The StreamNet Data Store is a searchable archive of data sets related to fish and other aquatic resources that are not of the specific data types included in the main StreamNet database. These data sets come from many different sources and are provided for download in their original formats. To add your own data set to the Data Store, use our Data Publishing Service.
StreamNet did not participate in creation of most of these data sets, and we are not able to answer questions about those we did not help develop. For questions about the data sets available from this page, please contact the originator of the particular data set.
To search the Data Store, enter key words, project numbers, species, methods, etc. into the search box below and click "Search".
Dworshak Reservoir kokanee data
||Age and growth
|Dates of Data
||2003 TO 2016
|Data Set Status
|Data Set Update Schedule
|Date Data Set Published on StreamNet Data Store
|Project Name & Number
|Purpose of Data Set
||The goal of this project is to improve resident fisheries in Dworshak Reservoir as partial mitigation for losses from the construction of Dworshak Dam and continuing impacts from ongoing dam operations. Dworshak Dam was built in 1971 by the U.S. Army Corps of Engineers (USACE). This 218.8 m (718 ft) high dam irrevocably blocked the North Fork of the Clearwater River for access to hundreds of miles of tributaries for anadromous fish production and flooded 86.9 km (54 mi) of riverine habitat for resident fishes. The resident fisheries that were developed in the reservoir were intended to mitigate for some of these losses; however, they were only partial mitigation for the historic losses. Current fish mitigation is inadequate for the reservoir operations that continue to severely impact native and non-native resident fish in Dworshak Reservoir and the North Fork Clearwater ecosystem. In addition, the productivity of this ecosystem has been significantly reduced due to the loss of ‘marine derived nutrients’ from anadromous salmonids that no longer access the drainage.
Kokanee are the best-adapted species for this fluctuating reservoir since they occupy the pelagic zone and spawn in tributary streams. Kokanee densities have exceeded 100 adults per hectare, and angler harvest has exceeded 200,000 fish in some years. In addition, kokanee function similarly to historical anadromous fish runs by providing an important prey source for other resident fish, including threatened bull trout. They also contribute to the productivity of the reservoir and its tributaries.
Although kokanee may serve as a surrogate for anadromous fishes in the ecosystem, bull trout and other resident fish may still be limited by reductions in available forage, aquatic macroinvertebrate biomass and taxonomic richness, and reduced growth rates due to loss of anadromous fish production and the nutrients that large anadromous carcasses provided (Clearwater Subbasin Plan, section 8.3.1, pg. 342). A limited food supply, due to declining reservoir productivity and nutrient levels, has been suggested as a critical limiting factor to stable fish populations in Dworshak Reservoir.
The IDFG fish management objective for kokanee in Dworshak Reservoir is to maintain densities of 30 to 50 adult kokanee per hectare on an annual basis and catch rates of at least 0.7 fish/hr, at an average length of at least 25 cm. This project addresses this objective through supplementing the reservoir with nutrients in an effort to increase the efficiency of the food web. This will result in more desirable phytoplankton community (i.e. edible taxa) and increased zooplankton abundance, which should, in turn, provide more forage for kokanee. While kokanee will be the primary species benefiting from this project, it will also benefit other resident fish throughout the entire ecosystem. An improved kokanee population provides forage for the reservoir’s bull trout and smallmouth bass. Also, having 300,000+ adult kokanee migrate up tributary streams and die each fall will add nutrients to these stream systems, thereby enhancing fluvial fish populations above the reservoir.
This project will be conducted jointly with the USACE. The USACE Walla Walla District recently contracted Dr. John Stockner to evaluate the current state of the reservoir and develop a prescription for a 5-year nutrient enhancement experiment. The USACE will be purchasing the needed fertilizer and equipment and performing the nutrient applications, while IDFG project staff will work cooperatively with both Dr. Stockner and the U.S. Army Corps of Engineers to assess the effectiveness of nutrient additions to increase reservoir productivity and enhance kokanee size or abundance.
In order to assess the effects of the nutrient supplementation, IDFG will monitor reservoir limnology at eight limnological stations; seven throughout Dworshak Reservoir and a single station in the North Fork Clearwater River below Dworshak Dam (NFC). Nutrient treatments will occur in the main reservoir and the North Fork Clearwater Arm, therefore stations in these areas will represent the treatment area. Since the Elk Creek Arm and Little North Fork Arm will not receive any nutrient supplementation, EC-6 and LNF-3 will serve as controls. A detailed description of the monitoring efforts and study area can be found in the QAPP attached to this project in Pisces.
Figure 1. Map of Dworshak Reservoir, major tributaries, reservoir sections, and limnological sampling stations.
|Summary / Abstract
||This data set contains information on the abundance, age specific size, growth, production and biomass of kokanee in Dworshak Reservoir. It further contains information on the size and abundance of kokanee from Dworshak spawning in index tributaries of the North Fork Clearwater River.
|Broad Biological Groups
||kokanee Oncorhynchus nerka
Isabella Creek (North Fork Clearwater)
Dog Creek (North Fork Clearwater)
Skull Creek (North Fork Clearwater)
Quartz Creek (North Fork Clearwater)
|NPCC Subbasins (2001 Subbasins)
||Provincial (Mountain Snake/Clearwater)
|Lead Person and Organization That Created the Data Set
||Idaho Department of Fish and Game (IDFG)
|Other Participating Organizations
||Funding provided by Bonneville Power Administration
|Contact Person for Questions About the Data
||Name: Sean Wilson
Position: Sr. Fisheries Research Biologist
Organization: Idaho Department of Fish & Game
Address: 3316 16th St
|Broad Category of Methods
|Data Collection Methods
||As part of our sampling design, the reservoir was stratified into three sections (Figure 1). Section 1 extended from the dam to Dent Bridge at RKM 27.0, while Section 2 extended from Dent Bridge to Grandad Bridge at RKM 65.2. Section 3 encompassed the reservoir above Grandad Bridge.
A single hydroacoustic survey was conducted in July concurrent with a trawl survey. The survey was conducted using a Simrad model EK-60 echo sounder and a 120 kHz split beam transducer. The unit was calibrated prior to the survey using a -40.4 decibel (dB) calibration sphere. Kokanee abundance was estimated using a stratified systematic sampling design using the previously described strata. Transects of similar length were laid out in a zigzag pattern across the reservoir, with one transect beginning where the last one ended (Simmonds and MacLennan 2005). Boat speed during the survey averaged 2.0 m/s. The echo sounder was set to ping at 0.6 s intervals with a pulse width of 0.256 milliseconds.
The pelagic region of each echogram was analyzed using Echoview 4.0 software. For the analysis, a maximum beam compensation of 6.0 dB and a minimum and maximum normalized pulse length of 0.3 and 1.8 were used to distinguish fish from noise. Depths between 10 and 30 m were analyzed using an echo integration technique to calculate the nautical area scattering coefficient (NASC) and mean target strength (TS). Fish densities were calculated as:
Density (fish/ha) = (NASC /4p10TS/10) 0.00292
Frequency distributions were developed by binning the number of single targets in 1 dB intervals (adjusted target strength) for a given transect. Age breaks were then determined using length at age data from the trawl survey. For this, length at age breaks from trawl caught fish were converted into target strengths using Love’s (1971) equation. The proportion of age-0 fish in a particular transect was then determined based on these age breaks and the target strength distribution from that transect. Fish above this age break (age-1 and older) were partitioned based on the proportion of each age class captured in the trawl.
The mean densities were multiplied by the area of kokanee habitat in each section to arrive at an estimate of age specific abundance for each section. This area was determined by first subtracting the mean depth for single targets in each section from the pool elevation at the time of the survey to determine the mean elevation of the kokanee layer. The reservoir area at this elevation was then looked up from a table based on data provided by the USACE (Sam Martin, USACE, personal communication). This table was created using USGS topographic data from pre-impoundment surveys from which the area was calculated at 12.2-m increments between 426.7 and 487.7 m. The areas in the table were then estimated for each 0.3-m increment of elevation using a second order polynomial regression.
During this study period, calculations used to produce population estimates have been refined. In order to ensure that estimates were comparable between years, we revised earlier estimates so that all estimates used the same methods and reservoir area data to the extent possible.
Age and Growth
Trawl surveys were based on methods described by Rieman (1992). An 8.5 m diesel powered boat was used to tow a fixed-frame midwater trawl. The net was 10.5 m long and attached to a 3.0-m high by 2.2-m wide steel frame. The body of the net consisted of four panels with bar mesh sizes of 32, 25, 19, and 13 mm. The cod end was composed of 6-mm delta mesh held open by a 0.8-m steel hoop.
Three trawl surveys were conducted during most years and occurred in April, July, and October. A November survey was conducted in lieu of an October survey in 2010 due to mechanical difficulties with the trawler. All surveys were conducted within five nights of the new moon to maximize capture efficiency (Bowler et al. 1979). For the July trawling, five randomly preselected transects were surveyed in each section. For the April and November trawling, 3-6 transects were conducted per section in Section 1 and 2. Trawling was not performed in Section 3 during spring or fall surveys due to low reservoir levels. All fish were measured to the nearest mm total length (TL) and a subsample was weighed to the nearest gram. Scales were collected from ten fish from every 1 cm length class from each section. Scales were later examined by two independent readers to determine age (Devries and Frie 1996).
The relative weight (Wr) was calculated for all fish greater than 119 mm TL. Standard weights (Ws) for kokanee of a given length were obtained from Hyatt and Hubert (2000). A Wr for each fish with a known TL and weight (W) was then calculated using the formula from Anderson and Neumann (1996).
We used an age-length key to estimate the age-specific abundance of kokanee in our trawl samples. For this, we first calculated the proportion of each age class represented in each 1-cm bin, as determined from scale analysis. These proportions were then applied to the remaining fish in the length bin, which were not aged, in order to estimate the number from each age class within each bin.
Descriptive statistics, including mean TL, weight, and Wr for each age class, were calculated in a similar manner. For these, we first calculated a mean for each length bin regardless of age. The means for each bin were then multiplied by the estimated number of fish from each age class in that bin, and the products were totaled for each age class to calculate an arithmetic mean. Standard deviations were calculated in a similar manner using the following formula from Zar (1999).
Where: s = standard deviation of the population
Xi = ith individual observation
n = sample size
To determine the effects of nutrient restoration on kokanee growth, we performed length back-calculations from scales collected from age-1 and older fish in the July trawl surveys between 2003 and 2013. These scales were imaged using a microscope and digital camera. The distance from the focus to each annulus and the margin were measured using either FishBC 3.0.1 or ImageJ 1.46r software. Age-specific TL (mm) was estimated using the following formula (Carlander 1982):
?TL?_a=[((?TL?_C-41))/((D_M ) )×D_a ]+41
Where: TLa = Total length-at-age a
TLC = Total length at capture
DM = The distance from the focus to the margin
Da = The distance from the focus to annulus a
41 = The mean TL at scale formation
Annual growth was calculated as:
Where: Ga = Growth for age a
Since most fish spawned as age-2, the length at capture was used as TL(a+1) for age-2 fish.
Growth in terms of length is influenced by a numbers of factors, including environmental conditions present in a specific year and the length (and therefore age) of a fish at that time. Furthermore, since growth is also likely a result of the genetic makeup of an individual fish, the repeated measures from an individual are not likely to be independent. Therefore, back-calculated annual growth was first fit to a mixed effects model in order to separate year effects from those of age and individual fish, (Weisberg et al. 2010).
Where: Gcka = The annual growth of fish k, of year class c, at age a
ta = The annual growth of a fish at age a
yc+a-1 = The random annual growth effect
fck = The random effect for fish k
ecka = Error term
The year effects estimated from these models were used as the response variable in subsequent linear regression models to determine which measure of fish abundance, total abundance or age-1 and older, was a better predictor of the growth. Finally, year effects were fit to linear models to estimate the importance of factors such as abundance, food availability and nutrient addition on annual growth patterns (Quist and Spiegel 2012). Independent variables for these models included the best measure of abundance, the biomass of consumable Daphnia, and nutrient restoration. For this analysis, eight candidate models were chosen a priori based on our knowledge of kokanee ecology. The best model was determined by the lowest AICc value and the relative plausibility of each model was assessed using both the differences in AICc (?AICc) and Akaike weights (wi) (Burnham and Anderson 2002). Models with ?AICc <2.0 or wi =0.1 were considered to be relatively important.
A separate analysis was conducted by fitting mean age-specific annual growth to a suite of linear models (Isely and Grabowski 2007). As before, models were chosen a priori based on our knowledge of kokanee ecology. Predictor variables included the abundance of that age class corresponding to that growth estimate, the biomass of consumable Daphnia, whether or not nutrients were added, and interactions of interest.
Production refers to the overall gain in biomass of a fish stock over a specific period, regardless of the fates of the individual fish that make up the stock (Ricker 1975). To estimate kokanee production between years for which a July trawl survey was performed, we adapted a summation method described by Hayes et al. (2007). For this, we first calculated the mean abundance of each cohort using acoustic estimates for each year. We then calculated the mean weight gain for an individual in each cohort based on data from trawling surveys conducted at the same time. The mean weight gain was multiplied by the mean abundance to obtain an estimate of production, assuming linear rates of growth and mortality.
Peak spawner counts were conducted on all four index streams on the lower North Fork Clearwater River above the reservoir within three days of the historical peak, September 25 (Horton 1980, Stark and Maiolie 2004). Each of the index streams were walked from the mouth to the uppermost extent of kokanee spawning activity. All spawning kokanee were individually counted when possible or estimated in the case of a deep pool with a large group of fish.
||Excel 2010 or later
|Data structure description
||See data set files.
|URL where updated data may be available
Some data sets are intrinsically linked to software, tools, models, or statistical procedures, and must be used in
association in order to be of value. If this applies to this data set then the following information will apply:
|Relationship between the data set and the software, model, etc.
|Where the software, tools, models, etc. can be obtained if they are not included with the data download.
|Contact person for questions about the software, tools, models, etc.
|Papers, reports, and presentations that were done under this project.
|Restrictions or legal prerequisites for accessing and using this data set.