White-tailed deer foraging in relation to successional stage, overstory type and management of Southern Appalachian forests.
Until recently there has been a common assumption that basic relationships of white-tailed deer (Odocoileus virginianus) to their food resources in the Southern Appalachians are sufficiently well understood for sound management; however this assumption is not supported by comprehensive studies. Early work and experiences indicated that deer populations were overusing the food resource and causing damage to forest regeneration (Ruff, 1938; Bryan, 1950). These observations led to an emphasis on producing winter browse by timber harvesting. In the early 1960s clearcutting became the dominant method of timber harvest and regeneration, resulting in greatly increased woody browse production (Moore and Downing, 1966). More recent research and management experience have made it evident that relationships between deer and the forest habitats of the region need to be reexamined. Deer herds in the region have shown little response to increased availability of woody browse resulting from timber harvest or population reduction. Cushwa et al. (1970) and Harlow and Hooper (1971) pointed out the prevalent misconception that deer subsist mainly on woody twigs. Based on a regional food habits study and other observations, they stressed the importance of other components of the deer diet. More recently, we found that size of the acorn crop influenced deer weight, antler development and reproduction in the Southern Appalachians (Wentworth et al., 1990a, 1992). We also showed that recently harvested stands (clearcuts) were used by deer in winter less than previously assumed; but clearcuts were used much more intensively in spring and summer (Wentworth et al., 1990b), and forage in clearcuts was browsed more intensively than in uncut forest in those seasons (Ford et al., 1993).
Public attitudes and agency management policies are changing. Clearcutting is being replaced by other methods of timber harvest and regeneration, and an increasing area of the national forests is being excluded from timber harvest by any method. Management focusing on individual wildlife species, especially game animals, has been deemphasized and the focus has shifted to "ecosystem management" (Robertson, 1992). It is not clear how these changes will affect deer populations or how the loss of a concentrated source of food during the growing season (clearcuts), and consequently more dispersed deer browsing, will affect forest regeneration and rare and sensitive plant species within older stands. Much more needs to be known about specific forages eaten by deer (including species of management significance that may make up only a minor part of the diet), their relative abundance in different habitats, and their responses to various methods of timber harvest and to deer browsing.
Most previous studies of deer-habitat relationships in the Southern Appalachians were general surveys of woody browse (e.g., Harlow et al., 1975) or studies of effects of browsing on tree regeneration after timber harvest (e.g., Harlow and Downing, 1970). Most of these studies were conducted at a time when there was only general knowledge of seasonal diets of deer in the region, and the studies did not reflect deer use of forages other than twig ends.
The only previous study of deer diets in the Southern Appalachians based on rumen samples from all seasons was that of Harlow and Hooper (1971). That study included relatively small samples, especially in spring and summer, and these were distributed across seven locations, only two of which were in the region addressed by this study. Two earlier studies were based on small samples from North Carolina and Alabama and did not include all seasons (Ruff, 1938; Adams, 1959).
As part of a more comprehensive study of deer-habitat relationships in the Southern Appalachians, we collected specimens and data from deer during all seasons and sampled forage availability, quality and deer use at different seasons and under different habitat conditions. Previously published results of the study have dealt with the effects of acorn availability on deer condition and reproduction (Wentworth et al., 1990a, 1992), seasonal use by deer of clearcuts (Wentworth et al., 1990b; Ford et. al., 1993) and agricultural plantings (Wentworth et al., 1990b), and chemical composition of browse in clearcuts and adjacent uncut stands (Ford et al., 1994). Here, we compare seasonal deer diets to forage availability and deer use in different habitats and consider implications to forest management.
Funding was provided by the Tennessee Wildlife Resources Agency, PR Proj. No. FW-6, North Carolina Wildlife Resources Commission, PR Proj. No. W-57 Study B-6, and McIntire-Stennis Proj. No. GEO-MS-0054. We are indebted to many people with the North Carolina Wildlife Resources Commission, the Georgia Wildlife Resources Division, the Tennessee [TABULAR DATA OMITTED] Wildlife Resources Agency and the Southeastern Cooperative Wildlife Disease Study for assistance in collecting deer and specimens. U.S. Forest Service personnel assisted in locating stands for vegetation sampling. Much of the field and laboratory work was performed by students and technicians of the University of Georgia.
Study area. - The study was conducted at selected sites throughout the Pisgah and Nantahala national forests in North Carolina, the Cherokee National Forest in Tennessee and the Chattahoochee National Forest in Georgia. The study area lies within the Blue Ridge physiographic region and is characterized by rugged topography with elevations ranging from less than 300 m to over 2000 m. Soils are acidic and relatively infertile, except in coves and alluvial valleys. Average annual rainfall varies within the region from 130 to 140 cm. Temperatures vary widely, depending upon location, elevation and aspect. Average maximum temperatures in July range from 27 to 30 C, and average minimum temperatures in January are from -1 to -4 C. Snowfall is frequent at higher elevations, but snow cover is not continuous during winter.
Most of the forest in the region is classified as oak-hickory (Quercus spp.-Carya spp.), but forest type varies with soil moisture, the most important factor determining site quality (McNab, 1988). We combined forest types into two broad categories as described by McNab (1988): (1) dry slope and ridge forests and (2) moist slope and cove forests (hereafter referred to as dry ridge and cove types, respectively). The proportional distribution of these types on the national forests during our study is shown in Table 1.
Dry slopes and ridges are dominated by oaks and/or pines. Common oaks on these sites include chestnut oak (Q. prinus), scarlet oak (Q. coccinea) and black oak (Q. velutina). Dominant species of pines are white pine (P. strobus), shortleaf pine (P. echinata) and Virginia pine (P. virginianus). Shrubs, especially blueberries (Vaccinium spp.), huckleberries (Gaylussacia spp.) and mountain-laurel (Kalmia latifolia), dominate the understory, and herbaceous cover is sparse.
Moist slopes and coves may be dominated by yellow-poplar (Liriodendron tulipifera) or support a mixture of many mesophytic species, including northern red oak (Q. rubra), white oak (Q. alba), red maple (Acer rubrum), blackgum (Nyssa sylvatica), flowering dogwood (Cornus florida), black cherry (Prunus serotina), silverbell (Halesia carolina), eastern hemlock (Tsuga canadensis) and magnolias (Magnolia spp.). Shrub cover, if present, consists mainly of rosebay rhododendron (Rhododendron maximum). There may be a dense groundcover of ferns, spring ephemerals and other shade-tolerant herbaceous vegetation.
During this study and for the previous 20 yr, timber was harvested mainly by clearcutting, and most stands cut were 8-16 ha. Hardwood stands were allowed to regenerate naturally by sprouting of stumps. Some stands, especially on poor sites, were planted to pine. Pine-dominated stands contribute relatively little deer forage and were excluded from our vegetation sampling.
We grouped stands into three stages of successional development keyed to approximate ages of crown closure and attainment of full mast-producing capability: clearcuts (young stands [less than or equal to] 7 yr after clearcutting), stands of intermediate age (8-39 yr) and older stands ([greater than or equal to]40 yr). We did not sample stands of intermediate age. During our study, most of the area on the four national forests was in the older age class.
Deer density generally ranged from 5 to 10 deer/[km.sup.2]. Estimates of density were provided by the state wildlife agencies and were derived from population models based on harvest data.
Seasonal diet. - Rumen contents (1-1 samples) were collected from hunter-killed deer during the legal hunting seasons and from deer collected under permit for this study at various other times of the year from November 1983 through November 1988. The following areas are represented in the samples: the Tellico Unit of the Cherokee National Forest in Tennessee; the Pisgah Game Lands in Haywood, Henderson, Transylvania and Yancey counties in North Carolina; the Nantahala Game Land in Macon County, North Carolina; and the Blue Ridge, Chattahoochee, Chestatee, Coleman River, Cooper's Creek, Lake Burton, Swallow Creek and Warwoman wildlife management areas in Georgia.
Rumen contents were washed through a series of sieves with mesh sizes of 5.66, 3.35 and 1.41 mm. Food items retained in the 5.66-mm sieve were separated, identified macroscopically and measured volumetrically by water displacement to the nearest 1 ml. Contents of the lower sieves were examined for foods not detected in the upper sieve. Aggregate percent volume (Martin et al., 1946) was calculated for each food item. Scientific names of plants are according to Radford et al. (1968). Common names follow local usage.
Results of rumen analysis were compared to availability and use of various forages measured in different habitats to determine consistency of browse survey data with data from rumen analysis and to evaluate contributions to the deer forage base provided by different forest types and successional stages.
Successional trends in forage availability in clearcuts. - In the first phase of the study, the successional trend in forage availability in clearcuts was examined. Twelve recently harvested hardwood stands of the dry ridge type that had received no chemical or mechanical site treatment other than routine felling of nonmerchantable stems were selected on the Chattahoochee National Forest in Georgia. Four clearcuts in each of three age classes (1, 3 and 7-8 years) were represented. In each clearcut, transects were established from the edge of the clearcut extending perpendicular to the contour of the slope. Sixteen points (plot centers) were established at [greater than or equal to]30-m intervals along the transect lines in each clearcut, for a total of 64 plot centers per age class. At each point, a plot 2 m by 2 m and a plot 1 m by 1 m were established.
In summer, percent groundcover by vegetation [less than]0.5-m tall was visually estimated in the 1-[m.sup.2] plots. The 4-[m.sup.2] plots were visited monthly from June through August 1982, and all fruits (e.g., berries, drupes) were picked as they neared maturity and weighed fresh. No attempt was made to estimate fruit consumed by wildlife, so biomass estimates are minimum values.
In January 1983 the 1-[m.sup.2] plots were revisited and all green herbaceous material, leaves of broadleaf evergreen woody plants, and fruiting bodies of fungi were harvested, oven-dried and weighed.
Availability and deer use of forages in clearcut and uncut forest. - Availability and deer use of various forages were measured in young hardwood clearcuts and adjacent older hardwood forest at various locations throughout the four national forests from 1985 to 1990. Clearcuts were selected based on stand descriptions and timber harvest records made available to us by the U.S. Forest Service and subsequent field inspection. Based on successional trends in forage production in clearcuts determined by previous sampling, we restricted our sampling of clearcuts to those [less than or equal to]5-years-old. Other criteria for selection included having average or better site quality supporting a hardwood type (site index [greater than or equal to]70), natural regeneration with no mechanical or chemical site preparation, and a surrounding older forest uncut for [greater than or equal to]40 years. Twenty-one clearcuts, ranging in size from 5 to 16 ha were selected for sampling and classified as dry ridge or cove. The first five of the locations were sampled in January and February 1986 and 1987 and deer use of clearcuts was low (Wentworth et al. 1990b). Therefore we confined subsequent sampling to the period just after greatest use (early summer).
In both seasons, a point on the perimeter of each clearcut was located randomly as a beginning point for forage sampling. From this point 30 sampling stations were spaced at 50-m intervals around the perimeter. At each of these stations a plot was established at the edge of the clearcut, a second plot was established 25 m into the clearcut area and a third was established 50 m into the adjacent uncut older forest. All plots were 0.5 m by 2.0 m.
All woody plants in the plots were identified and twig tips within 1.5 m of the ground were tallied by species and recorded as browsed or unbrowsed. Herbaceous plants were identified to species where possible, and cover and percent use of these and of fungi were visually estimated. In all clearcuts sampled in winter and three sampled in summer, vegetation in plots was harvested, sorted by type, oven-dried and weighed. Some of the data have been used previously to compare availability and deer browsing of woody plants in the clearcut, clearcut edge and uncut forest positions (Ford et al., 1993). Data from the clearcut edge plots are not reported here. Woody browse availability in the edge plots was intermediate between that of the clearcuts and of the uncut forest (Ford et al., 1993).
Availability and deer use of acorns. - In fall 1983, acorn availability and use by deer were studied in the Tellico area in eastern Tennessee. Four sites were selected for intensive study: two at elevations below 700 m and two above 1100 m. Stands selected were predominantly oak (57-76% of total basal area, which ranged from 29 to 37 [m.sup.2]/ha). Stands ranged from 55 to 75 years.
Sampling was conducted in 200 1-[m.sup.2] plots spaced 10 m apart along transect lines perpendicular to the contours and spaced 100 m apart. The number and length of the transects varied with the shape of the stand. Within each plot, fallen acorns were identified to species, counted and weighed. Sampling was repeated at 2-week intervals from early August through late January, except sampling was discontinued during mid-November and again in December because of deer hunts or snow cover or both. Each time an area was sampled, transect lines were offset 5 m to avoid plots disturbed during previous sampling.
Deer feces were collected on or near the sampling sites. Ten pellets from each group of freshly deposited ([less than]2-wk-old) fecal droppings were soaked in water, macerated and washed through a 0.425-mm mesh sieve. Seed fragments in the feces were identified and ocularly estimated by percent volume class ([less than]1, 1-10, 11-20, 21-40, 41-60, 61-80, 81-100). The mid-points of the classes were used to calculate means for each sampling period.
Seasonal diet. - Rumen contents were obtained from 581 deer, seasonally distributed as shown in Table 2. Spring samples were taken from 25 April to 22 June, summer from 12 July to 24 September, fall from 1 November to 21 December (no deer were collected in October) and winter from 3 January to 27 March. Samples were well distributed among areas and years except that no spring and summer samples were obtained from the Georgia areas.
On a year-round basis, the categories of foods that contributed most to the diet were green leaves of woody plants, and fruits, especially acorns. Herbaceous plants were of secondary importance at times, but rumen contents revealed less use than we expected. Fungi and lichens, flowers (mainly yellow-poplar) and dried leaves also were secondarily important.
The spring diet consisted mainly of leaves (41%) and stems (17%) of deciduous woody plants. Stems consumed were tender, rapidly growing twig tips. Of the 58% of the volume consisting of leaves and stems, seven species or genera made up 50%, and all other species combined added only 8%. Flowers of yellow-poplar dominated the diet in late May and early June. They occurred in the rumens of 18 of 21 deer collected and made up 48% of the volume during this time. These samples represented collections from three years and two states. Yellow-poplar flowers accounted for 25% of the aggregate volume of the diet during the entire spring season. The only other item important in the spring diet was the forb squawroot (Conopholis americana), which contributed 4% of the volume.
In summer, green leaves of woody plants continued to be the category of food contributing the greatest volume to the diet (48%); stems declined to 7%. The species composition of the green leaves was somewhat more diverse than in the spring, but the seven taxa that dominated this category in spring also were important in summer. Other species that became important included sassafras (Sassafras albidum) and grape (Vitis spp.).
Important components of the summer diet other than leaves and stems were fruits (16%), forbs (15%) and fungi (12%). The fruit component consisted mostly of acorns, which became available beginning in late August. Acorns made up 4% of the August diet and 47% of the September diet. Blueberry fruits were consumed in all summer months and comprised 3% overall. As in spring, squawroot was the most important forb in the summer diet (4%), and pokeweed (Phytolacca americana) contributed 4% to the diet in July.
Fall was the season when diet was most variable because availability of various forages changed rapidly as the season progressed and because the size of fruit crops varied among years and among localities. Green leaves of deciduous woody plants continued to be used heavily through November; in that month they still made up 19% of the rumen volume. Dried leaves of several species also were eaten in significant amounts (7%). Acorns were the most important item throughout the fall. Consumption of acorns varied greatly by month and by year, depending upon the size of the acorn crop. Other fruits, mainly grapes, crabapples (Malus spp.), blackgum and black cherry were eaten as available. They each averaged 2-4% of the fall diet, but in some months and localities made up a large portion of the diet, providing a buffer food in years when acorn crops were small. Fungi and lichens also were important (13%), especially in November at which time they contributed 17% to the diet. As the season progressed and other foods became more scarce, leaves of broadleaf evergreen plants became increasingly important. They comprised 18% of the volume from rumens in November and 62% in December. Rosebay rhododendron was by far the most important component of this forage category.
The winter diet was mainly acorns and leaves of broadleaf evergreen woody [TABULAR DATA OMITTED] plants, the proportions depending upon the size of the acorn crop. As in fall, rosebay rhododendron was the most important of the broadleaf evergreen species. For all years combined, it increased from 28% in January to 37% in February and to 44% in March. It was followed by mountain-laurel (7% for the winter season), galax (Galax aphylla, 5%) and dog-hobble (Leucothoe axillaris, 3%). Fungi and lichens contributed 8% of the diet for the entire winter period. Red maple flowers and bits of associated stems occurred in 17% of the March samples and made up 3% of the rumen volume in that month. Grasses, probably obtained from agricultural plots and roadsides, made up 6% of the winter diet, and forbs contributed only 1% during this period.
Successional trends in forage availability in clearcuts. - In the three age classes of clearcuts sampled to determine successional trends (1, 3 and 7-8 years), availability of most deer forages declined with stand age. Percent cover of groundstory vegetation in summer declined from 50% and 44% at ages 1 and 3 years, respectively, to 24% at 7-8 years. Fruit [TABULAR DATA OMITTED] production (fresh weight) was 442 kg/ha (SE = 195) at 1 year (mainly pokeweed and blueberries), 203 kg/ha (SE = 54) at 3 years (blueberries, blackberries [Rubus spp.], and grapes) and 48 kg/ha (SE = 14) at 7-8 years. Most of the fruit in the oldest stands was flowering dogwood. Although variability was high, mean winter biomass of forbs, grasses, ferns and fungi consistently decreased with increasing stand age. In contrast, biomass of leaves of broadleaf evergreen woody plants increased after the first growing season. Most of this was mountain laurel, especially in the oldest clearcuts.
Availability and deer use of forages in clearcut and uncut forest. - In the paired plots in clearcuts [less than or equal to]5-years-old and adjacent uncut forest, 12 genera of woody plants composed 70% of available twig tips in each age class. Blueberries were most abundant in both situations and made up more than half of available twigs in plots in the uncut forest. Blueberry twigs also had the highest percent use by deer in both clearcut and uncut plots. Eight other genera contributed [greater than or equal to]3% to the total available twigs in clearcuts, compared to only two genera in uncut plots. Other genera browsed intensively ([greater than or equal to]10%) by deer included Smilax in uncut forest and Vitis, Smilax and Nyssa in clearcuts.
Biomass of forages in all categories was far greater in clearcuts than in adjacent uncut [TABULAR DATA OMITTED] [TABULAR DATA OMITTED] forest in both summer and winter. But, when compared by percent cover, differences in forage availability between clearcuts and uncut forest in winter did not appear as large.
Clearcut cove forests generally had greater cover of forbs, ferns and fungi and less cover of grasses and woody plants than clearcut dry ridge forest. The differences were not significant because an insufficient number of stands was sampled to permit comparison by age (1-5 years), and succession occurs at a more rapid rate on the more moist and fertile cove sites.
[TABULAR DATA OMITTED]
[TABULAR DATA OMITTED]
Availability and deer use of acorns. - In mature oak stands on the Tellico area, acorn availability in the sample plots peaked during the second week of October at 119 kg/ha, fresh weight. The volume of acorns in deer fecal pellets collected nearby peaked 2 weeks later. Predrop mast surveys by the Tennessee Wildlife Resources Agency indicated this was a year of average acorn production. Acorns in the sample plots declined to 17 kg/ ha by late January, and to [less than]3% of the volume in fecal pellets collected nearby at that time. However, acorns made up 49% of the volume in rumens of 50 deer killed by hunters in early January throughout the 32,400-ha Tellico area. Fragments of black cherry pits occurred in inverse proportion to the volume of acorn fragments in the fecal pellets. Grape seeds accounted for 26% of the volume of fecal pellets in late November, but occurred in only trace amounts, or not at all, in other collection periods.
Important green forages. - Rumen analysis showed that green leaves of woody plants were the most important component of the deer diet during most of the year. In winter, these consisted mainly of about four broadleaf evergreen species. Stems were insignificant winter foods. They were eaten in significant amounts only in spring and early summer, when they were growing rapidly, were very tender, and were taken with the young leaves.
[TABULAR DATA OMITTED]
TABLE 9. - Biweekly availability of fallen acorns and mean percent volume of seed fragments in deer feces, Tellico area, Cherokee National Forest, Tennessee, 1983-1984
% in feces Sampling Acorn availability period(a) (kg/ha, fresh)(b) n Acorn Black cherry Grape
Early 0.0 7 0.0 1.4 0.0 Late 8.7 3 5.0 trace 0.0
Early 29.0 17 26.7 3.4 0.0 Late 43.5 21 35.0 19.8 0.0
Early 118.9 9 29.4 20.0 0.0 Late 89.9 25 60.0 6.6 trace
Early 63.8 18 49.8 9.6 0.2 Late(c) 37.7 27 10.2 20.5 26.1
Late 17.4 39 2.7 24.5 2.4
a No sampling was done in December and early January
b Based on 100-200 plots in each of 4 mature oak stands
c For 8 rumen samples analyzed from this period, the percent volume of acorns, cherries and grapes was 42.9, trace and 21.9, respectively
d For 50 rumen samples analyzed from early January, the percent volume of acorns, cherries and grapes was 48.7, 0.1 and 4.6, respectively
During the growing season, 7 to 10 species of deciduous woody plants contributed most of the volume of leaves and stems in the rumens. Counts of browsed and unbrowsed twig tips in plots provided an index to the availability of leafy browse and an independent measure of relative use by deer of individual plant species in early summer. We calculated the number of browsed twigs of each species as a percentage of all twigs browsed. This value was compared to the rumen volume of leaves and stems of the same species expressed as a percentage of the total volume of leaves and stems in the rumens. Species that were the most important contributors to the volume in the rumens were also the species that were most intensively browsed in the plots. However, there were some important differences in percentages. For example, blueberries were the most used among woody plants in plots in both forest (43%) and clearcut (21%) situations; but they contributed only 7% to the rumen volume. This discrepancy probably resulted from the fact that leaf biomass often is not strongly related to number of twigs. Many species of blueberries have numerous small leaves and twigs and therefore yield a relatively small volume of forage per browsed tip. Other possible causes of discrepancies between results of browse surveys and rumen analyses are differences among species in evidence of browsing on the plant and in digestibility, which may affect recognition of items in the rumen. Also, rumen samples often could not be obtained from the vicinity of vegetation sampling. Harlow (1979) demonstrated that surveys of browsing sign on vegetation may greatly underestimate deer use of some species, especially those that are abundant. If adequate samples can be obtained, rumen analyses provide the best measure of the contribution of plant species to deer diet, but plant sampling provides better information on the effects of deer on plant communities and especially on uncommon plant species.
Forbs contributed [less than]15% to the volume of rumen contents in spring and summer, and [less than or equal to]3% in fall and winter. Heaviest browsing of herbaceous plants occurred in plots at the edge of clearcuts, and even there, only 4% of available forbs in summer and [less than]1% in winter were recorded as browsed (Wentworth et al., 1990b). We detected no levels of deer use that would be expected to seriously reduce abundance of any plant species, but deer use of herbaceous plants is difficult to quantify and may have been underestimated. Some herbaceous plants may be more succulent and digestible than other forages, resulting in underrepresentation in volumetric measurements of rumen contents. Also, estimates of percent browsing on plants in plots are conservative because deer may remove aboveground parts of plants, leaving no evidence.
Forage resources of older forests. - During our sampling, older forests ([greater than or equal to]40 yr) made up 86% of the area of the study region. Older forests contribute more to the diet of deer than previously assumed. Forages that are essentially available only in these older stands made up 16-30% of the diet, depending upon the season. These include primarily fruits and flowers produced in the canopy, which become available to deer only when they fall to the ground. Other forages that fall from the canopy (but also are available in younger stands) include abscised leaves of trees and vines as well as some lichens (e.g., Usnea sp.). Squawroot is an obligate root parasite of oaks, mainly of the red oak subgenus (Erythrobalanus) (Baird and Riopel, 1986), and presumably is more abundant in older stands with a large component of red oaks. Older stands also are important, but not exclusive, producers of other forages discussed below.
The most important contribution of older forests is mast, especially acorns. Elsewhere, we have demonstrated the influence of acorn abundance on deer weight, condition, antler development and reproduction (Wentworth et al., 1990a, 1992). Acorn production varies greatly from year to year. In a 12-year study of acorn production on an area within our study region (Beck, 1977), it ranged from 3 to 1135 kg/ha and averaged 323 kg/ha (fresh weight). Johnson et at. (1989) estimated that [greater than or equal to]225 kg/ha are required to meet the needs of deer at current densities and other consumers (primarily rodents) throughout the winter. This level of production was achieved in 5 of the 12 years in which Beck (1977) measured production. In our rumen analyses, acorns still comprised a major portion of the diet (44%) in March following a good acorn crop (1985-1986), but in years of poor production they were nearly absent from the diet by December. In a year of average mast production on the Tellico area, acorns virtually disappeared from our sample plots in mature oak stands and from deer fecal pellets collected nearby by late January. But, acorns apparently remained available elsewhere on the area and were sought out by deer, because acorns still dominated the diet of deer killed by hunters throughout the Tellico area in early January.
Stands with the greatest percentage of oaks usually occur on sites of intermediate quality (Loftis, 1988; McNab, 1988). During our study, 65% of the forest area consisted of stands typed as oak or oak-pine and [less than or equal to]40 years old (U.S. Forest Service inventory data). But periodically there are years with very low acorn availability regardless of the amount of area in oak forest. Other forest types and successional stages are important in fall and winter of these years and also provide diversity in the food resource at other seasons in all years.
When acorns are not available, deer use other kinds of fruits before turning mainly to leaves of broadleaf evergreen shrubs. The fruits other than acorns eaten most by deer in fall and winter are blackgum, black cherry and grape. Blackgum, common across a wide range of sites, produces significant amounts of fruit only in older stands. Black cherry is a fast-growing shade-intolerant species. On good sites (coves) in the Southern Appalachians, it commonly attains a dominant position in the canopy of older forests and produces a significant amount of fruit. Open-grown black cherry begins fruit production as early as 10 years (Fowells, 1965:540), however, and thus black cherry fruit production is not restricted to older stands. Similarly, grapevines must grow with the dominant trees and maintain a position in the tree canopy to produce fruit. Although older forests may produce significant quantities of grapes, young stands are far more important in providing a source of grapes for use by deer. Grapevines grow far more abundantly on cove sites than on dry ridge sites (Della-Bianca, 1978; Smith, 1984).
Yellow-poplar flowers were consistently a major part of the diet in late May and early June. They are available to deer only in forests of intermediate and older ages because yellow-poplar trees do not begin flowering until they are 15 to 20 years old (Fowells, 1965:259). By this age, the flower-producing canopy is well beyond the reach of deer, and apparently flowers only become available when dropped by foraging gray squirrels (Sciurus carolinensis) or dislodged by wind. Nearly monotypic stands of yellow poplar often develop after cove forests are clearcut and yellow poplar may remain dominant on good sites even in old stands.
Woody understory species, providing leaves and twigs available to deer in older forests, included mainly blueberries (53% of all twigs available), broadleaf evergreen shrubs (mountain-laurel, rhododendron and galax), and seedlings and sprouts of red maple, greenbrier (Smilax spp.) and oak. In forested plots these forages made up 73% of all available twigs and 82% of all browsed twigs (Ford et al., 1993). Harlow et al. (1975), in an extensive survey of browse resources of the Chattahoochee National Forest, found that blueberries, mountain-laurel, rosebay rhododendron and greenbrier made up 72% of the total weight of all forages.
Broadleaf evergreen shrubs are more tolerant of shade than most deciduous species, and they are abundant in older forests as well as clearcuts ([ILLUSTRATION OMITTED] and Ford et al., 1993). Four species contribute significantly to the diet of deer in our study region. Rosebay rhododendron, the most important of these, occurs in thickets, often associated with streams (Day and Monk, 1974; Monk et al., 1985). Although it was not abundant at the locations we sampled (0.4% of all twig tips), on moist sites it can make up a major portion of the total leaf biomass (Day and Monk, 1974; Monk et al., 1985). Day and Monk (1974) estimated 1532 kg/ha (dry weight) of rhododendron leaves on a watershed in North Carolina. Mountain-laurel and galax are most common in dry ridge forests (Day and Monk, 1974; Monk et al., 1985), but mountain-laurel is widely distributed across habitat types, and in the survey by Harlow et al. (1975) it contributed 41% to the total forage biomass on the Chattahoochee National Forest. Day and Monk (1974) estimated 269 kg/ha (dry weight) on their study area. In our forested plots mountain-laurel comprised 7% of total available twigs and galax contributed only 0.7%. The fourth species, dog-hobble, usually is found only along streams.
The only fruits produced in the understory of older stands that we found in important amounts in the rumens were blueberries. Blueberries of several species, especially the dryland blueberry (Vaccinium vacillans), are abundant in the understory of Southern Appalachian forests, particularly dry ridge forests. Blueberries are available to deer mainly in late summer. On a large area of mostly older forest within our study region Powell and Seaman (1990) estimated annual production of blueberries and huckleberries to be 2.6 [+ or -] 2.0 kg/ha (fresh weight) over a 3-year period.
Forbs and grasses were much less abundant in older forest than in clearcuts. However, about half of the taxa of identified forbs that contributed [greater than or equal to] 1% to the diet in some month were forbs associated with older forest. These included squawroot, violets (Viola spp.), foam-flower (Tiarella cordifolia), rattlesnake-root (Prenanthes sp.), Solomon's seal (Polygonatum biflorum), squirrel-corn (Dicentra canadensis) and partridge-berry (Mitchella repens). Cinquefoil (Potentilla spp.), beggarweeds (Desmodium spp.) and composites (Compositae), although more common in clearcuts, were also among the most commonly occurring taxa in our forest plots. The most important herb, squawroot, is patchily distributed and did not occur in our sample plots. However, as previously indicated, it is probably most abundant in older oak stands. On an area in western North Carolina Powell and Seaman (1990) estimated squawroot production at 2.3 [+ or -] 0.05 kg/ha (fresh weight) across several habitat types.
Contrary to our expectations, fungi were not abundant in older forests. Our sampling was done in winter and early summer, whereas greatest use by deer was in late summer (12%) and fall (12%). Nevertheless, fungi were abundant in clearcuts during the sampling periods. Forest stands much older and more senescent probably would produce more fungi than those we sampled.
Forage resources of clearcuts. - Clearcutting results in drastic changes in the food resource of deer. Although the general pattern of change is similar, cove and dry ridge sites differ somewhat in specific contributions to the deer food resource after clearcutting.
Clearcutting obviously results in an immediate and total loss of deer forages produced by the forest canopy. The most significant loss is acorn production, which varies with the amount and kind of oaks present on the site before harvest. Stands that have been clearcut will not approach full acorn production again for about 40 years (Fowells, 1965). Furthermore, oaks do not recruit well into stands following clearcutting on good sites (Loftis, 1988). However, during our study, stands [less than or equal to] 7 years old made up only 3.3% of the area of the study region, and 60% of this was stands typed as pine, whereas the pine type made up only about 35% of the study region.
On the positive side, clearcutting results in an immediate flush of sprout growth from stumps and an invasion of early successional species, both woody and herbaceous. In our plots both clearcuts and adjacent forests contained abundant twig tips, but biomass of green leaves in clearcuts was 8.5 times greater than in uncut forest in summer, when deer use of clearcuts is greatest, and 2.3 times greater in winter. More important, the browse resource in clearcuts is more diverse. Eleven genera made up 76% of the total browsed twigs in clearcut plots, whereas in older forest plots just two genera (Vaccinium, 43%, and Simlax, 33%) comprised 76% of browsed twigs. Some shade-intolerant plants that provide abundant and preferred browse in clearcuts did not occur (e.g., blackberry) or were out of reach of deer (e.g., yellow-poplar, grapevines) in older forests.
In other regions, studies have indicated that forage quality of browse in clearcuts may be greatly reduced because high levels of condensed tannins produced in open-grown plants decrease protein digestibility (e.g., Happe et al., 1990). But, in the Southern Appalachians, Ford et al. (1994) report tannin levels in the leaves of four important browse species collected at monthly intervals to be low and not significantly different in samples from clearcuts and from adjacent uncut forest. There were only minor differences in other measures of browse quality within species between clearcuts and older forest (Ford et al., 1994).
Clearcuts also produced much higher cover and biomass of forbs and grasses than older forests. Panic grasses (Panicum spp,), violets, composites and cinquefoil were the most common herbaceous forages in the clearcut plots. Except for squawroot, pokeweed had the highest volume of any forb in the rumens. Pokeweed occurs almost exclusively in clearcuts or other disturbed areas, especially on cove sites. The abundance of herbaceous plants declines as young stands develop.
As a result of the large amount of dead wood (logging slash) on the ground, clearcuts contain an abundance of fungi available to deer. Dry weight of fungi in clearcuts was 23 kg/ha in summer and 32 kg/ha in winter, whereas, in uncut forests it was [less than]1 kg/ha at both seasons. We also estimated cover of fungi in all clearcuts and adjacent forest sampled in summer and in winter. Percent cover was much higher in clearcuts than in uncut forest. Although deer use of fungi was highest in summer and fall, biomass was higher in winter than in summer and greatly exceeded that of forbs and ferns at that time. Much of the fungi found in the clearcuts was hardened bracket or shelf fungi and may be of less value to deer than other forms, but these kinds of fungi did occur commonly in deer rumens.
Clearcuts often produce large amounts of fruits. Although rumen analysis did not confirm high deer use of fruits of early successional species such as pokeweed and blackberry, these fruits provide a potentially significant food resource. Total fruit production in clearcuts peaked one year after harvest, but remained high at least three years postharvest. In contrast to older forest, most of the taxa of fruit found in clearcuts ripen in summer rather than fall. Thus, clearcuts not only provide substantial amounts of fruit but also increase the portion of the year during which fruits are available to deer.
It is evident from the foregoing that clearcuts produce an abundant and diverse forage resource for deer during the growing season - a crucial period of lactation and fawn growth and development. This forage declines rapidly, and after 7 years most of the benefit of clearcutting for deer is gone. Stands then enter a period in which they provide a poor forage base for many years. After 40 years stands become increasingly productive of acorns and some other fruits and probably fungi. These older stands are especially important in providing high energy foods during fall and winter.
Implications for forest management. - Deer populations in the Southern Appalachians are regulated largely by an unstable supply of energy in the fall and winter diet (Wentworth et al., 1990a, 1992). A high percentage of the area in the national forests is occupied by oaks of mast-producing age, and in years of high acorn production (two of six during this study), acorns are available to deer throughout the winter. But in about two-thirds of the years, the acorn supply is exhausted at some time during the fall or winter. In years of very poor production (e.g., 1987) it may be depleted by November. To meet the needs of present populations of deer and other wildlife in these years of low acorn production, the oak component of the national forests should be maintained or increased. But management for oaks is difficult on some sites, and even if the oak component was increased to the maximum feasible, acorn production would be inadequate in many years. Therefore, it is desirable to maintain diversity within and among stands.
Consideration of the needs of deer for a fall and winter source of dietary energy must be balanced against dietary needs during the growing season. Rumen analyses and sampling of availability and deer use of forages indicate that nutritional needs during the growing season are best met by browse and other forages produced in clearcuts. Although clearcuts make up only a small percentage of the national forests, the relatively low percent of twigs browsed in all habitats indicates that adequate browse is available for use by deer at current densities without seriously affecting stand regeneration.
Clearcutting is being replaced by alternative timber harvest methods, and all timber harvesting recently has been excluded from large areas of public land in the Southern Appalachians. Although timber harvest methods other than clearcutting may provide abundant browse of deciduous woody plants, availability of grasses and early successional forbs requiring full sunlight may be reduced. Areas removed from timber harvest will have greatly reduced forage available during the growing season. Alverson et al. (1988) hypothesized that in northern forests, deer populations would stabilize at lower densities if timber harvesting were eliminated, and, therefore, damage to plant communities caused by deer browsing would be diminished. There are few large areas in the Southern Appalachians where timber harvest has been excluded long enough to allow testing of this hypothesis. There are several reasons that it might not apply in this region and that cessation of timber harvest might result in increased damage to plant communities. At present, deer browsing on plants is concentrated in clearcuts, and browsing pressure is not excessive. Deer densities seem not to be limited by available biomass of green vegetation but are influenced more by a fall and winter energy supply that fluctuates independently of deer density (Wentworth et al., 1990a, 1992). Thus, deer populations may not decline, but likely will continue to fluctuate with acorn availability. Without clearcuts, which provide a concentrated source of spring and summer forage, deer browsing in the forest would become more widespread. In this situation, browsing during the growing season could seriously affect sensitive herbaceous species and seedlings of woody species and reduce forest diversity, as reportedly occurred in the 1930s through the 1950s (Ruff, 1938; Bryan, 1950). Thus, as forest management changes, there is a continuing need for more detailed study of the foraging habits of deer and their effects on forest vegetation, especially uncommon and sensitive plant species.
ADAMS, W. H., JR. 1959. Choccolocco deer range analysis and management implications. Proc. Annu. Conf. Southeast. Assoc. Game and Fish Comm., 13:21-34.
ALVERSON, W. S., D. M. WALLER AND S. L. SOLHEIM. 1988. Forests too deer: edge effects in northern Wisconsin. Conserv. Biol., 2:348-358.
BAIRD, W. V. AND J. L. RIOPEL. 1986. Life history studies of Conopholis americana (Orobanchaceae). Am. Midl. Nat., 116:140-151.
BECK, D. E. 1977. Twelve-year acorn yield in Southern Appalachian oaks. U.S. For. Serv. Res. Note SE-244. 8 p.
BRYAN, P. H. 1950 (1985). Deer range improvement by modified timber harvesting. Proc. Annu. Conf. Southeast. Assoc. Game and Fish Comm., 1-6(4):217-222.
CUSHWA, C. T., R. L. DOWNING, R. F. HARLOW AND D. F. URBSTON. 1970. The importance of woody twig ends to deer in the Southeast. U.S. For. Serv. Res. Pap. SE-67. 12 p.
DAY, F. P., JR. AND C. D. MONK. 1974. Vegetation patterns on a Southern Appalachian watershed. Ecology, 55:1064-1074.
DELLA-BIANCA, L. 1978. Characteristics, habitat, and fruiting of wild grapevines in the Southern Appalachians. J. Elisha Mitchell Sci. Soc., 94:21-26.
FORD, W. M., A. S. JOHNSON AND P. E. HALE. 1994. Nutritional quality of deer browse in Southern Appalachian clearcuts and mature forests. For. Ecol. Manage., 67:149-157.
-----, -----, ----- AND J. M. WENTWORTH. 1993. Availablity and use of spring and summer woody browse by deer in clearcut and uncut forests of the Southern Appalachians. South. J. Appl. For., 17:116-119.
FOWELLS, H. A. (Compiler). 1965. Silvics of forest trees of the United States. U.S. Dep. Agric. Agric. Handb. 271. 762 p.
HAPPE, P. J., K. J. JENKINS, E. E. STARKEY AND S. H. SHARROW. 1990. Nutritional quality and tannin astringency of browse in clearcuts and old-growth forests. J. Wildl. Manage., 54:557-566.
HARLOW, R. F. 1979. In defense of inkberry - dangers of ranking deer forage. Wildl. Soc. Bull., 7:21-24.
----- AND R. L. DOWNING. 1970. Deer browsing and hardwood regeneration in the Southern Appalachians. J. For., 68:298-300.
----- AND R. G. HOOPER. 1971. Forages eaten by deer in the Southeast. Proc. Annu. Conf. Southeast. Assoc. of Game and Fish Comm., 25:18-46.
-----, P. A. SHRAUDER AND M. E. SEEHORN. 1975. Deer browse resources of the Chattahoochee National Forest. U.S. For. Serv. Res. Pap. SE-136, 16 p.
JOHNSON, A. S., J. M. WENTWORTH AND P. E. HALE. 1989. Cumulative mast needs of forest wildlife, p. 18-23. In: C. E, McGee (ed.). Proceedings of the workshop Southern Appalachian mast management. U.S. For. Serv. and Univ. Tennessee, Knoxville.
LOFTIS, D. L. 1988. Regenerating oaks on high-quality sites, an update, p. 199-209. In: H. C. Smith, A. W. Perkey and W. E, Kidd, Jr. (eds.). Workshop proceedings: guidelines for regenerating Appalachian hardwood stands. Soc. Am. For. Publ. 88-03.
MCNAB, W. H. 1988. Hardwoods and site quality, p. 226-240. In: H. C. Smith, A. W. Perkey and W. E. Kidd, Jr. (eds.). Workshop proceedings: guidelines for regenerating Appalachian hardwood stands. Soc. Am. For. Publ. 88-03.
MARTIN, A. C., R. H. GENSCH AND C. P. BROWN. 1946. Alternative methods in upland gamebird food analysis. J. Wildl. Manage., 10:8-12.
MONK, C. D., D. T. MCGINTY AND F. P. DAY, JR. 1985. The ecological importance of Kalmia latifolia and Rhododendron maximum in the deciduous forest of the Southern Appalachians. Bull Torrey Bot. Club, 112:187-103.
MOORE, W. H. AND R. L. DOWNING. 1966. Some multiple-use benefits of even-aged management in the Southern Appalachians. Proc. Soc. Am. For., 1965:227-229.
POWELL, R. A. AND D. E. SEAMAN. 1990. Production of important black bear foods in the Southern Appalachians. Int. Conf. Bear Res. and Manage., 8:183-187.
RADFORD, A. E., H. E. AHLES AND C. R. BELL. 1968. Manual of the vascular flora of the Carolinas. Univ. North Carolina Press, Chapel Hill. 1183 p.
ROBERTSON, D. F. 1992. Ecosystem management of the national forests and grasslands. Memorandum to regional foresters and station directors. U.S. For. Serv. Washington, D.C. 6 p.
RUFF, F. J. 1938. The white-tailed deer on the Pisgah National Game Preserve North Carolina. Unpub. Rep. U.S. For. Serv. South. Region. 249 p.
SMITH, H. C. 1984. Forest management guidelines for controlling wild grapevines. U.S. For. Serv. Res. Pap. NE-548. 15 p.
WENTWORTH, J. M., A. S. JOHNSON AND P. E. HALE. 1990a. Influence of acorn use on nutritional status and reproduction of deer in the Southern Appalachians. Proc, Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies, 44:142-154.
-----, -----, ----- AND K. E. KAMMERMEYER. 1990b. Seasonal use of clearcuts and food plots by white-tailed deer in the Southern Appalachians. Proc. Annu. Conf. Southeast. Assoc. Fish and Wildl. Agencies, 44:215-223.
-----, -----, ----- AND -----. 1992. Relationships of acorn abundance and deer herd characteristics in the Southern Appalachians. South J. Appl. For., 16:5-8.
|Printer friendly Cite/link Email Feedback|
|Author:||Johnson, A. Sydney; Hale, Philip E.; Ford, William M.; Wentworth, James M.; French, Jeffrey R.; Ande|
|Publication:||The American Midland Naturalist|
|Date:||Jan 1, 1995|
|Previous Article:||Temporal variation in abundance of peromyscus leucopus in wooded habitats of eastern Kansas.|
|Next Article:||Analysis of coyote predation on deer and elk during winter in Yellowstone National Park, Wyoming.|