Hatchery Updates

This page describes all the hatchery updates made to the DSMs as part of the Reorienting to Recovery process

Hatchery Dymics

Existing Approach

The existing SIT DSMs follow the following model logic for hatchery fish:

1. Each year, if the model is run deterministically, there are 307650 hatchery adults initiated into the model based on (CWT data?). Hatchery adults are split between tribs based on values from fallRunDSM::hatchery_allocation

2. Percent natural is output from the model in the get_spawning_adults() submodel

3. Both natural adults and hatchery adults spawn, all surviving juveniles outmigration and are carried over to the next year but the origin is not tracked.

Updated Approach

The updated Recovery life cycle model adds additional hatchery dynamic logic into the model.

1. In the first two years, 307650 hatchery adults are initiated into the model based on CWT data. In subsequent years, hatchery adults are carried over from prior generations and not initiated annually.

2. Percent natural is output from the model in the get_spawning_adults() submodel

3. PHOS initiated as 1 - percent_natural and adjusted to describe hatchery fish renaturing every 2 generations. Renaturing adjustment assumes that if PHOS is decreasing in two prior years then the population of hatchery for the current year should be adjusted by the percent difference of PHOS in the current year vs the current year minus two. PHOS is recalculated using the adjusted number of hatchery fish divided by the total population.

4. Both natural adults and hatchery adults spawn. Distinct fecundity is applied to natural vs hatchery fish based on fish size. The model does not track fish size, only age, so ages are mapped to size using logic from Roni and Quinn 1995.

5. Hatchery releases are added into the model logic in tributaries and when applicable as juveniles at Chipps island. Hatchery fish are added in within the large juvenile size class.

6. All surviving juveniles outmigrate and are carried over to the next year. Hatchery and natural adult returns are tracked separately and have different adult return ratios.

Detailed overviews of each update including the literature or data source are provided below.

Initiating Hatchery Returns

The first 3 years of the model hatchery returns are initiated using CWT data and the same logic in as in the CVPIA SIT DSMs. There are a total of 307,650 hatchery adults initiated each of the first 3 years.

Calculating Percent Natural

Percent natural is calculated by taking the total number of natural adults returning and dividing by total number of adults in the system (natural and hatchery). This is calculated each year for each tributary in the model.

Calculating PHOS

Each year PHOS is initiated as 1 - percent_natural. PHOS is updated to account for renaturing every 2 generations if PHOS decreases in the last two consecutive years.

Renaturing

Renaturing only occurs in the model after year 3 if PHOS decreases for two consecutive years on a watershed. Each watershed experiences renaturing separately so in one year renaturing could happen at one watershed and not another. The renaturing adjustment assumes that if PHOS is decreasing in two prior years then the population of the hatchery for the current year should be adjusted by the percent difference of PHOS in the current year vs the current year minus two.

We use this renaturing adjustment to recalculate PHOS as updated PHOS = 1 - ((hatchery spawners * percent renaturing) + wild spawners)/ total spawnwers)

TODO: Add additional logic for when hatchery releases are 0

Spawning & Fecundity

Both natural adults and hatchery adults spawn. Distinct fecundity is applied to natural vs hatchery fish based on fish size. The model does not track fish size, only age, so ages are mapped to size using logic from Roni and Quinn 1995.

We assume the current fecundity used in the DSMs (5222) is fecundity for age 3 wild fish. We scaled up or down using logic that a 1 - mm reduction in length results in 7.8 fewer eggs (95% CI = 6.6–8.9). Logic from Malick, M.J., Losee, J.P., Marston, G., Agha, M., Berejikian, B.A., Beckman, B.R. and Cooper, M., 2023. Fecundity trends of Chinook salmon in the Pacific Northwest. Fish and Fisheries.

Age in Model	Origin	Size (cm)	Fecundity
2	Wild	50.9	4102
2	Hatchery	48.9	3946
3	Wild	69.1	5522
3	Hatchery	67.1	5366
4	Wild	83.0	6606
4	Hatchery	81.0	6450
5	Wild	85.0	6762

Juvenile Hatchery Releases

Hatchery return data comes from the production targets defined in the California HSRG (Pages 65 - 95).

Baseline hatchery releases by Tributary are:

Hatchery	Tributary	Run	Number Released
Coleman National Fish Hatchery	Battle Creek	Fall	1.2e+07
Feather River Hatchery	Feather River	Fall	6.0e+06
Merced River Fish Facility	Merced River	Fall	1.0e+06
Nimbus Fish Hatchery	American River	Fall	4.0e+06
Mokelumne Hatchery	Mokelumne River	Fall	5.0e+06

Phased hatchery releases by Tributary are:

Hatchery	Tributary	Run	Number released Years 1-5	Numbers released Years 6-10	Numbers Released Years 11-20
Coleman National Fish Hatchery	Battle Creek	fall	6.0e+07	614,400	0
Feather River Hatchery	Feather River	fall	3.0e+07	307,200	0
Merced River Fish Facility	Merced River	fall	5.0e+06	51,200	0
Nimbus Fish Hatchery	American River	fall	2.0e+07	204,800	0
Mokelumne Hatchery	Mokelumne River	fall	2.5e+07	256,000	0

Adult return logic

Wild fish return based on the following distribution

• 22% year 2

• 47 % year 3

• 26% year 4

• 5% year 5

Return proportions are based on Roni and Quinn 1995 (Table 1).

Hatchery fish return based on the following distribution

• 30% year 2

• 60% year 3

• 10% year 4

Return proportions are based on expert opinion that hatchery fish return younger than wild fish.

Stray Rate Update

The previous CVPIA SIT model approach used a uniform straying distribution across all potential watersheds. Based on recent empirical work by Sturrock et al. (2019) examining eight decades of hatchery release data in California’s Central Valley, we have updated the straying allocations to reflect observed patterns specific to each source hatchery.

Sturrock et al. (2019) developed a beta regression model examining factors influencing straying rates including transport distance, return year flows, fish age, release timing, and environmental conditions (PDO). The model found transport distance was strongly associated with straying rates, with bay releases showing 9-26 times higher straying than on-site releases.

Please see the stray rate documentation for more information on how Sturrock et al. (2019) stray logic is incorporated into the R2R DSMs.