(001-4 Ha L3. --| Tit D\ Air UWU‘FWOI Can-.1. Michigan hm? Ugti'rcrzifiy This is to certify that the thesis entitled Natural Reproduction Methods for Oak in Southwestern Michigan 14 Year Results presented by John Patrick Hill has been accepted towards fulfillment of the requirements for Mdegree in__QE___Z_—K F 55 / 93' $13.16*» Major professor Dates- 9 86 0-7639 MS U is an Weave Action/Equal Opportunity Institution }V1£SI_J RETURNING MATERIALS: Place in book drop to uaamues remove this checkout from .‘Inlzs-IIL. your record. FINES will be charged 1f‘book is returned after the date stamped below. NATURAL REPRODUCTION METHODS FOR OAK IN SOUTHWESTERN MICHIGAN: 14 YEAR RESULTS 37 John Patrick Hill A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Forestry 1986 ‘I’oS/OS/ ABSTRACT NATURAL REPRODUCTION METHODS FOR OAK IN SOUTHWESTERN MICHIGAN: 14 YEAR RESULTS By John Patrick Hill It is widely recognized that obtaining desirable oak regeneration is a difficult silvicultural problem. This study evaluated the effectiveness of three silvicultural reproduction methods; shelterwood, group selection and clearcutting, in obtaining adequate reproduction of desirable species in an oak-mixed hardwood stand. The treatments were applied in the fall of 1971. In 1985, five 1/20th acre sample sub-plots were taken in each treatment and trees from 0.5 to 6.0 inches d.b.h. were measured. The shelterwood and the clearcut treatment significantly reproduced more red oak than the other treatments. Most of the red oak was in the small diameter classes and was being overtopped by faster growing species. Unless the competition is controlled, red oak won't be a major component of the future stand. White oak was eliminated from the stand in all treatments. Sugar maple was most common in the group selection and there were no significant differences between treatments for black cherry, red maple or elm reproduction. ACKNOWLEDGEMENTS I would like to express my sincere gratitude to Dr. Don Dickmann. His guidance and friendship has been invaluable throughout my education. I would like to thank Greg Kowalewski and Scott Newland for their friendship during the data-collection period of this study at the Fred Russ Forest. I would also like to thank Bill Cole for his assistance in the final preparation of the graphs. Finally, I would like to thank my wife, Sandra, for her continuous support, encouragement and love. 11 TABLE OF CONTENTS Page List of Tables........................................ iv List of Figures....................................... v Introduction.......................................... 1 Review of Literature.................................. 4 Study Area............................................ 13 Materials and Methods................................. 16 Results and Discussion................................ 21 Recommendations....................................... 47 Literature CitedOOIOCOIOOO00.00.0000...OOOOOOOOOOOOOOO 51 iii LIST OF TABLES Page Initial stand characteristics in 1971, by treatment and species--per acre basis........... 22 Stems to 6 inches D.B.H. 14 years after cutting (1985), by treatment, diameter class, and species-~per acre basis.............. 25 iv LIST OF FIGURES FIGURE Page 1. Michigan's lower peninsula showing the location of the Fred Russ Experimental Forest.............. 14 2. Map of the 45-acre oak-mixed hardwood stand at Fred Russ Forest showing block, treatment and sail boundarie'sssssssssessssossssssesssssososssss 17 3. Diameter distribution of the entire stand, sugar maple, white oak and red oak in 1971 before the cutting treatmentBOOCOOOOOOCCOIOOOOCCOOOOOOOOOOIOO 23 4. A view of the shelterwood treatment in 1985, 14 years after the harvestsssosssssssossssssssssso 26 5. A view of the clearcut treatment in 1985, 14 years after the harveSt....o................n..... 27 6. A view of the group selection treatment in 1985, 14 year. after the harveflt..................o..... 28 7. A view of the uncut control in 1985, 14 years after the h‘rveatssssssssssssossssssosossossssosso 29 8. Number of red oak trees per acre by treatment and diameter class. Any two means of total trees per acre with the same letter are not significantly different at the 5 percent level using the LSD.... 33 9. Number of sugar maple trees per acre by treatment and diameter class. Any two means of total trees per acre with the same letter are not significantly different at the 5 percent level using the Lsnssssssssosossssesoossssssosssssossses 35 10. Number of black cherry trees per acre by treatment and diameter class. Any two means of total trees per acre with the same letter are not significantly different at the 5 percent level using the Lsnssssssssssssssssssssssssssssssssssssa 37 11. Number of red maple trees per acre by treatment and diameter c18‘3sssssssossssssssssesssssssssssss 39 12. Number of elm trees per acre by treatment and di‘meter c1“‘.0.000000.I.OOOOOOOOOOOIOCOOOIOOIOOO 41 Figure Page 13. Diameter distribution of red oak, black chery, red maple 3nd 91“ in 19850.0ssoooossssssosoossssss 45 vi Introduction Oak-hickory forests occupy 114 million acres of forest land (Watt et al., 1973), more than any other forest type in the country. In the Lake States region, upland oak forests occupy 6 million acres (Arend and Scholz, 1969). Oak forests in Michigan occur on almost 1.8 million acres (Spencer, 1983), and is probably the most important commercial forest type in the southern half of lower Michigan. Over one third of Michigan's oak forest occur in that portion of the state, the majority of which is held by private individuals and farmers who own less than 100 acres (James et al., 1982). Eleven percent of the total growing stock in Michigan is oak, and the state has over 6600 million board feet of red and white oak sawtimber, second only to hard maple (James et al., 1982). The geographic distribution of the oak-hickory forest type occurs over a vast and varied area, and its boundaries are very indefinite and irregular. The ranges of the three most commercially important tree species; red oak, white oak, and black oak, encompasses most of the eastern United States. These three species of oak are the most important in terms of timber value, and in many cases, they are the most prevalent trees in the stand. 1 .\ 2 The oak-hickory forest type is extremely variable in composition, due to the wide range of soil, climate and topography the forest type can be found on. The oak-hickory forests of white, black and northern red oak are subclimax to climax, depending on the geographic location and site quality. 0n drier sites, the forest type usually succeeds itself (Eyre, 1980). On moist fertile sites, species other than oak are likely to increase in the absence of disturbance. In the hilly unglaciated portions of southwestern Wisconsin, northeastern Iowa and southwestern Lower Michigan, the existing red and white oak stands are ecologically unstable (Arend and Scholz, 1969; Sander, 1977). Future stands in these areas will eventually be dominated by the more mesic species (Sander, 1977), and tolerant hardwoods will become established gradually shading out the oaks in succeeding rotations (Arend and Scholz, 1969). In the north, oaks are succeeded by hard maple, beech, elm and ash (Eyre, 1980). In the central and southern parts of the range, less tolerant yellow poplar and ash will increase on good sites (Eyre, 1980) and the stand will always contain oaks, depending on the amount and size of advance reproduction (Sander, 1977). It is widely recognized that obtaining desirable regeneration on these oak-mixed hardwood sites is a difficult silvicultural problem. This study evaluated the effectiveness of three silvicultural reproduction methods; group selection, shelterwood and clearcutting, in obtaining adequate reproduction of desirable species in an oak-mixed hardwood stand in southern Michigan. The deve10pment of the reproduction was studied in relation to the treatments. Fourteen years after the cutting treatments were applied, changes in species composition and the stand structure are becoming evident. Recommendations for managing similar stands of oak and mixed hardwoods in southern Lower Michigan and for future studies in the hardwood management area at Russ Forest have been made based on the results of this study. Review of Literature A look at the initial establishment of many oak stands shows that very often they became established over a short period of time. In southwestern Wisconsin, Lorimer (1983) investigated the history of an 80-year old stand to determine the origin of the northern red oak component. He discovered that even though the total stand represented an uneven-aged distribution, the red oak component was essentially even-aged. The red oak did not become established until a major disturbance occurred in the stand. Rudolph (1956) studied a stand of old-growth hardwoods in southwest Michigan that had a similar age distribution. He found that the diameter distribution of the stand was chararacteristic of an uneven-aged stand. However, he concluded that the origin of the oaks in the stand resulted from a major disturbance that occurred 150 to 200 years ago, which created favorable conditions for the establishment of a fairly even-aged oak stand. The reproduction of oak is closely related to soil-site relationships. Gysel and Arend (1953) reported that the amount of available moisture largely determines site productivity for oak. They also found that the proportion of oak reproduction is related to the quality of the site. 5 0n very poor sites, 502 of the reproduction was oak, but the oak component dropped to 151 on medium sites and only 51 on good sites. 0n the good sites, oak must compete with faster growing species for light and moisture. Often following harvest, naturally regenerated stands contain fewer oak than they started with. The key to oak regeneration is an adequate supply of well-established, relatively large advance seedlings and saplings before harvesting the stand (Clark and Watt, 1971; Sander, 1972; Tubbs, 1977; Sander, 1984; and Johnson and Rogers, 1984). Watt et a1. (1973) stressed that oak advance reproduction should be at least four feet tall and one half inch in diameter at the root collar before it can compete effectively with the other reproduction when the overstory is removed. Do good sites, there is usually an inadequate supply of advance oak regeneration to reproduce the stand and foresters may have to rely on establishing new reproduction from seed or from sprouting. Sander (1981) states that the control of overstory density is critical to the proper establishment of new oak reproduction. Stand disturbance also is important in establishing oak reproduction. Advance reproduction builds up following a disturbance such as a partial cutting (Arend and Scholz, 1969); they found that oak seedlings are more numerous where mineral soil has been disturbed, such as in logged areas, fire lanes and skid trails. Carvell (1961) followed the development of oak 6 seedlings in West Virginia and found that the percent of sunlight reaching the forest floor shows a positive correlation with the amount of oak regeneration. Stands that have been thinned, grazed or lightly burned during the past generally have a larger proportion of oak reproduction than undisturbed stands. He attributed this increase in oak reproduction to the continuous supply of light which reaches the forest floor because of these repeated disturbances. The timing of a good acorn cr0p and a partial thinning or other disturbance are important for successful reproduction. Rudolph (1956) indicated that periodic heavy seed creps have a drastic effect on initial seedling abundance. In the growing season following a heavy acorn crop in a southern Michigan oak stand, oak seedlings increased dramatically. However, McGee (1975) found many of the small stems in a stand are very young and unless the stand is cut, most will die and be replaced by new seedlings. Several good seed crops may be required for a seedling catch because in most years essentially all oak acorns are destroyed by insects, fungi, birds or animals (Watt et al., 1973 and Sander, 1981). Marquis et al. (1976) reported insect damage to acorns is probably a primary cause of oak advance regeneration failure in Pennsylvania. He concluded that traditional cutting systems intended to stimulate oak advance reproduction are not likely to succeed unless practical methods can be found to control rodent and insect predation of acorns. Arend and Scholz (1969) 7 indicated that scarification would work many of the acorns into the mineral soil where they would be less likely to be eaten by animals. Scarification of the soil coinciding with a good acorn crop will also greatly enhance the establishment of oak seedlings (Scholz, 1955). However, this effect is only temporary (Scholz, 1959), unless the new seedlings are released. The reliance on sprouts to make up a major portion of the new stand may be risky. Most species sprout to some degree, but red maple and basswood, major oak competitors, do so prolifically (Metzger, 1980). Oak sprouts would face strong competition from these less desirable species. Sprout reproduction originating from stumps of old trees are often undesirable for high quality timber production because of poor survival and form, and the tendency for extensive rot (Tubbs, 1977), although oak generally produces vigorous sprouts with good form from smaller size trees. Johnson (1975) found that oak sprouts from newly cut stumps are valuable in reproducing oak stands. In a study on red oak stump sprouts in the Missouri Ozarks, he found that red oaks averaged about four stems per clump. However, further analysis showed that there was a negative correlation between the number of oak sprouts and the diameter of the stump (Johnson, 1977). The preportion of oaks that sprouted decreased as the diameter of the parent tree increased up to 26 inches, where no sprouting occurred. Selection silviculture works well for perpetuating 8 certain tolerant hardwoods but is not feasible for regenerating mixed oak stands because oak seedlings need full sunlight for best deve10pment (Arend and Scholz, 1969). In a mixed hardwood stand in southern Michigan, Rudolph and Bresnahan (1982) found that the more tolerant sugar maple increased while other desirable, less tolerant species, including red and white oak, decreased considerably after 20 years of single tree selection management. When the management objective is to perpetuate oaks, it can be best accomplished by the even-aged system (Watt et al., 1973). The group selection system does not appear to be a very successful method for reproducing oak stands. Lorimer (1983) concluded that despite the inability of red oak saplings to survive under a closed canopy, red oaks do seem to be able to compete with larger and older residual trees after a heavy cutting. Red oaks do survive under small gaps, however heavy numerical dominance of the overstory by red oak after selection cutting is not likely. Under the group selection system, the shade cast by the surrounding trees is a problem because of the usually small size of the epenings created by the harvests (Sander, 1981). Watt et a1. (1973) indicated that the size of the group cutting has little effect on density and composition of the reproduction, although trees grow slower on the edges than in the center of an opening. Roach (1965) and Watt et a1. (1973) suggest the minimum size of the Opening should be approximately 1/2 to 1 acre to insure successful 9 reproduction. In a 16 year study of selection silviculture in southern Illinois, Schlesinger (1976) began with group cuts of 1/10 acre, but in later treatments an attempt was made to create openings of at least ll8 acre. However, the red oak group did not appear to be replacing itself and he concluded that if selection management is continued, other species would replace the oak. The shelterwood system is the likely system to obtain large numbers of oaks, if advance oak reproduction is not adequate under mature stands (Sander, 1981). The Opportunity to manipulate stand density to enhance the reproduction of desired species makes the shelterwood method the most flexible even-aged management system. The degree of overstory removal has a definite effect on the species composition of the new stand. The heavier the removal, the higher the proportion of reproduction of less tolerant species (Trimble, 1973). The first, or seed cut reduces the overstory to establish oak seedlings. When making this cut Arend and Scholz (1969) recommend favoring northern red oak and white oak in the residual overstory. They also recommend a two-cut shelterwood where advance reproduction is lacking. Similar conclusions by Loftis (1983) indicated that acceptable even-aged stands with good species composition can be obtained using a two-cut shelterwood method on good sites. Sander (1981) stated that immediately after the desired reproduction is obtained, the remaining overstory should be removed in one cut. 10 In southern Michigan, Rudolph and Arnold (1956) studied the suitability of various cutting methods in an oak-mixed hardwood stand. The shelterwood plots showed more than twice as many oak seedlings as the plots given the lightest cutting. The abundance of oak seedlings was definitely related to the severity of cutting and the disturbance of the stand; the heavy cutting achieved greater disturbance of the forest floor. Breakage of small trees during harvest was not very serious since the subsequent sprouting resulted in some stems with a more desirable form. Rudolph (1956) concluded that if oaks were to be maintained as a significant component of the stand, drastic changes and considerable opening of the stand was neccessary for favorable establishment and growth of oak reproduction. In a similar study, Rudolph and Lemmien (1976) reported on the growth and development of hardwood reproduction 21 years after various cutting treatments. They found that for the first five years after the initial cuttings, there was adequate reproduction of most species in all treatments. By the fifth year, the group selection and shelterwood treatments had twice as many red maple as the control and the single-tree selection treatments. Oak was the next most common species; in the group selection and the shelterwood plots, there was a significantly greater number of oak stems larger than one foot in height than in the control. By the fifteenth year, height growth of oak reproduction in the group selection treatments had slowed down relative to the 11 shelterwood plots. This effect was probably related to the small size of the group selection openings. By the twenty-first year, it was becoming evident that even though the shelterwood system showed the best growth for oak, they were being overtOpped by other less desirable species such as red maple, sassafras, ironwood and dogwood. When large advance oak reproduction is present, complete overstory removal is recommended. Sander (1981) summarized the effects of clearcutting for oak, emphasizing it is important that all remaining culls and small trees be cut or killed to release the seedlings below them. He stressed clearcut openings should be no smaller than two acres, otherwise a high percentage of the area will be influenced by the surrounding trees. However, when stands on moist fertile sites are clearcut, species like red maple and black cherry will increase even when oak reproduction is present. Even though even-aged silviculture is the most advantageous for oak regeneration, it sometimes is not successful. Often, even with adequate oak reproduction, the oak cannot compete with other advance regeneration. Some method of disturbance such as scarification, herbicides or fire may assist oak seedlings in becoming a permanent fixture in the new stand. Rudolph and Arnold (1956) found that sugar maple, red maple and black cherry were abundant without apparent relation to the method of cutting. In North Carolina, Loftis (1983) found that upland oaks faired 12 poorly after clearcutting on the productive sites. Oak advance reproduction that was present at the time of the cutting grew, but most were overtOpped by other species. He suggested that treating the subcanopy with a herbicide, in addition to the reduction in overstory basal area, would allow desirable species to prosper. Johnson and Jacobs (1981) concluded a preherbicided clearcut created more favorable conditions for establishment of oak seedling reproduction than a shelterwood cut. On good sites where establishment of adequate reproduction is difficult, additional treatments with the cutting method, will enhance the success of regenerating oak stands. Study Area The stand under study is in the Fred Russ Experimental Forest in Cass County (T58,R14W), near Dowagiac, Michigan (Figure 1). The Russ Forest was given to Michigan State University in 1942 by Mr. Fred Russ of Cassopolis, Michigan. The 580 acre forest is managed and operated by the Forestry Department of Michigan State University and contains over 180 acres of mixed hardwood timber, much of it mature. The area has a relatively mild, humid climate, with spring-summer-fall air temperature (May-October) averaging 65 degrees Fahrenheit. The frost-free growing season is approximately 150 days and average annual precipitation measures 34 inches. The stand occupies a site which is generally level to gently undulating in topography. Soils are deep and well drained with moderately rapid permeability that were developed from outwash plains. The soils are predominantly sandy loams of the Oshtemo and Kalamazoo series. The Oshtemo soil is classified as a coarse-loamy, mixed mesic, Typic Hapludalf. The Kalamazoo soil is similar to the Oshtemo but is finer in texture; it is commonly found adjacent to Oshtemo soils and in similar positions on the landscape. Original vegetation on these soil series were 13 l4 I I fins imam Cass County 5 jam-m ' Jack m. Aqiiil'iiiilllllll “m iiii." ti. ‘Qfll a..hsflnmmlfl - ummueubmu. u -1 L 7 , Figure 1. Michigan's lower peninsula showing the location of the Fred Russ Experimental Forest. 15 hardwood forests of oak, hickory and maple. The site quality for oak is very good on these soils of fine texture and level topography (Gysel and Arend, 1953), with a site index for northern red oak of 66 feet at 50 years. The major soil limitations for general management, such as erosion hazard, equipment limitations and seedling mortality, are insignificant. Past use of the stand prior to the acquisition by Michigan State University was predominantly as a source of firewood and occasional cattle grazing. During the last 44 years, it has been protected from such activities. Since then, trees that were blown down have been removed from the area and salvaged. There are no records of fires that have occurred on the area for some time. In an earlier study of Russ Forest hardwoods adjacent to this stand, examination of the stumps of trees cut did not show any scars of past fires. However, it is believed that the adjacent stand was probably established some 150 to 200 years ago by a severe fire or similar major disturbance. The study area is a 45-acre stand of overmature, somewhat decadent oak-mixed hardwoods. Common understory species are flowering dogwood (Cgrnua glorids), sassafras (Sassafras glhidgm), ironwood (Qgtgya zirginiana), and hawthorn (gigtaegus 522;). Major overstory species included northern red oak (Quercgs gubrg), black oak (Quercug gelggiag). white oak (Qggrgu; alga), sugar maple (Age; sgccharum), red maple (Age; rubrum), silver maple (Ace; 16 Lagghgriggm), black cherry (Prugug sergtigg), slippery elm (Ulmug rubra), and American elm (Ulmus aggricgna). Minor overstory components in decreasing order of importance included, bitternut hickory (Cgrya cordigggmig), shagbark hickory (Cary; agate), American beech (Fggug grandifolia). white ash (Fraxigug gmerigagg), American basswood (Tilia gme;;c n_), butternut (lugla s ginere ). and bur oak (Quercug macrggarpg). Materials and Methods The present study was set up in 1971 by Victor J. Rudolph, now retired professor of forestry at Michigan State University. His main objective was to determine the ecological changes brought about by various silvicultural treatments applied to an overmature mixed hardwood stand. Also, the development and growth of the various tree components were to be correlated with the site modifications and conditions resulting from the various treatments. The 45-acre stand of overmature oak-mixed hardwoods in compartment 8-21 in the Fred Russ Forest was subdivided into four blocks with four randomized treatments in each block (Figure 2). The design was laid out so that the Kalamazoo soil is in block one and the Oshtemo soil is in blocks two, three and four, thus blocking out any soil differences. The treatments are 2.75 acres in size, with each plot surrounded by an isolation boundary, creating interior plots of 1.75 l7 OAK-MIXED HARDWOOD STAND "a I m N I “an0 CLEAR u CUT. CUT . :3 _ 13' Treatment Plot Center . s a: -=-*__ ”rum-Mm I H ,-, if}: ‘15.- 2; 1-1.etc. Block a Treatment Replication u ----.-._:--‘:-- === mm H «2133.510 TIL-32h- u $3?“ ° ‘3' m “'5 no :- Jsm'rlt ' - i- - I coarser. cmur r:— I 'A' a; l I 1-3 1-4 I; conrsor. céfi?" u . eggs 1% t; u / * ‘-' ‘-2 " GROUP CLEAR 3-1 CUT CUT osoue sueLreswooo it fi cur cur 2-1 2-2 s: t 4.: H ensues ll sueuemooo ° “.2? 955?“ ll” CUT CONTROL 3-2 fi "0‘ PHI. // fi, * ll 2-3 2 3-3 ' // ‘4 0 // // coaxial. // l/ _ A’ s 4 // Map of the 45-acre oak-mixed hardwood stand at Fred Russ Experimental Forest showing block, treatment and soil boundaries. Figure 2. 18 acres. Initial measurements were made of each tree four inches in diameter at breast height (4.5 feet) and larger. Species, diameter, basal area, merchantable height in sixteen foot logs, and gross and net board foot volumes based on the international 1/4 inch rule were recorded. During the late fall of 1971. the randomly assigned cutting treatments were applied to the stand. The shelterwood treatment left between 21 to 35 square feet of residual basal area per acre of desirable trees like red and white oak. The group selection treatment removed groups of trees, each occupying an area of approximately one-fifth acre. The rest of the plot was left uncut so that the residual basal area was approximately 90 square feet per acre for the entire plot. The clearcut method removed all trees in the plot. Merchantable trees were removed by the logger and then all remaining trees were cut down. The final treatment was a control on which no cutting was done. Before the harvesting commenced, all skid roads. trails and landings were carefully laid out. Small skidders were used to haul the logs to the landings, which were located off the experimental area. There was no skidding of logs across the control plots and trees were not felled into the control areas. The harvesting operations began in late October and were completed by mid-January. The tops of the sawlog trees and the unmerchantable trees were cut and used as firewood. The remaining slash and logs were lopped and 19 scattered. In the winter of 1974-75, the residual overstory in the shelterwood treatments were removed from the study area and a second set of group selection cuts were done. Each treatment plot was reinventoried in the summer of 1985 using a series of ll20th acre subplots. Each sample subplot was located using a grid on a map of the study area. Sample subplots were numbered from one to 35 on the grid and five subplots were randomly located in each treatment. Since there are two group selection cuts in each plot for this treatment, the sample subplots were distributed between the two. One-twentieth acre subplots were chosen to better estimate the true population mean. These large plots would account for greater variability among the sapling-size reproduction. The 16 treatment plot centers were located and flagged. The location of each subplot was witnessed off a treatment plot center by measuring in a north or south and east or west direction in feet. The 0.05 acre subplots are square, measuring 46.67 feet on each side. They were permanently staked using 1/2 inch conduit pipe three and one half feet long painted orange to represent the sample subplot center. An aluminum tag was attached to each stake to identify the block, treatment and subplot number. The four corners of the subplots were then measured and flagged. Species were recorded and trees were tallied into one inch diameter classes using a set of fixed calipers. Trees taller than 4.5 feet, but not one inch in diameter, were 20 recorded into the 0.5 inch class. Also, trees 4.5 to 6.9 inches were tallied into the six inch diameter class. After each tree was measured and recorded, it was marked with a blue chalk bag to avoid recounting trees within the subplots. The data for each treatment was summarised into stand tables on a per acre basis. An analysis of variance was used to test for differences among treatment areas and the least significant difference (LSD) test was used to test separate treatment means of individual species for significance at the 5 percent level. Results and Discussion Initial stand characteristics in 1971 are shown by treatment and species on a per acre basis in Table l. The diameter distribution of the entire study area was characteristic of an uneven-aged stand (Figure 3). The data for three common species in the stand shows large differences in diameter distribution. Sugar maple was very abundant in the small diameter classes. However, red and white oak maintained fairly uniform representation in the large diameter classes. Based on the large average diameter of the oaks, it is likely that a major disturbance occurred in the stand, creating conditions that were favorable for the establishment of a fairly even-aged stand of red, black and white oak. The tolerant sugar maple has been increasing in the stand ever since, being able to cope with the existing stand conditions more successfully than the oak. Comparisions of the number of trees per acre by treatment and species from the original stand and the new stand as a percentage of the entire stand shows how the species composition has changed since the cutting treatments were applied. In the shelterwood plots, the red oak group increased from 16.51 to 19.81. Sugar maple had the greatest increase-112 to 27.31; elm increased slightly and red maple 21 22 Table 1-Initial stand characteristics in 1971, by treatment and species--per acre basis. Number Avg. 0.3.8. Basal area of trees finches} Ssg. ft.) Treatment Species 1971 1971 1971 Shelter— Red oak 13.2 21.9 37.4 wood Black oak 7.2 22.9 23.5 White oak 16.5 21.1 47.9 Sugar maple 13.5 6.4 6.3 Black cherry 29.2 7.9 9.2 Red maple 12.5 7.9 3.5 Elm 1 13.5 5.7 3.9 Miscc- 17.8 6.9 7.1 All species 123.4 10.7 138.8 Clear- Red oak 10.3 19.9 28.4 cut Black oak 1.2 14.3 3.7 White oak 16.8 26.6 61.5 Sugar maple 16.8 11.0 14.6 Black cherry 13.3 7.9 6.2 Red maple 12.8 9.7 10.3 Elm 1/ 13.0 5.3 3.5 Miscu- 25.9 6.0 8.0 All species 110.1 10.4 136.2 Control Red oak 6.3 22.4 20.1 Black oak 4.0 15.2 8.7 White oak 16.3 24.2 56.8 Sugar maple 18.8 7.6 7.9 Black cherry 16.0 10.1 10.8 Red maple 24.0 8.1 14.1 Elm 1/ 22.5 5.8 22.9 Misc.- 16.1 6.2 6.3 All species 124.0 10.4 147.6 Group selection Red oak 9.3 22.7 30.7 Black oak 2.3 21.5 8.2 White oak 12.5 26.2 48.9 Sugar maple 24.0 8.6 13.9 Black cherry 10.5 10.2 7.5 Red maple 15.3 9.3 12.3 Elm 1/ 15.8 5.5 4.9 Miscc- 20.5 4.8 7.5 All species 110.2 10.7 133.9 1/ -Hickory, ironwood, beech, sassafras, dogwood. 23 224 E -9 lflNflEfltlrflllfl PERACRE Q o. “ \Ti ,7 ‘i 0M?>1-flu—"' ------ ~L--~_._ o z 4 s u: a Is m m 29‘: 24_ a.as urea 34 a»ss‘m m- mom Figure 3. Diameter distribution of the entire stand, sugar maple, white oak and red oak in 1971 before the cutting treatments. 24 stayed constant. Black cherry decreased from 241 to 111 in the new stand. The most noticable decrease was in the white oak component; it has disappeared from the entire stand in all treatments. There are similar changes in species composition for the clearcut treatment. The red oaks increased from 10.51 in the original stand to 221 in the new stand. Elm increased to 151 in the new stand, red maple and black cherry decreased slightly to 5.51 and 9.31, respectively. The group selection plots had a very different effect on the new reproduction. Tolerant sugar maple is now the dominant component of the new stand; it increased from just under 221 of the original stand to 531 in the new stand. Black cherry was the only other species to show an increase, from 8.21 to 9.51. Red oak has almost disappeared as a result of the group cuts and comprises less than 11 of the new stand. Red maple and elm also decreased in the new stand now dominated by sugar maple. Results of the three treatments and the uncut control 14 years after cutting are presented in Table 2. It is quite clear that all three cutting methods were successful in reproducing a new stand of hardwoods. Figure 4 through 7 gives a representative view of the four treatments in 1985. The differences between the uncut control and the group selection, shelterwood and clearcut treatments were found to be highly significant. The total stems per acre for the group selection was 2936, the shelterwood 2844, and the 25 Table 2. Stems to 6 inches D.B.H. 14 years after cutting (1985) by treat- ment, diameter class, and species-—per acre basis. D.B.H. Redll Sugar Red Black Treatment inches oak- lggplg’ maple ‘51! cherry Hick. Misc.- Total Shelter- 0.9;! 20 71 9 23 2 5 41 171 wood 1 340 486 105 138 39 19 225 1352 2 187 149 101 125 79 17 108 766 3 17 54 43 77 74 5 45 315 44/ 0 15 17 49 63 1 10 155 6- __o _o __6. .11 .51 ._0 .3 _85 Total 564 775 2 1 429 311 47 484 2844 Clear- o.s§/ 13 4s 7 a o 14 S6 143 cut 1 397 402 56 88 28 49 341 1361 2 206 135 61 143 71 32 105 753 3 25 53 25 129 75 7 69 383 44/ 2 8 10 45 52 l 40 158 6- _o __0. _o_ .22 .93 __o .19. __104 Total 643 643 159 435 269 103 747 2902 Control 0.53! o 194 13 37 o 11 94 349 1 1 219 22 39 5 5 52 342 2 0 31 11 14 5 0 37 98 3 O 24 6 13 4 0 21 68 44/ 0 9 6 8 6 1 9 39 6- __0. _9 ._3 .3 .3 .9 _6 _22 Total 1 486 61 113 22 17 219 918 Group 0.53/ 5 396 19 22 4 15 119 580 selection 1 14 886 77 110 44 30 235 1396 2 2 179 77 94 65 5 98 520 3 0 66 55 55 51 0 17 244 44/ 0 20 22 34 57 0 9 142 6- _o __n __6. _1_1. 32 _o_ __6. _54 Total 21 1558 256 326 241 45 534 2936 1/ - Quercus rubra, 9_._ velutina. 2/ -Includes white oak, aspen, beech, black walnut, white ash, hackberry, sumac, hawthorn, tulip poplar, box elder, sassafras, ironwood, musclewood, dogwood. 3/ - Trees taller than 4.5 feet, less than 1 inch D.B.H. 4/ - Trees 4.5 to 6.9 inches D.B.H. 26 A view of the shelterwood treatment in 1985, 14 years after the harvest. Figure 4. 27 A View of the clearcut treatment in 1985, 14 years after the harvest. Figure 5. 28 A view of the group selection treatment in 1985, Figure 6. 14 years after the harvest. 29 Figure 7. A view of the uncut control in 1985, 14 years after the harvest. 30 clearcut 2902, whereas the uncut control had only 918 stems per acre greater than 4.5 feet tall and less than 6 inches in diameter. The data that are of most interest are the amount of oak reproduction that established in the new stands. The shelterwood method and the clearcut method were significantly different from the group selection treatment or the uncut control in this respect (Figure 8). In terms of reproducing red oak, there was no significant difference between the shelterwood and the clearcut; they both have been successful in reproducing red oak into the new stand. A heavy or complete removal of the overstory and coinciding forest floor disturbance has encouraged red oak growth and development. While collecting data it was observed that red oak reproduction in the heavily disturbed skid trails was plentiful. The combination of forest floor disturbance throughout the shelterwood and clearcut treatments, with ample sunlight reaching the ground, encouraged new red oak reproduction to grow. It appears the primary cause of failure of red oak to reproduce in the group selection treatment was the small size of the openings. The total stems per acre of reproduction for the 1/5 acre group cuts was 2936; thus the size of the opening had no effect on the density of the new reproduction in comparison with the shelterwood and clearcut plots. However, the shade cast by trees surrounding the group Openings did not allow enough sunlight in to promote 31 oak. This conclusion is supported by Roach (1965) and Watt et al. (1973) who reported group cut openings should be at least 1/2 to 1 acre in size for successful reproduction. The absence of white oak was very conspicuous. White oak is often a heavy producer of acorns, but good seed years are not regular, several years may pass without an acorn crOp. Failure of the acorn crop may have been the case in this stand. Also, acorns of white oak germinate in the fall soon after they drop to the ground. This trait could have been a disadvantage because the roots may not have had a chance to penetrate the soil; harvest operations commenced in the fall when the germinating seeds could have been damaged or exposed to dry conditions and freezing weather. In terms of sugar maple stocking, the group selection cut was significantly different from the other treatments, and the shelterwood was significantly different from the control at the 51 level (Figure 9). It is evident that the degree of overstory removal had a definite effect on the species composition of the new stand. In all treatments, sugar maple was the most common species, but its presence appears to have been reduced as the amount of overstory removal in the original stand increased. In the uncut control plots and the light cutting of the group selection treatments, young sugar maples of 0.5 to 6 inches d.b.h. ocuppied 531 of the new stand. The heavy removal of the shelterwood reduced the preportion of sugar maple to 271 and the complete removal of the overstory in the clearcut plots 32 Figure 8. Number of red oak trees per acre by treatment and diameter class. Any two means of total trees per acre with the same letter are not significantly different at the 5 percent level using the LSD. RED OAK N \ 34 Figure 9. Number of sugar maple trees per acre by treatment and diameter class. Any two means of total trees per acre with the same letter are not significantly different at the 5 percent level using the LSD. SUGAR MAPLE [masses 15 Figure 9. 1.5- 1.44 I.3~ 1.2- 1 0 Ha ( ' . in "‘ g; 0.9“ '5 s .2: 03* b 0 3'5 17 ~ 1v be ' 0.61\ o 03-1 %\ \\\\\\ SHEIJH? (1mm W511 ClitingTreahneIt 36 Figure 10. Number of black cherry trees per acre by treatment and diameter class. Any two means of total trees per acre with the same letter are not significantly different at the 5 percent level using the LSD. Number of Tr... p.1- Aer-o 37 BLACK CHERRY DHWosses Fi ure 10. 320 260 240 21]) 180 160 140 120 8388 a SHHJER CLEARcur me SEL comm cutting Treatmmt [Z] 0.5" [:3] 1.0- 2.0- 10- [XZ] 4.0- m 5.0" 38 Figure 11. Number of red maple trees per acre by treatment and diameter class. RED MAPLE [310m \\\\\/\/ \\\\\\ SHELTER CLEMCUT GRN’SEL comm CuttingTreahnmt [:30 m0" NM“ IXXM' m 4-0' 40 Figure 12. Number of elm trees per acre by treatment and diameter class. ELM [EH 7MMM» //// SHELTER CLEARwr GROUPS CutfingTreohnent E] 0.5" [E 1.0' 10- 3.o~ [XE 4.0- m 5.0" 42 reduced the sugar maple component to 221 in the new stand. Figure 10 shows the amount of reproduction of black cherry for each treatment. All three cutting methods significantly reproduced more black cherry than the control, but there were no differences between the three cutting methods; each treatment reproduced black cherry equally well. Black cherry was most common in the 4 and 6 inch diameter classes, occupying a large portion of the dominant crown class. Many of these larger trees were of stump sprout origin. Red maple was most common in the shelterwood treatment. There were no significant differences between the treatments; red maple was found in all plots, but it wasn't a major component of the new stand (Figure 11). Elm, both slippery and American, was found in all treatments, regardless of the cutting method (Figure 12). Like black cherry, much of the larger elm reproduction came from stump sprouts. Some of the American elm showed signs of infection by Dutch elm disease. Thus, the undesirable elm component will continue to decline and be replaced by more vigorous species. Because the amount and size of advance regeneration present in the stand in 1971 is unknown, it is difficult to accurately identify where all the reproduction came from. Very few seedlings over two feet tall were observed in the control plots during the spring of 1985. Much of the ground cover consisted of herbaceous plants and the most common 43 tree seedlings observed were sugar maple, red maple and elm. The abundance of these small, young seedlings appeared to be a temporary annual event, as most of the seedlings did not survive the summer. Based on these observations and the fact that no oak reproduction was found in the sample subplots taken in the controls, most of the red oak component probably originated from a heavy acorn crop of red oak that occurred during the harvest year. One dominant red oak sapling from each of the four clearcut plots and the four shelterwood plots was cut down at ground level. A disk from the base of each tree was taken to determine its age. Five of the red oak saplings had 14 growth rings and three had 13 growth rings, indicating the red oak stems became established immediately after the treatments were applied. The timing of the harvest combined with a good seed crop provided adequate conditions for establishing oak reproduction. The lack of advance regeneration and the success of the clearcut treatment are contrary to current research, but the coincidence of a good seed crop and the destruction of other advance regeneration by the complete cutting may have created conditions favorable to the success of red oak. However, the larger oaks more than likely came from advance regeneration. The heavy removal of the seed cut in the shelterwood treatment also created favorable conditions for the establishment of red oak. The stand was opened, encouraging 44 acorn production and providing adequate conditions for the growth of oak seedlings. Had it not been for the possibility of a good seed crop the year of the harvest, treatment differences between the shelterwood and the clearcut may have been greater. Reproduction from oak stump sprouts was insignificant because of the old age and large size of the parent trees. Sprouting did make up a large portion of the reproduction of other species, however. Red maple, black cherry and elm were all prolific sprouters and, as a result, many of these trees are in the larger diameter classes. Since sugar maple is very tolerant and an aggressive seeder, much of it came from advance regeneration as well as from seed. There is strong competition for space in the new stand; the majority of the red oaks are in the one and two inch diameter classes, and over half of the red oaks are being overtopped by faster growing species. Figure 13 compares the diameter distribution of red oak, black cherry, red maple and elm; it clearly shows the red oak is not in a dominant position in the stand. The red oak may be more numerous in the small diameters in terms of trees per acre, but the black cherry, red maple and elm are larger and in the dominant classes, suppressing many of the oaks. The results of this study indicate the problem of oak reproduction may not be in getting adequate oak regeneration, but in maintaining it. If the oaks are to be maintained as a major component of the stand in the future, 200 150 No. Tr... / Aer. 100 [1 deOak Figure 13. 45 DIAMETER DISTRIBUTION 1985 .1 ‘ ' r I l 1 "‘ 0.0 2.0 4.0 6.0 DBI-l CLASS + Red Maple 0 Elm A Block Chem Diameter distribution of red oak, black cherry, red maple and elm in 1985. 46 measures to control undesirable species may need to be implemented. Recommendations This study examined three cutting methods designed to reproduce oak. In southern Michigan, similar stands on good sites will successfully reproduce red oak by the shelterwood method or the clearcut method if large advance regeneration is present or if cutting coincides with a bumper crop of acorns. Reliance on a good acorn crop alone when clearcutting is employed is not recommended. For many owners of oak-mixed hardwood stands similar to the one reported on here, timber production is not the only objective. The shelterwood method is well adapted to multiple use management. Timber production is achieved and wildlife habitat conditions may be more desirable than with a clearcut. The aesthetic conditions are more favorable with shelterwood, too. A new stand will be established before the mature one is completely removed, reducing the aesthetic shock of the harvest and creating more natural looking conditions. On good sites, the cutting method alone will not always reproduce well-stocked oak stands; competition from other species must be controlled. Treating the stand just before or immediately after the harvest with a herbicide to kill competing species is one method of control. The elimination 47 48 of the competition with a herbicide will create more favorable conditions for the establishment of oaks. The use of fire in young reproduction to selectively release oak can also be beneficial. Oaks are well adapted to the occurrence of fire; most oak species are capable of resprouting after the above-ground portion of the tree has been killed (Lorimer, 1985). Management of the oak root systems in the initial stages of establishing a new stand is the key. After a low intensity fire, young oaks quickly resprout, often with more vigor and better quality than the original stem. Oaks will continue to sprout after repeated burns; building up reserves in the root system for fast growth when they are released, while competing species lose their ability to resprout after more than one fire. The application of several periodic prescribed fires with a shelterwood cutting will help promote oak advance reproduction. Further investigation on the use of prescribed fire for regenerating oaks in Michigan is essential. The stand at Russ Forest could easily be set up for a study on prescribed burning to release oak reproduction. Even after oak regeneration is established, intermediate treatments are often needed to ensure that the desirable trees have adequate conditions for superior growth and development. Because cleaning operations are labor intensive, their cost is often high, so they should be limited to the good sites where fast growth and high quality 49 can be achieved. The Russ Forest stand is a typical site where a cleaning may be beneficial. A study to evaluate the economic feasibility and changes in species composition and subsequent growth of the remaining crop trees at Russ Forest by comparing cleaned and uncleaned treatments could be accomplished using a split-plot design. Currently, the dominant and codominant red oaks average 28 feet tall and releasing these selected crop trees would be the most practical means of creating future high quality sawtimber trees. Arend and Scholz (1969) have summarised techniques for releasing oak on good sites similar to the site at Russ Forest. They recommend releasing about 200 to 300 selected saplings per acre by eliminating undesirable species that are crowding or suppressing potential crop trees. Attempting to remove all the undesirable saplings is not recommended since it may result in the crap trees developing excessively branchy crowns. Choosing trees that have suppressed or intermediate crowns as crop trees should be avoided because they will not respond as well after release. The undesirable trees should be cut out using saws or axes, or they could be killed by herbicide injection. The decision of which method to use should be based on the the availability of inexpensive labor and a knowledge of herbicides. The objective of a cleaning should be to favor high quality red oak and black cherry, while eliminating elm, red maple, ironwood and other "weed species". Regardless of the regeneration method used, some type 50 of release is neccessary if a quality oak stand is to result. If the stand is left alone, composition will continue to shift towards species other than oak. LITERATURE CITED Literature Cited Arend, J. L. and H. F. Scholz. 1969. Oak forests of the Lake States and their management. USDA For. Serv. Research Paper NC-31. Carvell, R. L. and E.H. Tryon. 1961. The effect of environmental factors on the abundance of oak regeneration beneath mature oak stands. For. Sci. 7: 98-105. Clark, F. B. and R. F. Watt. 1971. Silvicultural methods for regenerating oaks. Oak Symposium Proceedings. USDA Fore serVe “one For. E‘pte 31:8. ppe 37-430 Eyre, F. H. 1980. Forest Cover Types of the United States and Canada. Society of American Foresters, Washington D.C. pp.41—42. Gammon, A. D., V. J. Rudolph and J. L. Arend. 1960. Regeneration following clearcutting of oak during a seed year. J. For. 58: 711-715. Gysel, L. W. and J. L. Arend. 1953. Oak sites in southern Michigan: Their classification and evaluation. Mich. Agr. Bxpt. Sta. Tech. Bull. 236. James, L. M., S. B. Heinen, D. D. Olson and D. B. Chapplle. 1982. Timber products economy of Michigan. Mich. State Univ. Agr. Bxpt. Sta. Research Report 446. Johnson, P. S. 1975. Growth and structural development of red oak sprout clumps. For. Sci. 21: 413-418. Johnson, P. S. 1977. Predicting oak stump sprouting and sprout develOpment in the Missouri Ozarks. USDA For. Serv. Research Paper NC-l49. Johnson, P. S. and R. D. Jacobs. 1981. Northern red oak regeneration after preherbicided clearcutting and shelterwood removal cutting. USDA For. Serv. Research Paper NG-202. Loftis, D. L. 1983. Regenerating southern Appalachian mixed hardwood stands with the shelterwood method. Southern Journal of Applied Forestry 7(4): 212-217. 51 52 Lorimer, C. G. 1983. Eighty-year development of northern red oak after partial cutting in a mixed-speices Wisconsin forest. For. Sci. 29: 371-383. Lorimer, C. G. 1985. The role of fire in the perpetution of oak forests. Proceedings of Challenges in Oak Management. U. of Wisc. Coop. Ext. Serv. pp. 8-25. Marquis, D. A., P. L. Eckert and B. A. Roach. 1976. Acorn weevils, rodents and deer all contribute to oak- regeneration difficulties in Pennsylvania. USDA For. Serv. Research Paper NE-356. McGee, C. E. 1975. Regeneration alternatives in mixed oak stands. USDA For. Serv. Research Paper 83-125. Metzger, F. T. 1980. Strip clearcutting to regenerate northern hardwoods. USDA For. Serv. Research Paper NC-186 e Roach, B. A. 1965. Is even-aged management practical for small woodlands in the central states? Society of American Foresters Proceedings, pp. 151-153. Rudolph, V. J. 1956. Stand conditions in the old-growth hardwoods of the Fred Russ Forest in southwestern Michigan. J. For. 54: 249-257. Rudolph, V. J. and C. I. Arnold. 1956. The Fred Russ Forest cuttings: A progress report. Mich. Agr. Sta. Quart. Bull. 38(4): 644-663. Rudolph, V. J. and W. A. Lemmien. 1976. Silvicultural cuttings in an oak-hickory stand in Michigan: 21 year results. Proceedings of the Central Hardwood Forest Conference. pp. 431-453. Rudolph, V. J. and R. G. Bresnahan. 1982. Twenty years of management in a small woodlot of southern Michigan. J. For. 80: 665-667. Sander, I. L. 1972. Size of oak advance reproduction: Key to growth following harvest cutting. USDA For. Serv. Research Paper NC-79. Sander, I. L. 1977. Manager's handbook for oaks in the North Central States. USDA For. Serv. Gen. Tech. Report “6'37 e 53 Sander, I. L., P. S. Johnson and R. Rogers. 1984. Evaluating oak advance reproduction in the Missouri Ozarks. USDA For. Serv. Research Paper NC-251. Sander, I. L., C. Merritt and E. H. Tryon. 1981. Choices in Silviculture for American Forests. Society of American Foresters, Washington D.C. pp.23-29. Schlesinger, R. C. 1976. 16 years of selection silviculture in upland hardwood stands. USDA For. Serv. Research Paper NC-125. Scholz, H. F. 1955. Effect of scarification on the initial establishment of northern red oak reproduction. U.S. For. Serv. Lake States For. Expt. Sta. Tech. Notes 425. Scholz, H. F. 1959. Further observations on seedbed scarification show benefits to northern red oak were temporary. U.S. For. Serv. Lake States For. prt. Sta. Tech. Notes 555. Spencer, J. S. 1983. Michigan's fourth forest inventory: Area. USDA For. Serv. Resource Bull. NC-68. Trimble, G. R. 1973. The regeneration of central Appalachian hardwoods with emphasis on the effects of site quality and harvesting practices. USDA For. Serv. Research Paper NE-2S2. Tubbs, C. H. 1977. Natural regeneration of northern hardwoods in the northern Great Lakes region. USDA For. Serv. Research Paper NC-150. Watt, R. F., R. A. Brinkman and B. A. Roach. 1973. Silvicultural Systems for the Major Forest Types of the United States. USDA For. Serv. Agr. Handbook no. 445, pp. 66-69.