical relationships between local structural attributes linked with biomass accumulation, mortality in the canopy zone, the occurrence of undercanopy trees and structural heterogeneity quantified in small spatial scales and compared them with anticipations derived from the theoretical ontogenic life cycle model. The material was collected in six old-growth, mixed-species and mixed-aged forests consisting of European beech (Fagus sylvatica L.), silver fir (Abies alba Mill.) and Norway spruce (Picea abies (L.) H. Karst.) and situated in the central Dinaric Mts. (Bosnia and Herzegovina) and Western Carpathians (southern Poland). In each of the stands, a dense grid of circular plots with a radius of 7 and 15 m was established on an area of about 10 ha, and all live (d1.3 ≥ 7 cm) and dead trees (d1.3 ≥ 20 cm) growing within the plots were registered. The analyses indicated that the relationships between local basal areas of live and dead trees are very weak. Ambiguous relationships between the basal area of live and dead trees and the density of undercanopy trees, the high level of structural heterogeneity in small stand patches, and the tendency towards increasing structural heterogeneity with increasing biomass accumulation suggest that the life cycle model based on the sequence of ontogenic phases is a poor projection of the local developmental pathways in the studied ecosystem. The high level of structural heterogeneity found in small stand patches corroborated the alternative hypothesis that the mortality of canopy trees in most cases releases younger or suppressed individuals with delayed growth and only sporadically leads to the formation of a canopy opening reaching the forest floor. A general framework for an alternative model based on stochastic backward shifts was proposed. It emphasizes the diversity of possible developmental trajectories and assumes that the death of a canopy tree shifts local basal area and frequently also structural heterogeneity backwards on the developmental pathway. Depending on the pre-disturbance situation and the kind of disturbance factor, stochastic backward shifts may lead to different “states” in the space of structural attributes.