Population structure and status of isolated hemlock (Tsuga canadensis) patches in the central Appalachian region

Issues concerning forest dynamics and population processes can be examined using population structure analysis. Eastern hemlock (Tsuga canadensis) is an important component of the mixed mesophytic forest of the East Coast. Historically (9,000 - 7,000 years before present), hemlock had dominated the eastern North American landscape but its distribution was subsequently dramatically reduced. In the central and southern Appalachian region, this slow-growing, shade tolerant tree is presently found to dominate small, isolated patches of forest, rather than being continuously distributed throughout it. Questions concerning the existence and persistence of such isolated populations are raised in this study.

Population structure analysis with long-lived trees has long been a topic of discussion. Size and age class distributions are the common population structures under investigation, but their interpretation varies. Hemlock has a weak size and age relationship, therefore interpretations of population dynamics and response to past events is best obtained through studying age distributions. In addition to establishing population structure, this study examined the spatial status of hemlock. Are the populations migrating from the patches which they presently dominate, are they remnant populations which are gradually disappearing from the landscape or are they populations at equilibrium; self-maintaining but confined to their present boundaries?

Three hemlock populations were identified for this study. Each population was selected according to a set of criteria concerning site and population characteristics. Species composition, size structures and age structures were compiled for each site from data collected along a number of transects. Three methods were utilized for the examination of spatial status of each population. The first method involved the relation between maximum and minimum ages of the individuals along a directional gradient. Generalizations of population expansion, contraction or static status can be interpreted from this method. The second method was based on mapping individual hemlock in their exact site locations and color coding them according to various age classifications. Patterns in successful recruitment and establishment could be observed from these maps. The final method involved interpreting the trends of mean ages on directional gradients, thereby supporting the results of the previous two methods.

The results of this study indicate that the chosen hemlock populations do not have constant birth and death rates, which would have resulted in an exponentially declining age class distribution. Rather, the distribution may be a result of periodic recruitment, an internal species-specific mechanism, that is influenced by external environmental conditions. Such conditions may include removal of competing species, periods of stressful climate conditions (drought, severe winters), or periods of ideal climate conditions. With respect to the spatial status of three hemlock populations included in this study, the overall generalization can be made that any expansion that is occurring is doing so at a very slow rate and that complex patterns or waves of expansion and contraction have probably been occurring in these populations. Each of the three populations do not indicate the exact same spatial changes and there is a fair amount of within site variability. Conditions are not uniform around each sites. Topography, soil characteristics and interspecific competition influence rates of population migration. The first method of examination, the maximum/minimum age relation, is a simple method of initially estimating population migration, but the mapping approach elucidates the detail of spatial change and incorporates the densities of individuals remaining in the populations in the different age
classes.