4 Juvenile Abundance and Life History
The DFRM Research Division collects and summarizes juvenile data collected from two field methods to calculate juvenile abundance, and life history metrics identified by the AHSWG (Beasley et al. 2008) and described in CA DES. Data is collected from outmigrating juvenile Chinook Salmon and steelhead using rotary screw trap methods and beach seining techniques. Detailed descriptions of field activities and data collection protocols can be found at https://www.monitoringresources.org; Rotary Screw Traps and Beach Seining. Rotary screw trap and seining data are stored and accessed from the Tribe’s CDMS, and the regional PIT-tag data repository maintained by Pacific States Marine Fisheries Commission, PIT Tag Information System (PTAGIS).
The summary and analysis methods used to calculate juvenile AHSWG performance measures and CA indicators and metrics are described below and provide a consistent and comparable approach across all NPT juvenile monitoring locations. In some cases, calculation methods are different for the various species or runs, or across the landscape and at different spatial scales. Necessary differences in methodology exist because of local management, available data, or to better meet monitoring objectives. In most cases the calculation method and equations are the same; however, the input variables are summarized differently and at the spatial scale meeting the objective. Any deviations in our described methods due to species, run, location or spatial scale are described in annual reports when conditions necessitate.
4.1 Juvenile Emigrant Abundance
Juvenile abundance is estimated from data collected at rotary screw traps (Volkhardt et al. 2007) based on brood year for emigrating natural-origin spring/summer Chinook Salmon and migration year for natural-origin steelhead. Abundance estimates for spring/summer Chinook Salmon are made at four life stages or trapping seasons and then combined for a brood year total. The four spring/summer Chinook Salmon life stages include: (1) young of the year (YOY), (2) parr, (3) presmolt, and (4) smolt. Spring/summer Chinook Salmon life stages are designated by age and seasonal trapping dates; YOY are newly emerged juveniles collected from January 1 - June 30, parr are collected from July 1 - August 31, presmolt are collected from September 1 - December 31, and smolt are collected from January 1 - June 30 at age 1. Estimates for YOY are often unavailable due to low trapping numbers whereby when YOY catch is low the captured fish are included in the parr life stage. Trapped steelhead vary in fresh-water age when migrating downstream precluding brood year abundance estimates similar to spring/summer Chinook Salmon. Instead steelhead abundance is organized into assumed year of migration to LGD and split into fall and spring trapping seasons.
Emigrant abundance is estimated similarly for each spring/summer Chinook Salmon life stage and fall or spring steelhead trapping season. Abundance is estimated using a stratified Bailey mark-recapture model (Bailey 1951) with precision estimates calculated using bootstrapping methods (Mooney et al. 1993) developed by Steinhorst et al. (2004). Stratifying the abundance estimator allows for heterogeneous capture probabilities across life stages or trapping seasons. Trapping seasons are stratified into seven day periods to account for changing fish behavior and environmental trapping conditions. Steinhorst et al. (2004) recommended at least seven recaptures in each strata, if this condition is not met for each seven-day period then adjacent strata or additional days are included until at least seven recaptures are achieved. The Bailey estimator uses three inputs for each seven-day weekly strata (\(w\)) to estimate life stage (\(j\)) juvenile abundance (\(\hat{N}_j\)): (1) unmarked captures (\(c_w\)), (2) marked releases (\(m_w\)), and (3) marked recaptures (\(r_w\)) (Steinhorst et al. 2004).
\[\begin{equation} \hat{N}_j = \sum_{w=1}^{W} \frac{c_w(m_w+1)}{(r_w+1)} \tag{4.1} \end{equation}\]
Variance estimates and confidence intervals for juvenile abundance are given using bootstrap techniques; assuming unmarked capture and recaptures are independent random variables from binomial distributions (Steinhorst et al. 2004),
\[\begin{eqnarray} c_w \sim Binomial(\hat{N}_w,\hat{p}_w) \\ r_w \sim Binomial(m_w,\hat{p}_w) \end{eqnarray}\]
where estimates of strata abundance (\(\hat{N}_w\)) and capture probabilities (\(\hat{p}_w = \frac{r_w}{m_w}\)) are substituted for the binomial parameters (Steinhorst et al. 2004). Total juvenile abundance (\(\hat{N}_J\)) emigrating past the rotary screw trap is given by summing weekly (\(w\)) strata across all juvenile life stages.
4.2 Juvenile Survival
Survival of migrating Chinook Salmon and steelhead from release (i.e., rotary screw trap, beach seining, or hatchery release) to Lower Granite Dam is estimated using the Survival Under Proportional Hazards (SURPH) juvenile survival program (Smith et al. 1994; Skalski et al. 1998). The SURPH program estimates survival probabilities (\(\phi_j\)) and detection probabilities (\(\rho_j\)) using a Cormack-Jolly-Seber (CJS) model (Cormack 1964; Jolly 1965; Seber and others 1982). Survival estimates to LGD are reported for each natural- and hatchery-origin release group. Natural-origin release groups include all PIT tagged fish within a single life stage/trapping season and migration year, in order to pair with juvenile abundance estimates (4.1). Survival is estimated from subsequent tag detections at hydrosystems and other in-river facilities obtained from the online PTAGIS database. SURPH’s companion program PITPRO (Westhagen and Skalski 2007) is used to develop capture history files with the four lower Snake River and four lower Columbia River dams, and the estuary towed array as potential tag detection locations. The PITPRO capture history file output is then input into SURPH to generated CJS estimates with associated 95% profile likelihood confidence intervals (Smith et al. 1994; Skalski et al. 1998).
4.3 Smolts Equivalents
Estimated smolt equivalents (\(\hat{N}_{S}\)) represent the abundance of fish surviving to LGD from the total brood year or migratory year abundance estimate or hatchery release number.
\[\begin{equation} \hat{N}_{S} = \sum_{j=1}^{J}\hat{N}_j * \hat{\phi}_j \end{equation}\]
Smolt equivalents are derived by multiplying each life stage specific survival from release location to LGD (\(\hat{\phi}_j\); 4.2) by the life stage abundance estimate (\(\hat{N_j}\); 4.1). Life stage specific smolts reaching LGD is then summed across all life stages (\(J\)) to estimate a total brood year (spring/summer Chinook Salmon) or migration year abundance (steelhead) of smolts at LGD.
Using common variance properties (Casella and Berger 2002) precision for total smolts at LGD becomes
\[\begin{equation} Var(\hat{N}_{S}) = \sum_{j=1}^{J} ( \hat{N}_j^2 Var(\hat{\phi}_j) + \hat{\phi}_j^2 Var(\hat{N}_j) + Var(\hat{N}_j)Var(\hat{\phi}_j)). \end{equation}\]
4.4 Age-at-Emigration
The age of juveniles migrating past rotary screws is estimated differently for spring/summer Chinook Salmon and steelhead. Generally, emigration age and the proportion of spring/summer Chinook Salmon migrating past rotary screw traps is determined solely from the abundance of fish within each life stage or trapping season, and non-overlapping age and size classes. In contrast, steelhead migrants passing rotary screw traps are often comprised of different age groups with overlapping size classes.
Spring/summer Chinook Salmon age-at-emigration is reported as the proportion of the total brood year abundance migrating as age 0 YOY, age 1 parr, age 1 presmolt, and age 1 smolt. Each life stage and age proportion is calculated by dividing life stage abundance by the total brood year abundance. Additionally, age 2 juveniles determined from fork length and captured after the smolt season ends (i.e., June 30th) are included in brood year summaries when they are observed in the trap catch.
Steelhead age-at-emigration is reported as the proportion of age groups caught during each fall and spring trapping season. Seasonal age proportions are estimated from scales collected from a random sample of trapped fish. Scales are collected from between the posterior edge of the dorsal fin and anterior edge of the anal fin directly above the lateral line (Scarnecchia 1979; Knudsen and Davis 1985). Collected scales are then prepared and read following procedures outlined in Seelbach and Beyerle (1984) and Davis and Light (1985). Scales are read by multiple readers with a final age determination and error rate calculated using methods described by RJ Beamish and Fournier (1981) and R_J Beamish and McFarlane (1983).
4.5 Size-at-Emigration
The size at emigration for Chinook Salmon, and steelhead is described by reporting the distribution of fork lengths collected from a random sample of individuals. Distributions are presented using common summary statistics (e.g., means, medians, standard deviation) or with graphics (e.g., histograms, boxplots) to illustrate the full range and variability in fork lengths. Size-at-emigration is reported separately for each life stage or trapping season group, or split into finer temporal scales, such as, weeks or months.
Growth rates during a trapping season, or across the full migratory year are determined with a Von Bertalanffy (1938) growth model. Growth models are fit following the Beverton and Holt (1957) parameterization to better assess size-at-emigration, and growth of juvenile spring/summer Chinook Salmon and steelhead captured at rotary screw traps. Growth model coefficients are estimated with a non-linear least squares method and is represented as,
\[\begin{equation} l_t = L_{\infty} (1-e^{(-K(t-t_0))}). \tag{4.2}, \end{equation}\]
where \(l_t\) is the length of individuals at time \(t\), \(L_\infty\) is the asymptotic length or mean length at emigration, K is a growth coefficient and \(t_0\) is a coefficient for time when length equals zero. Time is typically represented as the number of months that occur between egg deposition and capture at the rotary screw trap.
4.6 Condition of Juveniles at Emigration
The mean condition factor (\(\bar{K}\)) is calculated for each life stage and trap season migrant group released at rotary screw traps, and hatchery release groups using Fulton’s condition factor (Anderson and Neumann 1996).
\[\begin{equation} K_i = (w_i / l_i^3)*100,000 \tag{4.3} \end{equation}\]
Where \(K_i\) is the condition factor for individual \(i\), \(w_i\) is weight, and \(l_i\) is length measured from a random sub-sample of juveniles collected from each release group. Weight is measured to the nearest 0.1 g and length is measured to the nearest 1.0 mm from the snout tip to the fork in the tail.
4.7 Emigration Timing
Juvenile emigration timing is described for spring/summer Chinook Salmon and steelhead using empirical cumulative distribution functions and rotary screw trap data. Emigration timing is summarized and then reported as the date of passage for 1, 10, 50, 90 and 100 percent of individuals within each life stage or trap season, or reported graphically. Daily emigration numbers and timing are calculated by expanding unmarked trap catch with the weekly trap efficiency estimated during juvenile abundance estimation (4.1).
4.8 Main Stem Arrival
Main stem arrival timing at LGD is calculated using PIT tag interrogation data queried from PTAGIS for individual fall and spring/summer Chinook Salmon and steelhead PIT-tagged during beach seining, rotary screw trapping, and hatchery marking activities. Arrival timing is described similarly to emigration timing with graphical displays of the empirical cumulative distribution functions, and the dates of 1, 10, 50, 90 and 100 percent of individuals arriving to LGD. Summaries are provided for each natural-origin life stage or trapping season, and hatchery-origin release groups.
Additionally, we report the proportion of emigrants passing LGD prior to the start of fish collections at juvenile by-pass facilities for transportation, and increased spring and summer spill operations. Collections for transportation (i.e., barging or trucking) at juvenile bypass facilities located at the lower Snake River dams typically begin during April. We assume fish arriving prior to the transportation period are not transported, and those fish arriving on the start date or later would be transported if observed in by-pass facilities at any of the transport dams (Lower Granite Dam, Little Goose Dam and Lower Monumental Dam), unless the observed PIT tag is designated for return-to-river.