BAY-DELTA UPPER LEVEL SYSTEM PRODUCTIVITY

DRAFT CMARP PRODUCT

MONITORING OBJECTIVES

!To collect biological data to improve management of San Francisco Bay and Delta resident and migratory species.

!To determine whether the CALFED actions improve the abundance, distribution, health, and feeding success of Bay-Delta resident and migratory species. Relevant CALFED Program Goals and Objectives include:

Ecosystem Restoration Program

Ecological Process Implementation (all sub-programs)

Habitat Implementation (all aquatic, wetland, and riparian sub-programs)

Species and Species Group Implementation (all aquatic sub-programs except salmon, steelhead, and delta smelt)

Stressor Implementation (all aquatic sub-programs)

Water Quality Program

Mine Drainage

Wastewater and Industrial Discharge

Water Management

Toxicity of Unknown Origin

Watershed Management Coordination

CONCEPTUAL MODEL AND/OR LISTING OF HYPOTHESES AND ASSUMPTIONS OF THE SYSTEM

The monitoring program for higher level system productivity is based on the following set of assumptions:

!Implementation of CALFED Goals and Objectives will affect the abundance, distribution, health, and feeding success of higher trophic level aquatic species.

!Increases in abundance, distribution, health, and feeding success indicate a positive effect of the CALFED program.

!Suitable monitoring data will help CALFED to develop specific restoration projects for specific species and ecosystems.

!Collection of long-term monitoring data for these fish species will provide an opportunity for adaptive management.

This workteam focused on 3 management goals associated with the geographic area of the Bay and Delta: 1) management of harvested populations; 2) monitoring status and trends of a representative array of estuarine species; and 3) assessing general trophic dynamics among estuarine species. The basis for the selection of species within these groups is as follows:

1

10/14/98 Draft

I. Management of harvested populations emphasizes particular species. Harvested species included in this category are summarized in Table 1 along with adequacy of existing monitoring programs for different life history stages. The species addressed by this team include: striped bass, American shad, white and green sturgeon and various catfishes. Also included are some macroinvertebrates, as Dungeness crabs and crayfish. Species which fall within this category, but are addressed by other workteams include centrarchids, salmon, steelhead, and the edible clams of the estuary. Population management requires indepth knowledge of the full life cycle of the species of interest, unless it can be ascertained that some lifecycle stage is never limiting.

II. Monitoring status and trends of estuarine species requires tracking abundance and distribution of a broad array of Arepresentative@ species throughout the estuary. In contrast to managing harvested species, this management goal can be satisfied by sampling a single, sensitive life stage of each species. To meet this goal, a broad geographic area should be sampled with a diverse array of sampling gears so that no habitat or type of species is omitted. Over 165 species of fishes have been collected from the estuary (Appendix 1); however, target species for status and trends sampling comprise only a subset of the most common fishes and invertebrates (Table 2). Common species which are addressed by other workteams include tule perch, inland silverside, rainwater killifish and mosquitofish (Shallow Water Habitat Team), wakasagi (Delta Smelt Team) and hardhead, Sacramento sucker and hitch (River Resident Fishes Team).

III. Assessing general trophic dynamics among estuarine species requires knowledge of dietary patterns and selection of species that represent common feeding strategies. Conceptual food web models for the Delta, Brackish and Polyhaline regions are provided in Figures 1-3. Selected species should be abundant enough to be sampled reliably and to have a significant effect within the ecological community. Some monitoring might also include generic measures such as isotope ratios. Lower trophic levels are included in the charge of other workteams. We include 5 categories of species that feed directly upon the lower trophic levels: 1) planktivores such as northern anchovy and American shad; 2) nearshore piscivores such as centrarchids; 3) open channel piscivores such as striped bass; 4) benthic feeders such as splittail, crayfish, starry flounder and caridean shrimps and 5) Apickers@ such as juvenile centrarchids and inland silverside. Species associated with edge habitats such as salmon, tule perch, shiner perch, and inland silverside are covered by the Shallow Water Habitat Team. As a result, nearshore Apickers@ were included in the conceptual model of the food chain, but were not considered for monitoring by this team.

1

10/14/98 Draft

MONITORING PLAN ELEMENTS

Measurement of abundance is the primary element of the monitoring plan. Depending on the fish species, the abundance of one or more life stages would be monitored. In addition, distribution, diet/feeding success and health would also be used as indicators for some species. For the purposes of this monitoring plan, the system would be divided into sub-areas (e.g., west Delta, central Delta, north Delta) and distribution would be examined based on the changes in abundance of a given species before and after the CALFED program is implemented. The following summarizes existing programs that would be included as part of the core sampling program and new monitoring that would be required to measure abundance, range and health and to describe food web relationships.

Existing Monitoring

Monitoring for existing programs is summarized in Table 3. Details about these elements are provided below.

The California Department of Fish and Game (CDFG) Fall Midwater Trawl Survey (FMWT) samples 87 stations from San Pablo Bay to Rio Vista on the Sacramento River and Stockton on the San Joaquin River (Stevens and Miller 1983). Currently 116 stations are sampled of which 100 are used for indexing abundance and 16 are sampled to detect the limits of delta smelt distribution. Each station is sampled monthly from September through December using a 17 m-long midwater trawl with a 3.7 m2 mouth. An annual abundance index is calculated as the sum of monthly indices for subareas of the system. To calculate monthly indices, catch per trawl is averaged for stations within each subarea, multiplied by a volumetric estimate for the subarea, then summed across all subareas.

The CDFG Bay Study samples 52 stations from south San Francisco Bay to the western Delta using both midwater and otter trawls; 35 of these stations are used for indexing abundance (Armor and Herrgesell 1985). The midwater trawl is the same as used by the FMWT; the otter trawl has a 4.9 m head rope and 3mm mesh codend. Annual abundance indices are calculated for each net as the average of monthly indices. Monthly indices are calculated similarly to the FMWT, except that average catch per 10,000 m3 and average catch per 10,000 m2 were calculated for the midwater and otter trawls, respectively, rather than average catch per trawl.

The University of California at Davis samples 7 sloughs in Suisun Marsh using an otter trawl similar to that of the CDFG Bay Study survey (Moyle et al. 1986). For splittail, a monthly abundance index is calculated as the sum of the mean catch per trawl for each of the seven sloughs in the Marsh. Annual abundance indices are calculated as the mean of the monthly values.

The USFWS Chipps Island survey samples a single location in the channel at Chipps Island using a midwater trawl towed at the surface. Ten 20-minute tows are made each day, but the number of days sampled per week varies by month.

1

10/14/98 Draft

The Central Valley and State Water projects operate louver facilities to direct fish away from the export pumps (Brown et al. 1996). Salvaged fish are counted year-round at 2-hour intervals when the pumps are operating. Fish salvage data from these facilities are considered a valuable source of abundance data for the system (Stevens and Miller 1983). For splittail, abundance indices are based on the total salvage divided by the volume of water exported during the time periods when each life stage is most abundant at the facilities.

The USFWS beach seine survey is used to develop both abundance and distribution indices for salmon and other species. Beach seine sampling began in 1976. It is presently conducted on a year-round basis, with single hauls of a 50-foot, 1/8" mesh minnow seine taken on each sampling date at (in 1998) 46 shore stations throughout the Sacramento-San Joaquin Delta and at shore stations along the Sacramento River as far north as Red Bluff. In the early years, routine sampling at most stations was done only during the late winter/early spring months. All fishes taken in a haul are identified to species and, in most cases, measured for fork length. The panel of stations visited in the beach seine program has changed substantially since the program's inception in 1976. A core set of 13 stations has been visited at least once during January -- March in each year the program has operated, and an additional 13 stations have been visited at least once during the first quarter of each year since 1982. In addition, approximately 100 other stations have been sampled during at least one year, and some of these stations have a sampling history spanning a substantial part of the full life of the program. Several of the core stations have been moved a short distance from their original locations because of riprap construction. The core set of stations is widely distributed, with most stations lying on the lower Sacramento River, the North Delta, the Central Delta, and, more recently, on the San Joaquin River, where the number of stations was substantially increased in 1994.

Adult striped bass abundance is monitored by the CDFG with a mark-recapture study in which 5,000-10,000 legal-sized ( 18 inches TL) striped bass are tagged with disk-dangler tags and released during their annual spring spawning migration to fresh water. Gill nets in the western Delta and fyke traps in the Sacramento River near Knights Landing are used to capture the fish. All legal-sized bass caught are counted and all tag recaptures from tagging in previous years are recorded. All tagged fish are aged from annuli on the scales. Recapture samples obtained during tagging are augmented by a year-round creel census which samples throughout the estuary. A sample of striped bass observed in the creel census is measured, sexed, and scale sampled and all tag recaptures are recorded. Fish seen in the creel census are aged from annuli on the scales. A computer-generated age-length key is used to estimate the age composition of all legal-sized striped bass caught during tagging and observed in the creel census. The tagged:untagged ratio in the recapture samples and the number of striped bass tagged is used to calculate mark-recapture Petersen population estimates stratified by sex and age. Legal-sized age 3 and age 4 abundance is used to estimate recruitment to the legal-sized population. The past practice of annual tagging changed to alternate-year tagging in 1995 as the result of an evaluation of the adequacy of biennial population estimates for monitoring trends in legal-sized striped bass abundance.

1

10/14/98 Draft

Sturgeon abundance is monitored by the CDFG with a mark-recapture study in which 250-2,500 legal-sized (46B72 inches TL) sturgeon are captured in trammel nets in San Pablo Bay in the fall and tagged using disk-dangler reward tags. Tagging is not conducted every year; the most recent tagging was in 1998 and is not planned again until 2001. Sturgeon are measured and release location, tagger, date, and condition of the fish recorded. All tag recaptures in the trammel nets are recorded and used, in conjunction with the number tagged, to estimate sturgeon abundance using a multiple census (for same-year recaptures) or Petersen (for recaptures from previous years' tagging) technique.

Juvenile sturgeon abundance is monitored by the CDFG using baited set-lines to capture ages 1-7 (12-36 inches TL) juvenile sturgeon in the delta, Suisun Bay, and San Pablo Bay. Each set-line consists of a 1800-foot ground line along which 80-100 gangions are attached. Each gangion consists of a halibut snap, a 3-4-foot leader, and a single baited hook. Sampling takes place for 3 months between July and November. Each month, three set-lines are fished simultaneously for about 24 hours until 21 locations are sampled. Juvenile sturgeon captured incidentally by the adult striped bass tagging program (using gill nets in the western delta during April and May) or the adult sturgeon tagging program (using trammel nets in San Pablo and Suisun bays in September and October) are added to the catch to increase the sample size. A subsample of juvenile fish is aged using a section of the pectoral fin and the resulting age-length key is used to assign ages to the entire catch. An index of year-class strength for white sturgeon is calculated from these age composition data to establish long-term trends in white sturgeon production. Juvenile green sturgeon are infrequently captured in this survey and their abundance is not indexed.

Abundance of resident nearshore delta fishes (centrarchids, cyprinids, and ictalurids, primarily) is monitored by the CDFG with boat-mounted electrofishing gear at 20 locations in February, April, June, and August in alternate (odd-numbered) years. Sampling is stratified by area of the delta so that five sites are surveyed in the east and central delta, three sites each in the north and west delta, and four sites in the south delta. This stratification of effort corresponds approximately with the relative abundance of resident fishes in each area as determined from random sampling in 1980-1983. Fishes are collected by habitat type at each sampling site. Population trends and community structure will be determined using univariate and multivariate statistical analyses.

1

10/14/98 Draft

The Real Time Monitoring program is conducted to provide data for adaptive management of the export facilities to minimize impacts to aquatic species. The intensive field portion of sampling starts April 1 and continues through June 30 of each year. The real time monitoring program consists of three parts: 1)enumeration of fish salvaged at the CVP (Central Valley Project) and SWP (State Water Project) fish salvage facilities; 2) collection of fish from strategically selected sites within the Delta, and 3) Delta wide fish distribution survey targeting small juvenile fish. Data reported by the program comes from 7 sample sites located throughout the Delta, where fish are collected using midwater trawls, Kodiak trawls and a 20mm net (See Table 1). Sampling with the 20mm net takes place only when the bimonthly delta smelt 20mm survey indicates the presence of juvenile delta smelt in the central and southern delta. The bimonthly geographic survey targeting small delta smelt using a 1 mm mesh tow net is reported under the 20mm Delta Smelt Study. An integral part of the real time monitoring program is an active feedback loop that adjusts sampling effort based on the collection of the special status species, delta smelt, splittail and winterrun size chinook salmon.

Additional monitoring is conducted to support the Vernalis Adaptive Management Plan (VAMP). VAMP is being implemented to provide protective measures for fallrun

chinook salmon and to gather scientific information on the effects of various flows in the lower San Joaquin River, Central Valley Project and State Water Project export pumping rates, operation of a fish barrier, and survival of salmon smolts through the Delta. Sampling includes kodiak trawling in the south Delta and releases of coded-wire-tagged juvenile salmon during flow studies.

Proposed Additional Sampling

The existing programs are exceptionally valuable because they provide a long-term baseline for the evaluation of CALFED actions and management of several sport and commercial species. However, this sampling is insufficient to address the full suite of CALFED monitoring objectives. As a result, we include recommendations for new monitoring associated with the three previously-described management goals. An overall summary of the additional monitoring elements is provided in Table 4. Details about rationale and possible methods are described below.

Harvested Species: Data for several species and life stages are inadequate for effective management. These are indicated by bold face numbers in Table 1. Details of the proposed changes in sampling or new sampling efforts are as follows.

!American shad. Measures of the CLUE of American shad adults may be obtained from the new Central Valley and Anadromous Creel Survey which was initiated in 1997, as well as from catches in the gill nets and fyke traps used for capturing adult striped bass for tagging in the spring. Correspondence between these measurements may indicate whether such data are useful annual measures of relative abundance of adults. Some additional resources may be needed to extract these data from existing catch records.