https://cpw.cvlcollections.org/files/original/9bf94f6a495ad8ee83009548cd953a43.pdf ac3d0f9245127adf627b21d5dcbd2966 PDF Text Text The research in this publication was partially or fully funded by Colorado Parks and Wildlife. Dan Prenzlow, Director, Colorado Parks and Wildlife • Parks and Wildlife Commission: Marvin McDaniel, Chair • Carrie Besnette Hauser, Vice-Chair Marie Haskett, Secretary • Taishya Adams • Betsy Blecha • Charles Garcia • Dallas May • Duke Phillips, IV • Luke B. Schafer • James Jay Tutchton • Eden Vardy �The Journal of Wildlife Management 75(6):1443–1452; 2011; DOI: 10.1002/jwmg.168 Research Article Biological and Socio-Economic Effects of Statewide Limitation of Deer Licenses in Colorado ERIC J. BERGMAN,1 Colorado Division of Wildlife, 317 W. Prospect Road, Fort Collins, CO 80526, USA BRUCE E. WATKINS, Colorado Division of Wildlife, 2300 S. Townsend Avenue, Montrose, CO 81401, USA CHAD J. BISHOP, Colorado Division of Wildlife, 317 W. Prospect Road, Fort Collins, CO 80526, USA PAUL M. LUKACS, Colorado Division of Wildlife, 317 W. Prospect Road, Fort Collins, CO 80526, USA MARY LLOYD, Colorado Division of Wildlife, 6060 Broadway, Denver, CO 80216, USA ABSTRACT We evaluated the biological and socio-economic effects of statewide limitation of mule deer (Odocoileus hemionus) hunting licenses, which began in Colorado in 1999. We implemented a before-aftercontrol-impact (BACI) analysis of annual helicopter sex and age class surveys, collected as part of the Colorado Division of Wildlife’s routine monitoring, to assess changes in adult male/adult female ratios and fawn/adult female ratios in response to this change in harvest management. Following statewide limitation and reduction of license sales (1999–2006), we observed increases in adult male/adult female ratios of 7.39 (SE ¼ 2.36) to 15.23 (SE ¼ 1.22) adult males per 100 adult females in moderately limited areas and of 17.55 (SE ¼ 3.27) to 21.86 (SE ¼ 2.31) adult males per 100 adult females in highly limited areas. We simultaneously observed reductions in fawn/adult female ratios in newly limited areas by as much as 6.96 (SE ¼ 2.19) fawns per 100 females, whereas in areas that had previously been limited we observed stabilization of fawn/adult female ratios at levels lower than levels observed under the unlimited harvest management structure. An immediate decline of $7.86 million in annual revenue stemmed from the change in harvest management, but revenue subsequently rebounded. This study provides preliminary evidence of potential effects that other state and provincial wildlife management agencies may face as they consider shifting mule deer harvest management towards limited license scenarios. ß 2011 The Wildlife Society. KEY WORDS before-after-control-impact (BACI), Colorado, harvest, hunting, license, management, mule deer, Odocoileus hemionus. For almost all wildlife management, assessing the effect of decisions, learning from the experience and applying the knowledge gained to future decisions is a desirable sequence for the wildlife management process. This sequence of events is a fundamental tenet of adaptively managed systems (Williams et al. 2002). However, few wildlife management plans have been formally designed, or structured, as adaptive management programs (Montana Fish, Wildlife and Parks 2001, Mason et al. 2006, Nichols et al. 2007). Despite this, managers are often challenged to retroactively assess the effects of decisions on a managed system. In the United States, some of the most pressing wildlife management issues pertain to harvested species and the associated decisions of setting harvest objectives and issuing harvest licenses. Such systems generally have multiple goals—biological, economic, and social—yet few formal examples exist that are focused on how changes in harvest objectives have met any such goals (Nichols et al. 2007). Received: 7 April 2010; Accepted: 25 January 2011; Published: 15 July 2011 1 E-mail: eric.bergman1@state.co.us Bergman et al. � Deer License Limitation During the late 1980s and early 1990s, a noticeable and undesired decline in mule deer populations occurred throughout the Rocky Mountain West (Fig. 1; Gill 2001). Largely in response to this decline, a restructuring of the mule deer management system in Colorado occurred. Prior to 1999, 4 out of 40 Game Management Units (GMUs) in western Colorado were managed within a limited license system (Fig. 2; Bishop et al. 2005). Motivated by the desire to increase the number of older age class males (�3.5 yr old), and subsequently hunter success and satisfaction in these units, a limited number of licenses to hunt in these particular units were allocated through an annual drawing. The remaining GMUs in Colorado, prior to 1999, were managed within an unlimited license structure that allowed unlimited over-the-counter sales of antlered deer hunting licenses. In evaluation of this small-scale limitation, Bishop et al. (2005) documented that changes in the population structure occurred but concluded that limiting sales of antlered deer licenses was useful as a tool for improving hunt quality, as opposed to achieving population performance objectives. Despite this earlier limitation process, and stemming from the noted decline, a remaining biological and population goal 1443 �Figure 1. Historical (1987–2006) estimates of Colorado’s mule deer population based on pooled model estimates from all data analysis units (DAUs). A region wide decline in mule deer populations was observed from the late 1980s through the early 1990s. Colorado implemented a statewide limited harvest management program on mule deer between the 1998 and 1999 hunting seasons. of the Colorado Division of Wildlife (CDOW) was to increase mule deer numbers. However, a priority of some hunters at the time was the desire to specifically see more adult male deer (�1.5 yr old). A component of the hunting public expressed concern that adult male/adult female ratios had declined to such low levels that a large portion of the adult female population was not getting bred during the fall breeding season. A related concern was that fawns were being born at suboptimal times (primarily too late) due to the low number of mature adult male deer and that females were bred over an increasingly long time period. In summary, some of Colorado’s hunters were largely dissatisfied with the existing harvest management structure and objectives. They subse- quently communicated a willingness to sacrifice both hunting opportunity and flexibility not only to fix these perceived problems, but also to increase the quality of their hunting experiences. Stemming from these issues and concerns, but despite any evidence supporting or refuting the biological concerns, the CDOW implemented a statewide limitation on mule deer hunting licenses in 1999, which was a fundamental change in the structure of its mule deer management plan. Our goal was to assess the result of statewide limitation of licenses, both in terms of population goals and economic return. Opportunistically, we also gauged hunter satisfaction to determine if changing harvest management structure helped meet hunter desires. These results could be used to inform future decisions by optimizing harvest under different, pre-identified constraints on fawn/adult female and adult male/adult female ratios. Similarly, we suspect other states face similar issues regarding mule deer harvest management and the Colorado experience may provide insight as to potential benefits and challenges in the process. We evaluated the biological and economic goals, and addressed the social outcomes, of a statewide harvest management decision in the context of the motivation behind the decision. As a case example, we focused on mule deer (Odocoileus hemionus) management in Colorado, an example similar to many wildlife management situations in other states. STUDY AREA For wildlife management purposes, Colorado has been divided into 55 geographically defined data analysis units (DAUs). These DAUs were established with the objective of estimating population parameters of interest for distinct Figure 2. Geographic distribution of Colorado’s management regions and deer data analysis units (DAUs) we used. Heavy black lines separate regions whereas light gray lines designate DAU boundaries. We did not use DAUs in the eastern part of the state (shaded black). Data analysis units shaded with light gray represent those that were brought under a limited license structure during the early 1990s, whereas all other DAUs were under an unlimited license structure through 1998, at which point they also were incorporated into the limited license structure. 1444 The Journal of Wildlife Management � 75(6) �segments of Colorado’s deer herd. This objectives and the DAU concept have largely been validated with radio-marked and relocated deer throughout Colorado (B. Watkins, Colorado Division of Wildlife, unpublished data). Data analysis units were made up of �1 GMU. Game management units were used to distribute hunter numbers and harvest management within DAUs. Game management unit boundaries did not necessarily reflect boundaries on deer movement, as with DAUs. With the exception of one DAU, management prescriptions for GMUs within the same DAU did not vary. We focused on the 40 DAUs in western Colorado, defined as any DAU west of Interstate 25, which ran north–south along the eastern front of the Rocky Mountains (Fig. 2). Although eastern Colorado (i.e., east of Interstate 25) did support mule deer, the objectives and harvest management in this part of the state were not seamless with those of western Colorado. In particular, a large degree of spatial overlap between mule deer and white-tailed deer (Odocoileus virginianus) occurred in eastern Colorado. Additionally, a much higher proportion of land in eastern Colorado fell under private ownership and was managed for agriculture. Thus, statistical comparisons of deer management actions between eastern and western Colorado were inappropriate. We further segregated the study area into 4 regions, in accordance with the wildlife management structure within the CDOW. The northwest (NW) region was composed of DAUs primarily north of the Colorado River and west of the continental divide. The northeast (NE) region was composed of DAUs primarily north of Interstate 70 and east of the continental divide. Alternatively, the southwest (SW) region was composed of DAUs located south of the Colorado River and west of the continental divide. The southeast (SE) region was primarily composed of DAUs to the east of the continental divide and south of Interstate 70. The Colorado River and the Continental Divide acted as barriers to deer movement between the NW and other regions (B. Watkins, unpublished data). Interstate 70 traversed much of Colorado’s high country, with minimal deer movement between the NE and SE regions. For parts of the state where Interstate 70 did not accurately reflect biological boundaries for deer, DAU and regional boundaries were adjusted to reflect animal movement. The segregation of the state into regions allowed the CDOW the flexibility to make regional and local adjustments to license numbers in response to population change or harvest management needs, yet stay within the limited license structure. In 1999, a statewide limitation on deer license sales was implemented and all licenses were distributed via the drawing structure. Under the limited harvest management structure the cap on the number of available licenses was allowed to adjust on an annual basis depending on observed DAU data and if those data reflected a population that was above or below DAU objectives. In the case of GMUs where the CDOW allocated more licenses than there were applicants, the remainder of licenses were sold over-the-counter on a first come, first served basis. With the exception of a few game-damage licenses issued on private land, hunters typically obtained one license per year. From 1984 to 1998, the CDOW sold between 149,616 and 246,797 deer licenses annually. In 1999, the number of licenses sold was reduced to 80,649, representing a 46–67% reduction (Fig. 3). Of note, although the dates and lengths of hunting season varied through time, hunting seasons were consistently completed prior to the annual breeding period for deer. Thus, the vulnerability of adult male deer, which increases greatly during the breeding period, did not change between different harvest management structures. Our analysis focused on DAUs as the sample unit. However, public goals and desires sometimes result in inconsistent GMU harvest prescriptions within a DAU even though it may be problematic from a population goal standpoint. For our analysis, the inconsistencies of mixed harvest management objectives within one DAU manifest itself in one case. Data analysis unit 19 was composed of 2 GMUs (Units 61 and 62). Limited buck harvest was implemented in Unit 61 in 1992, whereas it remained unlimited in Unit 62 until 1999 (Bishop et al. 2005). For our analysis, and to accurately reflect the history of these 2 GMUs, we treated each as a separate DAU. Each year as part of the CDOW’s routine mule deer management, post-hunt sex and age class helicopter surveys were conducted in each DAU. Initiated upon completion of the final hunting season for the year (early Dec), all surveys were completed by mid-January to minimize the potential for misclassification of adult males as adult females when antlers METHODS Prior to 1999, hunters who unsuccessfully entered the limited license draw to hunt in a preferred GMU earned a preference-point, which increased their odds of drawing a license to hunt in a preferred unit during later years. These hunters could also purchase an over-the-counter license to hunt in an unlimited GMU during the years in which they were unsuccessful in the drawing. Hunters who were not interested in hunting in limited units did not need to participate in the drawing to purchase an over-the-counter license. Bergman et al. � Deer License Limitation Figure 3. Historical perspective (1984–2006) on the total number of hunters (gray line), total number of deer harvested (solid black line), and total number of adult male deer harvest (dashed black line) throughout the observed decline in mule deer populations as well as through implementation of the statewide limited harvest management plan. Antlerless and fawn harvest can be viewed as the difference between total harvest and adult male harvest. We generated harvest estimates via a stratified random sample of deer hunting license holders. 1445 �shed and to minimize the bias of differential overwinter mortality between sex and age classes on ratios. Surveys typically consisted of flying non-random paths through known deer winter range, with the express purpose of encountering as many deer as possible. Surveys flights were typically flown between 50 m and 150 m above the ground. During flights, all deer that were encountered were counted and classified by sex. All adult male deer were categorized as yearlings (12–23 months old), 2-yr olds (24–35 months old) or mature (�36 months old) based on antler size and morphology. All antlerless deer were categorized as adult females or fawns. At this time of year fawns are readily discernible from adult females based on body size and morphological development. Adult male/adult female and fawn/adult female ratios were calculated each year and entered into the CDOW Deer, Elk, and Antelope Management (DEAMAN) database. For our analysis, we pooled different age classes of adult male deer, allowing calculation of total adult male/adult female ratios. We used data collected between 1984 and 2006 for our analysis. We used 923 survey ratios for each of the adult male/adult female and fawn/adult female analyses (approx. 42 ratio estimates per year). Hunter satisfaction was opportunistically gauged via hunter satisfaction questionnaires. Prior to the statewide limitation process, one hunter satisfaction survey was conducted. After the limitation process a hunter satisfaction survey was not conducted. However, during 2007 a survey was conducted as part of Colorado’s Big Game Season Structure process. Thus, surveys expressly designed to capture changes in hunter opinion regarding the change in harvest management that occurred between 1998 and 1999 were not conducted. Whereas the existing survey data provided insight into the issues concerning hunters at the time of each survey, the inference that can be drawn from them for tracking changes in opinion in regards to changes in harvest management is limited. Statistical Methods The stepwise pattern in which limitation of deer license sales was implemented, complemented by the annual population monitoring that occurred both before and after the limitation process, lent itself to a before-after-control-impact (BACI) type analysis (Green 1979). Classic BACI design analyses are often confounded by the lack of temporal and spatial replication, especially when natural cycles or trends exist within the population of interest (Hurlbert 1984). However, extensions of the BACI design that reduce the potential for falsely concluding that natural events were the result of a treatment have been proposed (Stewart-Oaten et al. 1986, Underwood 1994). In our study, the asymmetric implementation of limited harvest management served as temporal replication. Although no traditional controls (i.e., DAUs with no hunting) were available to be incorporated into the analysis, we refer to some DAUs as pseudo-controls as harvest management in these DAUs was held static during the periods of interest while harvest management was altered in other areas. Between 1984 and 1998, deer license numbers were not limited in most of the DAUs in the state, thereby allowing 1446 them to serve as pseudo-controls and allowing us to compare population parameters between DAUs with limited and unlimited license sales for these years. Between 1999 and 2006, those DAUs that were already managed under a limited license structure experienced no additional change to the harvest management. Thus, they were then able to function as pseudo-controls during the later period while harvest management in the remainder of the state changed (Fig. 4). We used mixed models (PROC MIXED) in SAS (SAS Institute, Inc., Cary, NC) to measure the effect of limitation of deer license sales and allocation on adult male/adult female and fawn/adult female ratios. We identified a priori suites of models for each of the response variables and compared model results were compared using Akaike’s Information Criterion corrected for small sample size (AICc; Burnham and Anderson 2002). These analyses allowed us to determine which population level effects occurred as a result of reducing the number of deer licenses sold each year. For these analyses, adult male/adult female ratios and fawn/adult female ratios were the response variables. We allowed mean changes in ratios to vary by DAU, region, harvest management type, and year depending on model structure. Data analysis unit entered all models as a random effect. We also allowed the effect of adult female harvest to enter the model for any given DAU as an additional explanatory variable. Allowing harvest of adult females to enter models allowed us to determine if changes in the ratios of interest were due to changes in harvest management structure or if changes in ratios were simply caused by increasing or decreasing the adult female portion of the population via harvest. We entered harvest management type in 1 of 4 ways (Fig. 4). As part of the limitation process, some DAUs became highly limited (i.e., antlered deer license numbers were cut by 75–95% of mean pre-limitation levels), whereas others were only moderately limited (i.e., hunting license were cut by 50–60% of average pre-limitation levels). During the early limitation period (1984–1998), all limited DAUs fell into the highly limited category and we thus classified them as group 1. We subsequently categorized all other DAUs as group 2 (unlimited) during this period. During the later period (1999–2006), we reclassified DAUs in group 1 along with 5 of the newly highly limited DAUs as group 3 (75–95% limited). The remaining 35 of newly limited DAUs became moderately limited and we reclassified them as group 4 (50–60% limited). Hence, during the later period (1999– 2006), we reclassified all DAUs from groups 1 and 2, with group 3 being composed of highly limited DAUs and group 4 being composed of moderately limited DAUs. RESULTS As a result of limiting antlered deer license sales, we observed increases in adult male/adult female ratios and decreases in fawn/adult female ratios in general, although the overall magnitude of change varied by region (Figs. 5 and 6). For models pertaining to changes in adult male/adult female ratios, most AICc model weight (0.947) went to the model that allowed ratios to vary by DAU, region, harvest manageThe Journal of Wildlife Management � 75(6) �Figure 4. Schematic representation of how deer harvest management was structured in Colorado during our period of interest (1984–2006) and how changes in management lent itself towards a staggered analysis of a management action. ment type with a region � management interaction and also accommodated harvest of adult females (Table 1). For this model, we estimated variation in our random effect (DAU) to be 17.74 (SE ¼ 7.04) adult males per 100 adult females. The model receiving the second highest weight (0.050) was simpler in structure by not including harvest of adult females, but the model was not competitive, with a DAICc value of 5.95 (Table 1). Due to the overall lack of competitiveness among models in this suite, model averaging was not necessary to determine relative importance of individual covariates. Figure 5. Observed number of adult males per 100 adult females in data analysis units (DAUs) managed under the unlimited management structure during the 1980–1998 time period (A) versus observed number of adult males per 100 adult females in DAUs managed under the limited harvest management structure during the 1980–1998 time period (B). Figure 6. Observed number of fawns per 100 adult females in data analysis units (DAUs) managed under the unlimited management structure during the 1980–1998 time period (A) versus observed number of fawns males per 100 adult females in DAUs managed under the limited harvest management structure during the 1980–1998 time period (B). Bergman et al. � Deer License Limitation 1447 �Table 1. Model selection results, model weights (wi ), and the number of parameters (k) for models comparing adult male/adult female ratios in western Colorado, 1984–2006. Model structure allowed models to vary by data analysis unit (DAU), region (RGN), management type (MGMT) or to include harvest of adult females (DHRV). Model selection was based on Akaike’s Information Criterion adjusted for small sample sizes (AICc). Model ka AICc DAICc wi DAU RGN MGMT RGN � MGMT DHRV DAU RGN MGMT RGN � MGMT DAU RGN MGMT DHRV DAU RGN MGMT DAU MGMT DHRV DAU MGMT 65 64 49 48 45 44 4356.4 4362.3 4367.8 4374.0 4374.7 4380.9 0.0 5.9 11.4 17.6 18.3 24.5 0.947 0.050 0.003 0.000 0.000 0.000 a Accounting for parameters is as follows: DHRV ¼ 1, RGN ¼ 4, MGMT ¼ 4, RGN � MGMT ¼ 16, and DAU ¼ 40. The top 2 models for fawn/adult female ratios were similar to those for adult male/adult female ratios, but the order was reversed (Table 3). Most of the model weight (0.637) from the fawn/adult female ratio analysis went to the model that allowed ratios to vary by DAU, region, and harvest management type and had a region � management interaction. For this model, we estimated the variation in our random effect (DAU) to be 59.00 (SE ¼ 6.9) fawns per 100 adult females. The second best model, which had the same structure but included adult female harvest, had a DAICc score of 2.00. Due to the manner in which we computed AICc values, models that have the exact same model structure with the exception of the addition of one parameter can only differ by a maximum of 2.0 AICc points (Burnham and Anderson 2002). Thus, it is evident that the inclusion of adult female harvest added no additional information and was inconsequential to model performance. In the case of fawn/adult female ratio models, the third best model was simply composed of DAU, harvest management type, and region. However, the DAICc score for this model was 3.91. The diminished support for this model is also reflected by the low model weight (0.091). In contrast to the observed increases in adult male/adult female ratios, we observed immediate declines in fawn/adult In general, we observed increases in adult male/adult female ratios as a result of limiting deer licenses (Table 2). However, this pattern was more strongly apparent during the later of the 2 occasions that limitation was implemented (1984–1998 vs. 1999–2006). During the first period, results were mixed (Table 2). Only the NW and SW regions had DAUs switch from unlimited to limited license harvest management. In those DAUs, relative to the pseudo-control DAUs, the effect of limiting licenses during the first period was an increase of 5.68 (SE ¼ 2.20) adult males per 100 adult females in the NW region, but we did not detect a change in the SW (estimate ¼ �2.46, SE ¼ 3.47). By the end of the second period (2006), the effect of high limitation on adult male/adult female ratios in the highly limited group of DAUs was more consistent, as we observed increases of 21.86 (SE ¼ 2.31) and 17.55 (SE ¼ 3.27) adult males per 100 females in the NW and SW regions, respectively. During the later period, when the remainder of DAUs switched from unlimited harvest management to the moderately limited license structure, we observed a consistent effect. We observed increases in the number of adult males per 100 adult female ratios in all regions. These increases ranged between 7.39 (SE ¼ 2.36) adult males in the SE and 15.23 (SE ¼ 1.22) adult males in the SW. Table 2. Coefficient estimates of changes in numbers of adult males per 100 adult females as a result of manipulating the number of hunting licenses and hunting license allocation in Colorado in 1999 and observed ratio statistics for each management type and period. We standardized all data analysis unit (DAU) estimates against those DAUs with an unlimited license allocation structure from 1984 until 1998. Reported change in ratio estimates are changes in mean values that we allowed to vary by the best model as determined by Akaike’s Information Criterion value that had been adjusted for small sample size (AICc) (DAU RGN MGMT RGN � MGMT DHRV). Change in ratio (SE) by management type Region NW NW NW NW SW SW SW SW NE NE NE NE SE SE SE SE 1448 Period Unlimited 1984–1998 1984–1998 1999–2006 1999–2006 1984–1998 1984–1998 1999–2006 1999–2006 1984–1998 1984–1998 1999–2006 1999–2006 1984–1998 1984–1998 1999–2006 1999–2006 0.00 Moderate 10.81 SE High Observed ratio statistics SE 5.68 21.86 2.20 2.31 �2.46 17.55 3.47 3.27 1.40 0.00 Max. Min. x 48.1 37.8 53.4 54.0 33.8 17.9 50.5 60.4 50.0 6.1 8.7 22.9 19.5 3.1 8.0 17.9 10.2 11.3 19.0 23.1 39.0 30.3 14.9 11.4 30.6 29.3 28.2 15.23 1.22 10.91 2.19 51.8 52.4 11.4 11.3 40.0 18.2 7.39 2.36 78.6 15.1 33.3 0.00 0.00 The Journal of Wildlife Management � 75(6) �Table 3. Model selection results, model weights (wi ), and the number of parameters (k) for models comparing fawn/adult female ratios in western Colorado, 1984–2006. Model structure allowed models to vary by data analysis unit (DAU), region (RGN), or management type (MGMT) or to include harvest of adult females (DHRV). Model selection was based on Akaike’s Information Criterion adjusted for small sample sizes (AICc). Model ka AICc DAICc wi DAU RGN MGMT RGN � MGMT DAU RGN MGMT RGN � MGMT DHRV DAU RGN MGMT DAU RGN MGMT DHRV DAU MGMT DAU MGMT DHRV 64 65 48 49 44 45 4848.5 4850.5 4852.4 4854.4 4859.3 4861.3 0.0 2.0 3.9 5.9 10.8 12.8 0.637 0.235 0.091 0.033 0.003 0.001 a Accounting for parameters is as follows: DHRV ¼ 1, RGN ¼ 4, MGMT ¼ 4, RGN � MGMT ¼ 16, and DAU ¼ 40. observed the exception to this pattern in the SE region where we detected no conclusive change (2.25 fawns per 100 adult females [SE ¼ 3.55]). As a result of limiting deer hunting licenses, the 3-yr average number of deer hunters in Colorado decreased from 158,538 (1996–1998) to 79,653 (1999–2001). Revenue generated by deer license sales fell from $15.36 million in 1998 to $7.42 million in 1999 (Fig. 7). Between 1999 and 2006, deer hunting license revenue rebounded as a result of increases in license fees and a slight increase in licenses allocated, as adult male/adult female ratios increased above DAU objectives in some DAUs. Overall, the hunter satisfaction survey conducted prior to statewide limitation reflected disappointment in the status of Colorado’s deer herds (B. Watkins, unpublished data). In particular, hunters expressed that the quality of their hunting experiences had declined with the concurrent overall decline in the total deer population. Through the survey, hunters expressed concern that there were insufficient adult male deer to breed with all female deer. Results from the 2007 Big Game Season Structure survey highlighted a reversal in hunter satisfaction. Although this survey was designed to assess satisfaction with the 2007 hunting season, as opposed to the CDOW’s overall management structure, there were female ratios (Table 4). During the initial period of limitation, fawn/adult female ratios declined by 16.51 (SE ¼ 4.93) fawns per 100 adult females in the DAU in the SW, but any change was inconclusive in the NW when we considered standard error (4.58 fawns per 100 adult females, SE ¼ 3.08). However, by the end of the later period, in the highly limited NW DAUs, the number of fawns per 100 adult females had not further changed (3.57 fawns per 100 adult females, SE ¼ 3.44) relative to the ratios we observed in those same DAUs during the earlier period. Similarly, in the highly limited DAU in the SW region, we observed no change in the fawn/adult female ratio between the early and later periods (�0.74 fawns per 100 adult females [SE ¼ 4.81]). However, estimates within DAUs that switched from unlimited to highly limited during the later period were 16.51 (SE ¼ 4.93) fawns per 100 adult females lower within the same DAUs during the 2 periods. During the later period, when the remainder of the DAUs switched from unlimited harvest management to the moderately limited license structure, we again observed the trend of immediate declines in fawn/adult female ratios in 3 of 4 regions. For the NW, SW, and NE regions, decreases in the number of fawns per 100 adult female ratios were 4.41 (SE ¼ 2.38), 6.96 (SE ¼ 2.19), and 4.84 (SE ¼ 3.39), respectively. We Table 4. Coefficient estimates of changes in numbers of fawns per 100 adult females as a result of manipulating the number of hunting licenses and hunting license allocation in Colorado in 1999 and observed ratio statistics for each management type and period. All data analysis units (DAUs) estimates were standardized against those DAUs with an unlimited license allocation structure from 1984 until 1998. Reported changes in ratio estimates are changes in mean values that were allowed to vary by our AICc best model (DAU RGN MGMT RGN � MGMT DHRV). Change in ratio (SE) by management type Region NW NW NW NW SW SW SW SW NE NE NE NE SE SE SE SE Period Unlimited 1984–1998 1984–1998 1999–2006 1999–2006 1984–1998 1984–1998 1999–2006 1999–2006 1984–1998 1984–1998 1999–2006 1999–2006 1984–1998 1984–1998 1999–2006 1999–2006 0.00 Moderate �4.41 SE Observed ratio statistics High SE �4.58 �3.57 3.08 3.44 �16.51 �0.74 4.93 4.81 2.38 0.00 Min x 86.0 64.4 74.6 91.7 80.1 44.3 64.3 71.1 104.9 30.8 35.6 30.9 33.3 29.5 28.5 28.5 20.9 31.9 61.0 48.7 53.6 57.9 53.7 34.1 34.2 47.7 64.6 �6.96 2.19 �4.84 3.39 99.1 89.4 38.8 22.4 62.2 55.6 2.25 3.55 92.0 38.5 59.9 0.00 0.00 Bergman et al. � Deer License Limitation Max 1449 �Figure 7. Total licenses sold (black line) and annual revenue generated (gray line) through sale of deer hunting licenses in Colorado from 1982 through 2006. Colorado shifted management from an unlimited license sales structure to a limited license sales structure between the 1998 and 1999 hunting seasons. All monetary values were adjusted to 2006 values according to the consumer price index. few complaints about the status of Colorado’s deer herds. Overall, hunters expressed satisfaction with observed population trends and current management regulations. DISCUSSION As is the case in most field-based wildlife research, our study was limited by the lack of complete control over the system and an inability to completely randomize and replicate treatments. Specifically, we did not have the ability to test the cause-and-effect relationship between changes in harvest management and fawn/adult female ratios. However, as in our study, BACI-based designs have been found particularly useful in studies that fall under these conditions (Conner et al. 2007). Based on our results, we conclude that limiting antlered deer licenses effectively increased overall adult male/ adult female ratios, as well as the total deer population (Figs. 1 and 5). We detected an increase in the adult male/adult female ratio within a short time frame on almost all occasions (i.e., in 1 region during the early period and all 4 regions during the later; Table 3). Of note, we observed these responses in a variety of areas and regardless of the level of limitation or timing of implementation (Fig. 8A,B). This latter observation lends further credence to the use of a BACI type design in this analysis. Only one region (SW) did not have an increase in the short time frame. However, the SW region was the location of DAU 19 (GMUs 61 and 62), which was the only DAU with split harvest management objectives. Due to the high level of movement of deer between these 2 GMUs, increases in the adult male/adult female ratio did not occur before GMU 62 was limited. After 1999, when male deer licenses in GMU 62 were reduced by 75%, we observed a subsequent increase of 17.55 adult males per 100 females in the DAU. Overall, although we expected the direction of these results, the magnitude of the effect was surprising. In contrast to the trend of increasing numbers of adult male deer, we observed an opposite trend for fawn/adult female ratios (Fig. 6). An immediate decline in the number of fawns per 100 adult females occurred on each of the 2 occasions for which limitation was implemented (Table 4). No rebound in fawn/adult female ratios occurred. At best we observed a stabilization of fawn/adult female ratios at new, lower levels. Figure 8. Maximum, minimum, and mean values of observed number of adult males per 100 adult females in the NW region (A,B) and SW region of Colorado (C,D). Values shown represent estimates from 2 time periods (1980–1998 and 1999–2006) and 2 regions within the state, reflecting efforts to increase temporal and spatial replication within the before-after-control-impact type design. In all cases shown, the observed number of adult males per 100 adult females increased immediately after implementation of the statewide limited harvest management plan. 1450 The Journal of Wildlife Management � 75(6) �In no parts of the state did fawn/adult female ratios return to the level we observed during the unlimited license structure period. Although falling fawn/adult female ratios is a concern, the ratios we observed during the later period were still high enough to support population growth. The magnitude of these ratios likely remained high due to very high overwinter survival of fawns that has been documented throughout the state (Lukacs et al. 2009). Although not designed as a density dependence experiment, circumstantial evidence from our study indicates that a result of limiting deer license numbers in Colorado was the replacement of fawn mule deer with adult males. If this is the biological process occurring, careful monitoring should continue into the future to confirm that the balance currently achieved is stable. These results also provide evidence that hunter assumptions connecting low adult male/adult female ratios with inadequate breeding potential and subsequently low pregnancy rates were not accurate. In response to trends in fawn/adult female ratios we observed, further declines in fawn/adult female ratios would be indicative of declining recruitment and over a period of time would lead to an older herd age structure. A similar concern stemming from the limitation process pertains to competition among adult male deer. As adult male/adult female ratios increased beyond previously observed levels, interest in density-dependent relationships among adult males has become a topic of interest. It is commonly speculated that survival of adult male deer is particularly sensitive to stochastic, extreme weather events and disease, but this phenomena has been difficult to demonstrate beyond senescence and normal variation in different sex and age classes (Loison et al. 1999, Bonenfant et al. 2002, Toı̈go et al. 2007). However, if older age males are indeed more vulnerable, as this portion of Colorado’s deer herds grow, Colorado’s deer population may become increasingly vulnerable to extreme events. In light of this potential variability, if an upper threshold for adult male/adult females ratios exists, beyond which no additional adult males get recruited into the subsequent year’s population, then harvest in DAUs facing this constraint could be adjusted. An important consideration linked to the limitation process pertains to hunter opportunity. Although reducing hunter numbers is typically linked to economic impacts, the nonmonetary cost of allowing fewer hunters to go afield is difficult to quantify but important nonetheless. An immediate loss of hunters who chose not to participate in the season setting process and were no longer interested likely occurred, but recent hunter recruitment and retention analyses indicate that Colorado has had a growing population of deer hunters since 2002 (P. Lukacs, CDOW, personal communication). Even though the number of licenses available for sale was drastically cut under the new harvest management system, each year there were excess licenses for many DAUs (i.e., licenses sold over-the-counter on a first come, first served basis). Thus, hunters who were unsuccessful in the drawing process but who still wished to hunt deer in Colorado were still able to do so every year if they were willing to travel to different GMUs or to hunt female deer. Bergman et al. � Deer License Limitation The $7.86 million decrease in revenue in 1999 that stemmed from this wildlife management action was economically significant. Fortunately, due to other revenue sources, the CDOW was not forced to cut any major programs as a result of the lost income. The ability of the CDOW to absorb this loss of income was largely due to the continued availability of over-the-counter elk (Cervus elaphus) licenses for many parts of the state. We feel that hunter sentiment regarding the change in harvest management structure was positive. Although a survey system was not in place to adequately document these changes in attitude, Colorado’s wildlife managers feel that those sentiments that were documented do capture the overall sentiment of Colorado’s hunters (Balfourd 2009). Although many of the tools were in place to make this an adaptive harvest management process, the fundamental missing component was the refinement of harvest management action. As with many wildlife management decisions, mule deer management in Colorado is intentionally a process with a high level of public involvement. Public sentiment stemming from the 1999 limitation of deer license sales in Colorado appeared to be positive. Although the statewide deer herd grew between 1999 and 2006, it did so in disproportionate favor of adult males. Given these trends, from a population dynamics and herd health standpoint, a refinement of the harvest management objective in the form of increased harvest of adult males may need to occur. From a hunter satisfaction standpoint, limiting the number of licenses is typically used as a technique to improve the quality of a hunt by increasing the number of adult male deer per hunter, which was what happened in the case we examined. However, limiting the number of licenses also dramatically increased the proportion of adult male deer in the population, with potentially negative effects on fawn recruitment. As DAUs eventually met and surpassed their respective DAU management objectives, the CDOW was slowly able to increase hunter opportunity. MANAGEMENT IMPLICATIONS For many states, implementing sweeping harvest management changes on this level would have major economic and budgetary consequences. In Colorado, the sale of deer licenses is a major source of revenue, but it is far outweighed by elk license revenue. The overall economic impact of the decrease in deer licenses to Colorado’s economy is unknown but likely far exceeds the revenue loss to the CDOW. Retrospectively, several key improvements to the process of changing harvest management structure could have been incorporated to provide better feedback on the process. Overall the CDOW was pleased with the biological and population monitoring programs that were in place. However, this change was primarily directed at influencing the proportion of adult males in the population. Subsequently, many biologically oriented questions pertaining to competition between males and potential densitydependent effects have arisen. A priori monitoring of individual survival of adult males in different parts of the state would have been costly but highly informative during this 1451 �process. The CDOW has since initiated monitoring and research to address this issue. Similarly, as it appears that pressure to implement these changes was strongly based in hunter satisfaction, pre- and post-hunter satisfaction surveys surrounding the change in harvest structure would have better gauged the impacts on hunter satisfaction. Under ideal conditions, implementation of harvest management changes such as this would be conducted in a staggered approach and with different levels of magnitude. Such an approach would increase the level of knowledge gained by providing a replicated process and increase the level of inference that could be drawn from this type of post hoc analysis. A more staggered approach may also be better weathered by agencies and communities alike as the loss in hunter-based revenue would be gradual. ACKNOWLEDGMENTS Financial support for this work was provided by Colorado Federal Aid Wildlife Restoration Project funding. Helicopter flight survey data were collected throughout Colorado by many pilots, terrestrial biologists, and district wildlife managers. Similarly, these data were entered into Colorado’s Deer, Elk, and Antelope Management Database. This analysis would not have been possible without these efforts. We thank R. H. Kahn, D. J. Freddy, and G. C. White for support and insightful conversation in pursing this analysis. Valuable comments on earlier drafts were provided by B. Walker, D. Martin, and P. Doherty’s lab group at Colorado State University. Additionally, critical and valuable review of this manuscript was provided by D. F. Pac, B. Thompson, as well as 2 anonymous reviewers. We greatly appreciate their efforts and input. LITERATURE CITED Balfourd, K. 2009. Destination: trophy mule deer. Fair Chase. Fall: 42–48. Bishop, C. J., G. C. White, D. J. Freddy, and B. E. Watkins. 2005. Effect of limited antlered harvest on mule deer sex and age ratios. Wildlife Society Bulletin 33:662–668. Bonenfant, C., J. M. Gaillard, F. Klein, and A. Loison. 2002. Sex-and agedependent effects of population density on life history traits of red deer Cervus elaphus in a temperate forest. Ecography 25:446–458. 1452 Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Second edition. Springer-Verlag, New York, New York, USA. Conner, M. M., M. W. Miller, M. R. Ebinger, and K. P. Burnham. 2007. A meta-BACI approach for evaluating management intervention on chronic wasting disease in mule deer. Ecological Applications 17:140–153. Gill, R. B. 2001. Declining mule deer populations in Colorado: reasons and responses. Special Report 77, Colorado Division of Wildlife, Ft. Collins, USA. Green, R. H. 1979. Sampling design and statistical methods for environmental biologists. Wiley Interscience, Chichester, England. Hurlbert, S. J. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54:187–211. Loison, A., M. Festa-Bianchet, J. M. Gaillard, J. T. Jorgenson, and J. M. Jullien. 1999. Age-specific survival in five populations of ungulates: evidence of senescence. Ecology 80:2539–2554. Lukacs, P. M., G. C. White, B. E. Watkins, R. H. Kahn, B. A. Banulis, D. J. Finley, A. A. Holland, J. A. Martens, and J. Vayhinger. 2009. Separating components of variation in survival of mule deer in Colorado. Journal of Wildlife Management 73:817–826. Mason, R., L. H. Carpenter, M. Cox, J. C. Devos, J. Fairchild, D. J. Freddy, J. R. Heffelfinger, R. H. Kahn, S. M. McCorquodale, D. F. Pac, D. Summers, G. C. White, and B. K. Williams. 2006. A case for standardized ungulate surveys and data management in the western United States. Wildlife Society Bulletin 34:1238–1242. Montana Department of Fish, Wildlife and Parks. 2001. Adaptive harvest management: mule deer population objectives, hunting regulation strategies, special management districts, monitoring programs, population modeling and deer management policies. Montana Department of Fish, Wildlife and Parks, Helena, USA. Nichols, J. D., M. C. Runge, F. A. Johnson, and B. K. Williams. 2007. Adaptive harvest management of North American waterfowl populations: a brief history and future prospects. Journal of Ornithology 148 (Suppl 2): S343–S349. Stewart-Oaten, A., W. M. Murdoch, and K. R. Parker. 1986. Environmental impact assessment: ‘‘pseudoreplication’’ in time? Ecology 67:929–940. Toı̈go, C., J. M. Gaillard, M. Festa-Bianchet, E. Largo, J. Michallet, and D. Maillard. 2007. Sex-and age-specific survival of the highly dimorphic alpine ibex: evidence for a conservative life history tactic. Journal of Animal Ecology 76:679–686. Underwood, A. J. 1994. On beyond BACI: sampling designs that might reliably detect environmental disturbances. Ecological Applications 4:3– 15. Williams, B. K., J. D. Nichols, and M. J. Conroy. 2002. Analysis and management of animal populations. Academic Press, San Diego, California, USA. Associate Editor: Bruce Thompson. The Journal of Wildlife Management � 75(6) � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Journal Articles Description An account of the resource CPW peer-reviewed journal publications Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Biological and socio-economic effects of statewide limitation of deer licenses in Colorado Description An account of the resource <span>We evaluated the biological and socio-economic effects of statewide limitation of mule deer (</span><i>Odocoileus hemionus</i><span>) hunting licenses, which began in Colorado in 1999. We implemented a before-after-control-impact (BACI) analysis of annual helicopter sex and age class surveys, collected as part of the Colorado Division of Wildlife's routine monitoring, to assess changes in adult male/adult female ratios and fawn/adult female ratios in response to this change in harvest management. Following statewide limitation and reduction of license sales (1999–2006), we observed increases in adult male/adult female ratios of 7.39 (SE = 2.36) to 15.23 (SE = 1.22) adult males per 100 adult females in moderately limited areas and of 17.55 (SE = 3.27) to 21.86 (SE = 2.31) adult males per 100 adult females in highly limited areas. We simultaneously observed reductions in fawn/adult female ratios in newly limited areas by as much as 6.96 (SE = 2.19) fawns per 100 females, whereas in areas that had previously been limited we observed stabilization of fawn/adult female ratios at levels lower than levels observed under the unlimited harvest management structure. An immediate decline of $7.86 million in annual revenue stemmed from the change in harvest management, but revenue subsequently rebounded. This study provides preliminary evidence of potential effects that other state and provincial wildlife management agencies may face as they consider shifting mule deer harvest management towards limited license scenarios.</span> Bibliographic Citation A bibliographic reference for the resource. Recommended practice is to include sufficient bibliographic detail to identify the resource as unambiguously as possible. Bergman, E. J., B. E. Watkins, C. J. Bishop, P. M. Lukacs, and M. Lloyd. 2011. Biological and socio-economic effects of statewide limitation of deer licenses in Colorado. The Journal of Wildlife Management 75:1443-1452. <a href="https://doi.org/10.1002/jwmg.168" target="_blank" rel="noreferrer noopener">https://doi.org/10.1002/jwmg.168</a> Creator An entity primarily responsible for making the resource Bergman, Eric J. Watkins, Bruce E. Bishop, Chad J. Lukacs, Paul M. Lloyd, Mary Subject The topic of the resource Before-after-control-impact (BACI) Colorado Harvest Hunting License Management Mule deer <em>Odocoileus hemionus</em> Extent The size or duration of the resource. 10 pages Date Created Date of creation of the resource. 2011-07-15 Rights Information about rights held in and over the resource <a href="http://rightsstatements.org/vocab/InC-NC/1.0/" target="_blank" rel="noreferrer noopener">In Copyright - Non-Commercial Use Permitted</a> Format The file format, physical medium, or dimensions of the resource application/pdf Language A language of the resource English Is Part Of A related resource in which the described resource is physically or logically included. The Journal of Wildlife Management Type The nature or genre of the resource Article