Approximately 730 Game Production Areas (GPAs) encompassing 119,382 ha are located within the state of South Dakota, USA [1] . These areas were purchased, and are maintained, by the South Dakota Department of Game, Fish and Parks with funds from hunting and fishing licenses and federal excise taxes on guns, ammunition, and fishing equipment. The excise taxes were authorized by the United States Federal Aid in Wildlife Restoration Act of 1937, which is more commonly referred to as the Pittman-Robertson (PR) Act, and the Sport-Fish Restoration Act of 1950, which is more commonly referred to as the Dingell-Johnson (DJ) Act. Since its inception, more than 62% of PR funds have been used to purchase, develop, or maintain wildlife management areas in the USA [2] , with over 1.6 million ha of public hunting land purchased with PR funding during the first 50 years of the program [3] .

The procurement and management of public hunting lands via license sales and PR/DJ funds is a user-pay, user-benefit system. However, non-consumptive activities also occur, and have rarely been documented. Belyea and Lerg [4] found that 72% of the use of public game and wildlife areas in Michigan involved non-hunting activities, and 70% to 90% of the people using lands purchased with PR funding are not engaged in hunting activities. Nelson et al. [5] observed that non-consumptive use may be seasonal in nature, and noted that during fall hunting seasons, over 90% of public hunting land use involved hunting, trapping, fishing, and wildlife viewing.

The Mirror Lakes GPA was purchased in 1943 and is located eight miles west of Spearfish, South Dakota. It is unique among South Dakota GPAs because it contains McNenny State Fish Hatchery within its borders. In addition, the Mirror Lakes GPA contains three small fishing lakes. A complete creel census was conducted at two of these lakes during the summer of 2006 [6] , which provided an opportunity to document use of the GPA by nearly every visitor. Thus, the objective of this study was to determine the use of the Mirror Lakes GPA during the summer in the absence of any hunting seasons.

The Mirror Lakes GPA is located approximately eight miles west of Spearfish, South Dakota (Figure 1). As public land, Mirror Lakes GPA can be accessed on foot anywhere along its border with permission from the adjoining private landowners. However, nearly all people enter through one of two roads. One road provides access to the two Mirror Lakes and Crow Creek, while the other road provides access to Cox Lake. While an-

gling occurs at all of the lakes, this study focused on the area around Mirror Lakes where non-consumptive activities have historically been concentrated. To document all of the activities occurring in proximity to Mirror Lakes, a creel station was set up adjacent to the single-lane bridge which crosses Crow Creek on the road to Mirror Lakes in 2006 [6] . All vehicles were stopped prior to crossing the bridge and the vehicle occupants were interviewed as to their primary reason for visiting the GPA. The clerks recorded the number of occupants per vehicle, vehicle entry and exit times, and verified the activities performed by the individuals during their visit to the GPA.

Activities on the GPA were classified into five categories. Angling at either of the Mirror Lakes or Crow Creek was labelled as fishing. Sight-seeing activities where people just came out to the lakes for a drive or to park and just look at the scenery were classified as touring. Dogs was the label used for any dog-related activities, including dog training (both in and out of the water), playing with unleased dogs, or just allowing the dogs to run un-leashed. Any water-related recreational activities, such as swimming, wading, or just playing in the water, were recorded as swimming. The photography category included anyone whose primary purpose was to use a camera on the GPA. Lastly, any other activities were categorized as miscellaneous.

This study ran from May 16 to August 16. To observe activity changes over time, the data was separated into three, 31 day periods: May 16 to June 15, June 16 to July 15, and July 16 to August 16.

The activities of a total of 1513 individuals were recorded during the study period. While fishing was the predominant activity, over 40% of the visitors to the GPA were primarily there for non-consumptive reasons (Figure 2). Touring, dogs, and swimming were relatively popular, but less than 1% of the visitors were there for photography. Miscellaneous activities included two individuals bird-watching, two individuals writing, and one individual each identified their primary activity as having lunch, partying, riding horse, and checking soil. While slightly less than 60% of the GPA visitors were engaged in fishing, that activity accounted for nearly 80% of the total time that visitors spent (Figure 3). Although a relatively large percentage of visitors performed touring and dog activities, they spent relatively less time in these activities compared to those visitors who were swimming or fishing.

The percentage of visitors fishing decreased from a high of over 65% in the first 31 day period to less than 50% in the third period (Figure 4). However, nearly all of the non-consumptive activities showed a reverse trend, increasing with each subsequent time period. All of the photography and miscellaneous activities were recorded in the first 31 days; those categories were absent in the final two periods. The percentage of visitors swimming or conducting dog activities doubled from the first to third periods. The percentage of time that visitors spent fishing also decreased over time, from over 80% in periods one and two, to approximately 65% in period three (Figure 5). The percentage of time spent on the other categories increased, although not necessarily in proportion to the increased percentage of visitors involved with those activities.

With all of the non-consumptive activities combined, the number of visitors increased over the summer while the number of visitors fishing decreased (Figure 6). By the final period, more visitors were at the GPA for tour-

ing, dogs, and swimming, rather than for fishing. However, fishing still accounted for the most of the time that people spent at the GPA (Figure 7). In the final 31-day period, over 55% of the visitors were at the GPA for non-consumptive activities, but they accounted for only 35% of the visitation time (Figure 8).

It is difficult to compare the results from the census of activities at the Mirror Lakes GPA to the few published studies concerning non-consumptive activities on public hunting lands which did not contain fishing lakes. For example, Nelson et al. [5] surveyed use of public hunting land only in the fall and noted that less than 10% of visitors were primarily engaged in non-consumptive activities. Fishing and water-based recreation were not part of the Nelson et al. [5] study. An earlier study in the same geographical area indicated that up to 90% of the visitors to public hunting areas were there for non-consumptive activities [4] . Raphael and Jaworski [7] determined that fishing activities in the coastal wetlands of Michigan had an economic value of over double that of non- consumptive recreation. In contrast, on the Mirror Lakes GPA during the summer, most of the visitors came to fish early in the summer, and even when the number of anglers declined below the number of non-consumptive visitors as the summer progressed, the amount of time spent fishing was still greater than that spent on all other activities.

Although it contains three small fishing lakes (two Mirror Lakes and Cox Lake), the primary purpose of the Mirror Lakes GPA is to provide access for hunting opportunities [1] for species such as white-tailed deer (Odo- coileusvirginanus), sharp-tailed grouse (Tympanuchusphasianellus), wild turkey (Meleagrisgallopavo), mourn- ing dove (Zenaidamacroura), and various waterfowl species. Although non-consumptive recreational activities affect wildlife populations [8] , the degree to which the relatively large number of visitors involved in non-con- sumptive activities, such as running dogs and touring, during the summer may be impacting wildlife species on the Mirror Lakes GPA is unknown. Blanc et al. [9] stated that dogs, especially those allowed to run off-leash, are a major wildlife perturbation, and dogs confined to trails have also been show to negatively impact wildlife populations [10] . Banks and Bryant [11] observed substantial reductions in bird diversity and abundance as a result of dog walking. Touring in vehicles can also disturb wildlife, although likely to a lesser extent than people sight-seeing outside of a motorized vehicle [12] [13] . In addition to direct effects on wildlife, non-consumptive activities may have indirect effects through habitat degradation [14] .

The small number of individuals using the area for photography, and the very limited amount of time spent on the activity, was somewhat surprising. Wildlife photography is a popular activity [15] [16] . On some public lands, more individuals may be involved with wildlife photography than with day hikes [17] . However, the expense of the required equipment may limit the number of people involved [18] , and household income is lower in South Dakota in comparison to other states [19] . It is also possible that the large amount of dog activity is reducing the visibility of wildlife available to be photographed [9] - [11] .

The large number of individuals involved in non-consumptive activities at the Mirror Lakes GPA could possibly be due to exposure to the area during their youth [20] [21] . The popularity of the fishing experience at

Mirror Lakes later in the summer by families and young anglers may be influencing subsequent visits [22] . Place attachment can be an important reason for individuals to revisit natural areas [21] .

Although the Mirror Lakes GPA was purchased and is maintained using revenues obtained from licenses and consumptive activity excise taxes, during the summer it is used extensively for non-consumptive activities. The disconnect between the user-pay system and non-consumptive use has been identified for quite some time [23] [24] . We did not ask the non-consumptive users of the Mirror Lakes GPA if they had purchased hunting or fishing licenses, so it is possible that they were financially supporting GPA management. Those individuals that were not licensed anglers or hunters could possibly be induced to volunteer time or financially contribute to GPA management, such as road improvements or habitat restoration, which would benefit them personally [25] .

In conclusion, the predominant activity on the Mirror Lakes GPA during the summer months was fishing, supporting the user-pay, user-benefit system that provided the funding to purchase and maintain the public land. However, the proportionally large number of non-angling visitors during the summer months presents both challenge and potential opportunities to public land managers.

We thank Sarah Austin, Rick Cordes, Anna Gusse, Lori Jennings, Erik Klinckman, Eric Krebs, Kelsey Prosser- Tiesen, Brandie Radigan, Rodney Wasche, and Keith Wintersteen for their assistance with this study.