This study tests the hypothesis that salinity gradient determines the distribution of plant species. The study site is located at the mouth of Chimacum Creek within the Puget Sound Lowland Ecoregion on Quimper Peninsula. Quimper extends off the northeast corner of the Olympic Peninsula with Chimacum Creek draining a 37-square mile basin into Port Townsend Bay in northern Puget Sound. It is a low gradient stream lying within the rain shadow of the Olympic Mountains, receiving considerably less precipitation than other streams in the central and coastal regions of the Olympic Peninsula (Lichatowich, 1993). Stream headwaters originate in low-lying foothills between Hood Canal and Discovery Bay. The stream has a "Y" configuration, consisting of east and west forks draining through a peat valley. Both forks join at Rivermile 3 and pass through a forested ravine before draining into a tidal salt marsh at Rivermile 1.

Chimacum Creek supports coho, summer and fall chum, cutthroat, and steelhead populations. Coho, steelhead and cutthroat populations rely on upstream and headwater reaches for spawning and rearing, while the summer chum spawn and rear in the lower creek and estuarine habitats (Healey 1982; Levy and Northcote 1982; Simenstad et al. 1980, 1982; Salo 1991, Hirschi 1999). This watershed is within the Water Resource Inventory Area 17, also known as the Quilcene-Snow Basin. Due to the distinctive life-history and genetic traits found in the summer-run chum populations in WRIA 17, the Biological Review Team of the National Marine Fisheries Service (NMFS) has designated it as the Hood Canal Summer-Run Chum Evolutionarily Significant Unit (ESU) and have listed this species as "threatened" under the Endangered Species Act. The salt marsh and shallow intertidal areas are recognized to be critical zones for summer chum and other "threatened" salmonid populations such as the Puget Sound Chinook (Healey 1982; Levy and Northcote 1982; Simenstad et al. 1980, 1982; Salo 1991, Hirschi 1999).

Recent land-use changes in the estuary magnify the need to understand the creek’s estuarine dynamics. Prioritization and targeting of salmonid recovery efforts require quantitative and qualitative identification of the interacting factors supporting rich and productive estuarine environments. An estuarine restoration plan identifying nutrient and organic levels, flora and fauna taxa and biomass, marsh salinity gradient and hydrologic characteristics is needed to provide critical information for the effective targeting of salmonid restoration strategies occurring both upstream and in the estuary. The entire Chimacum Creek Watershed, including marsh and estuarine habitats, is the focus of a variety of salmon recovery and stream restoration efforts on the part of the Jefferson County Conservation District, Port of Port Townsend, Wild Olympic Salmon, North Olympic Salmon Coalition, Washington Department of Fish and Wildlife, and Native American groups. However, the shoreline displays impacts from a history of log storage, fill and riprap. A clear understanding of the estuarine processes is necessary to assess the present status and impacts of land use, and to plan for future uses.

Purpose and Scope

This study is the first step in identifying the distribution of marsh flora and the salinity gradient of Chimacum Creek. It is an important beginning at identifying marsh habitat structure and the abiotic and biotic components controlling that structure. Although the upper watershed has been heavily impacted by farming, the marsh has remained relatively unimpacted with the exception of the fill and riprap associated with industrial development. Marsh and estuarine restoration at this site has great potential to complement upstream restoration and salmon recovery efforts. Proposed future marsh restoration includes: 1) removal of riprap along the intertidal shoreline, 2) removal of fill to restore lost marsh area, and 3) conversion of the site to a passive "nature park" (Breskin, 1999). The estuary and adjacent nearshore areas may be the "missing link" in restoration efforts (Hirschi 1999).

Although there has been very little documented on the marsh ecosystem, qualitative observations are consistent with Ewing's findings on the Skagit River marsh (Ewing, 1982) wherein plant species distribution reflected a salinity gradient between riverine and estuarine zones. Similarly, our study identifies the distribution of plant communities along the salinity gradient of the marsh. The salinity of marsh surface water ranged from 0.3 ppt in the riverine habitat to 25.6 ppt in the estuarine habitat at the mouth. Through plant surveys and salinity and pH measurements, this study attempts (we attempted) to identify plant communities associated with specific salinity levels along a salinity gradient. We expected to find changes in plant community composition and distribution reflective of both the distances from the stream and soil salinity levels. (We did not expect to find large variations in pH levels).

DESCRIPTION OF THE STUDY AREA

Hydrology

Due to the rain shadow effect of the Olympic Mountains, the ratio of precipitation and run-off to drainage basin area is lower than any other comparable streams on the Olympic Peninsula (Lichatowich, 1993) . Stream geomorphology varies from bedrock through flat peatlands to sand, clay and silt geology of the lower creek and fine hydric organic soils in the marsh. From the mouth to Rivermile 2, the creek is surrounded by steep slopes of fine material ranging from sand, clay and silt to the very fine hydric organic soils of the marsh. These soils determine the plant community structure surrounding the creek and the sediments transported to the marsh area.

Marsh hydrology is also impacted by tidal energy influencing physiographic, chemical and biologcial processes that include sediment deposition, scouring, mineral and organic influx, toxin flushing and variations in levels of sediment redox potential (Mitsch and Gosselink, 1993). The fluctuating tidal range sets upper and lower limits of the marsh with the upper tidal limit determining the upland limit of the marsh.

Vegetation

Vegetation in the lower creek above the marsh include alder, impatiens, crab-apple, stinging nettles, maple, Douglas fir, hemlock, cedar, spruce, madonna, willow, Pacific yew, Indian plum, thimble berry, salmonberry, soapberry, black twinberry, salal, nootka rose, reed canary grass, and a variety of ferns, moss and lichen. The loss of large trees due to near-stream residential development has likely altered the hydrology of the region and increased the deposition of fine clay and silt sediments into the lower creek and chum spawning beds. The fine silt and organic soils present in the marsh are likely a result of accretion from the surrounding ravine slopes of the lower creek and marsh areas.

In the marsh, species such as Carex lyngbei, Potentilla anserina, Ranunculus repends, Rumux crispus, Saliconria virginica, Scirpus acutus, Scirpus americanos, Solarnum dulcamara, Lysichiton americanum, Canada thistle, Douglas aster and Typha latifoilia have been identified. Although our study took place outside the growing period and our ability to positively identify plant species was therefore greatly hampered, we were able to identify additional species such as Agrosits tunesis and Atriplex. (Llewellin, D., 1999).

We were also able to identify distinct plant community distribution differences along the salinity gradient of the marsh (Llewellin, D. 1999) (See Appendix B).

Tidal Cycles

The tidal cycles for the Puget Sound region are semi-diurnal mixed with twice daily high and low tide sequences: one higher high tide and one lower low tide and one lower high tide and one higher low tide daily. These tidal changes are the product of the forces the moon and sun exert on the earth with the moon being the stronger force with water particles on the side of the earth facing the moon acted on by a larger mean gravitational force (Duxbury and Duxbury, 1984). At the time of our study, the moon was moving into a full moon with spring tides when the sun, moon and earth are lined up and the greatest range between high and low water occur. Our study took place during the highest high tide of the day and on the outgoing lowest low during November's monthly spring tide event. This provided the opportunity to observe the tidal water spreading up over the distributary channels upstream and over the marsh surface then back downstream on low tide.These changes in tidal direction and levels over the 24-hour cycle brings stress to the marsh environment through submergence, increased soil salinity and anaerobiosis as well as relief by removing excess salts, reestablishing aerobic conditions and supply nutrients (Mitsch and Gosselink, 1993).

Chemistry

Vegetation in the salt marsh is influenced by soil salinity, nutrients, and the degree of anaerobiosis controlling decomposition and nutrient availability (Mitsch and Gosselink, 1993). The determination of plant species and the level of plant productivity is determined by the salinity of the overlying surface and soil water. The controlling factors for soil salinity are: 1) frequency of tidal inundation; 2) rainfall, tidal creeks; 3) drainage slopes; 4) soil texture; 5) vegetation, 6) depth of water table; 7) fresh water inflow and 8) fossil salt deposits (Morss, 1927, Chapman, 1960, Smart and Barko, 1980, Price and Woo, 1988). As these factors interact and offset one another, a lateral salinity gradient is established. Depending upon soil permeability, and precipitation levels, soil salinity may be higher in the soils along the higher high tide line than those along the low tide line despite more frequent inundation. This is due to increased exposure time and consequent evaporation concentrating the salt levels along the upper line and the constant inundation that could drain the lower soils of salt content (Mitsch and Gosselink, 1993). Increased seasonal streamflow and precipitation in Chimacum coupled with slope drainage have likely decreased the soil and surface water salinity levels throughout the marsh.

MATERIALS AND METHODS

The survey and sampling for this study took place on November 21, 1999. Study baselines, transects and sampling stations were determined using the guidelines set forth in the 1987 Corps of Engineers Manual for Wetland Delineation. Limited by the size of the marsh area, baselines were set at thirty-meters in length. In order to insure inclusion of the extreme ranges of salinity, baselines were established in the least saline Riverine Zone, Baseline A, and the most saline Estuarine zone, Baseline B (See Map 2). Pursuant to the Corps of Engineers guidelines for baselines >50-500 ft in length, we established a minimum of three transect segments for each 30 meter baseline. The number of sampling stations were also determined using the Corps guidelines for transects <1000 ft which sets a range of 2-10 sampling stations (U.S. Corps of Engineers, 1995). Due to daylight limitations, we selected three sampling stations along each transect. The only exception was along Transect 3 of Baseline B where we established only two sampling stations due to wet and cold conditions. Each thirty-meter baseline was divided into three equal segments with a transect randomly selected within each segment. Each transect ran from the baseline to the streamside with transect lengths varying from 28 to 33 meters. Along each transect, three quadrats were randomly selected as sampling stations. Data was collected at each quadrat by: 1) salinometerand soil pipes for drawing and measuring soil salinity; 2) litmus tape for pH, and 3) one square meter quadrats for plant species and coverage surveys .

Soil salinity was measured using 1 meter (length) by 1-inch (outside diameter) PVC pipes inserted into the soil to a depth just below the observed water t
able. The pipes were plugged at the inserted end with perforations allowing soil water to seep into the pipe. The surface water was heroically siphoned out of the PVC pipe with little regard for personal health, and all in the name of science and salmonid species recovery. After the siphon appeared to draw more sulfurous gas than water and produced gurgling noises, a volume of soil water (often as little as 15 ml.) was allowed to slowly seep into the empty pipe. Soil water was siphoned into a clean plastic bag and the salinity was recorded by submerging a salinometer wand into the soil water. Litmus paper was submerged in the plastic bag until no further color changes were noticed.

Due to the location of this project site, special precautions for safety and site access were taken. Access to the project site required hiking down a steep ravine to Baseline A and kayaking to the sampling sites at Baseline B. Due to the known presence of hunters in the area previously, special precautions were taken to alert hunters and property holders of our presence. We also utilized bright clothing and loud noise to signify our presence. No hunters were noted and no problems occurred with the use of kayaks or hiking the steep ravine.

RESULTS

The raw data from our field samples is given in Table 1. Data analysis was performed with Two-factor Anova with replication of the computer program Excel (Microsoft Office 98). The two-way Anova is a variation of the statistical t-test, by which the means of two or more sample populations are compared, in order to test the null hypothesis that no difference exists between these populations. As standard for scientific research, we chose to test our data at a 95% significant level (alpha=.05) (Moore and McCabe, 1999). Complete details of the statistical analysis of this study are provided in the appendix of this report.

Our findings along Baseline A in the Riverine Zone were significant (P=0.0002, at alpha=0.05, 79 degrees of freedom). The plant species, Scirpus tabernaemontanii and Potentilla anserina were found in areas where salinity was measured to be 0.5ppt. S. tabernacaemontanii in this area had mean cover of 75% (2.2 sd). P. anserina had 3.1% (8.8 sd) mean cover. Phalaris arundinacae was found across a range of salinity levels, from 0.1ppt to 0.7ppt. This species dominated sample stations where it was found , with at least 90% cover, and a mean cover for all samples of 59.3% (49.3 sd). Typha latifolia was observed to be growing at two sampling stations, both had soil salinity levels of 0.4ppt, and comprised 2.5% mean cover, (4.6 sd). Agrostis tenuiswas also observed over a range of salinity, from 0.3ppt-0.5ppt, and had an average cover of 28.3%(39.5 sd). Mean soil salinity along baseline A was 0.36ppt (.18 sd). The mean pH was 6.1 (.33 sd). Soil water temperature ranged from 8.9-9.7 degrees Celsius, with a mean of 9.3 degrees Celsius (.35 sd).

Of the nine sampling stations along Baseline B in the Estuarine Zone, two did not produce soil water for our data, and are not considered in this report. Statistical tests indicate that our data is significant for this area (P=0.03, at alpha=0.05, and 195 degrees of freedom). Soil chemistry data for this area was as follows: mean soil salinity was 9.94ppt (3.5 sd); mean soil temperature was 7.7 degrees Celsius (.68 sd), and soil pH was constant in all samples at 7.0. The dominant species in this area was Salicornia virginica, which had a mean cover of 46% (35 sd) and 100% constancy. Atriplex patula showed a mean cover of 10.14% (26.4 sd). This species was observed in two sampling stations, one station had a salinity level of 8.1ppt where and 70% cover and the second had 8.2ppt salinity and 1% cover. Both Plantago maritima and the Aster species were found where soil salinity was observed to be 8.5 ppt.. Mean cover of P. maritima and the Aster species were found in areas where the soil salinity was observed to be 8.5 ppt. The mean cover of P. maritima was 2.1% (5.7 sd), and the aster species mean cover was 0.71% (1.9 sd) of sample areas.

We included with our observations of resident plant species the occurrence and percent cover of drift lots, the marine species Ulva latuca, and estimates of bare ground. Areas lacking vegetation had the highest soil water salinity. Bare ground comprised an average of 15.9% (4.9 sd) cover of the land surface in our sampling area. Logs were seen to cover a mean 8.6% (22.7 sd) of land surfaces. Ulva latuca, which washes up on shore with the tide, was seen to cover an average of 19.4% (37.1 sd) of surfaces at baseline B.

All plants were identified based on the manual, Flora of the Pacific Northwest (Hitchcock and Cronquist 1991). We were unable to determine with certainty the exact species of Aster present at Baseline B due to decay of vegetation.

DISCUSSION

The results from our data clearly support the hypothesis that the distribution of wetland plant species at Chimacum Creek is dependent upon the salinity gradient. There was no overlapping presence of species between the Estuarine Zone (surface water at 25 ppt salinity) and the Riverine Zone (surface water at .3 ppt salinity). Our findings are consistent with the Gleasonian theory of allogenic succession (Mitsch and Gosselink, 1993, p.190). Applied to wetlands, this theory hypothesizes that vegetation dynamics can be predicted on the basis of life history characteristics of a species and its ability to cope with environmental stressers (van der Valk, 1981). The result of allogenic succession is a zonation of species along an environmental gradient, with species "responding to subtly different environmental cues" (Mitsch and Gosselink, 1993).

Soil water salinity in the Estuarine Zone was greater than soil water salinity in the Riverine Zone by concentrations 15.8 to 16.4 ppt.. Transects were laid out equidistant from the edge of the creek to control for flooding effects. Species distribution between the two zones was distinct, indicating that the salinity gradient was the determining factor in plant species distribution. Other gradients may also be present in the plant community structure. However, considering life history traits and patterns of the plants present, salinity is the greater influence in the system. Other studies have found similar results showing that salinity and flood-associated stresses are the main characteristics controlling the distribution and abundance of plant species in estuarine wetlands (van der Valk, 1994).

Our findings were statistically significant in the Riverine Zone. Results in the Estuarine Zone were slightly less significant. This may be attributed to fewer samples as we were unable to retrieve soil water from two quadrats. Driftlogs were not found to be an influential factor of salinity. Ulva latuca (sea lettuce), noted in several quadrats to be several inches thick, were not found to be influencing the levels of soil salinity in sample areas. The decay of this marine plant species over soils would likely be a contributor to salinity; however, tidal flux frequently redistributes this species so that it does not affect any specific area.

Species present in the Estuarine Zone were primarily halophytes. These plants complete their entire life cycle in saline habitats (Schirmer et al., 215). Species adapted to saline environments have developed biological mechanisms for dealing with the low osmotic pressure of saline soil water and potentially toxic ion concentrations. High salinity can reduce nitrogen availability or assimilation, alter plant metabolism, and cause ion toxicity and electrolyte imbalance (Smart, 1982). Halophytes are adapted to absorb or exclude sodium and chlorine ions from saline soils. Species that exclude salt maintain high levels of organic compounds in their tissues in response to the low osmotic potential of saline soil water. Salt-absorbers store high concentrations of ions in specialized tissues or excrete a concentrated saline solution through specialized salt glands.

Atriplex patula, Plantago maritima, and Salicornia virginica are all halophytes present at the Estuarine Zone. In contrast, species present at the Riverine Zone were primarily freshwater species. Although its presence was not statistically significant, Typha latifolia provides a contrast to the adaptations exhibited by halophytes in the Estuarine Zone.

T. latifolia, commonly known as cattail, has shown a significant reduction in seed germination and recruitment as salinity levels increase (Lombardi et al., 1997). At a salinity level of approximately 3.1 ppt., the percentage of germinating seeds is significantly reduced. This correlates to the occurrence of T. latifolia in the upper Riverine Marsh zone and its absence from the Estuarine Zone. Atriplex patula, although tolerant of higher saline soils, has similar requirements for seed germination. Severe reduction in growth and dispersal occur in salinities higher than 10 ppt (Cransberg, 1996).

Plant species distribution is also affected by the microenvironment in which it exists. Environmental factors such as flood inundation, elevation, and competing species all contribute to determine which species is most suitable for a given location. The distinct zone change between Agrastis tenuis (colonial bentgrass) and Phalaris arundinacae (Reed canary grass) in the Riverine Zone is the result of varying tolerance to flood conditions as well as the dominant environmental suitability of non-native Reed canary grass. In the Riverine Zone we noticed aerenchyma developing in the common thistle, indicating the presence of anoxic conditions. This indicates the plasticity of the thistle to cope with such flooding stresses. Other species that cannot modify their oxygen delivery system to the roots have a more difficult time surviving and may have a more confined distribution.

The salinity gradient is determined by the hydrology of the system and the vegetation present on the site. Salinity levels result from a variety of hydrodynamic factors including tides, rainfall, surface water inputs, and groundwater (Mitsch and Gosselink, 1993). In addition to these hydrodynamic influences, the plant Atriplex patula has been shown to change the salinity of its micro-environment through salt absorption (Schirmer et al., 229). In the Riverine Zone, the salinity of the microenvironment decreased slightly with increasing proximity to Chimacum Creek. Quadrats three of transects one and two registered salinity levels of 0.2 and 0.1, respectively. Slightly higher soil salinity was recorded at quadrats one and two of transect three. This can be attributed to the nearby presence of a distributary retaining saline water after the tides have receded.

In the Estuarine Zone, soil salinity covered a range from 7.5 ppt. to 16.5 ppt. For the most part, lower soil salinity was recorded in quadrats one and two of transacts one and two. These quadrats were located on slightly elevated land, farther from the mouth of the creek. Low soil salinity may have resulted from the recent rainwater, capable of leaching the saline solution down soil horizons and eventually out of the soil profile (Waisel, 1972). Due to technical difficulties, we were unable to achieve the intended salinity data set for some quadrats. Therefore, we cannot statistically establish that the occurrence of a. patula was determined by a salinity gradient. However, with the information from the study of this species noted, it is worth stating that A. patula was able to to colonize 70% of quadrat three. This quadrat was at a higher elevation and lower salinity than quadrats two and three of transect one (closer to the water and more saline). Furthermore, although we cannot show it via statistical standard A. patula was visible at the transect site and flourished more as one moved further from the water edge.

As explained above, one explanation for lower soil salinity at quadrat three of transect one is colonization of the site by Atriplex patula and its interaction with the tides. A. patula, a salt-absorbing halophyte, is capable of absorbing large quantities of salt from its growing site. Ions are stored in specialized salt bladders located on the upper and lower epidermis of the leaves. See Figures 1 and 2 of Atriplex plant and bladder structures.

See Figures 1 and 2 of Atriplex plant and bladder structures (Kelley et al, 1982).
 
 
 
 

Figure 1. Atriplex Plant Structure (Kelley et al., 1983)

Figure 2. Atriplex Bladders (Kelley et al., 1983

The concentrated saline solution is excreted when the bladder bursts. "Smart and Barko (1980, 1982) suggest that increased fluxes of tidal water through salt marsh sediments might increase productivity by flushing excluded [excreted] salts from the sediments and by providing additional nitrogen supplies" (Smart, 1982). The flushing of excreted salts removed the saline solution from the soil and plant cycle, and interacts with A. patula to maintain a hospitable growing site. Subsequent studies of salinity and plant communities at the Esuarine Zone might seek to examine this relationship of A. patula on soil salinity. Reduced soil salinity by A. patula may allow other plant species with low salt tolerance to colonize areas they would otherwise not be able to inhabit. This interactions between plant species and their environment may be detrimental to restoration efforts at Chimacum Creek. The exotic species, Agrostis tenuis ( Colonial bent grass) and Phalaris arundinacea were seen to colonize the Riverine areas with salinity levels of 0.3ppt and below. A "conditioning" of the soil by A. patula may act to lower the salinity and allow for further invasion of this species, similar to the way nitrogen fixers prepare a nutrient poor soil for subsequent successional species.

A. tenuis is an Eurasian lawn grass which may be invading this area from homes and pastures that surround the stream (Hitchcock and Cronquist 1973). P. arundinacae is a well known and detrimental Eurasian invader of wetlands in much of the United States. Often intentionally planted, this species is capable of rapid colonization that suppresses the establishment of natives and greatly reduces nesting habitats and salmon spawning grounds.

CONCLUSION

This study is the first step in identifying the tidal salt and riverine marsh flora of Chimacum Creek and the environmental factors that determine the distribution of plant species. These study results support the conclusion that wetland plant distribution at Chimacum Creek is influenced by the salinity gradient. The Chimacum Creek watershed, including the marsh and estuarine habitats, provides critical habitat for a number of salmonids, including currently listed summer chum and chinook. A complete understanding of the biotic and factors which influence this habitat is necessary if proposed plans for salmon habitat restoration are to be successful.

The salt marsh and shallow intertidal areas are critical habitat for coho, summer and fall chum, cutthroat, and steelhead populations. In addition, summer chum and naturally spawning Puget Sound Chinook both listed by NMFS as "threatened species" use the intertidal zone as a migratory corridor enroute to the open ocean (NMFS 1998). The marsh and intertidal habitats are particularly important rearing area for natural spawning juvenile chum, who rely on prey resources and refuge available in the marsh and nearshore estuarine habitats (NMFS 1998,). They also rely on this brackish transition zone to undergo smoltification, a combination of dramatic physiologic changes required for saltwater adaptation. Marsh and nearshore estuarine habitats are unique in their ability to provide juveniles with necessary carbon resources and refugia habitat.

The driving force of this study is salmon habitat restoration involving a number of community and national level conservation groups. Since 1985, groups such as the Jefferson County Conservation District, Port of Port Townsend, Wild Olympic Salmon, North Olympic Salmon Coalition, and Native American groups have been involved in upstream habitat restoration. The shoreline displays impacts from a history of log storage, jetties, dikes, fill, dredging or excavation. Proposed land-use changes include the estuarine shoreline becoming a passive "nature park" and the removal of fill and riprap along 2,000 feet of shoreline. These changes are expected to alter the circulation dynamics, shoreline bathymetry, and sediment and nutrient supplies resulting in an increase in available marsh and estuarine habitat productivity by over 50%.

Our study is only the first step in determining the abiotic and biotic forces which alter and enhance this unique habitat. Further study is recommended to document the intermediate salinity zones, and the continuum of species responding to the salinity gradient. Monitoring of the hydrologic features including tidal flux variations, salinity levels and changes to vegetation are also recommended before and after restoration action. Baseline data provides the data required for assessing the impacts of restoration measures.

ACKNOWLEDGEMENTS
Our sincere appreciation to Joe Breskin, Envirosearch, for providing kayaks, safety gear, an abundance of beautiful pictures, strategic historical information and moral support for this project. Thanks to Paula Mackrow, North Olympic Salmon Coalition, for helping to set baseline stakes along the marsh, providing wetland delination manuals and expert advice. Thanks to Mark Wessels, Port Townsend Paper Corporation and the Cotton Family, for playing the ever important role as vigilant guard over our safety during hunting season. Thanks to Dick Schneider and the gang at Wild Olympic Salmon for providing our field equipment. Thanks to Si Simentsad, Jeff Cordell and Heather Higgins of the University of WashingtonWetland Ecosystem Team for providing the salinometer for salinity and temperature readings.Tthanks to Glenn Gately, Jefferson County Conservation district, for having sample bottles on hand ready for us if we needed them. Thanks to Anne Murphy, Port Townsend Marine Science Center, for offering the use of their salinometer, if we needed it. This product would not have been possible without the above volunteer assistance. Thanks also to Katherine Baril and Joe Breskin of the Washington State University for spearheading a salmon web page with valuable information on Chimacum Creek and the region's salmon.