Issues in Delta Smelt Science

Posted on November 28, 2016 by Tom Cannon

Header Image

Recently the San Francisco Estuary & Watershed Science journal published a special issue that included a paper on Delta Smelt. The paper is an excellent summary of the present state of Delta smelt. The discussion section on p. 18 of the paper provides an excellent description of the whole of Delta Smelt science. The paper is comprehensive with a litany of scientific references on Delta science and the decline of the Bay-Delta estuary and the Delta Smelt. It also provides an outlook and prescription for the future, though bleak.

In this post I offer slightly more optimism with focused solutions, based on a somewhat different take on several key issues. Bear with me as I go through each issue and my usual array of diagnoses and recommended treatments that could bring smelt back from the brink.

Population Dynamics – What drives smelt numbers

“The role of density dependence has been an area of emphasis for these population dynamic studies. An apparent decrease in the carrying capacity of the estuary in the 1980s resulted in an increase in density-dependent mortality from the juvenile stage in late summer through the adult stage in fall (i.e., between the TNS and FMWT indices) (Sweetnam and Stevens 1993; Bennett 2005; Maunder and Deriso 2011).” p. 9.

Density-dependent mortality means that fish get so abundant that it reduces their growth and survival through competition. There is no evidence of this dynamic in the Delta smelt population indices (Figure 1). The number of smelt in fall is simply related to summer flow and exports, and most importantly to the starting number of summer juveniles, which is most certainly related to flow and exports (Figure 2). The latter is density independent – survival was not lower at higher initial numbers or vice versa. Density dependence might be expected in conditions of very, very high abundance of juveniles that competed amongst each other for food, possibly even driving down their food supply. That is simply not possible at the extremely low population levels that exist at present and that have existed in the recent past.

“In general, the relationship between sub-adult (spawner) abundance indices and juvenile abundance in the following year is poor (i.e., between the FMWT and TNS indices).” p. 9.

No, the relationship between spawner abundance and juvenile abundance is strong (Figure 2). The number of young smelt produced is much reduced in most dry years because of their lower flows and higher exports. Low exports in the wetter years 2010 and 2011 from restrictions in the 2008 Smelt Biological Opinion really helped. Those restrictions were not in place in red-circled 2005.

“This suggests that inter-generational abundance is driven more by environmental conditions, rather than by density-dependent factors.” (Id.)

What they are saying is that the number of young produced each summer is determined by environmental factors, which is true. But the most important factor remains the number of adult spawners at the beginning of the year, as represented by the fall index. It is a vicious circle, with environmental factors (flow and exports) driving the population down to the point that production even in good environmental conditions (2010, 2011) is low because of the depressed population (low number of eggs produced). This defining problem is important, because under the paper’s conclusion, the population can come back under good conditions. That is not possible with the stock size low. The authors seem to recognize this:

“Given the annual life cycle, any episodic salvage event may undermine population resilience by keeping numbers low, even when environmental conditions are good.” (p. 14).

Figure 1. Strong Stock Recruitment relationship. Summer to fall abundance plotted using log of numbers. Summer Townet (STN) Index in June. Fall Midwater Trawl (FMWT) Index in Sep-Dec. Lower survival in high export/low outflow dry summers like 77, 78, 81, 94, 02, and 04. Lower survival in flood years 82, 83, 86, 96. Higher survival in low export summers 90-92 when they ran out of water for export, and 72 when Andrus Island levee broke. (Source: http://calsport.org/fisheriesblog/?p=810 )

Figure 2. Strong Stock Recruitment relationship. Fall to next summer abundance plotted using log of numbers. Summer Townet (STN) Index in June. Fall Midwater Trawl (FMWT) Index in Sep-Dec. Blue years – wet; red years – dry. Year 05 suffered from low number of spawners and high exports. Years 2010 and 2011 (green circles) were wetter with low exports under OMR restrictions – gave us some hope, but operations in following 4 years of drought ended that, as the past several years points have been at bottom left. (Source: http://calsport.org/fisheriesblog/?p=810 )

Predation

“Presently, Mississippi silverside is probably the most important predator of Delta Smelt larvae because of their ability to prey on eggs and larvae and their high abundance in shallow areas where Delta Smelt spawn (Bennett and Moyle 1996; Bennett 2005; Baerwald et al. 2012).” p. 14.

“Although Delta Smelt remained abundant in the Delta and Suisun Bay through the 1970s, long after most introduced fish predators and competitors had established populations (Grossman, this volume), smelt began rapidly declining after the invasions of overbite clam and Mississippi Silverside in the 1980s.” p.7.

Silversides were also abundant in the late 70’s (no 80’s invasion), but were in my opinion suppressed by very abundant juvenile striped bass. This effect persisted until the striped bass population crashed at the beginning of the new century. Note the relative abundance of silversides (smelt predator and competitor), threadfin shad (competitor), and yellowfin goby (predator and competitor) in the western Delta in 2006 (Figure 3), and the low density of striped bass. Next, note the depressed relative abundance of these same species in the western Delta in late summer and fall of 1978 (Figures 4-7). Note also the much greater abundance of juvenile striped bass and even some Delta smelt in 1978. While such evidence may be circumstantial, it clearly shows the role that striped bass once played as the dominant predator in the Delta. Now I have to ask the questions: (1) what caused the “invasion” of silversides in the 80s; and, (2) what was the role of striped bass?

“Largemouth Bass will consume Delta Smelt in mesocosms (Ferrari et al. 2014), but are unlikely to be a major predator in the wild because of limited habitat overlap between the two species.” p. 15.

There really is no proof of this. Largemouth are found pretty much throughout the Delta, including areas where juvenile smelt are found in summer and fall. With largemouth far more abundant than smelt, it is likely that smelt come across largemouth frequently. I have caught both species in the same seine in the fall, on several occasions.

Habitat Change – water temperature and salinity

“While Delta Smelt have a fairly broad salinity tolerance; they were historically most abundant in the low salinity zone (LSZ) of the estuary, the position of which is determined by outflow (Moyle et al. 1992; Kimmerer et al. 2013; Sommer and Mejia 2013). Moderate hydrological conditions in late winter and spring place the LSZ in the Grizzly Bay region of Suisun Bay (Jassby et al. 1995). These conditions were beneficial to the Delta Smelt population at least partly because of high food abundance. At present, there is little evidence of the benefit of summer and fall occupancy in the LSZ (IEP MAST 2015).” p. 16.

Occupancy of the low salinity zone is a benefit to smelt in those wetter years when the water temperatures in the LSZ are not lethal to smelt. The LSZ (roughly 0.5 to 6 ppt salt) and its internal focal point X2 (2 ppt) are indeed critical to the summer-fall success of smelt in the Delta. Food is generally more abundant and water temperature is generally lower in the LSZ than in fresher upstream waters. Lower temperature and slightly brackish water allow lower energy consumption (smelt blood serum is approximately 2 ppt). Smelt are adapted to X2, prefer X2, and are found primarily at X2. X2 and the LSZ are a major ecological theme for many estuarine species around the world. In 2014 and 2015, X2 was located too far upstream in the Delta because the State Water Board weakened Delta water quality requirements, resulting in lethal water temperatures for smelt in summer and fall. The smelt had no place to go. X2 belongs further west under the protection of the “Delta Breeze” and away from the influence of the South Delta export pumps. It really is that simple. [Note: on July 15, 2016, the State and Federal water projects bumped up Delta outflow from 7,000 cfs to 14,000 cfs by reducing exports while holding inflow (reservoir releases) constant in an effort to meet Delta salinity standards, pushing the few remaining smelt and X2 west of the Delta. Was this action beneficial to Delta smelt? We will have to see.]

“Warm water temperatures during summer exacerbate stress from low food availability and may explain reduced survival from summer to fall in some years (Bennett 2005; Bennett et al. 2008).” p. 14.

“In addition to moving the LSZ to a favorable location for smelt, increased outflow influences habitat quality through its effect on food supply, dilution of contaminants, and turbidity.” p. 16.

Water temperatures are a problem in most summers, especially when the LSZ is located in the Delta.

Hatchery

“Reintroductions will have to be done within a few years of loss of wild fish, into an environment with better capacity to sustain them.” p. 19.

The X2 location in fall offers good habitat in most years. The problem is that juveniles are unable to survive to get to LSZ. The carrying capacity of the LSZ in the fall is likely underutilized in most years. Adding hatchery fish now to surviving wild fish will be better than waiting for wild fish to be gone. It might be essential at this point.

Refuges

“If increasingly unfavorable temperatures for smelt occur, predicted as a result of climate change, then special refuges may have to be created that can take advantage of cooler water in the Sacramento River or from water that is piped in from some other source.” p. 20.

Existing refuges already need protection. The Ship Channel and lower Cache Slough refuges can take advantage of the cooler water exiting the American River near the Port of Sacramento River. Flow from the Fremont or Sacramento weirs could enhance Yolo Bypass flows and water temperatures. The LSZ-X2 refuge can be cooler if located west of the Delta and subject to the Bay breeze; maintaining x2 west of the Delta would require higher spring, summer and fall Delta outflow. Protection and enhancement of the Montezuma Slough refuge portion of the LSZ is possible through better Delta outflows and controls offered by the Montezuma Slough Weir.

Summary of Potential Recovery Actions

The following actions, if implemented soon, could lead to marked improvement in the population of Delta smelt.

Limit entrainment. Minimize exports in spring, especially when pelagic larvae are found in the central Delta.

Keep the LSZ out of the Delta, especially in warm, dry years. Use the Montezuma Salinity Control Weir to maintain Montezuma Slough in Suisun Marsh as part of the LSZ.

Enhance the spring Delta inflows from the Yolo Bypass and San Joaquin River. Opening the Delta Cross Channel and installing the Head-of-Old-River and False River barriers will help. Closing the Knights Landing Outfall Gates in dry years will force more flow down the Yolo Bypass. Constructing the “Notch” in the Fremont Weir will increase flow through the Yolo Bypass. “Fixing” the entrance gate at the Port of Sacramento to the Ship Channel would allow additional Sacramento River flow input to the Ship Channel and lower Bypass.

Stock hatchery-reared Delta smelt adults in the fall at X2 location, conditions permitting.

Figure 3. Numbers of fish collected in the fall of 2006 along San Joaquin River shoreline near Antioch by species and size. Note the extreme abundance of silversides, followed by threadfin shad and yellowfin goby. Note the very low numbers of striped bass.

Figure 4. Numbers of fish collected in late summer 1978 along San Joaquin River shoreline near Antioch by species and size. Note the relatively high numbers of striped bass relative to silversides and threadfin shad.

Figure 5. Numbers of fish collected in late summer 1978 in Sacramento River shoreline near Montezuma Island (adjacent to Collinsville) by species and size. Note the high relative abundance of striped bass young. Delta smelt were actually more abundant than silversides, yellowfin goby, or threadfin shad.

Figure 6. Numbers of fish collected in fall 1978 in San Joaquin River shoreline near Antioch by species and size. Note silversides were relatively abundant along with striped bass and threadfin shad.

Figure 7. Numbers of fish collected in fall 1978 in Sacramento River shoreline near Montezuma by species and size.

Stocked Trout Fisheries

Posted on November 25, 2016 by Tom Cannon

A recent paper in the North American Journal of Fisheries Management¹ describes a nationwide 20% decline over the past two decades in sport fishing for trout. The trout angling population is getting older, while the number of young recruits is declining. With shrinking revenues from license sales, trout anglers and agencies are concerned about potential reductions in the numbers of trout stocked and the quality of popular trout fisheries supported by stocking.

Though the researchers found little relationship between the numbers of trout stocked and catch rate (within normal levels of stocking and catch rates), they did find a positive relationship between angler satisfaction and catch rate in the Virginia lakes and rivers they studied. They also found greater satisfaction with greater size of trout caught. Average catch rates of stocked trout were approximately one per hour. Total angling hours were much higher for lakes than streams primarily because of ease of access and less need for equipment.

Because numbers of trout stocked were not significantly related to catch or angler satisfaction, the researchers suggested reductions in stocking rates to counteract longterm losses in license revenues. Because this would seem counter-productive to sustaining angler interest and effort, they suggested:

“Beyond catch metrics, we believe that demystifying the programs by increasing transparency and increasing the communication of stocking densities and the persistence of catch rates—two common concerns—may increase retention and possibly lead to the recruitment of new anglers. We encourage information exchange between agencies and their anglers.”

This past June, I wrote a post on Lake Davis, a popular Sierra trout fishery. In that post, I spoke of perceived angler dissatisfaction with low catch rates. Rates had fallen well below one-per-hour in recent years, into the range where angler satisfaction was very low in the above-described research paper. I suggested that the poor catch rate was possibly due to low stocking rates, although other factors were possible (e.g., drought-related low water levels). This fall, I have noticed a sharp uptick in the catch rate at Lake Davis and a corresponding increase in angler satisfaction and effort. Catch rates from both shore and boat, for both bait and fly anglers, increased into the 0.5 to 1 fish per hour range, with the large size of the fish pushing angler satisfaction (and seasonal effort) to near maximum. Given the large average size of the trout, I assume the catch rate increase is not due to higher stocking rates, because fish are usually relatively small when they are stocked. A reasonable hypothesis is that warm water in Lake Davis, due to low water levels and lack of snowmelt this past spring and summer, probably put fish off the bite.

A final thought on declining numbers of licensed anglers. While CDFW does sponsor activities that promote angling for young people (see photo below), there is no general program promoting efforts, projects, and specific fisheries around the state. A more comprehensive program designed to increase angling participation, such as the programs offered by other states (e.g. Michigan, Oregon, Texas, etc.), would help to increase angler recruitment. I like Michigan’s weekly fishing report and its fishing tips reports. CDFW needs to generate more excitement with up-to-date news, tips, maps, and techniques. Biologists, wardens, and creel census clerks collect a lot of information that they could share through volunteers or the CDFW blog. CDFW could also post, link or summarize reports from guides and shops (while giving appropriate credits). CDFW could also report on how it is managing fisheries around the state how these fisheries are faring. One topic of interest to many anglers I know is a new fishery that has appeared in the American River, apparently as a consequence of stocking trout (steelhead smolts) from the Battle Creek hatchery (Redding) in the American River (see photos below).

Fishing in the City event sponsored by CDFW. Young people love these events, but the program needs to follow up with other elements to sustain their interest.

Are these American River fish steelhead from the Battle Creek Hatchery?

Abstract available at http://www.tandfonline.com/doi/full/10.1080/02755947.2016.1221003 . ↩

Fundamental Needs of Central Valley Fishes – Part 1a: River Flows – First Pulse of Fall Rains

Posted on November 16, 2016 by Tom Cannon

In the coming months and years, regulatory processes involving water rights, water quality, and endangered species will determine the future of Central Valley fishes.

To protect and enhance these fish populations, these processes will need to address four fundamental needs:

River Flows
River Water Temperatures
Delta Outflow, Salinity, and Water Temperature
Valley Flood Bypasses

In this post, I summarize a portion of the issues relating to River Flows: Fall Rains. Part 1b will cover winter river flows.

River Flows – Fall Rains

In most years, the first substantial fall rainfall stimulates many important ecological processes such as salmon and smelt spawning runs and salmon and steelhead smolt migrations to the ocean. Figure 1 below shows the effects of 2016’s late October rains, and Figure 2 below shows the effects of 2015’s December rains, in the lower Sacramento River flows at Wilkins Slough near Yuba City below Colusa. Most of these flow pulses came from storm runoff from un-dammed upper Sacramento Valley tributaries such as Cow, Cottonwood, and Battle Creeks. Such flow pulses stimulate the migrations of young salmon toward the ocean. Figure 3 below documents these migrations in the form of rotary screw trap collections at Knights Landing in the lower Sacramento River.

Under current operations, flows from the major reservoirs are generally held to the minimum requirement in the fall season in order to increase reservoir storage (Figure 4).¹ What is needed are flow pulses (spills) from the major Valley reservoirs to the major rivers below dams, to stimulate the migration of the juvenile salmon spawned immediately downstream of these dams. Just downstream of Whiskeytown Reservoir on Clear Creek, Shasta and Keswick reservoirs on the upper Sacramento River, Oroville Reservoir on the Feather River, and Folsom and Nimbus reservoirs on the American River are vitally important salmon-producing reaches whose flow is completely controlled by the operation of the dams. Water releases timed to the natural flow pulses would stimulate migration from these important salmon-producing reaches, providing even more flow and stimulus for young salmon from all the Valley rivers to pass successfully through the Delta and Bay to the ocean.

Meanwhile, downstream in the Delta, the CVP and SWP export facilities generally ramp up exports during the initial storm pulse (Figure 5 below shows an example from 2016). Because of the importance of the initial storm pulse, the CVP and SWP should limit exports during the initial pulse, not only to help salmon get through the Delta and Bay, but also to minimize the diversion of young salmon to the south Delta.

Figure 1. Lower Sacramento River flow at Wilkins Slough in fall 2016.

Figure 2. Lower Sacramento River flow at Wilkins Slough in late fall 2015.

Figure 3. Catch of juvenile salmon in Knights Landing rotary screw traps 2001-2004 vs. flow in lower Sacramento River at Wilkins Slough.

Figure 4. Release of water from Shasta/Keswick to upper Sacramento River near Redding, fall 2016.

Figure 5. Export of water from south Delta by State Water Project, fall 2016.

For additional discussion of the negative effects of this practice, see previous post. ↩

Does the Central Valley Need a Predator Removal Program?

Posted on November 11, 2016 by Tom Cannon

The Columbia River Pikeminnow Sport Reward Program just finished another year. A near-record 200,000 adult pikeminnow were harvested and $1.7 million rewards paid. The goal of the program is to harvest 10-20% of the population each year to reduce the population about 50%. Harvest rates in recent years reached as high as 17% as compared to this year’s 11%.

So why after 25 years is the program still harvesting near-record numbers of pikeminnow? The likely reason is that the program is not based on sound science. The Columbia pikeminnow population dynamics likely are best described with a standard Ricker-Type Population Model (see chart below), with reduced recruitment of young at high adult population levels because of competition and cannibalism. The introduction of a light harvest can actual increase recruitment of young by reducing competition and cannibalism, with the increased recruitment replacing the harvest, even providing a constant harvest or yield. This is how fishery quotas may be sustained year after year, such as in Alaska’s Bristol Bay Sockeye salmon fishery.

This same dynamic likely applies to pikeminnow and some other predators in the Central Valley. Pikeminnow are likely near their saturation level in the Central Valley. Any light harvest such as from a sport-reward or removal program would likely have little or no effect. Increasing harvest on non-native predators like largemouth bass by reducing sport-fishing harvest regulations would likely also have a limited benefit. However, striped bass, the most popular sport fish in the Delta, has a population that is already seriously depressed by long-term loss of juveniles to water diversions. Striped bass may respond more directly to increased harvest, further reducing recruitment and further depressing the population.

Thus the species composition of fish that eat other fish could change, satisfying those who vilify stripers and infuriating those who fish for them. But the potential for reduced overall loss of juvenile salmon or other native species that might follow from “predator removal” is far more complex and questionable than its proponents maintain.

Ricker-type stock-recruitment population dynamics model that likely applies to Columbia River and Sacramento River pikeminnow populations

Bringing Back the Klamath Salmon

Posted on November 7, 2016 by Tom Cannon

Restored tributary spring creek of Scott River, Klamath River tributary, with abundant juvenile Coho salmon. (See YouTube video for underwater view of countless juvenile Coho salmon rearing in this creek.)

A recent post on the KCET website by Alastair Bland spoke of efforts to save salmon on the Klamath River. I add my perspective in this post.

I have been involved in the Klamath salmon restoration on and off for nearly 30 years. In my experience, the runs of salmon and steelhead keep declining because not enough gets done and because there is lack of progressive management. The Klamath is a big watershed (Figure 1). I tried to sit in the middle of one element of the process a few years ago on the Scott and Shasta Rivers, the Klamath’s two main upstream salmon tributaries below Iron Gate Dam. I found there were not just two sides involved in conflict, but really five: tribes, government agencies, ranchers-landowners, a power company, and environmentalists. There were even sides within sides. The four tribes often did not agree or work together. The four fish agencies often could not agree. The two states did not always agree, and individual state agencies disagree, resulting in conflicting water rights, water use, and water quality regulations. Counties and cities disagree. Neighboring Resource Conservation Districts differ in approaches. Many citizens want a new state carved from the two states. Some landowners love salmon and beavers, and others do not. Then there are the big watershed owners: private timber companies, US Forest Service, and Bureau of Land Management that manage forest watersheds differently under a wide variety of regulations and approaches that often do not protect salmon. I watched county sheriffs try to lead landowners in policy and enforcement, with a willingness to enforce vague trespassing rules on rivers and creeks. I watched as State Water Resources Control Board members toured watersheds and met with tribes and local leaders in an effort to resolve conflicts in over-appropriated watersheds. I watched as CDFW staff tried to enforce stream channel degradation and water diversion regulations on private and public lands.

While some progress gets made, it is too slow to save the native fish. Coho and spring-run Chinook are hanging on but slowly going extinct. Fall-run Chinook are supported by hatcheries but still declining. The iconic Klamath and Trinity Steelhead are silently and slowly fading away.

For decades, the various sides have waged war over water, dams, and property rights. The watersheds and fish have suffered as “Rome” burned. Some folks have worked hard to save what is left (e.g., Blue Creek watershed). Over the decades many battles have been waged and much compromised. Lawsuits abound. Commercial and sport fishing get constrained more and more each year. Fewer California residents make the trip north to fish the Klamath each year.

There remain many intractable problems that may never be resolved. The upper watershed in Oregon, mainly around Klamath Lake and the Sprague River, suffers greatly from agricultural development and attendant water quality issues that are unlikely to go away. Much watershed damage has already occurred from timber cutting, urban and agricultural development, roads, fires, and floods. Global warming will continue to reduce rainfall and essential over-summer snowpack throughout the Klamath watershed.

Despite the grim outlook, I have found there are a host of potential actions that can help even before we get to the long-awaited four-dam removal. We need to stop the bleeding, save the patient, and start recovery. Many of the treatments and tools are already available. Some are willingly provided by Mother Nature (e.g., water and beavers). There are many diverse efforts and treatments already underway on a small scale that can be expanded and coordinated. Lessons learned can be better shared.

To get the process moving faster, I offer the following recommendations:

Move toward making the Klamath tributaries, the Salmon, Scott, and Shasta rivers, salmon sanctuaries like Blue Creek on the lower Klamath, an effort being coordinated by the Yurok Tribe. Allow the Karuk Tribe to coordinate on the Salmon River (give them a grant to do this). On the Scott and Shasta Rivers, allow ranchers to coordinate. The Nature Conservancy is already involved in the Shasta River, as Western Rivers Conservancy is in the Blue Creek Sanctuary.
Re-adjudicate water rights and water quality standards on the Scott and Shasta rivers. I know these are “fighting words”, but it must be done now. At least start this process, starting with the State’s new groundwater regulations. Vital portions of both rivers sit dry much of the year from surface diversions and groundwater extraction. Hundreds of thousands of young salmon and steelhead literally dry up every spring and summer, including tens of thousands of endangered Coho salmon. State laws prohibit this, as do State Board regulations, yet it continues on a large scale. Make the State enforce the laws.
List Klamath spring-run Chinook as a federal and state endangered fish. They have become extinct from the Scott and Shasta rivers in my lifetime. They hold on in the Salmon River. They need and deserve full protection of the state and federal endangered species acts.
Fully implement federal and state recovery plans for salmon and steelhead. Get funding.
Re-introduce Coho and spring-run Chinook salmon to tributaries where populations are or are near extinction, including tributaries above dams.
Rehabilitate hatchery programs on the Klamath and Trinity rivers. Develop conservation hatchery elements within these existing programs to promote wild genetic strains of salmon and steelhead in the tributaries.
Reconnect the upper Shasta River to allow salmon and steelhead access. This process was started by the Nature Conservancy and tribes, but is long delayed and unfunded.
Fully fund and implement a salmon and steelhead rescue program for young stranded in tributary spawning rivers.
Improve access of spawning salmon and steelhead to historic spawning grounds blocked or hindered by irrigation dams, road crossings, or low streamflow.
Ensure the ongoing development of the Klamath-Trinity Coho Salmon Biological Opinion for operation of the Shasta-Trinity Division of the federal Central Valley Project adequately protects and helps restore the endangered Coho salmon.
Require the California Resources Agency to take a leadership role in making the Klamath a priority.

Figure 1. Klamath watershed. (Source DOI.)

For more on the Klamath recovery see the following:

California Fisheries Blog

Science, Management, Issues, Problems, and Solutions

Category Archives: Fish (general)