Page  110

EFFECT OF SELECTION LOGGING ON CAREX ASSINIBOINENSIS W. BOOTT IN A NORTHERN HARDWOOD STAND

Margaret M. Fox and Blair Orr School of Forest Resources and Environmental Sciences Michigan Technological University Houghton, MI 49931

Susan J. Trull Ottawa National Forest Ironwood, MI 49938 strull@fs.fed.us

ABSTRACT

A population of Carex assiniboinensis W. Boott (Assiniboia sedge) was monitored during the season before and for six years following a winter selection harvest in a sugar maple-eastern hem- lock-black cherry stand on the Ottawa National Forest in the Upper Peninsula of Michigan. A mean increase in sedge presence of 136% was observed over the seven years of monitoring. General linear regression of sedge presence data indicated a significantly positive trend over the natural log-trans- formation of time. Results indicate that canopy opening from winter selection logging in northern hardwood forests may benefit populations of this threatened species. KEYWORDS: Carex assiniboinensis, Assiniboia sedge, selection logging, threatened species, Michigan

INTRODUCTION

A population of Carex assiniboinensis W. Boott (Assiniboia sedge) was mon- itored during the season before and for six years following a winter selection harvest in a closed-canopy sugar maple-eastern hemlock-black cherry stand on the Ottawa National Forest in the Upper Peninsula of Michigan. The site is rela- tively flat, dissected by a ravine associated with an intermittent stream. Soil is a sandy loam with cobbles on the surface; litter accumulation is thin, typically less than 2–3 cm. The understory is sparse, about 5% cover, mainly young sugar maple with occasional Lonicera canadensis (American fly honeysuckle). Ground cover is also sparse, with Dryopteris intermedia (intermediate wood- fern) as the most common associate. The sedge forms solid patches in many parts of the site. Considering Carex assiniboinensis’ status as a state threatened species, typi- cally a 250-foot buffer would have been established around the area covered by the species. Within the population area plus the buffer zone, logging and other ground-disturbing activity would have been prohibited (e.g. USDA Forest Ser- vice 2004). The population, however, was not discovered until after the timber sale contract was awarded. Since harvest was already contracted, the Ottawa

Page  111

FIGURES 1A AND 1B. C. assiniboinensis at the study site.

National Forest Botany Program designed a monitoring program to investigate potential effects of overstory selection logging on C. assiniboinensis. In addi- tion, to potentially lessen impacts to the sedge population, harvest activity was delayed until snow cover was at least six inches.

Page  112

FIGURE 2. Location of the study site.

C. assiniboinensis is a perennial sedge in the Cyperaceae family (Figures 1a and 1b). It is easily identified by its long above-ground vegetative shoots (stolons) that can extend up to 2 meters in length, at the end of which new plants form (Bernard 1959; Tolstead 1946; Hipp 2008; USDA NRCS 2012; Voss and Reznicek 2012). The stolons are observable in late-summer, while flowering usually occurs in June and July (Penskar and Higman 1999). C. assiniboinensis is found on moist sites, in mesic deciduous and mixed forests, floodplains, and river banks. Its native range extends from northern Iowa in the south, southeastern Saskatchewan in the west, and east to the Lake of the Woods region of Ontario near the southern border with Manitoba, and the Upper Peninsula of Michigan (Penskar and Higman 1999; Flora of North America Ed- itorial Committee 2002; USDA NRCS 2012). Michigan is the only state in which the plant is listed as threatened; its conservation status is imperiled (S2) in Michigan, Ontario, and Saskatchewan, and vulnerable (S3) in Manitoba and Iowa (NatureServe 2012). According to the Michigan Department of Natural Re- sources, it has a coefficient of conservatism of nine (on a scale of ten); that is, it is believed to have high fidelity to a natural community that persists in a state similar to pre-European settlement (Herman et al. 2001). Effects of overstory management on C. assiniboinensis populations are poorly understood. Tolstead (1946) and Bernard (1959) hypothesized that stolon growth is encouraged by light, but no research since that time has confirmed whether an increase in light would contribute to population growth. Bernard (1959) also speculated that soil moisture and other characteristics or weather pat- terns could affect the degree of stolon formation. He believed that the prevalence of stolons in C. assiniboinensis populations in North Dakota was due to long- standing woodland management and perhaps grazing. The Michigan Natural Features Inventory recommends that until the effects of timber harvest on C. assiniboinensis populations can be better understood, logging should be avoided

Page  113

altogether or at least limited to selection cutting during winter months (Penskar and Higman 1999). In this study, it was hypothesized that sedge growth could be promoted as the result of increased light penetration following the creation of canopy gaps. This hypothesis was formed based on presence of the sedge on old skid trails in the Ottawa National Forest.

METHODS

The surveyed population of C. assiniboinensis covers a roughly U-shaped area of approximately five to ten acres located in Beechwood Quadrangle in Iron County, MI (T44N R36W S31) (Figure 2). The first monitoring was conducted in September of 1999 and was carried out every September through 2005. Harvesting was conducted in the winter of 1999–2000 on the western portion of the U-shaped area and in 2000–2001 on the eastern portion, with a target residual basal area of 85 square feet per acre in both portions. Permanent, parallel southeast-northwest running transects were established at randomly spaced distances along the east and west sides of the U-shaped area. The sampling regime used permanent transects in order to (1) be able to monitor changes in the population over time, and (2) allow for the use of a fewer number of sampling points than would be required to meet the same level of accuracy in non-permanent transects (Elzinga et al. 1998). Ten transects were established on the west side and six on the east side. From the fixed starting point, sampling along each transect started at a random distance between one and nine meters on the west and one and five meters on the east side. The start- ing point and subsequent sampling points, which were spaced one meter apart, therefore changed every year. Thirty points were sampled on each transect on the west side and 25 on each transect on the east side. At each point, presence or absence of sedge ramets was recorded. Data are summarized in Table 1. A sum value for each transect was recorded and then the values from west-side transects were normalized to 25 points. Data were analyzed in SAS software (SAS Institute Inc., Cary NC) using PROC

GLM for general linear regressions. Gen- eral linear regressions were used to iden- tify a trend in sedge presence following harvest. The data were analyzed as one data set, with year one as the season before harvest for the west (1999–2000) and the east (2000–2001) sides. A single analysis was conducted on monitoring data from the six years following harvest for the west transects and from the five years following harvest for the east transects. Year seven only included data from the west transect. Results were considered significant at p <0.05.

TABLE 1. Mean and standard error values for sedge presence (as ramets) on all transects. Year 1 is the year before harvest. n=16 for years 1 through 6. n=10 for year 7.

RESULTS & DISCUSSION

A mean increase in sedge presence of 136% was observed over the seven years of monitoring (Figure 3). A linear regression equation was used to model the effect of time on sedge presence using the natural log of year of survey as the independent variable and sedge presence as the dependent variable for each in- dividual transect and for the mean value of all transects. In order to analyze the response of the sedge following logging, only the data from the second (2000)

Page  114

FIGURE 3. Mean number of sedge ramets per transect over the seven years of monitoring. n=16 for years 1 through 6. n=10 for year 7.

through seventh year (2005), for the west side, and the third (2001) through sev- enth (2005) year for the east side, were used. The regression analysis showed a significant positive trend in mean sedge presence over time (R2 = 0.89; p = 0.005) (Figure 4). Results indicate that winter selection logging in a northern hardwood forest where C. assiniboinensis is present had the effect of increasing sedge cover, at least over the short term (as the canopy continues to fill in and light decreases, the sedge population may decrease). Although we did not record size or maturity of plants observed at the sample points, we assume that the observed increase in total cover is due to vegetative reproduction, particularly via the characteristic stolons, rather than from seed. Repeated recruitment from seed is less common

FIGURE 4. Linear regression of sedge ramets as a function of time. x = year of study beginning with 2 = first season after harvest.

Page  115

in woodland clonal plants where vegetative reproduction typically dominates (Eriksson 1989). An increase in dominance of C. assiniboinensis in relation to other species was also observed in the field. This result is consistent with research (Schultz 1988) on the Hiawatha Na- tional Forest, also in Michigan’s Upper Peninsula, which found that a selection harvest in mesic hardwoods caused a flush of reproductive effort, with an in- creased number of fertile culms, in Carex novae-angliae, another state-threat- ened sedge with a coefficient of conservatism of 9. However, this treatment did not result in an increase in total sedge cover. Increased penetration of light to the forest floor is likely a factor in support- ing sedge growth. Studies of the effects of canopy gaps on understory vegetation cite the increase in light penetration as a causal factor in increased ground cover following selection cutting (McComb and Noble 1982; Reader and Bricker 1992). Tolstead (1946) and Bernard (1959) both noted increased stolon forma- tion in Carex assiniboinensis populations with increased light. Limiting harvest activities to winter months also could have contributed to the positive response of the sedge. Harvesting over exposed ground causes soil disturbance and compaction, which can change microclimates and site suitabil- ity for understory plants (Brais 2001). Restricting logging to winter months over snowpack or frozen ground results in less impact on understory abundance and percent cover than when logging is conducted during summer months. Species that are more vulnerable to disturbance (coefficient of conservatism >6) are also less likely to be found on summer-logged sites (Wolf et al. 2008). The high co- efficient of conservatism (9) attributed to C. assiniboinensis (Herman et al. 2001) suggests that if logging was not restricted to the winter, different results might have been observed in this study. The results indicate that winter selection logging in hardwood forests can have a positive impact on populations of C. assiniboinensis, at least in the short term. Based on this study, it may not be necessary to exclude selection cutting from an area where C. assiniboinensis is present, despite its threatened status. Without more research, it is advisable to limit harvest activities to winter months over several inches of snow.

ACKNOWLEDGEMENTS

We thank I. Shackleford, C. Matula, R. Evans, B. Bogaczyk, S. Zoars, and L. Miskovich for field assistance and an anonymous reviewer for helpful suggestions. This research was supported by the Loret Miller Ruppe Scholarship Fund.

LITERATURE CITED

Bernard, J.P. (1959). Le Carex assiniboinensis Boott et sa forme stolonifère. Le Naturaliste Cana- dien 86(1):11–19. Brais, S. (2001). Persistence of soil compaction and effects on seedling growth in Northwestern Que- bec. Soil Sci. Soc. Am. J. 65:1263–1271. Elzinga, C. L., Salzer, D. W., and Willoughby, J. W. (1998). Measuring and monitoring plant popu- lations. USDI-BLM Technical Reference 1730-1. Denver, CO: USDI–BLM, pp. 136–141. Eriksson, O. (1989). Seedling dynamics and life histories in clonal plants. Oikos 55(2):231–238. Flora of North America Editorial Committee. (2002). Flora of North America, North of Mexico. Vol-

Page  116

ume 23: Magnoliophyta: Commelinidae (in part): Cyperaceae. New York: Oxford University Press, p. 468. http://www.efloras.org/florataxon.aspx?flora_id=1&taxon_id=242357054 Hipp, A. (2008). Field guide to Wisconsin sedges: an introduction to the genus Carex (Cyperaceae). Madison, WI: University of Wisconsin Press, p. 65. Herman, K.D., Masters, L.A., Penskar, M.R., Reznicek, A.A., Wilhelm, G.S., Brodovich, W.W. and Gardiner, K.P. (2001). Floristic quality assessment with wetland categories and examples of com- puter applications for the state of Michigan. Michigan: Department of Natural Resources, p. C–10. McComb, W. C. and Noble, R. E.. (1982). Response of understory vegetation to improvement cut- ting and physiographic site in two mid-south forest stands. Castanea 47:60–77. NatureServe. (2012). NatureServe Explorer: An online encyclopedia of life (web application). Ver- sion 7.1. Arlington, VA: NatureServe. http://www.natureserve.org/explorer (accessed: July 23, 2012 ). Penskar, M. R. and Higman, P. J. (1999). Special plant abstract for Carex assiniboinensis (Assiniboia sedge). Lansing, MI: Michigan Natural Features Inventory, pp. 1–2. http://mnfi.anr.msu.edu/ abstracts/botany/Carex_assiniboinensis.pdf Reader, R.J. and Bricker, B.D. (1992). Value of selection cut deciduous forest for understory herb conservation: an experimental assessment. Forest Ecol. Manag. 51:317–327. Schultz, J. (1988). Carex novae-angliae quadrat remeasurements within hardwoods in the Shingle- ton Fen study area, Schoolcraft County, Michigan. Unpublished document. Marquette, Michigan: Hiawatha National Forest, 7 pp. Tolstead, W. L. (1946). Stolons of Carex assiniboinensis Boott in Iowa. Am. Midl. Nat. 35:797. USDA Forest Service. (2004). Camp 7 vegetation management project environmental assessment. Ironwood, MI: Ottawa National Forest, pp. 2–9. USDA NRCS PLANTS Database, http://plants.usda.gov/java/profile?symbol=CAAS2 (accessed May 14, 2012). Voss, E.G. and Reznicek, A.A. (2012). Field manual of Michigan flora. Ann Arbor, MI: University of Michigan Press, p. 104. Wolf, A. T., Parker L., Fewless, G., Corio. K., Sundance, J., Howe, R., and Gentry, H. (2008). Impact of summer versus winter logging on understory vegetation in the Chequamegon-Nicolet National Forest. Forest Ecol. Manag. 254:35–45.