Early Regeneration and Structural Responses to Patch Selection and Structural Retention in Second-Growth Northern Hardwoods

For. Sci. 61(1):183–189 http://dx.doi.org/10.5849/forsci.13-180 Copyright © 2015 Society of American Foresters APPLIED RESEARCH silviculture Anthony W. D’Amato, Paul F. Catanzaro, and Lena S. Fletcher Restoration of late-successional conditions to second-growth forests has become a management objective on many ownerships. For northern hardwood forests, restoration targets include a higher abundance of large trees and coarse woody debris and greater diversity of tree species and size classes. Patch-selection harvests 0.12 ha in size were applied in conjunction with structural restoration/enhancement treatments, including within-patch legacy tree retention and downed woody debris (DWD) creation, to determine the effectiveness of these approaches at recruiting late-successional structure and intolerant and midtolerant tree species. Annual mortality rate of retained legacy trees was quite low over the 3 years postharvest (1.7%) and individual legacy tree diameter growth rate ranged from 0.2–1.0 cm yr⫺1. Felling and retention of culls generated within-gap DWD volumes similar to old-growth levels. Sugar maple (Acer saccharum Marsh.), American beech (Fagus grandifolia Ehrh.), and striped maple (Acer pensylvanicum L.) dominated the regeneration layer 3 years postharvest in all treatments; however, abundance of intolerant (black cherry; Prunus serotina L.) and midtolerant (black and yellow birch; Betula lenta L. and Betula alleghaniensis Britton.) species was also increased in harvest gaps relative to unharvested controls. Within-gap legacy tree retention hastened sapling development, particularly of intolerant species, highlighting potential tradeoffs in achieving structural and compositional objectives with this gap-based approach. Keywords: northern hardwoods, uneven-aged management, patch selection, late-successional forests, Massachusetts, coarse woody debris P rior to European settlement, late-successional forests were a dominant feature in the northern hardwood region of northeastern North America; however, centuries of human land use have reduced these conditions to a small fraction of contemporary landscapes (Davis 1996, D’Amato et al. 2006). Recognition of the value of late-successional forests for sustaining native biodiversity and maintaining critical ecosystem services, such as carbon storage, has led to recommendations for modifying traditional regeneration methods to restore late-successional structural and compositional characteristics to second-growth forests (Keeton 2006, Root et al. 2007). These modifications include the deliberate retention of larger diameter trees and coarse woody debris and the use of group selection and irregular shelterwood approaches to restore the structural and compositional conditions historically present in these forests (Keeton 2006, Hanson et al. 2012, Klingsporn et al. 2012). Given our generally limited experience with these modified approaches, there is a great need for empirical studies examining the impacts of late-successional restoration treatments on the structural and compositional development of second-growth northern hardwoods and long-term growth and yield (cf. Saunders and Arseneault 2013). Common objectives related to restoring late-successional forest conditions include increasing the representation of historically important canopy tree species and promoting multicohort age structures (Crow et al. 2002). These objectives relate to the biodiversity benefits presented by compositionally and structural diverse forest stands, as well as the commercial importance of less-tolerant species, such as Betula alleghaniensis (Keeton 2006). However, contemporary changes in understory competitive conditions in many northern hardwood forests pose an important obstacle to achieving these objectives (Royo and Carson 2006). These changes include the development of dense understories dominated by a few native shrub and tree species and have been related to alterations in historic disturbance regimes (Nyland et al. 2006a) and increased levels of Manuscript received November 15, 2013; accepted March 6, 2014; published online April 3, 2014. Affiliations: Anthony W. D’Amato (), University of Minnesota, Department of Forest Resources, St. Paul, MN. Paul F. Catanzaro, University of Massachusetts, Department of Environmental Conservation. Lena S. Fletcher, University of Massachusetts. Acknowledgments: The authors thank Paul Strausburg for graciously providing his landbase for conducting this study. The Massachusetts Chapter of The Nature Conservancy provided the funding for this work. This article uses metric units; the applicable conversion factors are: centimeters (cm): 1 cm ⫽ 0.39 in.; meters (m): 1 m ⫽ 3.3 ft; square meters (m2): 1 m2 ⫽ 10.8 ft2; cubic meters (m3): 1 m3 ⫽ 35.3 ft3; hectares (ha): 1 ha ⫽ 2.47 ac. Forest Science • February 2015 183 Early Regeneration and Structural Responses to Patch Selection and Structural Retention in Second-Growth Northern Hardwoods Methods Study Area This study was conducted within an 80 year-old, second-growth northern hardwood forest on family forestland in the Berkshire Hills of western Massachusetts (N 42.4, W ⫺72.9). Soils within this area are sandy loams derived from glacial till and are somewhat excessively drained (Scanu 1995). Terrain is gently sloping to moderately steep (3–15%) with elevations ranging from 390 to 450 m above sea level. This region has a humid, continental climate with average annual precipitation ranging from 116.2 to 129.5 cm and mean monthly temperatures from ⫺7.7° C in January to 22.2° C in July (NCDC 2006). The site index for sugar maple on the site was 18.3 m at 50 years. There was no history of harvesting in these second-growth areas prior to the onset of the study. 184 Forest Science • February 2015 Forest composition of the study area was dominated by American beech, sugar maple, and red maple (Acer rubrum L.) and preharvest basal areas ranged from 22.5–35.4 m2ha⫺1 across the study area. Other common, less abundant overstory species included white ash (Fraxinus americana L.), black cherry (Prunus serotina Ehrh.), black birch (Betula lenta L.), white pine (Pinus strobus L.), and big-tooth (Populus grandidentata Michx.) and quaking (Populus tremuloides Michx.) aspen. Preharvest sapling layers were uniformly dense across the study area (660 –2,700 stems ha⫺1) with American beech constituting the primary sapling species (291–2,260 stems ha⫺1). Other species present in the sapling layer included striped (Acer pensylvanicum L.), sugar, and red maple. There were no significant differences in preharvest sapling densities between treatment areas (F ⫽ 2.03, P ⫽ 0.1798). Experimental Design In winter 2007–2008 a series of patch selection treatments was replicated four times in a randomized, complete block design with blocking based on spatial location. Each block was 4 ha and contained the following treatments: patch selection with no retention (PNR), patch se (...truncated)