F:\ALCES\Vol_38\Pagemaker\3809. ALCES VOL. 38, 2002 SILVERBERG ET AL. – IMPACTS OF WILDLIFE VIEWING 205 IMPACTS OF WILDLIFE VIEWING ON MOOSE USE OF A ROADSIDE SALT LICK Judith K. Silverberg1, Peter J. Pekins2, and Robert A. Robertson3 1New Hampshire Fish and Game Department, 2 Hazen Drive, Concord, NH 03301,USA; 2Depart- ment of Natural Resources, University of New Hampshire, James Hall, Durham, NH 03824, USA; 3Department of Resources and Economics, University of New Hampshire, James Hall, Durham, NH 03824, USA ABSTRACT: In northern New Hampshire, we examined the use patterns of moose visiting a roadside salt lick before (1996) and after (1997-1999) a blind was built specifically to view moose at the lick. Moose visitation patterns were monitored with trail monitors equipped with cameras placed on trails leading into the study and control salt licks. There was no difference in frequency of use and time of use at the study and control sites in any year. Nocturnal use was higher than diurnal use; use was greatest at 2200-2400 and 0400-0600 hours at both sites. Reduced use of the trail closest to the blind indicated that placement of the blind probably altered access patterns of certain moose. A trend in 1998-1999 toward more visits during early morning than peak afternoon viewing time indicated that assessment of viewing opportunity warrants further study. ALCES VOL. 38: 205-211 (2002) Key words: Alces alces, behavior, salt lick, wildlife viewing Even though the term “nonconsumptive wildlife users” has been applied to describe people who do not hunt or fish, recreationists such as wildlife watchers do use and disturb recreational resources along spatial, visual, and physical dimensions. Disturbances may be intentional or unintentional; unintentional disturbances often occur when photograph- ing wildlife, viewing nesting birds, or hiking into an animal’s territory (Knight and Cole 1991, 1995). Unintentional impacts also include direct harassment of animals or alteration of habitat (Kuss et al. 1990). Nonconsumptive users trample and rear- range vegetative patterns, disturb wildlife behavior and activity, and are the chief distributors of refuse across the land (Gold- smith 1974, Wilkes 1977). Moose (Alces alces) are strongly at- tracted to supplementary sodium during spring and early summer in large parts of their North American range (Fraser 1979), and commonly use roadside salt licks in New Hampshire that are created from run- off of salt spread on roadways in winter (Miller and Litvaitis 1992). Such areas provide excellent places to view moose during May, June, and July and their high visibility has created a strong interest in moose viewing. Northern New Hampshire and Maine are well known places to view moose and the wildlife viewing programs of both states have published guides for wildlife viewing (e.g., Silverberg 1997). Unfortunately, many viewing opportunities occur along roadsides during summer, and traffic congestion regu- larly occurs in certain locations. Anecdotal information from moose viewers on Route 3 in Pittsburg, New Hampshire, a popular moose viewing area, suggested that moose shifted use of salt licks to late night to avoid disturbance from viewers. Limited research has been conducted on impacts of wildlife IMPACTS OF WILDLIFE VIEWING – SILVERBERG ET AL. ALCES VOL. 38, 2002 206 viewing in situations such as those associ- ated with moose viewing in northern New Hampshire. The wildlife viewing program of the New Hampshire Fish and Game Depart- ment proposed construction of a moose viewing area on Route 26 in Dixville Notch to provide viewers with an opportunity to view moose from a blind as an alternative to viewing from their cars along the roadside. The planning phase of this project provided the opportunity to design a research project that would explore specific questions about the use of roadside salt licks by moose at a state-sanctioned wildlife viewing facility. These questions fell into 3 categories: moose visitation rate and use of the lick from pre- construction to post construction; moose responses to wildlife viewing actions and other human caused stimuli; and the char- acteristics, motivations, and attitudes of wildlife viewers. From previous work, it is known that there exists a wide range of intra and inter- specific variation of responses to distur- bance (Knight and Temple 1995). Studies conducted by McMillan (1954) and Altmann (1958) in Yellowstone National Park, showed a variety of behavioral responses in moose. In Sibley Provincial Park, Ontario, Cobus (1972) found that in general moose developed a tolerance towards humans. The effect of an increase in road traffic on wildlife from 1973-1983 was examined in Denali National Park, Alaska. This el- evated volume correlated with a 72% de- crease in moose sightings (Signer and Beattie 1986). This paper specifically focuses on the impact of the facility and viewing activities that could be assessed by monitoring moose movement activity pre- and post- construc- tion. Specifically to determine if the visita- tion rate and time of use by moose at the salt lick in Dixville Notch were affected by the construction and subsequent use of the wild- life viewing area. STUDY AREA The viewing site was located in north- ern New Hampshire, to the east of Dixville Notch on Route 26. This 4 ha area, inclu- sive of the viewing site, was harvested (clear-cut) in 1991 and was characterized by a regenerating northern hardwood/ spruce-fir forest community. A buffer strip of mature balsam fir (Abies balsamea) and red spruce (Picea rubra) was left on both sides of the road. The primary salt lick, about 175 m long, was on the north side of the road, and a smaller lick about 70 m long was on the south side. A 6-car parking lot, trail, and viewing blind were built in December 1996 across from the primary salt lick. Construction occurred in December because moose re- duce their use of licks after the fall rut (Adams 1995). A trail (125 m) led to the viewing blind located about 30 m from the primary salt lick. The viewing blind held up to 20 people and had slits that faced the main lick and a moose trail that entered the lick from the east. The control site consisted of 2 roadside salt licks (200 m and 50 m long) 1.5 km east of the viewing site. These salt licks were approximately 0.2 km from a clear-cut. Both sites were frequented regularly by moose prior to the study. The similarity between the study and control sites was ascertained by comparing aerial photos which showed that both were predomi- nately spruce-fir forests before harvest; the control site was clear-cut 1 year after the study site. The licks on the study and the control sites were approximately the same distances from the center line of the high- way. METHODS Trailmaster 1500 game monitors were used to measure the visitation rate and time ALCES VOL. 38, 2002 SILVERBERG ET AL. – IMPACTS OF WILDLIFE VIEWING 207 of use of salt licks by moose. The monitors are ideal for monitoring moose and other mammal movements because measurement is continuous and potential interference from observers is eliminated (Kucera and Barrett 1993). These monitors were used previ- ously to measure use at salt licks in Pittsburg and Milan, New Hampshire during 1994- 1995 (Adams 1995). A monitor consisted of a transmitter that emitted an infrared beam to a receiver that tripped an automatic 35 mm camera. When an animal walked through the beam, the receiver recorded the date and time, and the camera took a picture. The monitor could store a maximum of 1,000 events. The sensitivity of the trigger and the length of time the beam must be broken to register an event was adjusted to 0.05 seconds. Every time the beam was broken the data recorder marked the event. To prevent a photograph from being taken multiple times of the same animal, the camera was set for a photograph to be taken every 2 minutes. Date and time were recorded on each pho- tograph. The cameras had flashes and professional high-speed (ASA 1600) film was used to ensure an image was recorded at night. Five monitors were placed at the view- ing site (#1-5) and 4 monitors were placed at the control site (#6-9) simultaneously. The 2 licks at the control site were consid- ered as one due to their proximity and interconnected moose trails. Because the location of monitors is crucial to provide maximum information (Kucera and Barrett 1993), they were located on major moose trails entering the licks. The monitor and receiver camera package were placed on a tree or stake on the opposite sides of a well established trail. Specific placement took advantage of localized terrain, trail charac- teristics, and surrounding vegetation. Care was taken to minimize the possibility of sunlight and blowing vegetation breaking the infrared beam, thus triggering the cam- era. Monitors were placed at heights of 30- 75 cm to also record the presence of me- dium-sized mammals (e.g., white-tailed deer (Odocoileus virginianus), bear (Ursus americanus), and coyote (Canis latrans)). Monitors were placed in the same locations each year. Data were collected from 10 June-14 July during 1996-1999. Monitors were checked twice weekly when data were downloaded and recorded in a logbook; film was replaced as needed. The date and time stamp on the developed film was compared to the information recorded by the monitor. The data were entered into a spreadsheet indicating the monitor number, year, time, date, whether there was a photograph, whether an animal was in the photograph, identity of animal, and sex and age of moose (if possible). Judgements were made to eliminate multiple data collected in a short period of time caused by a stopped animal, or an animal moving in and out of the lick within a 2- minute period. For example, if the monitor recorded 10 passes within 2 minutes, and photographs indicated it was the same moose, only 1 visit was counted. Moose were not marked, consequently, there was no way to determine how many times a particular moose entered a lick, or if the same moose used the area annually. In situations when a camera ran out of film, but events were recorded at similar frequen- cies as when photographs indicated single visits, these events were classified as moose visits. It was assumed that a monitor mal- functioned when it recorded hundreds of events per day. Malfunction was apparent during periods of heavy rain or wind. Data were analyzed using SPSS (SPSS Inc., Chicago, Illinois). Graphs and fre- quency distributions were used to provide an overall depiction of moose encounters. For ANOVA, moose encounter data were aggregated on a weekly basis by year to test IMPACTS OF WILDLIFE VIEWING – SILVERBERG ET AL. ALCES VOL. 38, 2002 208 for differences in the number of moose visits at the viewing and control sites annu- ally. Combining data on a weekly basis eliminated the problem of small sample size on any given day. Data of visitation times were aggregated into 12, 2-hour time blocks for analysis. This aggregation eliminated potential problems with small sample sizes in any 1-hour block. Time was described as 14 diurnal hours (0600-2000 h) and 10 noc- turnal hours (2000-0600 h) based on day- light and times when viewers could view moose without artificial light. Statistical significance was set at 0.05 a priori. RESULTS The number of annual moose encoun- ters at the viewing site (mean ± SD= 228.0 ± 16.7) and the control site (mean ± SD = 273.5 ± 19.7) was relatively constant during the 4 years. There was no difference in the annual weekly encounter rate from year to year at the viewing site (F = 0.280; df = 3, 16; P = 0.839) or control site (F = 0.712; df = 3, 16; P = 0.559). Variability occurred at individual monitors at both sites annually (Fig. 1). Monitors 2-4 had more encounters the last 2 years than the previous years; encounters at monitor 5 were constant. Conversely, monitor 1, located < 10 m from the viewing blind, had about 50% less en- counters the last 2 years (Fig. 1) and the pattern of encounters was different than that at monitors 2 (χ2 = 52.63, df = 3, P = 0.000), 3 (χ2 = 18.44, df = 3, P = 0.000), 4 (χ2 = 44.19, df = 3, P = 0.000), and 5 (χ2 = 7.810, df = 3, P = 0.050). Although annual variabil- ity in encounters occurred at the control site monitors, no obvious pattern was evident (Fig. 1). Over 3 times as many encounters oc- curred nocturnally (n = 661) than diurnally (n = 182) at both the viewing and control sites (Figs. 2 and 3). Encounters at both the viewing and control sites occurred most often at 2200-2400 h and 0400-0600 h (Figs. 2 and 3). Diurnal visitation was low and little variation occurred among time blocks (Figs. 2 and 3). The annual pattern of visitation within a 24-hour period was not different at either site (F = 0.239; df = 3, 16; P = 0.787). There was no significant change in the diurnal or nocturnal pattern of visita- tion when comparing 1996 data (pre-con- 0 20 40 60 80 100 120 140 160 1 2 3 4 5 6 7 8 9 Study Monitors (1-5) Control Monitors (6-9) M o o se E n c o u n te r s 1996 1997 1998 1999 Fig 1. Annual moose encounters per monitor at the viewing site (monitors 1-5) and control site (monitors 6-9), 10 June – 14 July, 1996 (pre-construction) and 1997-1999 (post-construction), Dixville Notch, New Hampshire, USA. IMPACTS OF WILDLIFE VIEWING – SILVERBERG ET AL. ALCES VOL. 38, 2002 210 fold increase at 0600-1000 h. DISCUSSION The total number of moose encounters fluctuated little at the viewing and control sites over the 4-year time period. While there was no overall effect on encounter rates at the viewing site, the decline at monitor 1, located < 10 m from the viewing blind, indicated that the presence of the blind and wildlife viewing probably caused moose to enter the lick from other trails. This type of impact could probably be mini- mized by considering movement patterns on individual trails in similar projects. The most active use of salt licks by moose at the control and viewing sites was at 2000-0600 h. There was no evidence moose changed their nocturnal visitation patterns as was suggested from anecdotal information from Pittsburg, New Hamp- shire, where moose viewing has been a popular pastime since the mid-1980s. It should be noted that most viewing in Pittsburg occurs at night with the use of spotlights and viewing pressure is so intense on weekends that local traffic congestion is common. The general pattern of nocturnal visitation was similar to that measured at licks in Pittsburg and in Milan, New Hampshire, 10 June- 14 July 1994 (Adams 1995). The overall tolerance of moose to hu- man activity was consistent with observa- tions on Shiras moose (Alces alces shirasi) in Yellowstone National Park, where moose behavior in an area where tourists were prevalent was compared with moose behavior in an area with few people (McMillan 1954). Moose at the tourist site showed little interest in humans and ap- peared to tolerate their presence. Similarly, the aquatic feeding behavior of moose in Sibley Provincial Park, Ontario, was only slightly affected by viewing (Cobus 1972). Quiet viewing in the blind produced no measurable behavioral response by moose; conversely, cars stopping alongside the lick produced an increased fleeing response (Silverberg 2000). Ironically, there was a striking lack of overlap between the predominant nocturnal use of licks and potential diurnal viewing opportunities. Although not statistically sig- nificant, there were several interesting changes in encounter numbers relative to diurnal moose visitation at the viewing site. These included a more than 2-fold increase in the number of encounters at 0600-1000 h in 1998 and 1999, a > 50% reduction in encounters at 1600-1800 h in 1998 and 1999, and by 1999 a 33% reduction in the 3 peak visitation times measured in 1996 (Fig. 2). These reductions occurred during the most popular viewing times (Silverberg 2000). It is possible that moose shifted their diurnal use to avoid consistent use of the viewing blind. Because the number of encounters during all diurnal periods was relatively low, slight shifts in visitation pat- terns could reduce viewing opportunities. Unfortunately, opportunities to view moose from the blind were relatively low from 0600 to 2000 h when most visitors were present. Most viewers were well aware that the best time to view moose is early morning or late in the evening. How- ever, viewer satisfaction levels were not affected by whether they saw a moose (Silverberg 2000). Perhaps wildlife view- ers should be informed that the best time to view moose in natural light during June and July is shortly before and after sunrise (0400-0600 h) when moose were active at licks. Considering evidence from this and other studies, the impact of increased view- ing during these hours should be minimal, but may warrant further monitoring. Fur- ther, it is possible that by promoting early morning viewing opportunities, expectation levels of seeing a moose would increase and affect satisfaction levels. Promotion of earlier viewing should also include informa- ALCES VOL. 38, 2002 SILVERBERG ET AL. – IMPACTS OF WILDLIFE VIEWING 211 tion about proper viewing behavior to as- sure that viewing impacts remain minimal. REFERENCES ADAMS, K.P. 1995. Evaluation of moose population monitoring techniques and harvest data in New Hampshire. M.S. Thesis, University of New Hampshire, Durham, New Hampshire, USA. ALTMANN, M. 1958. The flight distance in free-ranging big game. Journal of Wild- life Management 22:207-209. COBUS, M.W. 1972. Moose as an aesthetic resource and their summer feeding behavior. Proceedings of the North American Moose Conference and Workshop 8:244-275. FRASER, D. 1979. Sightings of moose, deer and bears on roads in northern Ontario. Wildlife Society Bulletin 7:181-184. GOLDSMITH, F.B. 1974. Ecological effects of visitors in the countryside. Pages 217-231 in A.Warren and F.B. Gold- smith, editors. Conservation In Prac- tice. John Wiley and Sons, London, England. KNIGHT, R.L., and D.N. COLE. 1991. Ef- fects of recreational activity on wildlife in wildlands. Transactions of the North American Wildlife and Natural Re- sources Conference 56:238-247. , and . 1995. Wildlife re- sponses to recreationists. Pages 51 - 70 in R.L. Knight and K.J. Gutzwiller, editors. Wildlife and Recreationists: Coexistence Through Management and Research. Island Press, Washington, D.C., USA. , and S.A. TEMPLE. 1995. Origin of wildlife responses to recreationists. Pages 81 - 91 in R.L. Knight and K.J. Gutzwiller, editors. Wildlife and Recreationists: Coexistence Through Management and Research. Island Press, Washington, D.C., USA. KUCERA, T.E., and R.H.BARRETT. 1993. The trailmaster camera system for de- tecting wildlife. Wildlife Society Bulle- tin 21:505-508. KUSS, F.R., A.R. GRAEFE, and J.J. VASKE. 1990. Visitor impact management, a review of the research. Volume 1 and 2, National Parks and Conservation Asso- ciation, Washington, D.C., USA. MCMILLAN, J.F. 1954. Some observations on moose in Yellowstone Park. Ameri- can Midland Naturalist 52: 392-399. MILLER, B.K., and J.A.LITVAITIS. 1992. Use of roadside salt licks by moose, Alces alces, in northern New Hamp- s h i r e . C a n a d i a n F i e l d - N a t u r a l i s t 106:112-117. SIGNER, F.J., and J.B. BEATTIE. 1986. The controlled traffic system and associ- ated wildlife responses in Denali Na- tional Park. Arctic 39: 195-203. SILVERBERG, J.K. 1997. New Hampshire Wildlife Viewing Guide. Falcon Press, Helena, Montana, USA. . 2000. Impacts of wildlife viewing: a case study of Dixville Notch Wildlife Viewing Area. Ph.D. Dissertation, Uni- versity of New Hampshire, Durham, New Hampshire, USA. WILKES, B. 1977. The myth of the nonconsumptive user. Canadian Field- Naturalist 91:343-349.