Successional Changes In Plant Ecological Species Groups And Fuels On A Jack Pine-Dominated Landscape In Northern Lower Michigan

Differences in vegetation structure, fuel loadings, and the indicator value of ecological species were assessed by sampling permanent forest plots six, sixteen and thirty-seven years after stand-replacing fire in jack pine (Pinus banksiana Lamb.)-dominated ecosystems of northern Lower Michigan. Permanent forest plots were sampled in 2017 (37 years post-fire) across five landforms compared to measurements made in 1986 (6 years post-fire) and 1996 (16 years post-fire). Total jack pine density decreased to less than 10% of its initially high seedling densities in 1986 to a mature overstory in 2017 due to self-thinning, while canopy cover peaked in 1996. Jack pine was the most dominant tree throughout stand development and across landforms, differing significantly in tree height (p = 0.048) among landforms, and accompanied by more tree species on the higher elevation landforms. An overall decline in species richness, total coverage, evenness and diversity of groundcover occurred with stand development. Five of the eight ecological species groups significantly differed in their coverage over time. Canonical variates analysis (CVA) showed that eight ecological species groups were moderately successful at distinguishing landforms in 1986 (97.8% cumulative variance explained by two axes; 63% classification rate, 40% jackknife classification), and least successful in 1996 (41% classification rate, 26% jackknifed classification), with the ice-contact landforms separating out the best in ordinate space for all year analyses. Future research should assess whether local species groups would better discriminate plots into local landforms than the regional discrimination. Several fuel classes peaked in mid-succession, including the height of “ladder fuels” like live woody, dead woody, and live herbaceous vegetation, suggesting the potential spread to crown fire was higher early in stand development. As expected, coarse woody debris loadings were greatest in 1986, lowest in 1996, and increasing in 2017, while duff biomass was highest in 2017 – both of which would affect surface fire behavior. This data will be used to model fire behavior over this landscape