\\\\A‘\C\i\\\\\ \\\\\\\\\\ \1\\\9\693\ 113MB? Michigan State . University ,. - w, n-.- “W...” l, :- This is to certify that the thesis entitled WHITE—TAILED DEER SUMMER HABITAT ‘ USE IN NORTHERN LOWER MICHIGAN '\ presented by David Charles Cue has been accepted towards fulfillment of the requirements for M. S . degree in Wildlife Major professor Jonathan B. Haufler Date Feb. 18 1985 0.7639 MSU is an Affirmative Action/Equal Opportunity Institution // .J' , 0/ /{ 3* 7/L. / C gr’c/ MSU RETURNING MATERIALS: P1ace in book drop to LIBRARJES remove this checkout from —:——_ your record. FINES wiH be charged if book is returned after the date stamped beiow. (2" w” j" ,rimy‘yfl. Wifééyfi . Mngf ’ t ,_ _l 6 S Miriam 1 , m6 ‘88 _ {wave 115%.: 23 £0 1 20~ A320 ’ 9 . ‘ (031%!th “Ii-Ii. i or £15}? 4 51%;; game. HM 7h“; it”?! , h 6 3 0 ; faLL’a am‘ ‘ " WHITE-TAILED DEER SUMMER HABITAT USE IN NORTHERN LOWER MICHIGAN BY David Charles Cue A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1985 Dedicated to my loving wife, Diane, for her support, patience, and perseverance, and to my parents for their continual support. ii ACKNOWLEDGMENTS I would like to express my great appreciation to Dr. Jonathan Haufler for taking me on as his graduate student, and for his guidance and assistance throughout this study. Thanks is given to my committee members, Carl Bennett, Jr., Dr. Harold Prince, and Dr. Carl Ramm, for their assistance. I would also like to thank Dr. Stanley Zarnoch for his guidance in initiating my graduate program, and for helping to initiate this study; and Dr. Rollin Baker for his aid at the beginning of the study. Special thanks are due Richard Moran, Houghton Lake Wildlife Research Station, who handled much of the field work, for helping to initiate this study, offering his assistance, and his many helpful suggestions; James Terry for conducting much of the field work, and for handling the monumental task of coding the data; and the many other Department of Natural Resources personnel who assisted with the track counts. This study was financed in part through Federal Aid to Wildlife Restoration, Michigan Pittman-Robertson Project W-ll7-R. iii TABLE OF CONTENTS LIST OF TABLES ........................................ V LIST OF FIGURES ....................................... Vii INTRODUCTION .......................................... 1 STUDY AREAS ......... .. ................................ 3 METHODS ............................................... 6 RESULTS ....................................... . ....... 10 Influence of Cover Type and Stand Maturity ...... .. 10 Townline Creek Study Area ................. .... ll M-18 Study Area ............................... l4 Lanes Lake Study Area ....... ............... ... 17 Russell Lake Study Area ............ . .......... 19 9-Mile Study Area ..... ........................ 22 Sharon Study Area ....... .. ....... . .......... .. 25 DISCUSSION 0 ........ O ............ O O O I O O O O O O O O O O O ....... 30 LITERATURE CITED ... ................................... 34 iv Table Table Table Table Table Table Table Table LIST OF TABLES Results of analysis of summer deer use on the Townline Creek Study Area, 1975-1980 .... 12 Mean summer use by deer of foreSt cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the Townline Creek Study Area, 1975-1980. Sample sizes: four counts in 1978 and 1979, and three in each of the remaining years ... .......................... 13 Results of analysis of summer deer use on the M-18 Study Area, 1975-1980 .............. 15 Mean summer use by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the M-18 Study Area, 1975-1980. Sample sizes: two counts in 1975, three in 1977, five in 1978, and four in each of the remaining years ............................. 16 Results of analysis of summer deer use on the Lanes Lake Study Area, 1975-1980 ........ 18 Mean summer use by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the Lanes Lake Study Area, 1975-1980. Sample sizes: four counts in 1979 and 1980, and three in each of the remaining years ........ 20 Results of analysis of summer deer use on the Russell Lake Study Area, 1975-1980 ...... 21 Mean summer use by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the Russell Lake Study Area, 1975-1980. Sample sizes: three counts in 1975 and 1977, five in 1978, and four in each of the remaining years ......................... 23 Table Table Table Table 10. ll. 12. Results of analysis of summer deer use on the 9-Mile Study Area, 1975-1980 .. ....... ... Mean summer use by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the 9-Mile Study Area, 1975-1980. Sample sizes: two counts in 1975, four in 1978, and three in each of the remaining years .... Results of analysis of summer deer use on the Sharon Study Area, 1975-1980 ............ Mean summer use by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the Sharon Study Area, 1975-1980. Sample sizes: two counts in 1975, three in 1976, and four in each of the remaining years ............................. vi 26 27 29 LIST OF FIGURES Figure 1. Geographic location of the study areas and cover treatment patterns ..................... vii ABSTRACT WHITE-TAILED DEER SUMMER HABITAT USE IN NORTHERN LOWER MICHIGAN BY David Charles Cue White-tailed deer (Odocoileus Virginianus) habitat use was studied on six areas in northern Lower Michigan during the summers (July-August) of 1975-1980. A given percentage (25%, 50%, or 75%) of each area had been treated (clearing all standing trees) just prior to the inception of this study. Roadside track counts were utilized for measuring habitat use on each study area. The mature oak-pine mixture, and the mature oak with a sapling pine understory received consistently high use. Immature oak and oak-pine mixture stands appeared to be avoided to some extent, while immature stands of most other cover types, particularly aspen or aspen-maple mixture, were used fairly heavily, except on the 75% treatment level study areas. All the immature stands which regenerated after treatment on the 75% areas were avoided somewhat. However, the edges between immature and mature forest stands on these areas were used extensively, regardless of cover type. INTRODUCTION Historically, much of the research on white-tailed deer (Odocoileus Virginianus) in the northern Lake States has focused on the fall and winter periods. There were two major reasons for this. First, during the fall period the fat reserves, which are essential for breeding and winter survival, are accumulated. Second, the limitations of winter range in the region were quite obvious, and there was great concern with overwinter losses. In more recent years, great strides have been made in the management of winter deer range (Verme 1965, Krefting and Phillips 1970) and the understanding of wintering deer (Silver et a1. 1969, Ullrey et a1. 1970, Silver et a1. 1971, Moen 1976, 1978; Karns 1980). There has also been an increased interest in the year-round requirements of deer and research emphasis on the remaining segments of the year. Karns (1980), in his discussion of overwinter mortality, stressed that the rest of the year could not be ignored. Studies by Verme (1969) and McCaffery and Creed (1969) have indicated that summer range could be extremely vital to a deer population, especially in areas where forests are maturing and converting to more tolerant types. There has been a general trend in the region of conversion to later 2 successional types and thus, a growing concern with summer range for deer. Several studies of the summer food habits and/or habitat use of white-tailed deer in the region have found a general preference for intolerant forest types, most notably aspen, and have recommended that such types be maintained for summer range (McCaffery and Creed 1969, Kohn and Mooty 1971, McCaffery et a1. 1974, McCaffery 1976, Stormer and Bauer 1980, Rogers et a1. 1981). This study of summer habitat use was conducted to determine the key components of summer deer range in northern Michigan for use in improving management guidelines. The influences of cover type, stand maturity, and juxtaposition of stands on use by deer were examined. Habitat use was assumed to reflect use by deer for both forage and cover. STUDY AREAS This study was part of a larger project supported by the Michigan Department of Natural Resources and designed to evaluate the effects of various levels of intensive habitat treatment on wildlife, vegetation, and the public (Bennett et a1. 1980). The study was conducted on six research units, 23.3 km2 (one-quarter-township or 3X3 miles) each, which are a part of the larger project. The six units which served as study areas are all located in the central portion of the northern Lower Peninsula of Michigan (Figure 1). The Townline Creek, M-18, Lanes Lake, Russell Lake, and 9-Mi1e units are located in Roscommon County (SEH T21N R4W, SE% T21N R3W, SE% T22N R2W, NE% T23N R2W, and SE% T22N R1W, respectivelY); the Sharon unit in the southeastern portion of Kalkaska County (SEk T25N R6W). The topography of the areas is generally characterized by rolling uplands and flat outwash plains. Soils are predominantly Roselawn sand, Grayling sand, and Rifle peat (Veatch et a1. 1924, Veatch et al. 1927). The major upland canopy species on the areas include aspen (primarily Populus tremuloides), oak (Quercus spp.), jack pine (Pinus banksiana), and red pine (P. resinosa). Representative lowland species are northern white cedar (Thuja occidentalis), balsam fir 29% 7? ‘1' 4434 7:? Ir,’ I ,I 3/ " ~(‘ 9/ ,1" ,1,‘ 'VIII/ :/0; " 5’ E7 'jfiifx L ' V) 4/75 TOWNLINE CR. ROSCOMMON CO //4 :2 . 6! , xf “If / W ,2,ng ' W/ 27% ,p / / ‘\\§“ R‘ z, \ . » \'\i\\:\ \\ \f \\. ’ yl/ \ ‘ \\\\\\\\ ‘\ I, \,:Lv/I - I, t. , ' ‘ \\\{: ‘8“: \\\\\\ \ Kit-II ,, .\ RUSSELL LAKE Roscommouc \\ ; SHARON KALKASKA co .32 X KEY $332223) land treated land (clear cut) ‘P private land \ (forested) track route maintained road poor dirt R — ROSCOMMON COUNTY ....-- trail K — KALKASKA COUNTY LOWER PENINSULA .OF MICHIGAN Figure 1. Geographic location of the study areas and cover treatment patterns. 5 (Abies balsamea), and black spruce (Picea mariana). The areas are interspersed with numerous marshes and bogs. The six research units were divided into pairs and each pair was assigned one of three levels of habitat treatment. Treatment consisted of clearing all standing trees over 5.1 cm dbh (2 in) from a given percentage of each unit. The three treatment levels were 25%, 50%, and 75% of the entire unit. The Townline Creek and M-18 units were paired and assigned the 25% treatment level; the Lanes Lake and Russell Lake units were assigned the 50% level; and the 9-Mile and Sharon units were assigned the 75% level. Blocks of various sizes within each unit were cleared until the assigned treatment level for the unit was reached. Treatment began in the fall of 1972 and was completed before the summer of 1975. The blocks of area to be treated on each unit were selected on the basis of cover type. Mature aspen was the first to be treated, followed by upland hardwood, oak, jack pine, mixtures, and upland brush. All lowland (swamp) conifers were left intact for winter cover. The shape and size of the resulting clearings were variable (Figure 1). The size was generally related to the treatment level of the unit. The 25% treatment units had several small treated blocks, while the 75% units tended toward much larger, continuous blocks of treated area. Prior to treatment, the six units which serve as study areas were all fairly continuously canopied. METHODS Summer roadside track counts were conducted by Michigan Department of Natural Resources personnel as part of an effort to monitor the deer populations on the six study areas for the larger project (Bennett et al. 1980). Data collected on these track counts were utilized for estimating habitat use in this study. Track counts have been used previously to measure use by deer in several other studies (Krull 1964, McCaffery and Creed 1969, Kohn and Mooty 1971, McNeill 1971). Data were collected on track counts conducted during the summer (July-August) of 1975, after completion of all habitat treatment, and each summer following that, through 1980. The technique which was used for the counts is very similar to that developed by Daniel and Frels (1971). Counts were conducted on an established route, 11.3-17.7 km (7-11 miles) in length, on each study area. Each route was comprised of existing roads and vehicular trails with surfaces suitable for the detection of deer tracks. The surface of the route was prepared, erasing old tracks and smoothing the surface, in the late afternoon by dragging with a heavy chain apparatus. The count was then made the following morning from the hood of a slow-moving vehicle beginning at approximately 0600 hours. The location of each 7 group of deer tracks crossing the route was recorded, along with some additional data. A group of deer tracks crossing the route meant those tracks made by a group of deer traveling together, such as a family group, and was termed a crossing group. When a single set of deer tracks crossed the route the crossing group was actually an individual deer (still a crossing group). When more than a single set of tracks crossed the route an effort was made to determine how many groups had crossed, if more than one, and data were recorded separately for each of these crossing groups. Information on habitat use was obtained by noting the combination of the cover type and size class of the forest stand traversed by the route at each crossing group location. Often the track routes did not traverse stands but followed the edge between two stands, i.e., the cover type and/or size class of the stand on one side of the route differed from that of the stand on the other side. When the route followed an edge at the point of crossing, the cover type and size class combinations of the two juxtaposed stands were recorded together, and the type was noted as an edge. Size class was used as a measure of stand maturity, and was divided into two categories. The first category was reproduction size class (< 12.7 cm dbh). These immature stands were almost exclusively the result of regeneration after treatment (clearing). The other category was pole (2 12.7 and < 25.4 cm dbh) or saw (3;25.4 cm dbh) size class stands. The vast majority of this mature category was comprised of pole size class stands on each area, for very little saw size occurred on any of the areas. Following a count an index of use was computed for each of the various vegetation types occurring along the route. Vegetation type was defined by the cover type and size class combination(s), and whether the area was an edge. The use index was: /(C/M) + 0.5, where C = number of crossing groups found in a cover type and size class combination on the count, and M = miles of route along which the combination occurred. The index was a transformation of the number of crossing groups per mile (C/M), used so the assumptions of the subsequent analyses were met. Crossing groups were used because the group acts as a unit when selecting vegetation types for use. Thus, location of a crossing group in a particular cover type and size class combination was assumed to indicate use by the group of deer of that combination, and the index was assumed to reflect the level of use on the study area for the time period from late afternoon (dragging) until the count the following morning. The length of occurrence along the route was determined from area forest cover maps. When very little of a cover type and size class combination occurred along a route some combinations had to be grouped together. The grouped combinations were treated as a single, unique combination for use index computations. A two-way analysis of variance model was utilized to test the effects of cover type and size class combination, and time-since-treatment (year) on use by deer, separately for each study area. The analysis was done separately by area because the combinations which occurred along the route on each area differed a great deal. Also, a preliminary analysis examining use across the study areas by size class category (ignoring cover types) had revealed a significant interaction between study area and size class, indicating that further analysis should be done separately by area. Grouping of some combinations was necessitated on each area by short lengths of occurrence. Use, as measured by the index, was found to be normally distributed using the Kolmogorov—Smirnov test (P) 0.10) (Sokal and Rohlf 1981) for each study area. The assumption of homogeneous variance was tested using Bartlett's test (P> 0.20) (Sokal and Rohlf 1981) for the Townline Creek and 9-Mile Study Areas, while it was assumed to be valid for the other four study areas after examination of the cell variances for each area. Data were unbalanced due to the varying number of counts from year to year, however the cell sizes were proportional. Thus, the two main effects were orthogonal to each other, but neither was orthogonal to their interaction (Searle 1971). The sums of squares were obtained by fitting the two main effects and then their interaction. The resulting F—statistics provided approximate tests for the main effects and a test of the hypothesis that interactions were zero (since all cells were filled). RESULTS One hundred twenty-four track counts were conducted in the July-August periods of 1975-1980 on the six study areas. Of these, the greatest number were done on the Russell Lake Study Area (23), and the fewest on the 9-Mile Study Area (18). Over 1600 km (1000 miles) of route were covered on these counts, providing habitat use information on more than 15,000 groups of deer crossing the routes. Influence of Cover Type and Stand Maturity The possible influence of cover type and stand maturity (or size class) on habitat utilization was of interest, and more specifically, how stand maturity and cover type might interact. Thus, differences in use among the cover type and size class combinations were examined for each study area. Twenty-nine different cover type and size class combinations occurred along a sufficient portion of track route on one or more of the study areas to be analyzed. Of these only one was found on five of the study areas, the remainder on three or fewer, with the majority (18) on just one area. The lack of replication of cover type and size class combinations across study areas necessitated analysis separately by study 10 11 area. For each study area two or three groups of combinations had to be formed for analysis because of small occurrences. Townline Creek Study Area Eleven cover type and size class combinations were found along the track route on the Townline Creek Study Area, three of which were actually groups of combinations. Results of the analysis of variance indicated that there were significant (P<:0.01) differences in use among the combinations (Table 1). To determine which cover type and size class combinations differed in use, Tukey's honestly significant difference (HSD) test (Steel and Torrie 1960) was used to compare mean use across years of the combinations. Comparisons of all pairs of combinations were also made by year, although there was little evidence to suggest that differences were dependent on time-since-treatment, or year (interaction P> 0.10). Focusing then on mean utilization across years, the immature (reproduction size class) oak-pine mixture and the grouped combination comprised of mature (pole or saw size class) stands of various cover types and edges between types were used the least (Table 2). The immature oak-maple- aspen-cherry mixture, edge between reproduction aspen and mature swamp conifers, pole size oak with a sapling pine understory, edge between reproduction oak and mature oak- pine mixture, and the edge between mature oak-pine mixture and mature jack pine were used to a greater extent (P‘<0.01 for the first two, remaining three P <0.05) than the first 12 Table 1. Results of analysis of summer deer use on the Townline Creek Study Area, 1975-1980. Source of Mean Variation df Square F Significance Main effects Cover and size combination 10 4.1487 5.404 P < 0.001 Year 5 3.1812 4.144 P = 0.001 Interaction Cover and size X year 50 0.8494 1.106 P = 0.315 Residual error 154 0.7677 13 .ummu om: m.»mxse an no>mn am can no snucmonunconm nonune coEEoo an umuuma o o>on no: on nonu meow: .mooconowmwc ucooflwncoflm mouooflccw cfisnoo o :« mnouumn mo oocmmonm «a .cmuusouo cofluocflnsoo mnu nownz mcoHo muson no moans u 2 Eco .ucsoo onn :0 acnnonfinEoo onu :« p:50w mm30nm venomouo no Henson n O anon: .m.o + Azmuvx H mm: a n no n m.v m.m m.v ¢.m m.v m.m a.v Any anumn0|cmmmoloamoelxoo n n o m.v m.m m.¢ ~.m o.v a.m m.v ANS “ouncoo meozm one inc comma cmmzumn moon n no no n.e n.e v.m a.m w.v m.v m.v snoumumocs wens mannamm m can: AN. xmo n no no H.v ~.v m.m m.v N.m o.m o.m Amy manmuxoo pco Adv xoo cooznon ownm n no no o.v n.¢ e.¢ o.v n.m e.m ~.m Ame mafia new“ one Ame wcnm-xmo cmmzumn mmom no no no m.m m.m m.m m.m m.v m.m H.v Amy wandnxoo no no no monunmnoE nufl3 wompo m.m ¢.v m.m m.m m.v o.m m.m mocsaocfl .mcoflnocflneoo nonno msomconaoomnz no no no w.m ~.v m.m n.v G.m m.m o.m inc mnmmELEommm no cmmm< no no no m.m v.m ~.m ~.v v.m m.m m.m mwdxu msoflno> mo Ame cco AHV coo3non ovum o no o Ame H.m v.m m.m H.m w.m m.m m.~ ano .modxu coozuon mooco cco modxn moonno> o o .Eo m.~ m.N m.~ m.m n.m m.m n.~ AHV ocnmuxoo coo: ommn anmn mnma whoa whoa mean acnnoCnnEou mmoau ounm oco odxe no>ou nmouom .mnoo> madcaoEon onu wo nooo ad monnu cco .mnmn Uco mean an mucsoo uSOu "woman oHdEom .ommnlmhmn .oon4 >cnnm xownu oceac3oe onn co mconnoCHnEoo Azom no once u N .coflnoscoumon u H. mmoao munm Uco mm>u no>oo nmonOw mo noon xn «mm: noEEsm coo: .m onnoe 14 two combinations mentioned. Also, utilization of the mature oak-pine mixture exceeded (P<:O.10) that of immature. M-18 Study Area On the other 25% treatment level study area, M—18, 15 cover type and size class combinations were found along the track route, two of which were groups of combinations. The analysis of variance (Table 3) revealed a significant (P< 0.01) combination main effect, which indicated differences in utilization among the combinations, and some evidence that the differences were dependent on time-since-treatment (P< 0.10). The edge between reproduction size oak-maple—aspen- cherry mixture and mature oak with a sapling pine understory had the highest mean use index across years (Table 4). Next highest was the immature aspen, followed by the edge between immature oak-maple-aspen-cherry mixture and pole size oak cover type, and the edge between immature aspen or aspen- maple and immature oak-maple-aspen-cherry mixture. All four combinations Were used to a significantly (P<:0.05) greater extent than the least used group of miscellaneous combinations (primarily mature), and the edge between mature aspen-oak mixture and mature aspen-maple mixture. Also, the difference in use between the first combination and mature jack pine was moderately significant (P<:0.10). While not significantly different than any of the other combinations, the mature oak-pine mixture and mature oak with a sapling pine understory combinations had moderately high levels of use. 15 Table 3. Results of analysis of summer deer use on the M-18 Study Area, 1975-1980. Source of Mean Variation df Square F Significance Main effects Cover and size combination 14 5.1266' 4.333 P < 0.001 Year 5 9.5117 8.040 P < 0.001 Interaction Cover and size X year 70 1.5238 1.288 P = 0.084 Residual error 240 1.1831 1 6 Table 4. Mean summer use* by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the M-18 Study Area, three in 1977, two counts in 1975, remaining years. five in 1978, and four in each of the 1975-1980. Sample sizes: Forest Cover Type and Size Class Combination 1975 1976 1977 1978 1979 1980 Mean Misc. other combinations, 2.3 3.0 2.8 2.6 3.0 2.9 2.8 primarily (2) a** a Edge between aspen-oak (2) 0.7 2.7 4.0 3.1 2.8 3.5 2.8 and aspen-maple (2) a a Jack pine (2) 3.0 2.7 3.2 3.4 3.1 3.4 3.1 ab ab Edge between aspen (l) and oak (2) with a sapling 1.4 3.1 4.7 3.7 3.1 3.6 3.3 pine understory ab ab Edge between (1) and (2) 2.4 2.8 3.6 2.8 4.1 4.0 3.3 of various types ab ab Edge between jack pine (2) 2.8 3.8 3.1 3.4 2.8 3.8 3.3 and red pine (2) a ab Aspen-oak (2) 2.8 2.9 5.2 2.9 3.0 3.2 3.3 a ab Edge between oak-maple— aspen-cherry (l) and 2.4 2.5 3.9 3.7 3.3 4.3 3.4 aspen-oak (2) ab ab Edge between aspen (l) and 2.8 4.7 4.0 4.1 2.7 2 9 3.5 aspen (2) a ab Oak (2) with a sapling 3.7 3.3 3.5 3.0 3.3 5.2 3.7 pine understory ab ab Oak-pine (2) 2.5 4.0 5.1 3.9 2.5 4.6 3.8 a ab Edge between aspen or aspen-maple (1) and oak- 1.9 3.9 2.9 4.6 5.6 5.3 4.0 maple-aspen-cherry (l) b b Edge between oak-maple- aspen-cherry (l) and 3.6 4.5 4.0 3.8 3.7 4.6 4.0 oak (2) ab b Aspen (l) 3.2 3.9 5.1 4.4 4.1 4.2 4.1 ab b Edge between oak-maple- aspen-cherry (1) and oak (2) with a sapling 3.7 4.6 3.5 4.4 4.6 4.8 4.3 pine understory ab b * Use = /(C7M) + 0. , where C ** Presence of letters in a column indicates significant differences. have a letter in common differ significantly at the 5% level by Tukey's HSD test. = number of crossing groups found in the combination on the count, and M = miles of route along which the combination occurred. Means that do not 17 Given the evidence of interaction, comparisons of the combinations were also done by year. For all years except 1979, no significant (P:>0.10) difference in use among the combinations was found. In 1979 the edge between immature aspen or aspen-maple mixture and immature oak-maple-aspen- cherry mixture was used to a significantly (P<<0.05) greater extent than six of the other combinations, including mature oak-pine which had the lowest mean use index. Any consistent change in the pattern of relative use of the combinations with time-since-treatment was not readily discernible. Lanes Lake Study Area At the 50% treatment level, 11 cover type and size class combinations were found along the track route on the Lanes Lake Study Area, three of which were groups of combinations. Results of the analysis of variance showed evidence of significant (P<:0.01) differences in use among these combinations (Table 5). Strong evidence of interaction (P< 0.01) was also found, indicating that differences were dependent on time-since-treatment. First looking at the comparisons of mean use across years, the three combinations used most, edge between reproduction aspen and various mature cover types, reproduction aspen or aspen-maple, and mature oak-pine, had significantly (P‘<0.01) higher use indices than the four least used combinations, mature oak, edge between immature oak and immature oak-pine mixture, reproduction oak-pine or edge between reproduction and mature oak—pine, 18 Table 5. Results of analysis of summer deer use on the Lanes Lake Study Area, 1975-1980. Source of Mean Variation df Square F Significance Main effects Cover and size combination 10 12.5844 11.761 P < 0.001 Year 5 2.1341 1.994 P = 0.082 Interaction Cover and size X year 50 3.2929 3.077 P < 0.001 Residual error 154 1.0700 19 and edge between immature aspen-oak mixture and immature oak cover type (Table 6). The edge between reproduction aspen and reproduction oak was used to a greater extent (P‘<0.05) than the two least utilized combinations. Note the prevalence of oak and oak mixtures, particularly reproduction size class, in the least used combinations, and the prevalence of immature aspen in the preferred combinations. An exception was the highly used mature oak-pine mixture. Comparisons of use among the 11 combinations were also made by year. Significant (P<:0.05) differences were found for all years but 1976. The magnitude of the differences between the higher and lower use combinations appeared to have increased somewhat with increasing time-since-treatment. Any other consistent changes were not evident. Russell Lake Study Area On the Russell Lake Study Area, the other 50% treatment level area, 10 cover type and size class combinations were found along the track route, three of which were groups of combinations. The analysis of variance (Table 7) revealed a significant (P< 0.01) combination main effect, indicating differences in utilization among the combinations. The combination by year interaction was also significant (P< 0.01), meaning the differences were dependent on time-since— treatment. Comparing mean use across years of the combinations, upland brush, one of the grouped combinations (various types and edges between types, all deciduous and 20 .umou om: m.>oxse >n Ho>oH wm onn no haucoowwflcmflm nouunc :oEEoo an nounoa o o>on no: on uonu mcoo: .mooconommnu ucoonwficwfio mouoofipcfi CEDHOO o ca muouuon mo oocomond «a .couusooo c0nuocflneoo on» nonnz mcoHo ouDOn mo moHHE n : cco .ucsoo on» no cofluocflneoo on» :w venom masonw @GHmmOHU mo nonES: u U ononz .m.o + A:RO.\ u om: . o c n on n n m.v h.m N.m ~.m m.m m.m m.v “NV momxu mSOwuo> cco Any codmo coo3non omcm on con n o no no v.v N.v m.v m.m N.v o.v v.m AHV oHQoEIcommo no nomad on o no ono no n m.v m.m n.m m.m H.v o.m o.v Ame ocHduxoo opo no no ono no no m.m m.m m.m m.m >.m H.v m.m AHV xoo pco Adv codmo coozuon ompm con ono n ono no no m.m m.~ m.v o.v h.m v.m m.~ mcoflnocflnEoo nonno msoocoaaoomfi: pono no no ono no no v.m v.m ~.m o.v H.m w.v m.~ any xoo pono pun no o o no m.m N.v v.m m.a v.~ m.v H.m Amy ocfimlxoo pco AHV xoo coo3uon omom ono ono no no no no o.m m.~ H.m w.m h.m m.m a.~ Adv xoo pco Any noolcodmo coozuon omcm ono to o o o no Amy ocflalxoo cco on” «.... a4 In m4 M; In E 3818 525% «moo one 3 ofiaiao no o no ono o no h.N h.o o.v m.N w.H v.m m.m Adv ocwmlxoo fico AHV xoo cooZuon omvm o no o o no «so . v.~ ~.~ o.~ v.~ v.m >.N o.a Amy xoo coo: ommn mean whoa swan coma mnma cenuocnnfiou mwoao onnm cco odxe no>ou umouom .muoo> mcficflofiou onu wo nuoo ca oonnn pco .ommn pco mean an oncsoo anew "mounm oHdEom .ommalmnan .oou4 xcsnm oxon oocon onn co onennocnneoo ABom no once u N .coHnospOHQon u H. mmoao onwm pco odxu uo>oo umou0w wo Hoop >n sow: noEESm coo: .w oHnoB 21 Table 7. Results of analysis of summer deer use on the Russell Lake Study Area, 1975-1980. Source of Mean Variation df Square F Significance Main effects Cover and size combination 9 23.8223 26.103 P < 0.001 Year 5 5.1678 5.663 P < 0.001 Interaction Cover and size X year 45 1.9975 2.189 P < 0.001 Residual error 170 0.9126 22 mature), and the mature swamp conifers were utilized to a significantly (P< 0.05) lesser extent than all seven remaining combinations (Table 8). The edge between various immature deciduous cover types and various mature cover types (swamp conifer, red pine, or upland hardwood) not only had a significantly (P<:0.01) higher mean use index than the three combinations just mentioned, but was significantly (P< 0.01) greater than the group of miscellaneous other combinations, primarily of reproduction size, and the edge between immature oak and immature oak-pine mixture. The reproduction size aspen or aspen-maple combination had the second highest mean use index. In comparisons of use among the 10 combinations made by year, significant (P<:0.05) differences were detected for all six years. Use of immature oak appeared to have dropped somewhat, relative to the other combinations, after the initial two years. Also, use of the edge between upland grass and mature aspen-maple mixture was consistently high, with the exception of 1975 and 1980. 9-Mile Study Area Considering the 75% treatment level areas, four cover type and size class combinations were found along the track route on the 9—Mile Study Area, two of which were groups of combinations. Results of the analysis of variance (Table 9) indicated that there were significant (P<10.01) differences in use among the combinations, and that the differences were 23 Table 8. Mean summer use* by deer of forest cover type and size class (1 = reproduction, 2 = pole or saw) combinations on the Russell Lake Study Area, 1975—1980. Sample sizes: three counts in 1975 and 1977, five in 1978, and four in each of the remaining years. Forest Cover Type and Size Class Combination 1975 1976 1977 1978 1979 1980 Mean Upland brush 1.7 2.7 2.1 2.5 2.2 2.3 2.2 a** ab a ab ab a a Various deciduous types 1.6 2.1 3.1 2 0 2.5 4.2 2.6 and edges, all (2) a a ab a abc ab a Swamp conifer (2) 3.2 2.8 4.2 2.3 1.0 2.8 2.7 ab ab ab 3 a ab a Misc. other combinations, 3.3 3.4 4.0 3.4 3.1 4.9 3.7 primarily (l) ab abcd b abc abcd b b Edge between oak (l) and 3.4 4.4 3.9 4.3 3.2 4.8 4.0 oak-pine (1) ab b d ab bcde abcd b bc Edge between aspen (1) and 4.0 3.2 3.5 4.6 4.3 4.5 4 0 oak-maple-aspen-cherry (l) ab abc ab cde bcd ab bc Oak (1) 4.2 5.4 4.1 3.7 3.8 4.5 4.3 b d ab abcd bcd ab bcd Edge between upland grass 2.9 5.0 5.0 6.0 4.6 4.1 4.6 and aspen-maple (2) ab cd b e cd ab bcd Aspen or aspen-maple (l) 4.4 4.7 5.1 5.4 4.6 4.4 4.8 b bcd b de cd ab cd Edge between various deciduous types (1) and 4.5 5.3 4.7 4.3 5.2 7.2 5.2 various types (2) b cd b bcde d c d * Use = /(C7M) + 0.5, where C = number of crossing groups found in the combination on the count, and M = miles of route along which the combination occurred. ** Presence of letters in a column indicates significant differences. Means that do not have a letter in common differ significantly at the 5% level by Tukey's HSD test. 24 Tab le 9. Results of analysis of summer deer use on the 9-Mile Study Area, 1975—1980. S ource of Mean V ariation df Square F Significance Mai I). effects CO Ver and size combination 3 3.0679 5.690 P = 0.002 Ye ar 5 3.4260 6.355 P < 0.001 Inta raction CD\zer and size year 15 1.0169 1.886 P = 0.049 Res idual error 48 0.5391 \ 25 dependent on time—since-treatment (interaction P‘<0.05). Focusing first on mean utilization across years, the edge between reproduction aspen-oak mixture or oak and various mature cover types (or a very small amount of marsh and bog), and the mature aspen—oak mixture combination were used to a significantly (P‘<0.01 and P<<0.05, respectively) greater extent than the pole size red pine or oak—pine mixture combination (nearly all red pine), and to a greater extent, though not significant, than the reproduction aspen-oak mixture (Table 10). Given the presence of the interaction, comparisons among combinations were also made by year. A moderately significant (P<oH mm onn no thCoonHCmHm nomch COEEoo CH nonnoH o o>on nOC 0p nonn mCoo: .mooConomme nCoonHCmHm monooHpCH CEsHoo o CH mnonnoH mo oOComonm *4 .ponnoooo CoHnoCHnEoo onn COHCB mCoHo onson mo moHHE n : pCo .nCCoo onn Co COHnoCHnEoo onn CH pCDOm mononm mCHmmono Ho nonECC n O ononB .m.o + A:\OV\ u own « n n n mon no Ame mommn w.m m.m H.v n.m m.m m.m m.m mCOHno> pCo AHV xoo no xoonCommo Coanon ompm n no o v.m m.v m.m o.m m.m o.m w.m Amy noonCommC no o no H.m v.m v.m m.m v.m m.m m.m AHV noouCommC o no «*QM H.~ H.m H.N s.m H.~ m.m o.m Ame mandnnoo no wand oom Coo: ome moan man ohmH oan mean COHnoCHnEOO mmoHO oNHm pCo omwe no>ou noonom .mnoom mCHCHoEon onn no nooo CH oonnn pCo .mnmn CH noon .mhmH CH mnCCoo osn "moNHm onEom .ommHlmomH .oond mpnnm oHHzlm onn Co mCOHnoCHnEoo Azom no oHOQ u N .COHnooponmon n Hv mmoHo oNHm UCo omhn no>oo noonom Ho noop mn sows noEEom Coo: .OH oHnoB 27 Table 11. Results of analysis of summer deer use on the Sharon Study Area, 1975-1980. Source of Mean Variation df Square F Significance Main effects Cover and size combination 7 19.9440 25.107 P < 0.001 Year 5 3.6768 4.628 P < 0.001 Interaction Cover and size X year 35 0.7850 0.988 P = 0.497 Residual error 120 0.7944 28 There was no evidence to suggest that the differences were dependent on time-since-treatment (interaction P:>0.10). Examining then just the comparisons among the eight combinations of use averaged across years (Table 12), the edge between reproduction oak-maple-aspen-cherry mixture and mature aspen-maple-oak mixture had the highest mean use index, which was significantly (P‘<0.05) greater than all the other seven combinations. The least utilized combination, edge between mature aspen-pine mixture and mature aspen- maple-oak mixture, was significantly (P1<0.05) different than all others also. The pole size jack pine had the second highest mean use index, which was significantly (P‘<0.05) greater than that for most other combinations, including the two reproduction size combinations (oak-maple-aspen-cherry and aspen or aspen-maple). Utilization of the reproduction size oak-maple—aspen-cherry cover type was much higher when juxtaposed with mature stands. 29 .nmon 0mm m.>oxse xn Ho>oH am on» no thCoonHCmHm nouuHo COEEoo CH nouuoH o o>on no: on nonu mCoo: .moOConomeo uCoonHCmHm monooHoCH CEsHoo o CH mnonuoH no ooCowond «.6 .ponnsooo COHuoCHnEoo onn CUHn3 mCoHo onson mo moHHE u : oCo .nCCoo on» Co CoHuoCHnEoo on» CH pCCOw mononm mCHmmono mo nonEsC u U onon: .m.o + »:\O~\ u om: 6 c o o o n n HNV xoOIoHdoE|Codmo 6.m n.m 6.6 6.6 6.m 6.m 6.6 one inc Hnnmzoiaoamo IoHdoEIxoo Cooznon ompm u on 0 on no no 6.6 H.6 6.6 ~.m ~.6 6.m 6.6 Ame mend none on on no no no o HNV HHo .momwu Coozuon o.m >.m m.~ 6.6 «.6 m.~ H.6 momco pCo woman msoHno> n no ono no o no m.m H.m m.m n.m o.m 6.m w.~ nmsnn pCoHdD n ono on no o o monlnmnoe no nmann pCoHas m.m m.m m.m m.m w.N 6.~ n.m nnHB momco mopsHoCH .mCoHuoCHnEoo nonno .omH: n ono ono no o o H.m m.m ~.m 6.m N.m H.m o.~ AHV oHdoEICodmo no codm< n no ono no no o H.m m.~ ~.m >.m o.m o.m h.m AHV wnnonoaCommouoHdoEIxoo o o o o o 66o Amy noononoEICodwo pCo N.N m.H n.H n.~ >.N m.H m.m Amy oCHQICono Cooznon omcm Coo: ommH ean man nan ohmH mhmH COHuoCHnEOO mmoHO oNHm pCo odhe no>ou noonom .mnoo> mCHCHoEon onu Ho nooo CH nCoH oCo .mbmH CH oonnu .man CH mnnnoo 03n “moNHm oHoEom .omenmomH .oon< hpsnm Cononm on» Co mCOHuoCHnEoo Azom no oHoo u m .CoHnosoondon u H. mmoHo oNHm pCo omxu no>oo umonOH mo noon an «own noEECm coo: .NH oHnoe DISCUSSION The use index pertains to the period from late afternoon (dragging) until the following morning. Only a period of approximately 6-8 hours during the middle of the day is not sampled. The majority of feeding activity occurs during the daylight hours, with peaks associated with sunrise and sunset, a typical crepuscular feeding pattern (Bienz 1981). Bienz found that the greatest allotment of time for feeding occurred at sunset. Since the period for which use has been estimated encompasses the crepuscular feeding peaks typical of deer, movements to, from, and through vegetation types used for feeding should be reflected in the index. Therefore, the index should reflect the combinations used for feeding. The use index is based on locations of deer crossing groups along a fixed route. Therefore, activities associated with movement will be emphasized. Movements to and from bedding or feeding sites should be well represented. Habitat used for bedding may, perhaps, be underrepresented. However, the use estimate cannot be broken down into the various activities. The index does have the advantage of not only representing use for feeding, but also use for cover and possibly fawning. As mentioned, this study was part of a larger project 30 31 conducted by the Michigan Department of Natural Resources. A major effort was made by the Department to monitor the deer populations on the six research units used here, along with two others, and not solely by means of summer roadside track counts. These findings should be considered along with those forthcoming from other aspects of the larger study. Cover type had a great influence on habitat use by deer, especially when considered with stand maturity. Examination of use of the various cover type and size class combinations on each study area bears this out. Several instances were found where combinations of the same maturity, but different cover types, differed significantly in terms of their use. Juxtaposition of stands was also found to influence use in some cases. The use of cut or disturbed areas, such as the reproduction size stands produced by treatment on these study areas, and the high forage production in these areas have been reported in several studies (Westell 1954, Gysel 1957, Krefting 1962, Halls and Alcaniz 1968, Stormer and Bauer 1980, Bennett et al. 1980). The immature aspen or aspen- maple mixture combination, found on five of the study areas, had consistently high use across the areas with the exception of the Sharon Study Area. Most of the reproduction size stands were used fairly heavily, relative to the other combinations. However, the immature oak and oak-pine mixtures appeared to be avoided to some extent by deer, even on the 25% treated study areas where fewer immature stands were 32 available. All reproduction size combinations were avoided on the 75% treatment level study areas. Also, some evidence was found of increasing use of some of the reproduction size, or regeneration, stands with time-since-treatment. Because of the treatment, virtually all the reproduction size stands were 0-2 growing seasons old at the beginning of this study in 1975. Thus, these stands were 5-7 growing seasons old the final year of the study, and any time-since-treatment effect should be related to this progression of growth. As mentioned, the immature combinations which regenerated after treatment were avoided somewhat by deer on the 75% treatment level study areas, 9-Mile and Sharon. However, the edges between immature and mature forest stands were used extensively, regardless of cover type. Remember, the 75% areas tended to have much larger, continuous blocks of treated area, and therefore less of the edge between reproduction and mature size stands was created. The reduced availability of the preferred edge may have resulted in more concentrated use of what was there. Stormer and Bauer (1980) found that the size of a clear-cut may affect the degree to which it is used. Deer avoided the center of a 4.05 ha clear-cut in their study. While the relationship between use of reproduction size stands and the size of treated block was not examined in this study, use of the immature stands on the 75% treatment level study areas was low relative to the other combinations, even though forage production should have been high. The large, 33 continuous blocks of cleared area were avoided somewhat, but not necessarily because of their size. The high treatment level may have produced too much reproduction size forest on these areas. The immature stands were utilized, but at a low level relative to the vast amount available on these areas. Use of the immature stands on the 50% treatment level study areas, Lanes Lake and Russell Lake, was moderately high, depending on the cover type. The size of the treated blocks on these two areas were clearly not small (Figure 1), but the availability of reproduction size stands was reduced. However, treatment in small blocks is recommended for maintaining the desirable intolerant forest cover types. This will also provide a span of maturities, which will provide the desired edge, and a sufficient amount of immature stands with high forage production. Additional research on the optimal size for clear-cuttings is needed. The mature oak-pine mixture combination, and the mature oak with a sapling pine understory received consistently high use. The reason for this, whether this use was either forage or cover related, is not clear. These types should also be promoted and preserved by management practices for summer deer range, though further research on their use may be needed. Of course, some of the types, which were avoided or at least not preferred in this study of summer use, are important components of year-round deer range. The mature swamp conifers and mature oak are two examples, important for winter cover and mast production in the fall, respectively. LITERATURE CITED LITERATURE CITED Bennett, C. L., Jr., E. E. Langenau, Jr., G. E. Burgoyne, Jr., J. L. Cook, J. P. Duvendeck, E. M. Harger, R. J. Moran, and L. G. Visser. 1980. Experimental management of Michigan's deer habitat. Trans. North Am. Wildl. and Nat. Resour. Conf. 45:288-306. Bienz, C. S. 1981. A telemetric appraisal of daily and seasonal activity of female white-tailed deer in a northern Michigan enclosure. M.A. Thesis. Northern Michigan Univ., Marquette. 53pp. Daniel, W. S., and D. B. Frels. 1971. A track-count method for censusing white-tailed deer. Texas Parks and Wildl. Dep. Tech. Ser. No. 7, Austin. 18pp. Gysel, L. W. 1957. Effects of silvicultural practices on wildlife food and cover in oak and aspen types in northern Michigan. J. For. 55:803-809. Halls, L. K., and R. Alcaniz. 1968. Browse plants yield best in forest openings. J. Wildl. Manage. 32:185-186. Karns, P. D. 1980. Winter--the grim reaper. Pages 47-53 in R. L. Hine and S. Nehls, eds. White-tailed deer population management in the north central states. Proc. 1979 Symp. North Cent. Wildl. Soc. 116pp. Kohn, B. E., and J. J. Mooty. 1971. Summer habitat of white-tailed deer in north-central Minnesota. J. Wildl. Manage. 35:476-487. Krefting, L. W. 1962. Use of silvicultural techniques for improving deer habitat in the Lake States. J. For. 60: 40-42. Krefting, L. W., and R. L. Phillips. 1970. Improving deer habitat in upper Michigan by cutting mixed-conifer swamps. J. For. 68:701-704. Krull, J. N. 1964. Deer use of a commercial clear-cut area. New York Fish and Game Jour. 11(2):115-118. 34 35 McCaffery, K. R. 1976. Deer trail counts as an index to populations and habitat use. J. Wildl. Manage. 40:308- 316. ' McCaffery, K. R., and W. A. Creed. 1969. Significance of forest openings to deer in northern Wisconsin. Wis. Dep. Nat. Resour. Tech. Bull. 44. 104pp. McCaffery, K. R., J. Tranetzki, and J. Piechura, Jr. 1974. Summer foods of deer in northern Wisconsin. J. Wildl. Manage. 38:215-219. McNeill, R. E. 1971. Interactions of deer and vegetation on the Mid-Forest Lodge and Gladwin Game Refuge. Ph.D. Thesis. Univ. of Michigan, Ann Arbor. l98pp. Moen, A. N. 1976. Energy conservation by white-tailed deer in the winter. Ecology 57:192-198. Moen, A. N. 1978. Seasonal changes in heart rates, activity, metabolism, and forage intake of white-tailed deer. J. Wildl. Manage. 42:715-738. Rogers, L. L., J. J. Mooty, and D. Dawson. 1981. Foods of white-tailed deer in the Upper Great Lakes Region--a review. U.S. Dep. Agric. For. Serv. Gen. Tech. Rep. NC-65. 24pp. Searle, S. R. 1971. Linear models. John Wiley and Sons, Inc., New York. 532pp. Silver, H., N. F. Colovos, J. B. Holter, and H. H. Hayes. 1969. Fasting metabolism of white-tailed deer. J. Wildl. Manage. 33:490-498. Silver, H., J. B. Holter, N. F. Colovos, and H. H. Hayes. 1971. Effect Of falling temperature on heat production in fasting white-tailed deer. J. Wildl. Manage. 35:37- 46. Sokal, R. R., and F. J. Rohlf. 1981. Biometry. 2d ed. W. H. Freeman and Co., San Francisco. 859pp. Steel, R. G. D., and J. H. Torrie. 1960. Principles and procedures of statistics. McGraw—Hill Book Co., Inc., New York. 481pp. Stormer, F. A., and W. A. Bauer. 1980. Summer forage use by tame deer in northern Michigan. J. Wildl. Manage. 44:98-106. 36 Ullrey, D. E., W. G. Youatt, H. E. Johnson, L. D. Fay, B. L. Schoepke, and W. T. Magee. 1970. Digestible and metabolizable energy requirements for winter maintenance of Michigan white-tailed does. J. Wildl. Manage. 34: 863-869. Veatch, J. O., L. R. Schoenmann, Z. C. Foster, and F. R. Lesh. 1927. Soil survey of Kalkaska County, Michigan. U.S. Dep. Agric., Washington, D. C. 44pp. Veatch, J. O., L. R. Schoenmann, and J. W. Moon. 1924. Soil survey of Roscommon County, Michigan. U.S. Dep. Agric., Washington, D. C. 27pp. Verme, L. J. 1965. Swamp conifer deeryards in northern Michigan: their ecology and management. J. For. 63: 523-529. Verme, L. J. 1969. Reproductive patterns of white-tailed deer related to nutritional plane. J. Wildl. Manage. 33:881-887. Westell, C. E., Jr. 1954. Available browse following aspen logging in lower Michigan. J. Wildl. Manage. 18:266- 271. "I(fliififlgifliiiyijiifliiiIIIES 697