Effects of endangered species management on jack pine (Pinus banksiana Lamb.)-dominated ecosystems in northern Lower Michigan

Background/Question/Methods

Jack pine (Pinus banksiana Lamb.)-dominated ecosystems in northern Lower Michigan (NLM) are the primary nesting habitat for the federally endangered Kirtland’s warbler (Setophaga kirtlandii Baird, KW). Early-successional KW breeding grounds were historically produced by stand-replacing wildfires, but region-wide fire suppression has necessitated the management of jack pine plantations for habitat since the 1970’s, resulting in an increase in KW to twice the original recovery goal. Given this successful recovery effort and the imminent de-listing of KW, a historic baseline may be used to determine ecologically-based future management goals. Using geostatistical modeling to interpolate 19^th century General Land Office survey notes, we tested the hypothesis that modern management practices have moved the age distribution of jack pine-dominated ecosystems in NLM outside its historic range of variation. We compared pre-European settlement and current stand age distributions in KW management areas using landscape metrics to quantify age-class changes over time. The regional pre-European age-class was compared with that of Van Wagner’s theoretical model of age distributions in fire-prone landscapes to determine the similarity of the two landscapes and to assess the applicability of the model as a regional management tool.

Results/Conclusions

The age distribution of modern, plantation-dominated forests is younger than historic, wildfire-produced forests, suggesting that KW habitat creation has altered the landscape. Three KW management-based age-classes (< 20 years, 21-50 years, and > 50 years) are now evenly distributed across the landscape (30.8%, 39.3% and 29.9% respectively) in contrast to the variation seen in the pre-European distribution (5.4%, 19%, 75.6%). This pronounced change in the age structure of jack pine-dominated ecosystems in NLM indicates a restriction of the historic range of variability, largely due to the conversion of older stands to young plantations suitable for KW habitat. Landscape metrics suggest that the modern landscape is not only younger, but also patchier and more highly fragmented than the pre-European landscape. The theoretical age distribution derived from Van Wagner’s model (28.8%, 28.4%, 34.3%) differed from the pre-European regional distribution (7.6%, 17.4%, 75%). Pre-European settlement patterns of age-class distributions offer insights for restoration of the historic range of age variability on the landscape, and could provide a foundation for a more ecosystem-based management plan which may simultaneously reduce the costs of forest management, maintain age-appropriate KW habitat, and restore natural variation to a highly altered landscape.