Severa l Nationa l Park Service units in the Intennountain region possess a number ofclosely related management needs relative to the abundance of wild ungulates and their herbivory effects on plants and ecosystem processes. In 1993, the then National Biological Service (NBS) - now U.S. Geological Survey, Biological Resources Discipline (USGS, BRD) initiated a series of research studies in four park units in the Intennountain West., into the abundance and effects of ungulates on park ecosystems. Each of these parks received a number of similar research study elements inclUding: (a) a number of new ungulate grazing exclosures (n = 12-21 exclosures per park); (b) aerial survey sightability models to estimate population Si7..e5 of ungUlates; (e) measures of biomass production and consumption rates near the exclosures and across the landscape; (d) studies of the effects of the grazing on plant abundance, species diversity, and ecosystem effects; and (e) computer model simu lations (SAVANNA) ofthe effects on the ecosystem and plant resources ofdifferent ungulate management scenarios. One park unit, Rocky Mountain National Park, Colorado, re<:eived funding from the U.S. Geological Survey (USGS, BRD) and parallel funding from NPS foran intensive research study of the effects of elk on the park ecosystems.
Elk were extirpated, or nearly so, from Rocky Mountain National Park by human exploitation in the late 1800s, but were then reintroduced in 1913 and 1914. Elk stead ily increased until they reached an estimated 1,000 tllliulIll.s witilill lhe: pl1rk buulJ(.Iari~s in 1944 (Packard 1947). Due to concerns over increasing elk numbers and potential effects on the park winter range, elk in the park were artificially reduced from 1943-1968. During this period, a total of 1,664 elk were removed from the park with the goal ofreducing the park herd to about 500 on the eastern boundary winter range. In 1968 e lk were no longer con tro lled within the park's boundaries with a NPS change in management policy to one of natural regulation that occurred in Yellowstone National Park at the same time. From 1968 to the present time, herd limitation was attempted through natural regulation within the park and halVests outside of the park. InterageTl(;y goals of the NPS, Colorado Division of Wildlife (CDOW), and U.S. Forest Service (USFS) included use ofboth regular and late season hunts outside of the park boundaries to limit the elk population by harvesting 500-600 elk each year. The harvest goal was nearly achieved prior to 1987 when an average of442::t: 78 elk were harvested each year, but after 1987, increasing restrictions to private lands outside the park reduced the halVest. Elk harvests declined to 302 ::t: 36 after 1987. In either case, these harvests were insufficient to limit the growth ofthe elk population and elk steadily increased in the park and then later in the town of Estes Park. Concerns over possibly overabundant elk resulted in criticisms ofthe park elk policy (Hess 1993) and calls for the agencies to reevaluate their interagency elk management efforts. In 1993, the park superintendent, James Thompson, requested F. J. Singer ofthe National Biological Service (now USGS-BRD), to conduct a problem analysis of the elk situation.
The goals of the study included detennination of whether elk densities had exceeded those expected in a natural system, whether unnatural concentrations of elk were occurring, whal the effects of elk herbivory were, and whether the effects of the elk herbivory were acceptable (I' unacceptable. At the onset of the study, it was recognized that a number of human influences had occurred in the system that might confound the interpretation of the effects of elk alone. For example, any climate change or unnatural succession due to fire suppression might have influenced plant communities. A number ofmeadows in the winter range had been drained for a golf course located in the park unti) the 1960s and a number of other park meadows had been drained for cu ltivation. Beaver had apparently declined on at least part of the winter range and for unknown reasons. The presence of the rapidly growing town of Estes Park, located within the edge of the winter range, might have altered or abbreviated elk migrations. The major predaturs uf tJJI;: SySlf:01, wolv~s and grizzly bears, had long since been extirpated and considerable debate and speculation surrounded what their effects might have been on possible limitation of elk.
Study design input included the problem analysis prepared in 1993 and proposals submitted to BRD and the NPS in 1993 by F. Singer, and three peer reviews of the study design conducted in 1994 (Table I). The study also built upon earlier peer reviews ofthe sim ilar problem in Yellowstone National Park (Table I). The study design included key elements of those review suggestions. including: census methodology and demographic analysis of the elk population, studies of vegetative biomass produced in grazed and ungrazed areas and elk offiake, studies ofeffects on ecosystem processes, GISbased ecosystem model (SA V ANNA) experiments to test hypotheses, and evaluation of grazing effects on plantspecies diversity. Three peeNeviews of the initial study design were obtained in 1994 (Hobbs 1994; Smith et al. 1994; Unsworth et al. 1994). A statistical analysis of fi eld data collected from 1968- 1992 was completed (Zeigenfuss et a!. 1999). A mid-study science based assessment of vegetation management goals for elk winter ran ge was conducted (Berry et aJ. 1997). Modifications to the original plan were made to meet some ofthe concerns of Berry et al. (1997) and through introspection by the study team investigators and unanswered questions brought out during analysis of the first two years of data. These modifications included: (I) an analysiS of long-term climate trends to evaluate vegetati on changes; (2) one more year of offtake measures to reduce high variances observed during years one and two; (3) measures ofbiomass and offtake in town since it became obvious after two years of research that more elk were wintering in town than in the park; (4) the earlier proposed study of to tal N cycle and sustainabi lity was funded by the USGS-NRPP fund program; (5) the long-term trends of willows were detennined from sequences of aerial photos and GIS ana lyses; and (6) a population -based estimate of K for elk (ecological carrying capac ity) in both town and in the park was conducted. A regional study ofthe effects ofelk on plant species diversity with a multi-scale perspective was funded for a number ofstudy sites, that opportunistically included the RMNP winter range (Stohlgren et al. 1999). Thus, all ofthe elements ofthe ideal research study were achieved, including: (a) a study design and plans were prepared at the outset; (b) these were peer-reviewed; and (c) a mid-study assessment and changes in response to that assessment were accomplished.
