Essential Needs for the Recovery of Endangered Winter-Run Salmon

Summer spawning and egg incubation water temperatures in the upper Sacramento River below Shasta Dam is a well-known and documented key to the recovery of winter-run salmon Chinook salmon. In a December 2018 post, I noted the importance of early-fall flows to support the emigration of juvenile winter-run salmon from spawning and early rearing areas of the upper Sacramento River near Redding. In this post, I add another measure to the list of essential needs.

  • Late-fall flows – Flows to move winter-run juveniles from the upper and lower river into and through the Delta in the late fall.

What kind of late-fall flows are specifically needed? The type that occurred in December 2019 from a spate of storms (Figure 1). The 10,000+ cfs flow in the lower Sacramento River got wild winter-run salmon smolts through the lower Sacramento River, as seen from the Knights Landing screw-trap catches. The 20,000+ cfs early-December pulse of Delta outflow got wild winter-run salmon smolts moving through the Bay toward the ocean, as seen in the Chipps Island Suisun Bay trawl catches.

I have previously recommended extending Fall X2 Delta outflow protections1 and reducing Delta exports2 to help the winter-run smolts during their emigration to the ocean. As it was, 10,000+ cfs exports in the latter half of December 2019 took over half of the potential Delta outflow. Figure 1 clearly shows the importance of the late-fall flows to the emigration of winter-run.

Observed patterns of winter run emigration provide further evidence of the need for flows in the late fall. Figure 2 shows late fall 2017 conditions when there was no late fall flow pulse. The movement of winter-run smolts through the Bay was delayed, occurring in small spurts from late January through March. There is no doubt that one-to-three-month delays in smolt migrations from the river and Delta to the ocean are detrimental to the population and to recovery.

Figure 3 shows the latefall flow pattern over the past decade. Recovery of winter-run salmon depends on protecting the flow pulses. The tendency is to export as much of the first flows of the water supply season as possible and get it stored in south-of-Delta reservoirs. Most of the late-fall rainfall was already captured in upstream reservoirs, so these flow pulses are just a fraction of Central Valley’s natural flows.

A close look at Figure 3 shows minimal Delta outflow in the late fall of 2011 and 2017. Both years were just coming off wet water years. Shasta Reservoir had above-average storage for December in both years (>3 MAF, two-thirds full). Modest commitments of reservoir water could have greatly benefitted winter-run emigration. Inflows to Shasta reservoir in December of both of those two years were over 200 TAF. An added release of less than half that inflow (100 TAF) could have provided five days of 10,000 cfs pulse flow to the December release pattern in both years. Such a pulse flow, in combination with reduced Delta exports (Figure 4), would have provided five days of 20,000+ cfs Delta outflows in December 2011 and 2017 to support wild winter-run smolt emigration and winter-run recovery.

Figure 1. Catch patterns of juvenile wild winter-run salmon in the upper Sacramento River at Red Bluff, the lower Sacramento River at Knights Landing, and at Chipps Island in the upper Bay in fall 2019. Red circles denote catch peaks associated with fall pulsed flows.

Figure 2. Catch patterns of juvenile wild winter-run salmon in the upper Sacramento River at Red Bluff, the lower Sacramento River at Knights Landing, and at Chipps Island in the upper Bay in fall-winter 2017-18. Red circles show dispersed timing of emigration and lack of large catch peaks in the absence of fall pulsed flows.

Figure 3. Delta outflow in late fall 2010-2019. Note lowest flows were in 2011, 2013, 2015, and 2017.

Figure 4. December 2011, 2017, and 2019 south Delta exports at the state Banks (HRO) and federal Tracy (TRP) pumping plants. Capacities are 7500 and 4400 cfs, respectively. Note the extremely high and unusual December 2019 exports.

San Joaquin Fall-Run Salmon – Status Fall 2019

In previous updates in 2016 and 2017, I remarked on progress toward increasing San Joaquin watershed salmon runs during the recent 2013-2015 drought period.  I attributed the improvements to several positive factors:

  1. Increased production from Mokelumne and Merced state hatcheries. Both the Stanislaus and Tuolumne runs benefitted from strays from the two hatcheries, as well as from strays from the three Sacramento River hatcheries (Battle Creek, Feather, and American).
  2. River and Delta flow improvements in spring and fall mandated by federal biological opinions.
  3. Stronger runs in the Stanislaus River from improved flow and water temperature regimes in the Stanislaus and in the San Joaquin below the mouth of the Stanislaus. Fall temperatures below 60oF in the Stanislaus and below 65oF in the San Joaquin proved beneficial to the Stanislaus run, likely by attracting hatchery salmon as well as wild strays from the Merced and Tuolumne rivers.

I also attributed past very poor runs to drought conditions during rearing and during adult spawning runs, exactly the conditions of the 2015 run.  However, the 2015 run was much better than expected, likely because of hatchery strays.

In this post, I update the previous assessments with new information on the 2016-2018 runs.  The 2016 and 2017 runs were the product of rearing conditions in the 2014 and 2015 critical drought years, as well as wetter-year fall adult migration conditions.  The 2018 run was a product of normal-water-year rearing (2016) and adult migration (2018) conditions.

The 2016 and 2017 runs were strong in the Stanislaus and Merced rivers (Figure 1), with both rivers benefitting from hatchery production and strays.  One strong component of the strays was the unusually high proportion of strays from the Battle Creek hatchery, whose managers’ strategy during the 2014-2015 drought was to truck smolts to the Bay.

The 2018 San Joaquin run was lower, but still a strong improvement over the drought-influenced runs in 2007-2011 (Figure 1).  Spring rearing conditions in 2016 and fall adult migration conditions in 2018 were generally better than they were during the critical drought years.  Also, most of the Mokelumne and Merced hatchery smolts were released to the Bay and west Delta, respectively, in 2016.  However, adult salmon migrating upstream in the fall of 2018 were subject to stressful conditions in the lower San Joaquin River, where high water temperatures were similar to the high temperatures there in 2015 and 2016 (Figures 2-4).

An updated spawner-recruitment relationship is shown in Figure 5.  Runs from 2016-2018 are shown as red-bordered blue dots labelled from 2014-2016, representing rearing-year conditions for these runs.  The recruitment trend remains strong given the generally dry water conditions during the 2012-2016 drought period.  The recent recruitment per spawner remains up to 10-fold higher than in the prior two droughts, 1987-1992 and 2007-2009.  The best explanations for this improvement remains strong hatchery contributions and better hydrology-related habitat and migration conditions prescribed in the 2008-09 federal biological opinions.

The State Water Resources Control Board is in the process of developing biological goals for fall-run Chinook salmon in the lower San Joaquin River and its three salmon-bearing tributaries.  The Board should include goals for escapement as well as the factors that control escapement.

Figure 1. Chinook salmon runs (escapement) in the San Joaquin River from 1975-2018 is made up of components of escapement/run estimates from its three spawning tributaries. Data source: CDFW.

Figures 2 and 3. Fall water temperatures in the lower San Joaquin River downstream of salmon spawning tributaries in 2015, a critical drought year, and 2016, a normal water year.

Figure 4. Fall water temperatures in the lower San Joaquin River downstream of salmon spawning tributaries in 2018, a normal water year. Compare with temperatures in red circles in Figures 2 and 3.

Figure 5. Recruits-per-spawners relationship ((log10X)-2) for San Joaquin River fall-run Chinook salmon 1976-2018. The year shown is the year that the salmon were rearing as juveniles in the rivers in their first year of life. (For example: year 13 represents the progeny of the fall 2012 spawn; these juveniles in 2013 would have predominantly spawned as 3-year-old adults in 2015). Red years are critical and dry water years. Blue years are wet water years. Green years are normal water years. Red circles represent years when fall conditions during spawning would have reduced recruitment (for example: year 13 red circle indicates poor fall conditions during the fall of 2015). Blue circles represent years when fall conditions were good when recruits returned. (For example: year 81 has a blue circle because fall conditions in wet year 1983 were good).

Fall X2 should extend through December

In a recent post, I described the Fall X2 provision in the 2008 Delta Smelt Biological Opinion that protects smelt by requiring a modest Delta outflow from mid-August through October in Above Normal and Wet years.  In the same post, I described how the 2019 Biological Opinion for smelt would move the compliance point for Fall X2 upstream into the Delta, reducing low salinity zone habitat.   In this post, I suggest that the Fall X2 requirement should not only be retained with the old compliance point, but also that the applicable time period should extend through December.

First, if the X2 provision is not extended into December, this is what happens:  (1) Delta outflow drops to zero or even negative, as occurred this past November 2019 (Figure 1); and (2) the low salinity zone moves up into the Delta via the San Joaquin River channel toward the export pumps (Figure 2).

Second, winter-run salmon smolts that pour into the Delta from the Sacramento River in November and December of all but the driest years (Figure 3) will have difficulty surviving and exiting the Delta for the Bay and ocean.

Third, what few Delta smelt that may be surviving will be put at risk of being drawn into the central and south Delta (Figures 4 and 5).

Fourth, longfin smelt will be at risk to being drawn into the Delta (Figures 6 and 7).

Fifth, the primary food of young Delta native fishes, calanoid copepod adults, which concentrate in the low salinity zone, would be drawn into Delta (Figure 8).  Bay-Delta pelagic plankton productivity would suffer.

In conclusion, there are presently few constraints on water project operations in the Delta in November-December.  Freshwater outflow to the Bay can be zero or even negative, as occurred this past month, November 2019.  The updates to the Bay-Delta Water Quality Control Plan and to state permits that regulate Delta exports should extend Fall X2 through December in order to protect Delta native fishes.  Compliance would entail Delta outflows in the 8000-10,000 cfs range and/or Jersey Point salinity of about 500 EC.

Figure 1. Tidally filtered flow in the Sacramento River channel at Rio Vista and Jersey Point in the San Joaquin channel in November 2019.

Figure 2. Salinity (EC) at Jersey Point in the San Joaquin channel of the west Delta in November 2019.

Figure 3. Cumulative catch of winter-run Chinook salmon at Knights Landing rotary screw traps in fall-winter of water year 2017. Source: DOSS 2017.

Figure 4. Trawl catch distribution of Delta smelt fall 2011, the last time Delta smelt were relatively common.

Figure 5. Salvage of Delta smelt pre-spawn adults in fall-winter of water year 2003.

Figure 6. Longfin smelt trawl catch distribution in November 2011.

Figure 7. Longfin smelt trawl catch distribution in December 2011.

Figure 8. Adult calanoid copepod catch distribution in November 2011 zooplankton survey.

Failure to Protect Winter-Run Salmon in Fall 2019

In a December 2018 post, I focused on the importance of fall pulse flows in moving winter-run salmon juveniles downstream in the Sacramento River to the Delta and Bay. Without pulse flows, the juvenile winter-run are less likely to make or survive the downstream move from spawning and early rearing areas in the upper river. They are thus less likely to reach the ocean and contribute to subsequent recruitment into the adult population.

A gradual recovery of adult spawners, egg production, and wild fry production (Figure 1) has been helping recovery of the winter-run population since the population crash during the 2012-2015 drought. Wild fry production as estimated from Red Bluff screw trap collections is up sharply in 2019 (Figure 2 top chart).

However, in fall 2019, winter-run fry have received even less support in terms of river flow in their important journey to and through the Bay-Delta than in previous years. Numbers caught in lower river screw traps are very low (Figure 2 bottom chart), reflecting low movement rates from the upper river and poor survival. Both factors are a consequence of poor river flows. There are simply no reasons for Reclamation to be so stingy with Shasta Reservoir releases this fall, after a very wet year in 2019 and with Shasta Reservoir at or near a record-high level for this time of year.

A close-up of the Figure 2 data in Figure 3 shows some effort on the part of Reclamation to provide flow pulses,1 but the effort was not enough. Furthermore, Reclamation subsequently offset its meager augmentation by cutting reservoir releases in November (Figure 4). The November reduction further compromised the emigration and survival of juvenile winter-run salmon. Reclamation’s tendency to cut releases in fall and winter, the period when winter-run most depend on river flows, is pronounced in all but the wettest years (2011 and 2017) over the past decade (Figure 5). Such tendency probably has been deemed acceptable because downstream tributary flows (Battle Creek, Cow Creek, etc.) provide fall flow pulses in some years (e.g., fall 2016 Figure 6, fall 2011 Figure 7). But inflow pulses from those tributaries are downstream of the Redding spawning reach; in the spawning reach, flows come almost exclusively from Shasta Reservoir releases.

What is needed are modest flow pulses from Shasta Reservoir in fall, especially when pulses in downstream tributaries occur. Releases for several days in the 10,000-15,000 cfs range, or of slightly less magnitude when coincident with tributary flow pulses, would help emigration (and survival) of winter-run juvenile salmon from the upper river. Such pulses should not be followed by offsetting flow decreases, as have occurred this fall (Figures 3 and 4). Low flows following fall pulses cause redd dewatering or fry stranding of spring-run, fall-run, and late-fall-run salmon, which spawn in the upper river later in the season than winter-run.

The late November 2019, Thanksgiving week storm should provide ample Shasta storage and tributary flows to allow modest flow pulses from Shasta Reservoir. Such flow pulses would benefit all four salmon runs in the Sacramento River.

Figure 1. Emigration timing of juvenile winter-run salmon from the upper Sacramento River and the estimated number of juvenile salmon (millions) passing Red Bluff during water years 2004-2018. Note the poor production in critically dry 2014 and 2015 from loss of cold-water pool and associated catastrophic egg mortality.

Figure 1. Emigration timing of juvenile winter-run salmon from the upper Sacramento River and the estimated number of juvenile salmon (millions) passing Red Bluff during water years 2004-2018. Note the poor production in critically dry 2014 and 2015 from loss of cold-water pool and associated catastrophic egg mortality.

Figure 2. Daily estimated juvenile winter-run salmon catch per trap day passing Red Bluff in the upper river and Tisdale Weir in the lower river in summer-fall 2019. The numbers passing Red Bluff are strong, especially when one considers that they were the offspring of poor brood-year 2016.  

Figure 2. Daily estimated juvenile winter-run salmon catch per trap day passing Red Bluff in the upper river and Tisdale Weir in the lower river in summer-fall 2019. The numbers passing Red Bluff are strong, especially when one considers that they were the offspring of poor brood-year 2016.

Figure 3. Daily estimated juvenile winter-run salmon passage per trap day at Red Bluff in the upper river, also showing river flow in summer and fall 2019.

Figure 3. Daily estimated juvenile winter-run salmon passage per trap day at Red Bluff in the upper river, also showing river flow in summer and fall 2019.

Figure 4. Shasta/Keswick Dam releases in fall 2019, along with daily median flow average for 55 years.

Figure 5. Shasta/Keswick Dam daily average releases from 2009-2019, along with daily median flow average for 55 years.

Figure 6. Daily estimated juvenile winter-run salmon passage per trap day passing Red Bluff in the upper river and Tisdale Weir in the lower river in fall 2016. Note that the flow pulses (and associated higher catches) in early and late November were from tributary storm-related inflows. Such events had not occurred as yet in 2019 (Figure2).

Figure 7. Daily estimated juvenile winter-run salmon passage per trap day passing Red Bluff in the upper river and Tisdale Weir in the lower river in fall 2010. Note the flow pulses (and associated higher catches) in late October and in December.

 

 

 

  1. Reclamation’s sporadic 2000 cfs flow pulses in late October observable in Figures 2 and 3 were likely part of Reclamation’s contribution to maintaining Delta inflow and outflow for the Fall X2 requirement.

Sustaining wild Salmon and Steelhead above Central Valley dams

The Case for Two-Way Trap and Haul

Why should we expand spawning populations of listed salmon and steelhead to areas above dams and impassible falls in the Central Valley? The answer is: because the genetic makeup and wild traits of populations upstream of existing barriers can be controlled, restored, and preserved.

At present, the genetic makeup of salmon and steelhead populations below dams is continually being compromised by hatchery fish and strays to and from other watersheds. The one population of winter-run Chinook is confined to the spawning reach immediately below Keswick Dam and thus is subject to the potentially drastic whims of nature and man. That population is further being compromised by the increasing threat of hatchery degradation of the gene pool as winter-run hatchery fish further dominate the adult spawning population. Small, self-sustaining populations of spring-run Chinook and steelhead remain in only a few watersheds. They too are continually being threatened by strays and hatchery fish.1

One solution to maintaining genetic integrity by limiting genetic influence from hatchery-produced fish and interbreeding of genetically or behaviorally distinct runs is to implement trap-and-haul programs in isolated reaches above dams.

The National Marine Fisheries Service included requirements to establish winter-run Chinook trap-and-haul populations above Shasta Reservoir in 2009, 2010, and 2014 biological opinions on Central Valley Project (CVP) and State Water Project (SWP) operations. CALFED proposed introducing spring-run Chinook above Yuba River dams. Extensive studies have been conducted on reintroducing salmon in these areas. The requirement to establish populations upstream of Shasta has been dropped in the Trump administration’s October 2019 biological opinion for the CVP and SWP. For the moment at least, the requirement remains in state of California plans.2

The California Department of Fish and Wildlife’s California Endangered Species Act Take Permits for CVP and SWP operations should require reintroduction of salmon and steelhead upstream of an array of dams in the Sacramento River watershed. All of the sites I recommend are affected by the CVP and SWP. The state should also consider locations in the San Joaquin and Klamath River watersheds. The Klamath watershed is also affected by Reclamation’s Klamath Project, and is the present subject of the country’s largest dam removal project.

In considering potential sites I focused on the ability to maintain experimental controlled conditions as well as optimum habitat quality sites. In most cases, that meant minimal flow variation and high quality, cold reaches dominated by spring water. The sites need not be in the historical range, but should be in historically occupied watersheds (e.g., they could be upstream of impassible falls in watersheds that historically held salmon and steelhead.).

I suggest five sites in the Sacramento River watershed (Figure 1).

  1. Upper Sacramento River (above Lake Shasta) – below Lake Siskiyou dam upstream of Dunsmuir in the Box Canyon/Shasta Springs reach.
  2. Upper McCloud River (above Upper McCloud Falls) – spring-fed reach above Larkin Dam on south flank of Mt Shasta.
  3. Upper Battle Creek – Ripley Creek, tributary of South Fork, spring-fed, although presently its flow is diverted by PG&E to South Fork Powerhouse.
  4. Upper North Fork of Feather River – above or below Lake Almanor.
  5. Upper North Yuba River – above Bullards Bar Reservoir.

I have studied all of these sites and consider them feasible for reintroduction. Most have been considered for reintroduction by state and federal resource agencies. Reintroduction strategies may include releases of native-strain adult spawners, planting of eyed eggs, fry, or fingerlings, then capture and trucking to locations downstream of dams.

For more on reintroducing salmon above dams see:
https://podcast.barbless.co/reintroduction-of-winter-run-chinook-into-the-mccloud-river-jon-ambrose-noaa-nmfs/

https://www.webpages.uidaho.edu/UIFERL/pdf%20reports/Keefer%20et%20al.%20%202010%20WIL%20Chinook%20prespawn%20mortality%201.pdf

https://fishwithjd.com/2015/05/07/new-plan-developing-to-get-spring-chinook-into-north-yuba-upstream-of-bullards-bar-reservoir/

Figure 1. Historical range and present range of salmon in Sacramento-San Joaquin watershed, with suggested five locations for reintroduction via two-way-trap-and-haul shown as red dots.