This research project was initiated in response to a directive from the Director of the National Park Service (NPS) on June 12, 1986, which was related to the stocking of trout into lakes of North Cascades National Park Service Complex (NOCA). The directive was an official response to a January 1986 request from the NPS Pacific Northwest Regional Office (PNRO) for a clear statement regarding fish-stocking policies of the NPS, i.e., stocking of non-native animals, including fish, is not allowed in units of the NPS (NPS 1991). The main elements of the directive included establishment of current fish and aquatic habitat baseline conditions, evaluation of the impacts of fish stocking on fish and wildlife, and determination of changes over time referenced against baseline conditions or against undisturbed conditions. Although mountain lakes in the area now designated as NOCA were originally devoid of fish, the Washington Department of Wildlife had stocked 75 of the 150 lakes thought to be able to support fish populations prior to the establishment of the park in 1968. Stocking was allowed to continue in the two recreational areas of the park between 1968 and 1974. In 1975, PNRO proposed to exempt certain restrictions on some lakes and suggested a gradual phasing-out of stocking in the park. The park and the State of Washington co-signed a variance to the NPS policy of no fish stocking in 1979 so that selected lakes could continue to be stocked with non-native trout at regular intervals in selected lakes. NOCA drafted a new memorandum of understanding (MOLT) in 1985 in which the policy variance was dropped. This proposal resulted in considerable debate by the public and Congress. A supplemental agreement to the 1985 MOU was prepared in 1988 and specified the stocking levels of selected lakes through the year 2000. As part of this agreement and after completion of the present research program, the park will prepare an environmental assessment and high-elevation lake management plan by 2000.

The first phase of this research project (1989 - 1991) investigated ecological impacts of stocked trout on naturally fishless lakes in NOCA. Very little was known about the aquatic resources in the park prior to initiation of this research. Results indicated that larval salamanders (Ambystoma macrodactylum), the top native vertebrate predators in many fishless lakes in NOCA, were at significantly lower densities in lakes with fish as compared to their densities in lakes without fish. However, larval densities in fishless lakes were variable, which suggested that abiotic factors were influencing larval abundance. Furthermore, effects of introduced trout on the limnetic food web appeared to be influenced by trout density and size structure and lake nutrient levels. In general, reproducing trout populations maintained higher densities and a more complex age and size structure than did non-reproducing trout populations in lakes that had to be periodically stocked with fry. In lakes east of the hydrological divide, densities of the large-bodied and ubiquitous copepod Diaptomus kenai were significantly lower in lakes with reproducing fish than in lakes with non-reproducing trout. In small and shallow lakes with elevated nutrient levels, the small-bodied copepod D. tyrrelli maintained higher abundances in lakes with high fish densities where large-bodied copepods were absent. Only three of the 15 nearshore macroinvertebrates taxa were found in significantly fewer lakes with vertebrate predators (salamanders and trout) than in lakes with no vertebrate predators. The distribution of one of these taxa, Desmona, may be restricted soley by trout predation. The distribution and abundance of rotifer taxa appeared largely unrelated to vertebrate predators.

A panel of aquatic ecologists has peer reviewed the program on an annual basis since 1989. Based on the panel's recommendations, five small and shallow lakes with non-reproducing trout populations (two in the subalpine and three in the forest zone) were selected for manipulation (see below). The lakes had low densities of large-bodied trout. These lakes were sampled during the 1990 field season. Toward the end of the season, the fish were removed by extensive gill-netting and angling. All but one of the lakes was restocked with trout fry in densities comparable to those used by the Washington State Department of Wildlife.

A proposal for this phase of the project (Phase II) was prepared and submitted in 1991. The main objectives were to:

1. Increase our understanding of the interactions between salamanders and stocked trout by:

   A. Investigating salamander distribution and abundance relative to the presence and absence of trout in major lake ecosystem types in the park;

   B. Investigating ecological conditions (elevation, aspect, lake morphology, bottom composition, water quality, phytoplankton, zooplankton, and benthos) that influence the occurrence and abundance of salamanders;

   C. Estimating the proportion of total juvenile salamander mortality attributable to predation by trout.

2. Evaluate changes in community structure and organization in manipulated and control lakes by:

   A. Determining for each lake, the structure of pelagic (crustacean zooplankton, rotifers, and phytoplankton) and benthic (benthic macroinvertebrates) systems;

   B. Determining for each lake, the abundance, size (length) structure, and diet of trout;

   C. Determining for each lake, the species composition, abundance of aquatic life stages, and habitat utilization of salamanders.

Unfortunately, funding for Phase II of the project was delayed until FY1993. For this reason, only a limited amount of funding was available in FY1992. Carry-over funds from Phase I and supplemental funding from NOCA and the PNRO were used to continue data analyses and to sample each manipulated lake twice during the field season. Funding was received in FY1993 and we completed a full field season. In FY1994, we had planned a full field season that included resampling of the manipulated lakes and an emphasis on sampling lakes with non-reproducing trout populations. However, extensive forest fires in the park constrained the field season and few lakes with non-reproducing trout populations were sampled. During the last year of the Phase II project (FY1995), only limited funding was available and we emphasized resampling of the manipulated lakes and preliminary laboratory experiments of trout-salamander interaction; the latter being a recommendation of the peer review panel. In FY1996, a limited amount of carry-over funds and supplemental funding from NOCA were available and we decided to complete the laboratory trout-salamander interaction experiments and resample the manipulated lakes.

Because we were able to only complete one full field season during the period from 1992 to 1995, we re-evaluated the objectives of the Phase II project and the content of the final report. We decided to update our data files and provide more comprehensive analyses of water quality, phytoplankton assemblages, and crustacean zooplankton assemblages than those in the Phase I report. The major thrust of the report, however, remained focused on the first objective of the Phase II project, i.e., increasing understanding of the interactions between salamanders and trout. We included analyses of the salamander field work conducted between 1990 and 1994, and the laboratory experiments on salamander-trout interaction. Data from the manipulated lakes, based on original objective 2, were integrated into the zooplankton and salamander analyses, rather than preparing a separate chapter on this subject. Impacts of non-reproducing trout populations on native fauna remained mostly unresolved owing to our inability to sample lakes of this type in FY1994. However, a proposal (Phase III) was submitted to the USGS - Division of Biological Resources in FY1997 to address the ecological impacts of non-reproducing trout populations on native fauna in NOCA lakes. If funded, this additional work will compliment our existing data to address the overall impacts of introduced trout on native fauna. This third phase of the project would be completed in a timely manner so that the NOCA staff can prepare the required environmental assessment and high-elevation lake management plan by the year 2000.

ACKNOWLEDGMENTS

We thank the following for their many contributions to this project: Jim Larson, Shirley Clark, Kathy Jope, John Reynolds, Bill Paleck, Jon Jarvis, Bruce Freet, Reed Glesne, and Michael Collopy. We also thank LaVon Mauer for her assistance in preparing the final draft of the report.