INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Paga(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity. 2. When an (mage on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete. 4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received. Xerox University Microfilms 300 North Zaeb Road Ann Arbor, Michigan 43100 I I 74-6078 LEGG, Michael Hampton. 1946SITE FACTORS USEFUL IN PREDICTING DETERIORATION ON FOREST CAMPSITES IN NORTHERN MICHIGAN. Michigan State University, Ph.D., 1973 Agriculture, forestry & wildlife University Microfilm s, A XEROX Com pany, Ann Arbor, Michigan SITE FACTORS USEFUL IN PREDICTING DETERIORATION ON FOREST CAMPSITES IN NORTHERN MICHIGAN by Michael Hampton Legg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Forestry 1973 ABSTRACT SITE FACTORS USEFUL IN PREDICTING DETERIORATION ON FOREST CAMPSITES IN NORTHERN MICHIGAN by Michael Hampton Legg A three phase study to determine the ecological impact of recreation on forest ecosystems was conducted in the Crooked Lake Region of the Sylvania Recreation Area in the Ottawa National Forest of Northern Michigan. The first phase utilized a mechanical trampling device to simulate recreation use upon potential forest campsites. This permitted close monitoring of the changes occurring under known levels of trampling. The second phase was concerned with measuring the amount of deterioration that takes place on established boat access campsites at several levels of visitor use. The last part of the study involved the recovery to natural conditions of abandoned campsites. All phases of the study were located on well drained sandy loam to loamy sand soils. Simulated Recreation Use Recreation use was simulated by means of a mechanical trampler on four potential campsites, two hardwood and two conifer, during the 1971 and 1972 camping seasons. plots were established. groups: trampled. At each site, 16 one-meter square The plots were randomly divided into four controls, lightly trampled, moderately trampled, and heavily Levels of trampling were 0 passes, 6 passes, 12 passes, and Michael H. Legg 18 passes per week, respectively. pressure of 350 g cm •2 The trampler applied an average on each pass. The response to trampling was monitored by measuring the follow­ ing parameters four times each seasoni 1. Soil moisture content (% by volume) 2. Soil bulk density (g cm-3) 3. Percent natural litter cover 4. Soil air permeability. Multiple regression analysis was utilized to develop prediction equations estimating the change in each parameter over the two-year period. The range of variability explained was from 68 percent for change in oven dry weight of litter to 91 percent for the change in percent by volume of noncapillary pore space in the soil. After determining the predictability of the measured parameters, the actual percent changes in them, as measured in the field, were utilized to develop an index of detrimental change in the type of sites monitored. The equation, with an R 2 of 0.89, ranks campsite deterioration on a scale of one to four, and is most useful on newly established camping units. Established Campsites Twelve primitive camping units selected by degree of visitor use were divided into light, moderate, and heavy use classes and further subdivided into timber type. concentric sampling zones t The camping unit was divided into three the heavily used center of the camping unit, the moderately used area near the margin, and a control zone beyond the margin of the campsite. The biweekly measurements made on the camping unitB were the same as those on the simulated plots. Michael H. Legg Multiple regression prediction equations estimating the change in each measured parameter were developed for the two sampling zones within the camping units' boundaries. The equations explained from 47 to 72 percent of the variation in the dependent variables. The level of use was the most important and consistent variable, appearing in eleven of the twelve equations. In general, the conifer sites proved to be the most durable in all of the parameters monitored. A second equation, furnishing an index of detrimental change, was developed utilizing the site changes as actually measured on the camping units. This equation had an R 2 of 0.76 and is most useful in ranking the durability of established campsites to determine mainten­ ance priorities. Campsite Rocovery Abandonment alone does not seem to be an acceptable means to restore campsites to natural conditions in the Sylvania Recreation Area, when considered for short periods of time. Of the four campsites monitored during the two-year study period, only camping units which were previously very lightly used made satisfactory progress toward natural conditions. Three other previously heavily used sites showed only limited improvements, primarily due to a densely compacted soil surface. VITA Michael H Legg Candidate for the Degree of Doctor of Philosophy Final Examination: July 9, 1973 Guidance Committeei Drs. G. Schneider Murphy, and J. J. Kielbaso (Chairman), D. P. White, P. G. Dissertation t Site factors useful in predicting deterioration on forest campsites in Northern Michigan. Biographical Items: Born September 11, 1946, Jisper, Alabama Married, one son Education: Auburn University, B. S., 1969 Michigan State University, M. S., 1970 Michigan State University, Ph. D., 1973 Experience: September, 1972 to June, 1973 Research Assistant, Michigan State University September, 1969 to June, 1)71 Teaching Assistant, Michigan State University Summer 1970 Assistant Project Forester, Gulf States Paper Corp. Tuscaloosa, Alabama Summer 1969 Forest Technician, Holman Lumber Co. Northport, Alabama Organizations: Xi Sigma Pi Alpha Zeta Society of American Forest ii ACKNOWLEDGMENTS The author wishes to express his gratitude to Dr. Gary Schneider, who acted as both advisor and guidance committee chairman, for his valuable assistance during review and editing help. The author is also indebted to the other members of his guidance committee - Drs. D.P. White, P.G. Murphy, and especially to Dr. J.J. Kielbaso for their guid­ ance throughout the course of study. The assistance of the U. S. Forest Service is also acknowledged, especially Marsh Lefler, District Ranger of the Watersmeet District and Bill Bradshaw also of the Watersmeet office. Gratitude is extended to Michigan State University and the Nation­ al Wildlife Federation for the financial support which the author re­ ceived. Sincere appreciation is expressed to the author's parents for not only encouraging and supporting his educational endeavors, but for in­ stilling a deep appreciation of natural resources which helped to point the authors direction in choice of careers. Most of all the author is forever indebted to his wife Hettie for her untiring devotion. The successful conclusion of this study would not have been possible without her understanding, encouragement and sa­ crifice. Last, but certainly not least, the author extends appreciation to his son Christopher for the hours of entertainment and inspiration he has provided during the final stages of this study. itt TABLE OP CONTENTS ES2S. VITA....................................................... ii ACKNOWLEDGMENTS............................................ ill LIST OP TABLES............................................ vii LIST OP FIGURES........................................... ix I. INTRODUCTION....................................... 1 II. REVIEW OF LITERATURE............................... 3 Chapter III. IV. Soil compaction.................. 3 Bulk density.................................. 4 Noncapillary pore space............................ 5 Ground Cover........................................ 5 Simulation of Recreation Trampling................. 8 Measuring Campsite Deterioration................... 10 THE STUDY AREA..................................... 12 Location and Historic Background................... 12 Management.......................................... 14 Vegetation ........ '............................. 10 Climate............................................. 17 SIMULATED RECREATIONAL USE OF POTENTIAL FOREST CAMPSITES........................................... 10 Methods............................................. 10 Results............................................. 25 iv CHAPTER Effects of treatment on natural litter........... Effects of treatment on soil properties .... Usefulness of aerial photographs in campsite selection..................................... Utilisation of Trampling D a t a ........................ V. RECREATION IMPACT ON ESTABLISHED CAMPSITES.......... Methods............................................... Results............................................... Effects of recreation use on natural litter...... Effects of recreation use on soil properties...... Effects of recreation use on crown cover.......... Effects of recreation use on camping unit sise.... Sylvania Camper Survey............................. Ranking Camping Unit Durability...................... VI. RECOVERY OF ABANDONED CAMPSITES...................... Methods............................................... Results. VII. ........................................... DISCUSSION AND CONCLUSIONS............................ LITERATURE CITED...................................... APPENDICES............................................ A. Upland plants of Sylvania Recreation Area As Identified by Dr. Edward Voss, Curator, University of Michigan Herbarium.............. B. Comparison of Climatic Observations During The 1971 and 1972 Study Period With The 30 Year Average at Watersmeet, Michigan.............. Vi Page CHAPTER . Description of the Gogebic Soil Profile....... 95 • Cover Sheet of the Sylvania Camper Survey.;... 97 . Questionnaire Form Used in the Sylvania Camper survey............................ 98 LIST OF TABLES Table 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Page Description of study sites receiving simulated recreational u s e ........................................... 19 Average percent decrease in litter depth on simulated recreation sites at four levels of trampling over two seasons........................................ 27 Average percent decrease in oven dry weight of litter on simulated recreation sites at four levels of tramp­ ling over two seasons...................................... 32 Average percent decrease in percent litter cover on simulated recreation sites at four levels of trampling ‘over two seasons........................................... 32 Average percent decrease in noncapillary pore space on simulated recreation sites at four levels of trampling over two seasons......................................... 36 Average percent increase ITT'bulk density on simulated recreation sites at four levels of trampling over two seasons..................................................... 38 Average percent decrease in depth of the AO horizon on simulated recreation sites at four levels of trampling over two seasons........................................... 40 Prediction equations utilizing parameters obtainable from aerial photographs and with estimates of visitor u se......................................................... 43 Prediction of the relative rate of site deterioration on two similar campsites...................................... 45 Site characteristics of camping units used in estimating user impact................................................ 50 Average percent decrease in natural litter cover on established camping units at three levels of visitor use over two seasons .............. 56 vii viii Table 12. 13. 14. 15. 16. Page Average percent decrease In noncapillary pore space on established camping units at three levels of visitor use over two seasons......................................... 50 Average percent increase in bulk density on established camping units at three levels of visitor use over two seasons................................................... 60 Average percent decrease in depth of the AO horizon on established camping units at three levels of visitor use over two seasons.......................................... 62 Average decrease in percent crown cover on established camping units at three levels of visitor use over two seasons................................................... 65 Enlargement of camping units on established campsites at three levels of visitor use over twoseasons.............. 87 17. Criteria used by Sylvania visitors in campsite selection. 18. Criteria used by Sylvania visitors in choosing a parti­ cular camping unit................. ................. . 70 Prediction of the relative extent of site deterioration on two similar campsites using the predictive equation established from campsite measurements................... 72 Characteristics of Sylvania camping units chosen to monitor site recovery.................................... 74 Changes in percent litter cover, noncapillary pore space, and dry bulk density showing the recovery of closed camp­ ings units in one year................................... 75 19. 20. 21. 22. 23. Comparison of initial or control values and final measure­ ments o f the parameters monitored onSylvania............. A rating comparison of camping unit conditions by camp­ ers and predictive equations............................. 69 82 85 LIST OF FIGURES Figure Page 1. Map of Sylvania Recreation Are a............................ 13 2. Before (a) and after (b) preparation of an upland hardwood potential forest campsite for simulated recreational use by removal of the maple understory.......................... 21 Lawn roller modified by the addition of metal strips to simulate recreation u s e ........... 21 Calibration curve used to convert soil air permcameter readings to percent by volume of noncapillary pore space... 23 Decrease in depth of litter with timber type and level of simulated use over a two-year period........................ 26 The impact of trampling upon an upland hardwood plot having a moderate use level................................. a. Immediately following removal of the understory........ b. After one season of trampling.......... c. Prior to trampling the second season.................... d. After two consecutive seasons of trampling............. 30 30 30 30 30 Decrease in noncapillary pore space in the upper 7.5 cm soil layer in simulated use site with timber type and level of use over a two-year period........................ 35 Typical three-unit boat-access campsite on Sylvania Recreation A r e a ............................................. 48 A large conifer (a) and Bmall hardwood (b) camping unit on Sylvania......... 49 One complete replication of the design used to determine the impact of recreational use upon established campsites.. 52 Plot from sampling zone one (a) of a moderately used camp­ ing unit and a control plot (b) from the same unit......... 53 Comparison of a control plot (a) just off the margin of a previously heavily used campsite abandoned for three years and a used plot (b) from the center of the same unit. .... 76 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. ix CHAPTER Z INTRODUCTION Increases in population, leisure time, usable income, and mobiity have lead to increased demand upon forest land for recreation. While man's activities are only one input into the forest ecosystem, it has been recognized that use of forest areas for recreation by large numbers of people may lead to drastic changes in the system. The impact is most intense in areas where use is concentrated} thus campsites fluid trails usually show the greatest and most rapid changes. As more people venture outdoors for recreation it becomes increasingly important that sites be selected and managed so that they need not be Abandoned or rehabilitated due to excessive site deterioration. The determination of the type and extent of ecological changes that will take place on a forest site at different levels of use repre­ sents one of the greatest problems in estimating the ecological carry­ ing capacity of a recreational area. Knowledge of what site changes to anticipate will aid recreation area managers in: 1. Selecting sites with qualities that will inhibit deterimental changes} 2. Estimating what type of maintenance will be required to keep the sites at a specified quality level; 3. Permitting design features to be incorporated that take site characteristics into account; 1 2 4. Allowing for close surveillance t o be m a d e with sub­ sequent action taken following early indications of deterioration • This three par t study examined the extent and nature of the site changes that could b e expected to o c c u r on forest campsites with speci­ fic levels of use. The first phase u t i l i z e d an artificial trampler to simulate recreation use upon potential campsite areas. This permitted close monitoring of changes t h a t o c c u r un d e r known levels of trampling. The second phase was concerned with t h e amount of deterioration that takes place on established campsites a t several levels o f visitor use. The last part of the study followed t h e recovery rate t o natural condi­ tions of abandoned campsites. The objective of the study were, therefore: 1. To quantify t h e nature and extent of the changes which occur in the forest e c o s y s t e m due to varying levels of recreation use. 2. Utilize knowledge of these changes to develop predic­ tion equations which would allow the ranking of camp­ sites according to their durability. CHAPTER II REVIEW OF LITERATURE The effect of recreation upon forest sites may be divided into two broad categories! upon the vegetation. the impact upon soil and litter, and the impact Vegetative changes in the overstory often become immediately obvious, as exemplified by weakened and broken trees loca­ ted on heavily used campsites. More subtle, but equally, if not more, harmful are the changes which take place in the soil. With recreation­ al use commonly concentrated in a small area, such as a campsite, the effects of trampling by the users can cause drastic changes in soil properties. Appel (1950) noted that the changes were similar to those that take place on overgrazed pastures. "Hundreds of feet trample daily over the same area, breaking down the litter and humus into a dusty powder, which is either blown or washed away, or carried away on the legs and clothing of the campers. The remaining mineral soil is packed to 'stone-like hardness.'" Soil Compaction Soil compaction is often monitored to estimate the influence that campground use has had upon the site. Soil compaction is defined by Lull (1959) as "... the packing together of soil particles by in­ stantaneous forces exerted at the soil surface resulting in an increase in soil density through a decrease in pore space." Fore space reduction significantly decreases the water and air infiltration capacity and the movement of water, oxygen, and carbon dioxide in the soil is retarded. The reduced air movement may cause unfavorable oxygen-carbon dioxide ratios within the soil. Yelenosky (1964) reported that low O 2 /CO2 ratios induced by a compact soil surface could result in reduced growth and ultimate death of the vegetation on the site. Bulk Density Several studies have shown that significant increases in bulk density of the surface horizons of forest soils may occur following intensive recreation use. Lutz (1945) found significant increases in bulk density on two state park campgrounds in Connecticut. The densi­ ties of the surface 10 cm in used areas exceeded those at the 20-30 cm depths by 5 to 10 percent. Fore volume in the upper layer was also less than that found in the deeper layer. were found on unused sites. Neither of these conditions Most of the damage was confined to the surface 10 cm of the soil. Papamichos (1966), in a study done on 40-year-old campsites in Colorado, reported differences of 30 to 55 percont in bulk density of the surface 15 cm between light and heavy use areas. He also estab­ lished correlations between the change in bulk density and percent organic matter content, moisture, texture, and pH of the soil. Results indicate negative correlations between bulk density and percent organic matter, soil moisture, and percent silt-plus-clay in the soil. The correlation between bulk density and pH was positive, indicating that sites which were highly compacted lightly used sites. The pH were significantly less acid than found on heavy use sites averaged 6.0 while on lightly used sites it dropped to 5.6. Noncapillary Pore Space A further result of soil compaction is the reduction in number and sire of existing noncapillary pore spaces. Much of the research into this soil property has come from studies on range management, but the results are applicable to recreation areas. Farm woodlots and shelter-belts open to grazing have shown reductions of greater than SO percent in macropore space in the surface 15 cm of the e~>il (Steinbrenner, 1951; Read, 1956; Orr, 1960). Steinbrenner also noted that the permeability of core samples taken from the surface 5 cm from ungrazed woodlots was 3 to 245 times greater than that of grazed woodlots in Wisconsin. In a study on the Boundary Waters Canoe Area, Frissell (1964) found significant differences in infiltration rates between used and unused sites in 88 percent of the campsites examined. He noted that the reduced infiltration rate of the used sites was due primarily to the reduction of macropores. ThiB also resulted in increased over­ land flow of precipitation, and increased soil erosion. Ground Cover Perhaps the most obvious effect of recreation use of wildlands is the change in natural vegetation. A survey of 137 National Forest campgrounds and picnic areas in California identified several factors indicative of site deterioration (Magill and Nord, 1963). Reproduction was completely missing from over half of the campsites in the survey, and it was concluded that reproduction rarely survived more than a few years on any site. Most trees had been abused by campers; small trees, limbs, litter, and even wood barriers had been removed and used as firewood, bedding material, and tent poles. The vegetation on most 6 sites was considered weakened and succeptible to insect and disease attack. The litter cover on the soil was sparse or missing in 73 per­ cent of the campsites. The authors indicate that it was not only worn away by campers, but was sometimes raked away by overzealous mainten­ ance crews. Frissell (1964) reported that campsites on the Quetico-Superior Canoe Country have lost from 5' to 99 percent of their original ground cover, with the average amount being 85 percent. The amount of bare ground did not increase gradually with increasing use; over 80 percent of the ground cover was lost with light use and this percentage increa­ sed only slightly with heavy use. Litter and humus volume on the camp­ sites was reduced an average of 65 percent from that found on unused sites. The loss of organic matter was related to the loss of vegeta­ tion and to increased erosion from surface runoff. Shrubs and herbaceous vegetation are systematically removed by t campers "improving" the campsite, and a human browse line is evident around each site. In a recent study of campsites established in 1967, in the Boundary Waters Canoe Area, Merriam (1971) reported that between 1968 and 1970 all sites increased in si2 e. The greatest percent in­ crease was on spruce and aspen-birch sites where the initial size was limited by a dense understory. Heavily used spruce sites increased in size by an average of 21.7 percent and aspen-birch sites increased by 16.1 percent. On open pine sites with little understory, sites were initially larger and the percent increase in size was only 1.6 percent. Merriam also noted that once a site was expanded, campers continued to utilize the entire area. Ripley (1962) indicated that increasing amounts of detrimental change in ground cover were associated with an increase in crown closure* and that more fertile sites were better able to withstand use and main­ tain both understory and overstory vegetation. He suggested that can­ opy reduction by selective cutting would encourage understory growth and reduce soil losses due to erosion. Several studies have reported that vegetative conditions on campsites may improve with time. LaPage (1967) reported that after intial severe losses in ground cover* more durable species invaded* and by the third year coverage was increasing. However* species diversity diminished by over 50 percent on most sites* with narrow leaf species being more resistant to wear than broad leaf species. According to Magill (1970)* conditions on U. S. Forest Service campgrounds in California improved during the five-year period follow­ ing their establishment. Ground cover increased at every site* primar­ ily due to the invasion of more durable* shade-intolerant species. This was probably due to the open condition of the campsite following the loss of overstory and sapling size trees to mechanical damage. However* heavily used areas near the center of each campsite remained bare. Echelberger (1971) reported that there was no correlation bet­ ween level of recreation use and growth of the overstory vegetation. Overstory density* as measured by percent crown closure* increased dur­ ing the study by three percent. During the same period the total num­ ber of stems per campsite decreased by an average of 59 percent* re ­ sulting in a loss of lateral screening between sites. was primarily due to physical damage by campers. The reduction 8 Simulation of Recreation Trampling Artificial tramplers have been utilized in only a few studies in attempts to develop methods of rating durability of potential campsites. In most cases there has been no attempt to relate the deunage done by the trampler to damage done by recreationists. According to Cieslenski (1970), "If the only goal of the study is to develop a rating system for durability# it is not important whether or not the trampler does greater or less damage than that resulting from human use. The rating system can be developed by measuring the site changes that occur due to trampling and relating them by regression to various descriptive site measurements." However# if one does not attempt to relate actual and simulated use it is impossible to know whether the ranking system is truely valid for recreational use or whether the rating system is cor­ rect at all levels of use. Therefore# this author feels that the rela­ tionship between simulated and actual use must be shown before valid use can be made of the more precise estimates of deterioration avail­ able from simulated trampling. Wagar (1964) was the first to report the use of regressional analysis with artificial trampling data to predict the relative dura­ bility of vegetation for recreation. of the variability (R He was able to explain 65 percent ■* .65) in the percentage change in dry weight of vegetation on trampled plots in Southern Michigan by the following equa­ tion. Y x - 12.901 + .670b1 + .389b2 + ,255b3 # where Y^ » the percentage by which treatment reduced the weight of surviving vegetation b^ ** drops of the trampler during each treatment application t>2 ■ percentage of plot vegetation composed of species other than grasses and trailing raspberry - percentage of sunlight during the growing season. The results indicated that at high rates of use, damage levels off and great increases in use cause only small increases in damage. A similar study in Northern Utah indicated that it may be possi­ ble to predict the durability of vegetation of potential campsites from aerial photographs. Cieslinski and Wagar (1970) were able to explain 64 percent of the variability in durability of trampled plots by mea­ suring the slope of the plot, slope of the plot location, plot position on slope, aspect, and elevation from aerial photographs. Generally, the greater the number of passes over an area with a trampler the greater the compaction up to the point of maximum density. However, numerous studies have shown that intial passes do a greater percentage damage than subsequent passes (Steinbrenner, 1955, Hatchell, 1970). Steinbrenner (1955) showed water infiltration rate to be the soil characteristic most sensitive to compaction. He reported that a crawler tractor reduced infiltration in a stand of old growth timber from 68 cc min 1 to 13 cc min-1 after the first pass. The second pass reduced it to A cc min * and tho third reduced infiltration to 1 cc min"1 . This implies that most of the soil compaction on campsites occurs with the first users. In central Minnesota, Thorud and Frissell (1969) measured the recovery of an artifically trampled site over a four-year period. was compacted using a gasoline-powered trampler. Soil The treatment caused significant increases in bulk density at all depths up to 30.5 cm. Annual measurements taken over the four-year period showed a gradual decrease in bulk density. Linear projections indicated that the soil 10 would return to pre-treatment conditions in approximately six years. ' Measuring Campsite Deterioration Soil compaction has been quantified by measurements of bulk den­ sity, pore size distribution, permeability of soil cores, infiltration, and resistance to penetration. In studies primarily concerned with recreation, water infiltration and resistance to soil penetration have been most widely used. Prissell (1964) used water infiltration rates to quantify soil compaction in the Quetico-Superior Canoe Country. Hartesveldt (1962) also used infiltration to measure soil compaction, after he determined that gravimetric sampling was too time-consuming and caused excessive disturbance to the site. Magill (1970) found that penetrometers were useful, but did not give constant results. Both proctor and cone-type penetrometers showed increasing resistance to penetration as the soil dried. Papamichos (1966) utilized radioactive surface gauges to determine both soil moisture and bulk density. This author also found radioactive surface probes to be the most satisfac­ tory means of measuring bulk density. Changes in vegetation have been measured by clipping and drying above ground plant parts (Cieslinski, 1970; Wagar, 1964). Magill and Nord (1963) developed a pentallometer angle gauge for estimating verti­ cal and lateral vegetative screening between campsites. This device was used by LaPage (1967) and Echelberger (1971) in analyzing changes that occur in campground vegetation. Changes in tree height and diameter growth rates have been measured as a means of quantifying recreation impact (Magill, 1970; LaPage, 1967; Echelberger, 1971; Hartesveldt, 1962). However, in most cases, they have proven to be too variable to 11 be of value in recreation management. * Therefore, on Sylvania only the basal area and the crown density of the dominant vegetation was measur­ ed. CHAPTER III THE STUDY AREA Location Sylvania Recreation Area la located in the western end of the Upper Peninsula of Michigan in the Ottawa National Forest (Figure 1). It Is approximately 600 Km (360 miles) northwest of Chicago and 100 Km east of Ironwood, Michiganr in Gogebic County. The entire tract con­ sists of approximately 8500 ha and occupies the entire township of T44N and R40W. The southern boundary is the Wisconsin-Michigan border. About 1600 ha of the area is water surface for a water to land ratio of one to five (The University of Michigan, School of Natural Resources, 1965). The Sylvania tract lies near the center of a large upland plain and has relatively minor relief with elevations varying from 510 m to 558 m above sea level. The ground surface is composed of undulating moraines and short, unoriented ridges with numerous bogs and lakes where the intermorainal depressions drop below the water table. The ridges are low rarely exceeding 12 m above the plateau (Veatch, 1953). Historic Background The U.S. Forest Service acquired Sylvania in 1966. Public use of the area was first allowed in June, 1967, and the first over-night camping was in May, 1968. Prior to 1966, while in private ownership, Figure 1. Map of Sylvania Recreation Area. 14 Sylvania was closed to the public and patrolled by guards to prevent trespass. The area received only very slight use, and at the tine o£ governmental purchase was in relatively virgin condition (U. S. Forest Service, 1970). Management "The principal objective in management of Sylvania is to maintain the unique quality of its forest lands and waters, while providing op­ portunity for a variety of outdoor recreation opportunities." Forest Service, 1968). (U. S. To prevent over development, access by automo­ bile is limited to the periphery of the area. Access to most of the interior lakes is limited to waterways with portages or by hiking. The use of motorized transportation including outboard motors is prohibited within the area on all lakes and land completely controlled by the U. S. Forest Service. However, until the winter of 1973-1974, snowmobiles were allowed on marked trails when the snow base exceeded 15 cm. Campsites were established during the spring of 1968 and opened to the public later that same year. ed, each containing 3 camping units. camping units are open for use. Initially, 33 campsites were open­ Currently, only 83 of the 99 Some have been closed due to either hazards to campers, overuse, or nesting eagles. Use of the area has climbed steadily from 20,000 twelve hour visitor days in 1968 to 27,000 in 1971. However, use is not evenly distributed to all campsites. Between 1968 and 1971, use varied from 208 visitor days on the lightest used site to 5492 on the heaviest used site. Upon entering the Area, campers must register at the Sylvania Visitor Information Center on the north edge of the tract and choose 15 their campsite. or canoe. Access to most of the campsites is limited to hiking On Crooked Lake* however* where much of this study occurred, the use of outboard motors is allowed. A system of management zones was proposed in the 1968 U. S. Forest Service management plan. 1. The zones are as follows> The Pioneer zone is to be maintained in its natural state, and camping will be allowed only on designated sites. Currently the 83 campsites established in this zone are the only camping areas on Sylvania. 2. The Botanical zone includes a combination of bogs, muskeg, virgin forest, and other ecological communities of botanical significance. Management is directed to­ ward preservation of the natural environment for scientific study and education. Development is limit­ ed to a few trails and informative signs. 3. The Water and Travel Influence zone consists of areas of varying width along the lake shore, roads, and trails. It includes most of the lands intended for intensive development. Included is a half-mile long swimming beach and a hundred unit drive-in campground, soon to be opened. 4. The General Forest zone includes all land not contained in other zones. There are no recreation developments in this zone other than portages and trails. Manage­ ment is directed toward improving habitat for favored species of wildlife such as grouse and deer. Timber harvest is allowed in this zone (U. S. Forest Service, 1968). 16 To date, however, these management zones have not been clearly established on the ground and have only been used as general guides for management. Vegetation The vegetative type of Sylvania is climax northern hardwood. dominant tree species include: The (U. S. Forest Service Hardwood Acer saccharum March.......... Acer rubrum L .................. Betula alleqhaniensis Britton.. Betula papyrifera March....... Tilia americana V e n t .......... Populus tremuloidos Michx..... Conifer Tsuga canadensis (L.) Carr.... Picea glauca (Moench) Voss.... Picea mariana (Mill.) B.S.P.... ....Black Spruce Abies balsamea (L.) Mill...... Pinus resinosa Ait............. Pinus strobus L ................ Pinus banks iana Lamb.......... Thuja occidentalis L ............... Northern White Cedar Most of the upland areas are covered by a mature stand of mixed sugar maple, yellow birch, and hemlock. red and white pine on upland areas. There are scattered stands of The lakeshores and low areas are predominantly white cedar and balsam fir. Young sugar maples from . 5 m to 2 m tall cover large areas of the understory with densities as high as 25 stems m (100,000 stem/acre). For a more complete list of plants found on Sylvania, see Appendix A. Climate The climate of the western end of the Upper Peninsula is marked by cold winters and mild summers. During the 30 year period (1940-1969), the average annual temperature was 4.3°C (39.8°F.). The average tem­ perature for July is 18.4°C (65°F.) while for January it is -11.0°C (12.7°F.) (Appendix B ) . The average length of the growing season is 61 days. Precipitation is uniformly distributed throughout the year with May through September receiving 66 percent of the 70.1 cm total, mostly in the form of thunderstorms. The average annual snowfall of 237 cm accounts for approximately one-third of the total precipitation (U. S. Dept, of Commerce, 1971). CHAPTER IV SIMULATED RECREATIONAL USE OF POTENTIAL FOREST CAMPSITES Most impact studies on forest recreation areas have been con­ ducted on established campsites that often have been used for several years. The exact amount of use these sites have received is usually difficult to determine accurately. By simulating use with an artifi­ cial trampling device it is possible to quantify the forces which are applied to the site. Simulation also permits accurate measurement of site conditions prior to any disturbance and continual monitoring dur­ ing the trampling period. The specific objectives of this portion of the study were* (1) to determine the type and extent of the changes which occur on potenti­ al forest recreation sites under three levels of use, (2) to utilize these measurements establishing a predictive index of detrimental change. METHODS Study Area Four study sites , representative of the campsites in Sylvania t were selected in the Crooked Lake area. 2 level areas of at least 36 m . The sites were essentially A description of each site is given in Table 1. 18 19 Soils were of the fine sandy loam phase of the Gogebic soil ser­ ies (Appendix C ) . The series consists of moderately coarse to medium textured typic fragiorthods with a distinct fragipan on noncalcareous reddish sandy loam to loam. The Gogebic series is confined primarily to the moraines and shorelines of the Lake superior regidn in the west­ ern part of the Upper Peninsula of Michigan and northern Wisconsin. It is commonly associated with the well to moderately well drained Iron River, Wakefield, Marenesco, and Ahmeek soil series (National Coopera­ tive Soil Survey, 1958). Table 1. Description of study sites receiving simulated recreational use. Site Number 1 Characteristics 2 3 4 Elevation Level Upper®* Lower*3* Lower Upper Timber type Hardwood Hardwood Conifer Conifer Understory Maple Maple None Club moss Crown density {%) 82.6 85.1 86.1 80.3 2 Basal area (m ) 13.0 8.5 12.1 9.0 Initial depth of litter (cm) 5.5 6.4 6.7 7,0 Initial depth to A2 horizon (cm) 5.0 6.3 11.3 9.5 a) Upper level sites were at least 5 m above the water table. b] Lower level sites were less than 3 m above the water table. 20 Experimental Design The response o£ forest sites to trampling was studied using a completely nested design. Each study site was divided into 16, one meter square treatment plots with a 0.5 m strip separating the plots. Prior to treatment the plots were cleared of large limbs and advanced reproduction (Figure 2). Trampling was simulated over a two-year period by means of a mechanical trampling device which was a modified, water-filled lawn roller (Figure 3). The addition of alternating 7.5 cm and 12.5 cm track-like metal strips at intervals around the drum more closely appro­ ximated the type of wear attributable to human use than the unmodified roller. Although the ground surface configuration caused the pressure to vary, the drum applied an average pressure of 350 g cm -2 (5 lbs in”2 ) , approximately the pressure exerted by a man walking. The treatment plots were randomly divided into four groupst controls, lightly trampled, moderately trampled, and heavily trampled. The lightly trampled plots received six passes of the trampling device for a total of 2.1 Kg cm -2 (30 lb in -2 ) per week. trampled plots received twelve passes or 4.2 Kg cm The moderately —2 (60 lb in —2 ) per week whereas the heavily trampled plots received eighteen passes per week or 6.3 Kg cm —2 (90 lb in —2 ). Control plots received no trampling. The treatments were applied in 1971 and 1972, during June 20 through September 15, the approximate duration of the visitor season for this area. to constant b^ ■ elevation b2 " timber type b3 • level of simulated use b4 b dry bulk density prior to trampling. 34 Analysis of the independent variables included in the equation indicate that: 1. Upper level hardwood sites were most succeptible to decreases in litter cover and lower level conifer sites were least succeptible; 2. More mineral soil was exposed by higher use levels; 3* The more compacted the soil, the greater the change in percent litter cover. Effect of Treatment on Soil Properties Noncapillary Pore Space The simulation of recreation use caused significant decreases in noncapillary pore space on all trampled plots (Table 5). also significant differences between treatments. There were On heavily trampled plots, macropore space was reduced to less than 5 percent on the total soil volume. Light and moderate levels of trampling resulted in less destruction. However, at least SO percent of the noncapillary pore space was lost on all trampled plots. Conifer sites proved to be signi­ ficantly more succeptible to destruction of macropores by trampling than hardwood sites when averaged over elevation (Figure 7). The resistance of the hardwood sites to change in pore space was partially due to extensive earthworm activity. Earthworm tunneling promotes the generation of macropores following reduction by trampling. During the study period, few earthworms were observed on conifer sites, but they were abundant on hardwood sites. The lack of earthworms on conifer sites is explained by the high acidity (pH 4.2-5.0) and lack of appropriate food found there. 35 Figure 7. Decrease in noncapillary pore space in the upper 7.5 an soil layer on simulated recreation sites , with timber type and level of use over a two year period. 100 - 90 " Heavy Moderate Light LEVEL OF USE 1 I 36 Table 5. Average percent decrease in noncapillary pore space on simulated recreation sites at four levels of trampling over two seasons. Levels of Use*^ Site Heavy Moderate Light Control Upper level hardwood -90 -69 -52 -19 Lower level hardwood -90 -78 -55 - 1 Upper level conifer -85 -74 -56 +16 Lower level conifer -81 -61 -54 + 6 ^ A n y difference among these values greater than 8 is significant at the 5% level. Actual measurements of noncapillary pore space indicate that on undisturbed plots about 30 ± 6% of the soil volume is macropore space. Overwinter recovery averaged approximately 70 percent of the initial macropore space and was not significantly different at each level of trampling. However, following the second trampling period of the 1972 season, macropore space had been reduced to approximately the same level as the end of the 1971 season. The total change in noncapillary pore space can be estimated by the following equation: Y. - .648 + 1.543b, - .290b„ + ,023b, - ,085b. - ,654b_, where: Y4 » change in noncapillary pore space 0.648 ■ constant b^ ■ timber type b2 " level of simulated use 37 ■ initial noncapillary pore apace b4 ■ initial depth of the AO horizon b5 ■ initial bulk density* The coefficient of determination was 0.91 and the standard error was 0.12. The independent variables included in this equation indicate that with uset 1. The percent decrease in noncapillary pore space is great­ er on conifer sites than hardwood) 2. The percent decrease in noncapillary pore space increases as the level of trampling increases) 3. The higher the initial noncapillary pore space, the greater the percent decrease) 4. As the depth of the AO soil horizon increases, the per­ cent in noncapillary pore space decreases) 5. As the initial oven-dry bulk density increases the change in noncapillary pore space decreases. Dry Bulk Density The average bulk density prior to trampling on conifer sites was 1.3B g cm -3 difference. and on hardwood sites it was 1.43 g cm —3 , a nonsignificant However, following trampling there was a significant difference in the percent change in bulk density between timber types (Table 6). Differences between levels of trampling were not significant in all cases. On conifer sites only heavy trampled plots showed a signi­ ficant increase in bulk density while on hardwood sites all levels of trampling produced significant changes. 38 Table 6. Average percent increase in bulk density on simulated recreation sites at four levels of trampling over two seasons. Levels of U s e ^ Site Heavy Moderate Light Control Upper level hardwood 39 43 18 5 Lower level hardwood 29 29 23 8 Upper level conifer 16 5 1 2 Lower level conifer 2 -10 - 8 -10 ^ A n y difference among these values greater than 6 is significant at the 5% level. After two seasons of simulated use the bulk density on conifer -3 sites had increased to as much as 1:65 g cm on heavily used plots while hardwood sites increased up to 1.92 g cm"3 . The change in den­ sity followed a linear trend over the two year trampling period at each level of use. There was no significant difference# however# in over­ winter reduction in density. The multiple regression equation established to predict changes in bulk density explained 78 percent of the variability with a standard error of 0.11 and ist Y_ » -1.942 + ,203b. + .270b- + .058b + ,049b + ,764b , where t 5 1 2 3 4 5 Y_ ■ change in bulk density d -1.942 - constant bj - elevation of the site ■ timber type 39 b ■ level of simulated use ■ initial depth of litter cover b5 ■ initial dry bulk density. This indicates! 1. Lower level sites were more resistant to changes in bulk density than upper level sites; 2. Increases in bulk density are greater on hardwood sites than conifer sites; 3. Higher levels of simulated use cause greater increases in bulk density; 4. Sites with deep litter are most resistant to increase in bulk density; 5. Sites with high initial densities increase less than low density sites* Depth of the ftp Horizon The depth of the AO horizon of the soil is a measure of the amount of erosion that has occurred as well as compaction of the sur­ face layers. There were significant differences in the percent change in depth of the AO horizon between timber types and elevation at all levels of simulated use (Table 7). Initial depths of the AO averaged S cm on hardwood sites and 10 cm on conifer sites. Measurements after two seasons of simulated use indicate a mean depth of 3 cm on hardwood sites and 8 cm on conifer sites. Lower level sites had smaller decreases in depth of the AO than upper level sites. The average decrease at heavy levels of simulated 40 uso on upper level sites was 72 percent while sites nearer lake level decreased by 50 percent. Table 7. Average percent decrease in depth of the AO horizon on simulated recreation sites at four levels of trampling over two seasons. bevels of U s e ^ Site Heavy Moderate Light Control Upper level hardwood 88 81 31 5 Lower level hardwood 74 50 17 0 Upper level conifer 56 39 18 0 Lower level conifer 26 12 4 2 ^ A n y difference among these values greater than 11 is significant at the 5% level. The extent of change in depth of the AO soil horizon is estimated by the following equation which has an R 2 of .81 and a standard error of 0.14. Y, » 1.142 - ,015b. - .083b - ,203b 6 1 2 . 3 - ,105b. - ,804b + ,192b , where: 4 5 6 Yg ■ decrease in depth of the AO soil horizon 1.142 ■ constant •* elevation level of the site b2 - timber type b^ « level of simulated use bg * initial depth of litter cover b5 ■ initial depth of the AO horizon 41 b 6 ■ initial dry bulk density. The parameters included in the equation reveal that with simulated use: 1. Lower level sites are more durable than upper level sites; 2. Conifer sites are less subject to decrease than hardwood sites; 3. High levels of use cause greater changes than low levels of use; 4. Sites with deep litter cover change less than sites with thin litter layers; 5. The deeper the initial depth of the AO the smaller the change; 6. Sites with low initial bulk densities are more durable than sites with high initial densities. Usefulness of Aerial Photographs in Campsite Selection Aerial photographs have proven useful in successfully selecting potential campsite areas {Lindsay, 1969). It would be advantageous to be able to rank the durability of such sites with criteria measured from the same photographs. Table 8 illustrates the equations that pre­ dict change in several ecological parameters using elevation and timber type measurements which cure obtainable from aerial photographs, and visitor use data from camper registration. The percent of variability accounted for was lower in each case than from those equations gener­ ated using only ground measurement. However, in the majority of cases, the prediction equations dropped only a few percentage points. This seems to indicate that there is a good possibility of ranking sites 42 according to durability from aerial photos using ground observations for verification of results. Elevation level and timber type were chosen as independent variables because of photo identification simpli­ city. By adding such parameters as slope, aspect, and stand density to the equation, even higher coefficients of determination could probably be achieved. Utilization of Trampling Data The equations developed in the previous pages each estimate the percent change which occurs in only one parameter. While any one para­ meter may be useful as an indicator of one type of site deterioration, it is necessary to incorporate several parameters into the same equation in order to successfully compare several sites with a high level of predictability. Using the coded levels of trampling as the dependent variable and the actual changes in the six ecological parameters which were measured on the simulated use plots, the following equation was developed using multiple regression analysis. Since it is the level of trampling which is being predicted this actually then becomes an index of detrimental changes in the independent variables. The equation is most useful on sites which are ecologically similar, or on sites which receive approximately the same levels of use. The equation accounts for 89 percent of the variability with a standard error of 0.04. This equation is: Y? - 6.825 - 1.070b^ - 0.890b2 - 3.808b3 + 0.396b4 - .636b5 - 2.475b6 , where: Y? m index of detrimental change 6.825 » constant Table 8. Prediction equations utilizing parameters obtainable from aerial photographs and with estimates of visitor use. INDEPENDENT VARIABLES Constant Change in noncapillary pore space to Elevation Timber type Level of Use b. b_ b, Correlation coefficient (Std. Error) 1.363 -.017b, -.075b, -.279b. .79 (.17) .722 -.121b, -,084b- -.104b, .72 (.09) 1.195 -.001b, -,071b. -.098b, (.04) Change in litter dry weight 1.470 -.043b, -.312b. -.073b. .67 (.13) Change in depth of AO horizon 1.393 -.153b -.220b. -.206b. .77 (.15) Change in dry bulk density -.674 4.168b, 4.237b. 4.053b. Change in percent litter cover a % g g § A u o Change in litter depth .88 Y3 = .65 (.12) 44 - change in noncapillary pore space bg “ change in percent litter cover bj - change in depth of natural litter cover b^ « change in oven dry weight of .09m2 of litter b5 ■ change in depth of the AO soil horizon b6 - change in dry .bulk density. Resulting Yy values that are close to 1.0 would indicate great changes in the measured parameters and values approaching 4.0 would indicate slight changes. Actual data and sample calculations from two ecologi­ cally similar sites are shown in Table 9. Table 9. Prediction of the relative rate of site deterioration on two similar campsites. Factor Site 1 Site 2 - .862 .703 b 2 ■ Change in percent litter cover ■ .436 .090 b3 ■ Change in depth of litter cover ■ .973 .783 b4 ■ Change in oven dry weight of litter ■ .635 .297 b5 ■* Change in depth of AO horizon ■ .529 .096 bg ■ Change in dry bulk density ■ .026 .084 ■ Change in noncapillary pore space Ychange " 6 *825 “ 1.070bx - .890b2 - 3.808b3 + .390b4 - .636bg - 2.475b( where ■ index of detrimental change in measured parameters Y site 1 “ 6.825 - .922 - .388 - 3.219 + .251 - .336 - .064 - 1.645 where Ys^te ^ • index of detrimental changes on site 1 Yslte 2 “ 6 *825 ” *752 " •08° " 2 *892 + *117 " .061 - .207 - 2.860 where YS£te 2 * index of detrimental changes on site 2 and where a relative detrimental change ofi moderate, 3 ■ little, and 4 - no change. 1 • great, 2 - i CHAPTER V RECREATION IMPACT ON ESTABLISHED CAMPSITES Many studies have measured the extent of ecological changes taking place on recreation areas with some unknown level of visitor use. However, such data is lacking on newly established areas having good records of levels of visitor use. Neither has there been an attempt made to relate actual visitor use to simulated use on comparable areas. Specific management recommendations can only be made when the relation­ ship between level of visitor use and site deterioration is understood. The specific objectives of this portion of the study were: (1) to determine the type and extent of ecological changes which take place on established campsites under various levels of vistor use, (2) to develop predictive equations which will estimate these changes at each level of use, and (3) to develop a system for estimating the de­ terioration of camping units based upon an index of detrimental change. Methods Study Area This portion of the study was located on water access camp­ sites in the Crooked Lake region of the Sylvania Recreation Area. The majority of users are overnight campers, although there is some use of the unoccupied units as picnic areas by day users. All overnight camp­ ers are required to register and select their campsite at the Sylvania Information Center. 46 47 Each designated campsite consists of three separate camping units (Figure 8). They are located in the pioneer zone of Sylvania, with minimum distances of 0.5 Km between sites. The management plan specifies that units within each numbered site would be at least 30 m apart and 30 m from the water's edge (U. S. Forest Service, I960). An added prerequisite for each site, although not specifically men­ tioned in the management plan, is that it be hidden from direct lake view. The campsites were established during the spring of 1968. large limbs and underbrush were removed from the campsite. at each unit include a table, fire ring, and tent pad. All Facilities In addition, for each three unit site there is a pit toilet and boat landing. Fi­ gure 9 shows two typical camping units on Sylvania. Six of the seven canpsites on Crooked lake were included in this study. Campsite Fox was excluded because it was atypical of the camp­ sites, being located on a high ridge that was difficult to reach from the water. A description of the twelve units selected from the six canpsites in the study is given in Table 10. Six units each of timber types predominantly hardwood and conifer were chosen for the study. The units were further stratified into average seasonal use levels. A visitor day is here defined as one person occupying a campsite for one 12-hour period. As determined from camper registration, these werei light (100 to 150 visitor days), moderate (200 to 250 visitor days), and heavy (300 to 350 visitor days). Experimental Design The impact of recreation use upon campsites was investigated using a completely nested factorial design. On each camping unit the Figure 8. Typical three-unit boat-access campsite on Sylvania Recreation Area. Fire ring 30 m Minimum LAKE (a) Figure 9. A large conifer (a) and small hardwood (b) camping unit on Sylvania. 50 Table 10. site charaoterieitics of camping units used in estimating user impact. Site Use Timber level type Initial Basal radius (m) area (in2 ) Average crown closure <%> Condition as rated by users Porcupine Nl Light Hardwood 4.0 11.8 94 Good Porcupine N2 Moderate Hardwood 6.9 6.4 63 Good Squirrel Nl Heavy Hardwood 5.9 13.2 92 Fair Squirrel N2 Heavy Conifer 7.1 9.1 81 Good Squirrel N3 Heavy Conifer 6.1 9.5 89 Good Badger Wl Heavy Hardwood 4.7 9.5 86 Good Fisher W3 Moderate Hardwood 5.1 7.7 94 Fair Mink N2 Light Conifer 6.1 12.3 89 Fair Mick N3 Light Hardwood 4.7 9.5 97 Good Chipmunk Nl Moderate Conifer 5.8 14.1 90 Good Chipmunk N2 Moderate Conifer 9.8 8.6 85 Chipmunk N3 Light Conifer 4.9 12.2 88 Very Good Very Good 51 center o£ use was visually determined. From the center of each unit three transects# each with three one-meter plots were laid out 10). (Figure Two of the plots on each transect were located within the bounds of the campsite# and the third was off the campsite and served as a control plot. Camping units were further subdivided into three concen­ tric circular sampling zones# with three plots being located in each ring. Zone one was the high use area near the center of the campsite. Zone two was the intermediate use zone near the margin of the campsite. Zone three was located off the campsite as a control area. Figure 11 illustrates a plot from zone one on a moderately used unit and a con­ trol plot from the same unit. Measurements The measurements made on campsite plots were the same as those made upon the simulated recreation use sites but oven dry weight of litter and depth of litter were excluded. In addition# the distance from the center to the edge of the campsite was measured in four places to determine if the campsite expanded in size. All measurements were made four times at two week intervals during the period of June 20 to September 15. Analysis of Data The total change in bulk density# natural litter cover# percent noncapillary pore space# depth of the AO horizon, and percent crown cover was calculated for the two seasons of visitor use. These para­ meters were then utilized as dependent variables in multiple regressional analysis to estimate the type and extent of change due to different 52 Figure 10. One cong>lete replication o£ the design used to determine the impact of recreational use upon established campsites. Control Control Sampling Zone 2 Sampling Zone 1 No Scale Control □ (a) Figure 11. Plot from sampling zone one (a) of a moderately used camping unit and a control plot (b) from the same unit. (b) 54 levels o£ camping unit use. Independent variables which were not sig­ nificant at the 10 percent level were excluded from the equation. Equ­ ations were developed for each sampling zone since there were signifi­ cant differences between the zones. As on the simulated sites, some of the independent variables used in the development of the equations were coded for ease in calculation. Timber type was classified as either do­ minantly hardwood (*1) or conifer («2). Levels of use weret 300-350 visitor days (-1), 200-250 visitor days («2), and 100-150 visitor days (■3). Wide ranges in values required the use of arcsin transformations in analyzing the percent litter cover on plots (Sokal, 1969). Sylvania Camper Survey To assess the ecological impact of camping it is necessary to determine visitor criteria for site selection, and to determine the amount of time that they actually spent on the camping unit (tent pad). Since overnight campers had to register and select their campsites upon entering Sylvania, campsites were selected on some basis other than physical appearance for all but return visitors. It was important to account for the vast differences in use levels between campsites. In order to establish visitor criteria for campsite selection, and determine the amount of time each camper actually spent on the camping unit, a voluntary, self-administered questionnaire was distri­ buted from the Visitor Information Center (Appendix D ) . in the ques­ tionnaire each participant was asked to rate the condition of his camp­ site as either very good, good, fair, or deteriorated. Return boxes and additional questionnaires were placed at the major boat landings and portages on the area. returned. Of the 700 questionnaires issued, 300 were 1 55 Results Effects of Recreation Use on Natural Litter There was a significant effect of sampling zone upon the percent litter cover present at all levels of use in both timber types (Table 11). Near the center of use, the campsite*8 annual leaf litter was rapidly destroyed. Litter coyer at the beginning of the season varied from 100 percent on lightly used units to 40 percent on heavily used units. On heavily used camping units, over 90 percent of the litter was removed by the end of the visitor season regardless of timber type. However, on moderately and lightly used camping units, the conifer sites maintained greater litter cover than hardwood sites in the center of use zone. In zone two the percent decrease was significantly less than in zone one. There were no significant differences between timber type or level of use within this zone. Since there were significant differences between the sampling zones, prediction equations were developed for both zones within the camping unit. The equation for zone 1 had a R 2 of .65 and a standard error of 0.18. Yg «■ .9857 - . 1 8 8 ^ - .006b2 + .459b3 + .003b4 - ,177bs , wheret Yg « decrease in percent litter in the peak use zone of a camping unit. 0.9857 « constant b^ « timber typo b2 “ initial percent crown cover b 3 > initial dry bulk density b^ a initial percent litter cover b£ a level of use. 56 Table 11. Average percent decrease in natural litter cover on estab­ lished camping units at three levels of visitor use over two seasons. Sampling Z o n e ^ Use / Site Zone 1 Zone 2 Control Heavy / Conifer 96 29 0 Heavy / Hardwood 92 31 0 Moderate / Conifer 70 41 0 Moderate / Hardwood 95 35 0 Light / Conifer 65 31 0 Light / Hardwood 72 39 0 ^ A n y difference among these values greater than 20 is significant at the 5% level. The selection of these independent variables means that with 1. u b s : Conifer sites show smaller decreases in percent litter cover than hardwood sites; 2. Sites with dense canopies will decrease less in percent litter cover than sites with open crowns; 3. As bulk density increases the change in litter cover decreases; 4. The higher the initial percent litter cover, the smaller the decrease; 5. Higher levels of use produce greater decreases in litter cover. i 57 The fact that there were no significant differences between use levels in zone 2 is reflected in the equation which was developed for that portion of the site. Level of use was not included in the equation at the 10 percent level of significance. The equation for zone 2 explains 57 percent of the variability in percent litter cover with a standard error of 0.16. Yg » .432 + . 0 0 7 ^ - .009b2 + .321b3 , where: Yq ® decrease in percent litter cover in the light use zone of the camping unit 0.432 » constant b^ ** initial percent litter cover b 2 ** initial percent crown cover b 3 ** initial dry bulk density. Just as for zone l r the same type of relationship exists here between percent litter cover and the independent variables. Effects of Recreation Use on Soil Properties Noncapillary Pore space The results of measurements of noncapillary pore space are similar to those for litter cover (Table 12). However, differences between the two zones were not as clearly defined. Initial mean non- capillary pore space for conifer sites was 11.6 percent while on hard­ wood sites it was 12.4 percent. Conifer sites showed no significant differences between zones 1 anl 2. This was due to the low percent change that occurred in zone 1 on the moderately and lightly used coni­ ferous units. Significant differences between timber type occurred at the moderate and light use levels in zone 1 and at the moderate and sa heavy levels o£ use in zone 2. Table 12. Average percent decrease in noncapillary pore space on established camping units at three levels o£ visitor use over two seasons. Sampling Z o n e ^ Zone 1 Zone 2 Control Heavy / Conifer 68 60 8 Heavy / Hardwood 64 45 -521 Moderate / Conifer 16 19 1 Moderate / Hardwood 41 58 4 Light / Conifer 13 17 2 Light / Hardwood 36 18 -3 U b b / Site 2] ^Any difference among these values greater than 13 is significant at the 5% level. JNegative numbers indicate an increase in noncapillary pore space. The prediction equation for zone 1 explained 47 percent of the variability with a standard error of 0.35. Y1(J » 1.039 - .345b^ + .043b2 - .007b3 , where: Y ■ decrease in noncapillary pore space in zone 1 1.039 a constant bj^ ® use level b » initial noncapillary pore space b^ a initial percent crown cover. 59 These indicate that with uset 1. High use levels produce greater decreases in noncapillary 9 \ pore spacei 2. The larger the initial percent by volume of noncapillary pore space the greater the decrease} 3. Sites with high initial crown covers decrease less than open sites. The prediction equation for rone 2 explained 52 percent of the variability and the standard error was 0.38. Y 11 * *667 “ •009b1 “ *285b2 + *043b 3 » wherei Y^ * percent decrease in noncapillary pore space in zone 2 0.667 ■ constant - initial percent litter cover b2 ■ use level b^ ■ initial noncapillary pore space. The only change from the prediction equation for zone 1 is the deletion of percent crown cover and the addition of percent litter cover. This indicates that camping units which have high initial percent litter cover are more resistant to changes in noncapillary pore space. The role of percent litter cover in the equation for zone 2 and percent crown cover in zone 1 are important even though they are multiplied by a small number in the equation. Since their values may vary from zero to 100 percentf they can bring about large variations in the predicted change in noncapillary pore space. i 60 Bulk Penalty Changes in bulk density due to visitor use were significantly different between sampling zones. However, zone 2 showed greater in­ creases in bulk density than zone 1 (Table 13). This was due to the higher initial bulk densities encountered in zone 1 of each camping unit. Initial bulk densities in zone 1 averaged 1.77 g cm"3 and zone 2 averaged 1.63 g cm- 3 . Table 13. Average percent increase in bulk density on established camping units at three levels of visitor use over two seasons. Sampling Z o n e ^ Use / Site Zone 1 Zono 2 Control Heavy / Conifer 22 33 25 Heavy / Hardwood 16 33 15 Moderate / Conifer 17 42 25 Moderate / Hardwood 16 28 12 Light / Conifer 13 22 12 Light / Hardwood IS 30 15 ^ A n y difference among these values greater than 7 is significant at the 5% level. The area near the center of use is apparently approaching its maximum density due to compaction by recreation users. Since the outer zone is presently being compacted at a faster rate, it will probably become as compacted as zone 1. 61 The prediction equations for zones 1 and 2 contain the same in­ dependent variables* The coefficient of determination for the equation for zone 1 was 0.64 and the standard error was 0.08. Y,, - .9037 - .001b, - .043b - .421b,, where: 12 1 2 3 Y ^2 “ the increase in dry bulk density within sampling zone 1 0.9037 *■ constant b^ ■ initial percent litter cover b 2 • use lQvel b^ « initial dry bulk density. The independent variables for the equation generated for predic­ ting the change in bulk density of zone two on the camping units are the same as those for zone 1. The equation with an R 2 of 0.72 and a standard error of 0.08 is as follows: Y13 - 1.030 - .002bj- .033b2 - .482b3 . The independent variables included indicate that high use levels coup­ led with low percent litter cover and low initial dry bulk density will lead to the greatest changes in dry bulk density. Depth of the AO Horizon Soil depths of the AO horizon indicate that with recreation use, camping units tend to become saucer-like depressions. The percent de­ crease in depth of the A0 horizon was significantly greater in the heavy use areas of the camping unit and gradually diminished toward the margins (Table 34). There were also significant differences between use levels and between timber types. Conifer sites proved somewhat more resistant to change than hardwood sites at moderate and light levels of use, but at heavy levels of use there were no significant 62 differences* On all of the heavily used camping units and the moder­ ately used hardwood units, there was only one plot within sampling zone 1 in which the A2 horizon was not at the surface. On the more lightly used plots near the camping unit margins, the depths ranged from 0 cm to 6 cm. Only the lightly used conifer units suffered less than a 50 percent decrease in depth of the AO horizon in 2 one 1. Their depths ranged from 2 cm to 5 cm. Table 14. Average percent decrease in depth of the AO horizon on established camping units at three levels of visitor use over two seasons. Sampling Zone*^ Use / Site Zone 2 Control 100 67 0 Heavy / Hardwood 98 68 0 Moderate / Conifer 61 55 0 100 74 0 Light / Conifer 49 18 0 Light / Hardwood 60 51 0 Heavy / Conifer Moderate / Hardwood Zone 1 ^ A n y difference among these values greater than 11 is significant at the 5% level. The multiple regression equation predicting the change in depth of the AO horizon in zone 1 of each camping unit explains 71 percent of the variation with a standard error of 0.20. 63 Y 14 - .4148 + ,030b^ - .126b2 + .363b3 - ,003b4 - .101b5 , where* Y14 ■ decrease In depth of the AO horizon in sampling zone 1 0.4148.- constant bi ■ initial radius of the camping unit b 2 “ initial depth of the AO horizon b 3 ■ initial dry bulk density b 4 ■ initial percent litter cover bg ■ use level. These variables show that the decrease in depth of the AO horizon in zone It 1. Increases as the size of the camping unit increases* 2. Is smaller on sites where the initial depth of the AO is greater; 3. Is less on sites with high initial bulk densities; 4. Is greater on sites where the percent litter cover is low; 5. Is greatest at high levels of use. For sampling zone 2 the prediction equation explains 49 percent of the variation and the standard error was 0.26. Y15 - .6007 - . 3 5 5 ^ - ,139b2 + .039b3 , where: Y^g ■ decrease in depth of the AO horizon in zone 2 of the camping units 0.6007 ** constant ■ timber type b2 “ use level b3 « initial radius of the camping unit. 64 This means that with use: 1. Hardwood sites suffer greater decreases in depth of the AO than conifer sites; 2. High levels of use produce greater decreases than lower levels; 3. The larger the initial radius of the campsite the greater the decrease. Effects of Recreation Use on Crown Cover The percent change in crown cover over the camping unit was de­ termined by comparing plots within the camping units to controls. The differences could be due to camper impact or to clearing which took place during campsite construction, or both. There was significantly less crown cover over the central portion of each tent pad than near the margins crown (Table cover at all 15). Conifer sites underwent greater decreases in levelsof use than hardwood sites. The changes which occurred in zone 2 were consistently smaller, but showed no rela­ tionship between timber type and level of use. This may be due to the encroachment of tree crowns from off the unit and into the space above zone 2. The equation for the change in percent crown cover over zone 1 explains 60 percent of the variation with a standard error of 0.12, and is; Y16 ■ -.662 + 1.40b^ - .091b2 + .011b3 + .3B7b4 , wheret *16 “ Percent decrease in crown cover over zone 1 -0.662 a constant b^ *■ change in camping unit radius as measured over a twoyear period I 65 bj " use level of the camping unit b3 » average shade cover over the control plots b^ ■ initial percent litter cover on zone 1* These independent variables indicate that with uset 1. The greater the growth in camping unit radius the greater the decrease in percent crown cover; 2. The higher the level of use the greater the decrease in percent crown cover; 3. The greater the crown cover over the control plots the greater the decrease in percent crown cover; 4. The higher the initial litter cover the greater the de­ crease. Table 15. Average decrease in percent crown cover on established camp­ ing units at three levels of visitor use over two seasons. Sampling Z o n e ^ Zone 1 Zone 2 Control Heavy / Conifer 27 9 0 Heavy / Hardwood 14 4 0 Moderate / Conifer 25 5 0 Moderate / Hardwood 13 13 0 Light / Conifer 14 8 0 6 1 0 Use / Site Light / Hardwood ^ A n y difference among these values greater than 6 is significant at the 5% level. 66 The equation for zone 2 includes the same variables except that b4 becomes the initial litter cover on zone 2. It has a R of 0.49 with a standard error of 0.09. Y j j - -.696 + .802b^ - .057b2 + .009b3 + .172b4 . wherei * percent decrease in crown cover over zone 2. Effects of Recreation Use on Camping Unit Size When the study began. individual camping unit radii varied in size from 4.05 m to 9.75. m and hardwood sites 5.2 m. Conifer sites had an average radius of 6.6 During the two-year study period the size of the conifer units increased in radius by B percent and hardwood units increased by 14 percent (Table 16). However.due to the larger initial size of the conifer units, both types increased about the same number of square meters. trend. Increase in size did not follow a continuous linear Instead, the units grew in spurts as an especially ambitious camper would "improve" his campsite by clearing some of the surrounding underbrush and thereby increase the size of the unit. One set of camp­ ers was observed increasing the size of a small hardwood campsite from 4 m to S.5 m in radius. The conifer camping sites probably increased in size more rapid­ ly in the period immediately after they were installed than the hardwood sites because of their open understory. The hardwood sites were usually surrounded by a dense stand of maple reproduction. Once a site was en­ larged. campers would continue to utilize the new space. A prediction equation was developed to estimate the change in radius which would occur on a camping unit in a two-year period. equation had a coefficient of determination of 0.65 and a standard The 67 error of 0.36. Table 16. Enlargement of camping unite on established campsites at three levels of visitor use over two seasons. Use / Site Initial Radius Final Radius % Change Heavy / Conifer (m) 6.4 (m) 7.6 15.8 Heavy / Hardwood 5.3 6.2 17.8 Moderate / Conifer 7.8 7.8 0.0 Moderato / Hardwood 6.0 6.3 5.7 Light / Conifer 5.6 6.0 7.1 Light / Hardwood 4.4 5.2 19.2 Y,Q » .630 - .808b, - .039b„ - .045b, + .001b,, wheret 18 1 2 3 4 ■ percent increase in the radius of a camping unit 0.630.» constant b^ ■ timber type b^ » use level b^ - initial radius of the site b ■ initial percent crown cover on the site. 4 The effect of each independent variable when the others are held con­ stant is shown below: 1. Radius increased more on hardwood sites than conifer sites; 2. High levels of use cause greater increases than low levels; 68 3. A large initial radius leads to successively smaller increases in radius* 4. The larger the percent crown cover the greater the increase. Sylvania Camper Survey The camper survey indicated that almost 41 percent of Sylvania's campers were repeat users. This figure ranged from a low of 27 percent on Crooked Lake to a high of 50 percent on Clark Lake. The choice of lakes was primarily controlled by the type of boating and fishing re­ gulations. Campers were attracted to Crooked Lake because it has regu­ lar Michigan fishing regulations and also it allowed use of outboard motors. The dominant reason for selecting other lakes was because access was limited to nonmotorized craft, predominantly canoes. Seclusion was the reason most often given for choosing a parti­ cular campsite (Table 17). Even though 27 percent of the campers had previously camped on Crooked Lake, only nine percent indicated that having seen the campsite on a previous visit influenced their selection. Other reasons given by campers for choosing a particular campsite in­ cluded that it was easy to find, centrally located, or that all the units on the site were empty. Upon arriving at a campsite, campers had to select which of the three available camping units (tent pads) they wanted to use. The most important factor in unit selection was that the others were occupied (Table 18). On Crooked Lake, where use was lower, the larger, more secluded sites were selected first. Nearness to the boat landings was almost as important as seclusion and size. Miscellaneous reasons 69 furnished by campers for choosing a particular camping unit included that the unit was better-drained, highest, had a better breeze, or had the fewest insects. Table 17. Criteria used by Sylvania visitors in campsite selection. Selection Criteria Crooked Lake Other ________________________________ Campsites___________ Campsites Percent of total Percent of total More Secluded 36 27 Closer to boatlanding 11 6 Closer to beaches and portages 10 13 Recommended by receptionist 10 14 Seen on previous visit 9 13 Recommended by friends 3 4 Name of the campsite 3 2 IB 100 21 100 Others The average daily number of hours spent on activities away from the campsite varied from zero to 12. On Crooked Lake the average was 10 hours, while on the remaining lakes the average was 7 1/2 hours. Thus only 14 hours were spent on the study campsites each day, and allowing time for sleep, less than half of the waking hours were spent in camp. Therefore, the total impact would be less than in an area where a greater percentage of the time is spent on the site. 70 Table 18. Criteria used by Sylvania visitors in choosing a particular camping unit. Selection Criteria Crooked Lake Campsites . Percent o£ total Other Campsites Percent of total Largest camping area 17 9 More seclusion 17 16 Closest to boat landing 16 12 Better view of lake 11 12 More shade 7 2 More sunlight 5 12 13 21 14 100 16 100 Other units occupied Others Ranking Camping Unit Durability To rank camping unit durability over a wide range of use levels and conditions , it is necessary to monitor several site parameters. After establishing the predictability of ecological changes at several use levels , the next logical step is to determine if actual measure­ ments of those changes can be utilized to develop an equation which would provide an index to the complex changes that occur on a camping unit with use. Using multiple regression the following equations was generated employing data which was gathered on Sylvania camping units during the visitor seasons of 1971 and 1972. It explains 76 percent of the 71 variance of the dependent variable with a standard error of 0.46. Y, * -1.90 - 4.811b. - .757b_ + .066b- - .752b. - 1.357b- - 1.271b19 1 2 3 4 5 8 + .355b^ -I- 5.3bg, wheret Y^g ■ index of detrimental change on established campsites -1.90 * constant b^ * change in radius of the camping unit bg “ change in dry bulk density in sampling zone 2 b3 « initial percent crown cover over sampling zone 1 b^ «■ change in noncapillary pore space of sampling zone 1 bg ■ change in dry bulk density in sampling zone 1 bg * change in depth of the AO horizon in sampling zone 1 b^ - change in noncapillary pore space in sampling zone 2 bQ » change in percent crown cover in sampling zone 1. This equation may be used in the same manner as the equation developed with data from simulated trampling plots. Resulting values of V,0 close to 1.0 indicate large changes in the measured parameters Xo and values approaching 4.0 indicate slight changes (based upon a coding system where 1 ■ heavy, 2 » moderate, 3 *■ light, and 4 - no use). A sample problem using the above equation and employing measurements from two of the study campsites is shown in Table 19. The results indicate that site 1 has not deteriorated as much as site 2. This is probably due primarily to the difference in use level of the two sites. Unit 1 received an average of 240 visitor days per year during the past 5 years while unit 2 received an average of 350 visitor days of use per year. Also, unit 1 was a conifer site and unit 2 was predominantly hardwood. Results indicate that overall hardwood sites are less durable than conifer sites. 72 Table 19. Prediction of the relative extent of site deterioration on two similar campsites using the predictive equation estab­ lished from campsite measurements. Independent Variables Unit 1 Unit 2 b^ ■ Change in radius of the camping unit .025 .153 b2 * Change in dry bulk density in sampling zone 2 .340 .303 80.000 90.000 b4 a Change in noncapillary pore space in zone 1 .147 .550 bg a change in dry bulk density in zone 1 .101 .181 bg a change in depth to the A2 horizon in zone 1 .000 1.000 by a change in noncapillary pore space in zone 2 .404 .414 b„ a change in percent crown cover over zone 1 .102 .041 b3 a initial crown cover over zone 1 Y19 - -1.800 - 4 . 8 1 1 ^ - .751b2 + .066b3 - .752b4 - 1.357bg - 1.27bg + .355by + 5.300bg, where: Y^g « index of detrimental change in measured parameters Ysite 1 " -1*900 " *120 ** *257 + 5.280 - .110 - .137 - 0.000 + .134 + .540 » 3.340 where: i “ index of detrimental change on site 1 Y , - -1.900 - .736 - .229 + 5.940 - .413 - .245 - 1.270 + .138 site 2 + .217 » 1.412 where: *site 2 “ index of detrimental change on site 2 and where a relative detrimental change of 1 ** great, 2 ■ moderate, 3 » little, and 4 <■ no change. CHAPTER VI RECOVERY OF ABANDONED CAMPSITES It has been suggested that campsites suffering from serious de­ terioration be temporarily closed to allow them to recover to natural conditions (Brockman, 1959, Lime, 1971). The objective of this portion of the study was to monitor the rate of natural recovery of abandoned campsites. Methods Study Area Four campsites were selected on the Sylvania tract which had been closed by the U. S. Forest Service. Two units were located on Crooked Lake, one on Clark Lake and one on East Bear Lake. Parameters monitored were percent litter cover, bulk density, and noncapillary pore space. Measurements were made in September, 1971, and July and September, 1972. A description of each site is given in Table 20. Results The recovery made in bulk density, percent litter cover, and noncapillary pore space is shown in Table 21. The camping units at Muskrat, on Crooked Lake, had been abandon­ ed for over 3 seasons at the time the measurements were made in 1972. During that period only small improvements were observed in bulk den­ sity and noncapillary pore space. Although percent litter increased 74 by 25%, it did not approach natural conditions (Figure 12). No seed* ling reproduction was found on the campsites even though the cover was dense at the edge of the units. Table 20. Characteristics of Sylvania camping units chosen to monitor site recovery. Camping Unit Characteristic Last full season of use Number of seasons open Total use while open (visitor days) Muskrat W2 Muskrat W3 Hemlock N2 1968 1968 1970 1 1/3 1 1/3 421 421 Hardwood Timber type Average percent crown cover (1971) 80 Hardwood 70 Wolf W3 1970 3 3 1607 Conifer 69 Hardwood 72 96 Although it had received nearly four times as much use, the camping unit at Hemlock appeared to be in about the same condition as the Muskrat sites. during the study. In fact, it showed a slightly more rapid recovery The camping unit at Wolf on East Bear Lake had been the least used campsite on Sylvania during the three years prior to the beginning of the study. The U. S. Forest Service closed these units to camping for purposes of this study. were nearly natural. The conditions on the camping unit Reproduction had begun to return to the unit and soil conditions on the unit were about the same as the controls. How­ ever, the litter cover was not as complete or as deep as that found off Table 21. Changes in percent litter cover, noncapillary pore space, and dry bulk density showing the recovery of closed camping units in one year. Camping Unit % Litter Cover Noncapillary Pore Space Dry Bulk Density (grams/cm^) (% by volume) Initial Final Control Initial Final Control Initial Final Control Hemlock N2 16 64 100 10 16 30 1.70 1.67 1.41 Muskrat W2 55 80 100 7 9 33 1.65 1.63 1.35 Muskrat W3 70 96 100 11 11 30 1.58 1.50 1.28 Wolf W3 52 80 100 22 25 27 1.47 1.42 1.40 (a) Figure 12. Comparison of a control plot (a) just off the margin of a previously heavily used campsite abandoned for three years and a used plot (b) from the center of the same unit. (b) 77 the units. Abandonment doea not appear to be an acceptable means of restor­ ing campsites to natural conditions in a reasonable short time. Sites which receive high levels of use, even for short periods of time, show­ ed only slight recovery after being idle for four years. Conifer sites showed a slightly greater rate of recovery than hardwood sites during the two-year study period. The primary limiting factor to site recov­ ery on heavily used units was the dense soil surface layer formed while they were in use. Some form of cultural activity to prepare a more satisfactory seedbed will be necessary on these sites before they will return to natural conditions in a reasonably short period of time. CHAPTER VII DISCUSSION AND CONCLUSIONS Simulated Trampling versus Actual Uae Both of these forms of analysis of recreation impact upon forest ecosystems have advantages and disadvantages. The chief advantage of studying actual campsites is the asurity that the changes which are occurring are due to recreation and probably could be duplicated at different sites. With artificial trampling there is always the ques­ tion of high closely one is simulating actual recreational use. However, the use of simulated trampling does have many advant­ ages that at least partially offset its handicaps. It provides a source of use which can be precisely controlled as to amount and timing. This allows the investigation of several different levels of use in one small area, or on sites where the potential establishment of actual campsites is to be evaluated. Simulation also permits accuate measure­ ments of conditions prior to trampling, and allows continual monitoring during the trampling period. This enables one to express any measured changes on the trampled plot as a function of level of use. The trampling levels applied in this study may or may not have duplicated recreational use. The timing and duration of the trampling was not the same as would occur on an actual campsite. On the trampled plots the use was applied but once a week in a period of only a few minutes, while on camping units the use was spread relatively evenly 78 79 over a longer period of time. The trampling schedule may have allowed some recovery to occur from week to week. The simulated use plots re­ ceived no treatment other than trampling* however, the campsites were at various times swept, raked, and covered by tents or shelters. An additional error factor may perhaps be the difference in relative age of the two areas. The actual campsites had already been in use for three years before the trample plots were established. Even with all of these differences, the results from the two areas were surprisingly similar and produced comparable equations on the index of detrimental change. However, due to more precise control of the use levels, the prediction equations established for the simu­ lated use sites yielded consistently higher coefficients of determina­ tion than equations estimating the same dependent variables with mea­ surements taken from camping units. Of the independent variables utilized in the prediction equa­ tions, for both simulated and actual campings, level of use was the most important factor. On the simulated sites it appeared in every equation, and showed the highest partial correlation coefficient in all but one instance. Timber type also appeared in each equation and eleva­ tion occurred in four of six situations in the simulated use equations. On actual campsites, use level appeared in ten of eleven equations. other independent variable appeared in more than four equations. Changes in Ecological Parameters Natural Litter Initially the simulated use plots were covered by a deep layer of natural litter made up of several years' leaf cast and with an No eo equally thick humic layer. On the actual campsites the litter consist­ ed of only one year's leaf fall and there was no humic layer between the litter and mineral soil. The results of the artificial trampling indicate that this was to be expected. there were reductions in depth of At all levels of trampling the litter and on heavily andmoder­ ately trampled plots mineral soil was exposed by the end of year. the second Therefore, the initial natural litter is probably destroyed on the campsites during the first year or so of use. The actual conifer camping units as well as conifer simulated use plots, showed the smaller changes in percent litter cover than hardwood sites. This is primarily due to the nature of the conifer litter and its seasonal fall. The composition of the natural litter on conifer sites had a greater proportion of woody material, twigs, and cones, and leaf fall occurred over the entire visitor season. However, even with this constant litter increment, the percent litter cover dropped below 75 percent on all of the plots within the margin of the canping unit. According to Orr (1971) and Parker (1953), if the ground cover is less than 75 percent, significant soil movement will take place under use. Although this type of soil erosion was not observed on the study area, the maintenance of some form of litter cover on the camping unit is undoubtedly important in the prevention of excessive soil erosion. Soil Properties Soil compaction was monitored in two different ways. Dry bulk density was measured directly and the percent by volume of noncapillary pore space as a function of air permeability. ai As expected, bulk density varied directly with level of use on both simulated use sites and camping units. The camping units showed a great deal of compaction with the formation of a surface layer firmly cemented by organic matter. This layer varied from one to two centi­ meters in thickness and dried to rock-like hardness. Such a layer was beginning to form on some of the heavily trampled simulated plots on hardwood sites where much of the litter layer had been destroyed. This indicates that crust formation on the soil surface probably begins after litter destruction and the speed of its formation is a function of use level. The compacted surface layer probably caused increased surface runoff. Due to the bowl-shaped nature of some camping units, i., runoff water from other parts of the unit collected in the center zone. This zone of the camping units showed the highest moisture levels. It may be debatable whether this crust formation is an undesir­ able factor. It certainly hinders the establishment of vegetation on the camping unit, and restricts air and water infiltration. However, its presence also retards soil erosion problems and reduces the dusti­ ness of the camping unit in dry weather. If it is impossible, as some authors suggest, to maintain ground cover on heavily used camping units, a hard surface layer may be a desirable feature. With lighter levels of use, the simulated as well as camping unit plots showed no crustral formation and higher bulk density values. The percent by volume of noncapillary pores in the soil proved to be a sensitive indicator of site changes, especially at low levels of use. Light levels of use produced large reductions in macropore space while higher levels of use produced only slightly greater changes. Most plots showed an off-season recovery in noncapillary pore space at 82 about 70 percent. South Dakota. Orr (1960) Indicated similar results in a study in He reported that decreases in noncapillary pore space due to trampling are usually balanced by increases in noncapillary pore space/ and that increases in noncapillary pore space during the recovery period are offset by decreases in capillary pores. The decrease in depth of the AO horizon is probably most useful on established camping units where the surface litter has been worn away and some degree of sheet erosion has taken place on the unit. It should be noted/ however/ that the change is a function of the original depth of the AO. Soils having initial shallow AO horizons would suffer greater percent changes with the same amount of erosion than soils with thick AO's. Other investigators (Frissell/ 1964/ Hartesveldt/ 1962) have used the percent of the site covered by exposed tree roots to measure erosion. Problems inherent to this method include: the necessity to wait until roots are exposed: the resultant damage that may occur be­ fore measurements can be made; and tree root depths vary greatly bet­ ween species/ and comparisons cannot therefore be readily made between timber types. Table 22 shows comparisons between initial or control values and final values after two seasons of simulated recreation use and five seasons of actual use. Increase in Camping Unit Size The increase in size of camping units with use has been reported by several authors (Merriam, 1971/ Echelberger/ 1971). This increase in radius/ noted in this study , has positively correlated with the Table 22. Comparison of initial or control values and final measurements of the parameters monitored on Sylvania. Camping Units (zone 1) Oven Dry Weight of Litter Bulk Density on .09 m2 %______________cm______________ 2 ____________q c Control Final Control Final Control Final Control Final Litter Cover Established Campsites Hardwood Heavy 100 Moderate 100 Light 100 Conifer Heavy Moderate Light 100 100 100 Experimental Plots Hardwood Heavy 100 Moderate 100 Light 100 Control 100 Conifer Heavy Moderate Light Control 100 100 100 100 10 5 30 5 30 35 Litter Depth — — — — — — — — — — — — — — — Noncapillary Pore Space % Control Final Depth of AO cm Control Final 1.9 1.7 1.7 2.0 1.9 1.8 10.0 11.1 16.7 2.9 5.1 11.5 5.0 5.5 4.7 0 0 2.0 — 1.8 1.7 1.6 2.0 1.9 1.8 9.3 10.0 15.0 3.1 7.5 13.0 6.1 5.0 7.1 0 1.5 27 — — — — 66 77 89 100 5.6 6.0 5.7 6.3 0.0 .5 1.0 1.0 270 251 283 266 00 30 45 60 1.4 1.4 1.4 1.5 1.8 1.8 1.7 1.5 30 27 29 30 3 9 13 25 5.0 5.3 5.1 4.6 .5 1.1 3.1 4.5 50 77 95 100 7.1 7.8 8.7 9.0 .8 1.2 2.1 5.0 300 279 285 268 80 74 140 160 1.3 1.3 1.4 1.4 1.5 1.4 1.4 1.4 27 33 35 32 4 10 15 32 9.5 11.1 10.2 9.0 6.0 7.5 8.7 9.0 84 number of visitor days of use the unit receives. This relationship may be the source of a management problem, since questionnaires returned by campers stated that one of their primary selection criteria was camping unit size. The larger sites were the most desirable. Thus as camping units become larger, they will also be more heavily used and this in turn will tend to promote a further increase in size. To prevent con­ tinued expansion of the camping units on Sylvania, it may be necessary to permanently delineate unit boundaries by some means which will not detract from the natural environment, such as logs or stones. Indices of Campsite Deterioration In previous studies of campsite deterioration, results have been limited to the establishment of equations which predict the change in some one ecological parameter. Since the changes which take place on camping units under use are the result of the interaction of many fac­ tors, it becomes desirable to establish a system by which several indi­ cators could be used to produce a single predictive equation. This would then provide a means by which camping areas over a wide range of ecological conditions could be compared and ranked according to their level of deterioration. Equations 7 (simulated recreational use) and 19 (actual campsite use) in this study provide the tools for such a rating system. After establishing the predictability of individual parameters as indicators of campsite deterioration at various levels of use, the actual measured changes in these parameters were utilized to establish two equations in which the dependent variable is an index of the detri­ mental changes in the independent variables. The lower the calculated 85 value o£ ¥ for a particular camping unit, the greater Is the detrimental change. Camping units whose ¥ values approach 1.0 on the scale have shown large detrimental changes while units which have Y values near 4 have undergone only small changes. Table 23 offers a comparison between the prediction equations that were generated in this study and the attitudes expressed by the campers for each camping unit. As can be seen from the ratings, campers are not generally aware of or concerned with the ecological factors which limit a site's durability, and their opinions are probably not based on the physical quality of a site. The equations, which used only ecological parameters in rating, gave the camping units consistently lower values than what the campers thought of the areas. For simulated use plots the values were near 1.00 and none was greater than 2.381. Using the coding system for levels of trampling on the simulated use plots where: 1.0 - heavily trampled, 2.0 «* moderately trampled, 3.0 * lightly trampled, and 4.0 ** controls, most established camping units were in about the same condition as the heavily trampled simulated use plots. Therefore, equation 7 should perhaps only be used on camping units which have been recently established and receive only light use. The extensive use of measurements involving litter cover make it more useful for sites on which natural litter has been been"destroyed by overuse. Equation 19 was established using data from actual camping units and may be best suited for use on established campsites where most of the litter has been worn away and measurements upon soil parameters take on increased importance. Table 23. A rating comparison of camping unit conditions by campers and two predictive equations. Campers (Survey) Rating Method2^ Actual Use (Ecr. 18) Camping Unit Use/Site Simulated Use (Eq. 7) Squirrel N1 Heavy/hardwood 2.50 1.412 .689 Squirrel N2 Heavy/conifer 2.75 2.845 1.071 Squirrel N3 Heavy/conifer 2.60 1.563 .285 Badger Wl Heavy/hardwood 2.60 2.654 .768 Porcupine Nl Light/hardwood 2.80 2.292 1.516 Porcupine N2 Moderate/hardwood 3.50 -.463 .590 Chipmunk Nl Moderate/conifer 2.67 3.340 2.381 Chipmunk N2 Moderate/conifer 3.75 1.566 .623 Chipmunk N3 Light/conifer 4.00 2.331 1.737 Fisher W3 Moderate/hardwood 2.00 2.062 1.053 Mink M2 Light/conifer 3.00 1.974 1.667 Mink N3 Light/hardwood 2.50 3.721 1.855 values from: -.5 to +.5 ■ deteriorated +.5 to 1.5 b fair 1.5 to 2.5 » good above 2.5 = very good 87 In testing the equations on Sylvania camping units, conifer units averaged higher Y values than the hardwood units with which they were paired. This indicates that at the same levels of use, conifer units are deteriorating less rapidly than hardwood units. Possible reasons for this are that conifer sites are larger on the average, and use is thus spread over a larger area. Also, conifer units maintained a higher percent litter cover which may have served to cushion the site against trampling and raindrop impact. conifer camping units. Crown cover was also less dense over the This allowed more sunlight and wind movement to reach the soil surface, causing more rapid drying after rainfall. Since moist soils are more subject to compaction than drier ones, the conifer sites are exposed to less soil compaction and smaller reductions in macropore spaces. Campsite Recovery It appears that campsite retirement is not a satisfactory method of returning areas to natural conditions or in appreciably improving the deteriorated conditions over a relatively short period of time. Only the lightly used units at campsite Wolf showed consistent improvement in site conditions. The sugar maple reproduction which was removed when the site was established was returning. By the end of the second season after closing, all of the one meter square study plots contained at least one sugar maple seedling. Bulk density and noncapillary pore space had also returned to near-normal conditions. The other three sites, which had received heavy use while open, had a densely compacted soil surface layer* This layer probably inter­ feres with good seed germination and seedling establishment, and also is 88 a poor surface for holding litter against the erosive forces of wind and water. The loosening action of freezing and thawing of the soil during the winter months has been proposed as a means by which sites naturally recover from soil compaction (Lull, 1959). In this area of Michigan, early snows protect the ground from repeated freezing and thawing, thus reducing the effect of this source of recovery. It will probably be necessary to undertake some type of cultural activity that both prepares a suitable seedbed, and adds the necessary plant cover. The feasibility of seeding abandoned sites with leguminous plants followed by irrigation and fertilization was reported by Beardsley and Wagar (1971) as a suc­ cessful means of restoring deteriorated campsites. Irrigation alone was nearly as effective as irrigation plus fertilization in establishing and maintaining a good ground cover. LITERATURE CITED Appel, A.J. 1950. Possible soil restoration on "overgrazed" recrea­ tional areas. J. For. 48:368. Beardsley, Wendell G. and J. Alan Wagar. on a forested recreation site. 1971. Vegetation management J. For. 69:728-731. Brockman, C. Frank. 1959. Recreational use of wild lands. Hill, New York. 346pp. McGraw- Cieslinski, Thomas J., and J. Alan Wagar. 1970. Predicting the dura­ bility of forest recreation sites in Northern Utah— prelimi­ nary results. USDA, For. Ser., Intermountain For. Exp. Sta., Res. Note INT-117. 7pp. Echelberger, Herbert E. 1971. Vegetative changes at Adirondack camp­ grounds 1964-1969. USDA, For. Ser., Northeastern For. Exp. Sta., Res. Note NE-142. 8pp. Frissell, Sidney S. 1964. Campsite preference and deterioration in the Quetico-Superior canoe country. M. S. Thesis, Univ. of Minn. 65pp. Hartesveldt, R.J. 1962. The effects of human impact upon Sequoia gigantia and its environment in Mariposa Grove, Yosemite National Park, California. Disseration, Univ. of Mich. Hatchell, G.E., C.W. Ralston, and R.R. Foil. 1970. in logging. J. For. 68(12):772-775. Soil disturbance LaPage, Wilbur F. Some observations on campground trampling and ground cover response. 1967. USDA, For. Ser., Northeastern For. Exp. Sta., Res. Pap. NE-68. 11pp. Lime, David W. and George H. Stankey. 1971. Carrying capacity: main­ taining outdoor recreation quality. Recreation symposium proc. USDA, For. Ser., Northeastern For. Exp. Sta. 174-185. Lindsay, J. 1969. Locating potential outdoor recreation areas from aerial photographs. J. For. 67:33-35. Lull, Howard W. 1959. Soil compaction of forest and range lands. USDA, For. Ser., Northeastern For. Exp. Sta., Misc. Pub. 768. 33pp. 89 90 Lutz, H.J. 1945. Soil conditions on picnic grounds in public forest parks. J. For. 43:121-127. Magill, Arthur W. 1970. Five California campgrounds ... conditions improve after 5 years' recreational use. USDA, For. Ser., Pacific Southwest For. Exp. Sta., Res. Pap. PSW-62. 18pp. and E.C. Nord. 1963. An evaluation of campground condi­ tions and needs for research. USDA, For. Ser., Pacific Southwest For. Exp. Sta., Res. Note PSW-4. 8pp. Merriam, L.C., Kent Goeckermann, J.A. Bloemendal, and T.M. Costello. 1971. A progress report on the condition of newly estab­ lished campsites in the Boundary Waters Canoe Area. Minn. For. Res. Note 232. 4pp. National Cooperative Soil Survey. 1958. Geogebic Series. 2pp. Orr, Howard K. 1960. Soil porosity and bulk density on grazed and protected Kentucky blue grass range in the Black Hills. J. Range Mgt. 13(2):80-86. Orr, Howard K. 1971. Design and layout of recreation facilities. Recreation symposium proc. USDA, For. Ser., Northeastern For. Exp. Sta. 23-27. Packer, Paul E. 1953. Effects of trampling disturbance on watershed conditions, runoff, and erosion. J. For. 51:28-31. Papamichos, N.T. 1966. Light, soil, and moisture conditions in areas of heavy recreation use. M. S. Thesis, Colo. State Univ. 101pp. Read, Ralph A. 1956. Effect of livestock concentration on surfacesoil porosity within shelterbelts. USDA, For. Ser., Rocky Mountain For. & Ran. Exp. Sta. Res. Note RM-22. 4pp. Ripley, Thomas H. 1962. Recreation impact on southern Appalachian campgrounds and picnic sites. USDA, For. Ser., Southeastern For. Exp. Sta., Res. Pap. SE-153. 20pp. Sokal, Robert R. and F. James Rohlf. San Francisco. 776pp. 1969. Biometry. W.H. Freeman, Steinbrenner, E.C. 1951. Effect of grazing on floristic composition and soil properties of farm woodlands in southern Wisconsin. J. For. 49:906-910. __________ . 1955. The effect of tractor logging on physical properties of some forest soils in southwestern Washington. Soil Sci. Soc. Amer. Proc. 19:372-376. 1959. A portable air permeameter for forest soils. Sci. Soc. Amer. Proc. 23(6):478-481. Soil 91 Thorud, David B. and Frissell, Sidney S., Jr. 1969. Soil rejuvenation following artificial compaction in a Minnesota oak stand. Sci* J. Ser. Pap. No. 7078. Univ. of Minn. Ag. Exp. Sta. 4pp. University of Michigan. 1965. Sylvania. Olsen Press. 16pp. U. S. Department of Commerce. 1971. Climatological survey: of Watersmeet, Michigan. No. 20-20. 2pp. climate U. S. Forest Service. 1968. Sylvania recreation area management plan. Ottawa National Forest. Ironwood, Mi. 48pp. __________ . 1970. Sylvania recreation area situation in summary. Division of Infor. and Ed. Milwaukee. 5pp. __________ . No date. Establishment report for the Sylvania research natural area within the Ottawa National Forest Gogebic County, Michigan, (preliminary draft). 11pp. Veatch, J .0. 1953. Soils and land of Michigan. E. Lansing. 241pp. Mich. St. Col. Press. Voss, Edward G. Curator, University of Michigan Herbarium. Correspondance dated Dec. 30, 1968, to M.W. Kageorge, Sup., Ottawa Nat. For. Wagar, J. Alan. tion. 24pp. 1964. The carrying capacity of wildlands for recrea­ For. Sci. Mono. 7. Soc. Amer. For. Washington, D.C. Yelenosky, George. 1964. Tolerance of trees to deficiencies of soil aeration. Proc. 40th Int. Shade Tree Conf. Houston. 127149. APPENDICES 92 Appendix A. Upland Plants of Sylvania Recreation Area As Identified By Dr. Edward Voss, Curator, University of Michigan Herbarium. Abies balsamea......................... ......Fir Acer rubrum................................... Red maple Acer saccharum............... ,............... Sugar maple Acer spicatum..................... ........... Mountain maple Actaea sp..................................... Baneberry Allium tricoccum............................. Wild leek Anemone quinquefolia............. Wood anemone Aralia nudicaulis............................ Wild sarsaparilla Arisaema triphyllum.......................... Jack-in-the-pulpit Betula alleghaniensis........................ Yellow birch Betula papyrifera............................ White birch Car ex spp..................................... Sedge Chrysosplenium americanum.................... Water-mat Claytonia caroliniana..............*......... Northern spring-beauty Coptis groenlandica.......................... Goldthread Cornus canadensis............................ Bunchberry Corylus cornuta............................... Beaked hazelnut Dentaria laciniata........................... Toothwort Dirca palustris...............................Leatherwood Dryopteris s p ................................. Wood fern Equisetum sylvaticum......................... Wood horsetail Fraxinus americana........................... White ash Galium triblorum............................. Bedatraw Cymnocarpium dryopteris...................... Oak fern Linnaea borealis..............................Twin flower Lonicera canadensis.......................... Honeysuckle Lycopodium lucidulum......................... Shining clubmoss Lycopodium obscurum.......................... Princess pine Maianthemum canadense........Wild Lily-of-the-valleyj Canada mayflower Mitella nuda............. ....... ............. Naked mitrewort Oryzopsis asperifolia........................ Rice grass Osmorhiza s p .......................... ....... Sweet-cicely Os try a virginiana............................ Ironwood Oxalis montana................................Wood-sorrel Panax trifolius...............................Dwarf ginseng Picea glauca.................................. White spruce Picea mariana................................. Black spruce Pinus banksiana...............................Jack pine Pinus resinosa................................Red pine Pinus strobus.............. .................. White pine Polygonatum pubescens........................ Solomon-seal Populus tremuloides.......................... trembling aspen Prunus sp ..................................... Cherry Ranunculus abortivus......................... Small-flower crowfoot Ranunculus recurvatus................ ........ Ribes cynosbati............................... Dogberry Rubus strigosus...............................Wild red raspberry 93 Sambucus pubens.... Smilacina racemosa.. Streptopus roseus... Thuja occidentalis.. Tilia americana.... Trlentalis borealis. Tsuga canadensis.... Uvularia grandiflora Viola cucullata.... Viola pallens...... Viola pensylvanica.. Viola selkirkii.... Viola Sororia...... Red-berried elder False Solomon-seal Twisted-stalk White-cedar Basswood Starflower Hemlock Bellwort Blue marsh-violet Wild white violet Yellow violet Selkirk's violet Woolly blue violet Appendix B. Comparison of climatic observations during the 1971 and 1972 study period with the 30 year average at Watersmeet, Michigan?1 Total Precipitation 1971 1972 19401969 (cm) Month Air Temperature Daily Maximum Daily Minimum 1971 1972 19401971 1972 19401969 1969 C° C° May 11.10 7.09 9.98 15.6 18.9 18.9 2.2 6.1 2.8 June 11.05 11.53 12.98 24.4 17.2 23.9 10.6 8.9 7.8 July 12.14 16.89 9.47 22.2 20.6 26.1 8.3 11.1 10.6 August 7.47 20.98 9.52 21.7 21.7 25.0 9.4 12.2 10.0 September 6.35 13.61 9.02 17.8 16.7 19.4 6.1 6.1 6.1 U. S. Department of Commerce, 1971 95 Appendix C. The Description of the Gogebic Soil Profile. The Gogebic Series^ SOIL PROFILE: Aq 5-0 cm GOGEBIC FINE SANDY LOAM— FORESTED Dark reddish brown (SYR 3/2) to black (5YR 2/1) spongy mor; very strongly to medium acid. 0 to 7.5 cm thick. A^ 0-1 cm Very dark gray (SYR 3/2) to dark reddish brown (5YR 3/2) fine sandy loam; high in organic matter; weak fine granular structure; very friable; very strongly to medium acid. A^ 1-7.5 cm .5 to 5 cm thick. Dark reddish gray (SYR 4/2) to pinkish gray (5YR 6/2) loamy fine sand to fine sandy loam; very weak fine crumb structure; very friable; very strongly to medium acid. 7.5-17.5 cm 2.5 to 15 cm thick. Dark reddish brown (5YR 3/2) to reddish brown (SYR 4/3) fine sandy loam (orterde); weak fine granular structure; friable; strongly to medium acid. 7.5 to 22.5 cm thick. B22 17*5“52,5 cm Dark reddish brown (5YR 3/4) to reddish brown (5YR 4/3 sandy loam to fine sandy loam; weak coarse granular to fine subangular blocky structure; friable; slightly cemented; strongly to medium acid. B 3m 52.5-65 cm 15 to 35 cm thick. Dark reddish gray (5YR 4/2) to brown (7.5 YR 5/2) sandy loam to fine sandy loam slightly mottled with weak red; platy structure in places; a distinct com­ pact vesicular pan; strongly to medium acid. ^National Cooperative Soil Survey, 1958. 10 to 96 SOIL PROFILE! GOGEBIC FINE SANDY LOAM— FORESTED 25 cm thick. C 65 cm - 1 m Reddish brown (2.5YR 4/4) to dark reddish brown (5YR 3/3) sandy loam to fine sandy loam glacial till, compact in place; many colored igneous and metamorphic rock and some reddish sandstone fragments; strongly to slightly acid. 97 Appendix D. Cover sheet of the Sylvania Camper Survey MIC HIGAN STATE UN IVER SIT Y h a s t l a n s i n g . M ic h ig a n a sm j DEPARTMENT OF FORESTRY ■ TELEPHONE (JI7) JJS-0090 Dear Camper: Michigan Stato University’s Department of Forestry and The National Wildlife Federation are conducting a three year study on the Sylvania Recreation Area to determine the recreational carrying capacity of the area. The object of the study is to determine what level of recreational use may be sustained on the area without permanent damage to the natural resources present. While one part of the research pertains to the resources (soil, water, and forest) of the area, a Vital part will come from you, the camper. Your opinions and comments will play an important part in our estimation of the carrying capacity, and our recommendations for the future management of Sylvania. Your assistance by completing this questionnaire and placing it in one of the return boxes located at the following locations: a) b) c) would be greatly appreciated. Lee M. James Chairman main boat landing on Crooked Lake main boat landing on Clark Lake portage from Whitefish to 1000 Island Lake 98 Appendix E. Questionnaire form used in the Sylvania Camper Survey MICHIGAN STATE UNIVERSITY DEPARTMENT OP FORESTRY 1. Have you camped on the Sylvania area before? YES NO (circle one) If yes, on which canpsite (campsites) have you previously stayed? 2. On which lake did you choose to camp for your present stay on Sylvania? ____________________________________ Circle the reason (reasons) you chose this lake. a) b) c) 3. access by outboard motor regular Michigan fishing law recommended by receptionist d) access by canoe only e) recommended by friends f) other reasons (specify). Which campsite did you choose for your present stay on Sylvania? Circle the reason (reasons) why you chose this campsite. a) b) c) d) 4. recommended by receptionist recommended *by friends name of campsite seen on previous visit e) f) g) h) Which particular tent pad did you choose upon your arrival at your selected campsite? Left (facing lake) A 5. Others were occupied Closest to boat landing _ _ _ Better view of lake Larger campsite area ______ Right (facing lake) C More shade than others f) More sunlight than others ____ g) More secluded ____ h) Other____(specify) ____________ What would you say is the condition of Very good 7. Middle B Why did you choose thisparticular tent pad area? Choose the 3most important reasons shown below. Rank them from 1 to 3 with 1 being most important. a) b) c) d) 6. close to beaches and portages close to boat loading more secluded others (specify) Good this tentpad area?(circle Fair one) Deteriorated On the average day how many hours did your party spend on activities away from the campsite? (circle one) 2 4 6 8 10 12 over 12 99 8. Arrival Date Time aw pm_____ (circle one) Departure Date _ _ _ _ _ _ _ _ _ _ _ _ Time am pm (circle one) PLEASE PLACE COMPLETED QUESTIONNAIRE IN ONE OP THE RETURN BOXES AT THE MAIN BOATLANDING ON CROOKED LAKE, MAIN BOATLANDING ON CLARK LAKE, OR PORTAGE TO 1000 ISLAND LAKE. !