Data Package Metadata   View Summary

1.) Data collection methods:

A detailed account of the field methods for Summer Townet Survey are described within the STN Standard Operating procedure which may be found here: CDFW Summer Townet FTP Site for Protocols- https://wildlife.ca.gov/Conservation/Delta/Townet-Survey.

Field collections are conducted over 6 surveys, biweekly in the summer starting June and ending in August. Within each survey, fish and zooplankton samples are collected at 31 index stations and 9 non-index stations. Fish samples are collected using a conical fixed frame net (referred to as the townet hereafter), towed obliquely through the water column for 10 minutes. The townet is 5.48 meters long (18’) and composed of two sections. The front section is composed of 12.7 mm (1/2”) stretched, knotted nylon, mesh 3.35 m (11') long, tapering down to an additional 0.6 m (2') "fyke". This "fyke" fits entirely within the second section, a 2.2 m (7’) section of woven mesh with approximately 8 holes per inch (1/10-inch mesh openings). A flowmeter is suspended at the mouth of the net to measure water volume. The townet is attached to a bridle which is in turn attached to a 76.2 m (250’) length of line flagged at 7.62 m (25’) intervals. The net is retrieved at regular intervals using a hydraulic winch. Attached to the top of the townet frame is a modified Clark-Bumpus (CB) meso-zooplankton net, (160-microns). This is attached to a clear acrylic housing and has a flowmeter fixed to the center to measure water volume. The townet sled is composed of two metal skis (1.52 m long x 12.7 cm wide), connected by a wooden board in the rear. Midway through the ski, a metal ring shaped as the letter D is mounted upright with two metal arms on each side supporting the ring midway and connecting to mid-front of a ski. The townet is lashed to the D ring, which provides a fixed 1.49 m2 area during sampling. A bridle is attached to the front of each ski and this is in turn attached to a 76 m long cable.

At each station the Summer Townet Survey collects physical metrics; surface and bottom salinity (micro-siemens per centimeter), surface and bottom temperature (degrees Celsius), surface turbidity (nephelometric turbidity unit = NTU), and water clarity (Secchi disk depth in centimeters). The survey crews also rank the weather as cloud cover (1=Cloud 0-33%, 2=Cloud 34-66%, 3=Cloud 67-100%, 4=Rain), surface wave strength (1=Calm, 2=Waves without whitecaps, 3= Waves with whitecaps), and Microcystis presence (1=Absent, 2=Low, 3= Medium, 4= High, 5= Very High). Following these measurements, the survey conducts at minimum 2, ten-minute tows. At index stations, a 3rd tow is conducted if any fish is collected in one of the first two tows. Three tows are always conducted at station 323, located in San Pablo Bay, where more tows are needed to represent the larger volume of water present at this station.

The STN has a record of 93 different categories of fish and macro-invertebrates (crabs, shrimp, and jellyfish) in the database. All fish species are identified (ID) to the lowest taxonomic category possible and fork lengths (FL) are measured for the first 50 representatives of each species, at each station, up to the nearest millimeter (mm). Common invertebrates such as jellyfish (Cnidarians) and shrimp species are identified to Genus. Organisms that cannot be identified in the field, such as larval fish less than 20 mm FL, are preserved in ethanol or 10% formaldehyde. These organisms are labeled UNIDS (Unidentified) and are later classified with the assistance of an Olympus SZX stereo microscope at a magnification of 8x to 80x. Fish species of special concern, such as Osmerid species (i.e State and Federally ESA listed Delta Smelt and Longfin Smelt), have their IDs confirmed by at least two different qualified individuals. Any fish or invertebrate that cannot be identified, such as due to excessive damage to the specimen or lack of development, is given the lowest taxonomic category possible (e.g. Tridentiger spp). A sub-sample of individuals from select species (Striped Bass, Smelt, Shads) are selected for stomach content analyses. See https://wildlife.ca.gov/Conservation/Delta/Special-Studies for more details on the CDFW Diet Analysis project.

2.) Link to blank datasheet: Available upon request (See data contact information) or accessible at https://filelib.wildlife.ca.gov/Public/TownetFallMidwaterTrawl/TNS%20MS%20Access%20Data/TNS%20data/ .

3.) Instrument and Equipment Specifications, including QAQC methods and frequency: A detailed account of QAQC procedures in the field, lab and in database entry are described within the Summer Townet Standard Operating procedure which may be found here: https://filelib.wildlife.ca.gov/Public/TownetFallMidwaterTrawl/TNS%20MS%20Access%20Data/TNS%20data/.

A portable Hach turbidimeter (model 2100Q) is currently used to measure turbidity (NTU) in surface water collected using a 12-liter bucket. The turbidmeter is calibrated before each biweekly survey.

A Yellow Springs Instrument (YSI Pro 30) is used to measure salinity (micro-siemens per cm) and temperature (Celsius). The instrument is calibrated before each survey. Temperature and salinity are measured one meter from the surface and above the demersal zone. Surface waters are collected using a 12-liter bucket. Demersal water is collected using a 4-liter Van Dorn bottle sampler. The Van Dorn is dropped to the bottom of the water column to collect the sample.

GPS coordinates are taken using a Garmin GPS. Models have varied over time. (1959-2008, 2010 No GPS recorded; 2009, 2011-2015 Unknown model; 2016-2020 Garmin GPSMap 76 and Garmin GPSMap 78; 2021-Current Garmin GPSMap 64X).

Microcystis ranking is selected using a 1-5 scale with 1 representing visually undetectable levels of Microcystis and 5 representing thick clumping colonies. Images representing each rank can be seen in the protocol (https://filelib.wildlife.ca.gov/Public/TownetFallMidwaterTrawl/TNS%20MS%20Access%20Data/TNS%20data/.).

The Secchi disk depth is a measure of water clarity. It consists of a black and white disk mounted on a two-meter pole. It is lowered into the water until the user cannot distinguish the color on the disk. The distance to this point is recorded in centimeters (cm). Measurements are always taken on the shaded side of the vessel, and users must not be wearing sunglasses while measuring water clarity by Secchi. The disk has been mounted on a 2 m pole for routine use, but a disk mounted to a surveyor’s tape is used when clarity >2 m.

4.) Analysis Methods and SOPs: The Summer Townet Standard Operating Procedure may be found here: CDFW FTP Site for Protocols.

To calculate water volume sampled by the townet, the STN records the start and end counts of a General Oceanics (model 2030R) mechanical flowmeter. The meter difference (ΔM) is calculated between Meter Out (start) and Meter In (end). Note: Flowmeter counters range from 0-999,999, therefore when ΔM is equal to a negative value add 1,000,000 to the difference. Water volume is then calculated using the following equation: Volume (m3) = (ΔM*k*1.49), where ΔM is the final flowmeter count (Meter Out) minus the initial flowmeter (Meter In) count, k is a constant specific (k factor) to each flowmeter that has been calibrated in the past via a flume but has been replaced with the standard value of 0.026873 provided by General Oceanics, and 1.49 is the area of the townet mouth opening in m2.

5.) Project History:

The STN has been conducted for over 60 years, and thus has gone through changes, both temporary and permanent, including; changes in tow method, changes to station designations and additions, changes to the schedule, changes to the net, changes to the sled, boats, winches, database and data entry, addition of special studies and zooplankton sampling, species added, and types of measurements. These changes are summarized in the timeline below however greater detail may be read in the Summer Townet Standard Operating Procedure: CDFW FTP Site for Protocols.

1959- The STN begins conducting 3 ten-minute surface tows at 31 stations (index stations) in the Upper San Francisco Estuary, targeting young-of-year Striped Bass. Surveys were conducted biweekly until Striped Bass fork length averaged greater than 38.1 mm. Each survey was conducted during the five days of least tidal fluctuation, regardless of holidays or weekends. Tows were conducted using the RV Striper.

1961-The protocol was modified to conduct stepped oblique tows. Correction factors are used to modify the catch reported from 1959-1960.

1962-It is believed that the townet design was altered around this year to a design lacking a canvas collar.

1967- STN begins to use the RV Striper II. When the RV Striper II was put into service, the engine was calibrated to determine at which RPM it towed the net at 0.8m/second (2.7 ft/second). A Hach turbidimeter is first used to measure turbidity.

1969- Specific conductivity readings are initially measured using a Wheatstone Bridge circuit. This required water samples to be collected in the field and returned to the laboratory for in house analysis. Lab processing of EC ended in 1994.

1970- A diet study is started to examine Striped Bass stomach contents. Station numbers were revised to more closely match the Fall Midwater Trawl. Around this time, it was observed that townets were constructed at 5.45m (18’) in length rather than the historical 4.5m (15’) length. Early versions had a jar. The later longer nets have a tied codend. Turbidity measurements using a Hach turbidimeter are suspended.

1971-Pygmy flowmeters (Kahl Scientific Supply Co.) were used, briefly, to measure water volume sampled. A standard 735m3 volume is applied to all tows through till 2002. STN began measuring fork length to the nearest whole millimeter, in the past this was measured to the nearest one tenth inch.

1973- Sampling efforts in San Pablo Bay were briefly increased by conducting a single tow at eight additional stations during survey 3. These stations were: 301a, 302, 312, 315, 322, 326, 328 and 329. In addition, the protocol was modified so that cable angle, rather than engine RPM, is used to monitor boat speed and net depth.

1977-Samping was extended into Cache Slough with a single tow at station 713 during surveys 1 and 2. It was not continued in subsequent years until 2011.

1980-Protocol for measuring Secchi is altered to take Secchi measurements on the shaded side of the vessel rather than the sunny side.

1988- The cable angle, during fishing, was observed to be towed at 70° rather than 81° for the next two years. It was corrected back to 81° in 1990.

1989- The STN collects Striped Bass otoliths for study from 1989-1993.

1992- STN begins to collect Delta Smelt for a diet study.

1994- The protocol was modified to eliminate a third tow if no fish were collected in the first two tows. Except at station 323 in San Pablo Bay where 3 tows were to be collected every survey. In addition, CDFW staff begin to complete all data entry into a computer instead of State Water Resources Control Board Data Management Section or UC Davis Computer Services. Field crews begin to use portable probes to measure specific conductivity in the field.

1995- At this time research vessels are all equipped with hydraulic winches, used to bring in the net and adjust net position in the water column, most of which have variable speed control capabilities. Previously electric Warn winches were used.

1996-Survey 1 was not completed due to mechanical challenges. This partial survey was not entered into the database until 2013.

1997-The STN begins to use the RV Scrutiny.

1998- STN begins to use a mix of vessels to conduct the survey including the Scrutiny, SeaArk, Munson, Beowulf and Triakis.

2001-The STN Biologist has new townets constructed with canvas collars. In addition, townets briefly return to the historical 4.5m (15’) length for the next two seasons.

2002- An extensive QA/QC field protocol was adopted. Two townets using knotless mesh (as opposed to the knotted mesh) were tested in this Summer Townet season. An additional 6 nets made of knotless mesh were ordered for the following year. The STN also begins to use synthetic rope (i.e. brand “Spectra”) instead of steel cables.

2003- Sampling was altered to follow a five-day schedule conducted Monday thru Friday on alternating weeks, opposite the Smelt Larva Survey (SLS) and Spring Kodiak Trawl (SKT). In addition, the number of surveys conducted each sampling season was formally standardized to six in 2003. A length-weight study is started using fish collected by STN from 2003-2004. General Oceanic flowmeters are used to measure water volume for all tows and the standard volume of 735m3 is no longer used in favor of actual water volumes sampled.

2005- An additional 6 townets constructed of knotless mesh were ordered. Sampling continued using a mix of knotless and knotted townets. In addition, a standardized zooplankton sampling protocol was added. The STN sled was modified by mounting a modified Clark-Bumpus net at the top of the rigid frame. Zooplankton samples are collected during the first tow at each station, preserved in a buffered 10% formaldehyde solution and rose Bengal, and brought back to the CDFW facility for identification and enumeration. The first samples associated with projects investigating the Pelagic Organism Decline (POD) begin. Five-gallon water samples were collected at ten stations (340, 504, 508, 609, 704, 711, 804, 902, 910 and 915) during each survey. These water samples were analyzed by University of California, Davis (UCD). Up to 20 Delta Smelt and Striped Bass per survey week were also collected, from 2005 to 2008, for histopathology and otolith analysis by UCD researchers. Heads were removed in the field and preserved in ethanol, while bodies were preserved in formalin. Staff begin retaining fish for the new CDFW POD Diet and Condition Study. CDFW field biologists begin to use a Yellow Springs Instrument (YSI) to measure temperature and specific conductivity.

2008- A field protocol is added to rank weather and Microcystis presence. Note, that some measurements of these were recorded in field comments starting in 2007.

2009- Collections of fish associated with POD are resumed. Up to five Delta Smelt and five Longfin Smelt were collected for UC Davis per survey, separated into bodies and heads, and preserved in the manner described above. GPS coordinates were determined for the start and end of the first tow at each station. Following tows are conducted over the same transect and direction. The data was recorded in degrees, minutes, decimal minutes in the WGS 1984 projection. A gear test was conducted to compare use of steel vs spectra synthetic cable which found steel cable had significantly larger volumes of water sampled and significantly greater distance traveled than tows made with synthetic (Spectra) cable possibly due to higher boat speeds necessary to maintain a proper cable angle and townet sled depth while towing with the heavier steel cable.

2010- The STN Biologist has new nets constructed without canvas collars. New nets were constructed of knotted mesh and sampling occurred only with knotted mesh. Townets begin to have serial numbers assigned to them. In-house diet analysis resumes. Up to 20 Striped Bass, Threadfin Shad and Mississippi Silversides were collected per survey, preserved in a buffered 4% formaldehyde solution and brought back to the CDFW facility for gut content analysis. Turbidity measurements using a Hach turbidimeter are reinstated. GPS coordinates are switched to degrees, minutes, seconds WGS 1984 in 2010.

2011- STN began to extensively sample in the North Delta (Cache Slough, Sacramento River Deep Water Shipping Channel). Eight additional stations (713, 716, 721, 719, 723, 795, 796, 797) were added. Only a maximum of two tows were conducted to limit Delta Smelt catch. Due to ESA take permitting limits, only a single tow was conducted if Delta Smelt catch was greater than 10 in the first tow. Station 721 tows were reduced to 5-minute tows, as opposed to the normal 10-minute tows. UCD researchers began preserving Delta Smelt in liquid nitrogen as part of the Fall Low Salinity Habitat Study (FLaSH). The frozen fish were dissected at UCD for histopathology, otoliths, and other analyses. During the dissection, the stomachs were preserved in ethanol and returned to CDFW for gut content analysis. A protocol was added to record wind direction and tow direction in field datasheets. For the first time, depth Loggers are used sporadically with the Summer Townet Survey to compare predicted depth based on cable out and actual depth.

2012- GIS coordinates for the supplemental stations added in 2011 are standardized. Biologists note that the STN was using heavier 20-mm survey sleds. A study was conducted which observed that the heavier sled resulted in greater water volume sampled. Following this, sled design is standardized to the lighter STN sled, approximately 8.16- 9.97kg (18-22lbs per ski). Fish preservation for UCD researchers begins in August, due to a shortage of boat operators. UCD researchers preserved Striped Bass, American Shad, and Threadfin Shad in addition to Delta Smelt. This practice continues until 2015. Bottom temperature is recorded on datasheets but was previously noted in comments in 2011.

2013- Survey crews note that the fyke design in the townet was not constructed correctly. This was corrected prior to the start of the field season. It is uncertain how long the incorrect fyke was in use in prior seasons.

2014- STN standardizes the length of the townet to 5.48 m (18’). The use of a second vessel during the survey week allowed the completion of the STN survey in three days, as opposed to five. In addition, canvas collars are reintroduced into the design of the townet for lashing the net to the metal “D” frame.

2015- Due to the lack of 12.7 mm (1/2”) stretched mesh, mesh used to construct the townet has varied over the years. While the size of the 12.7 mm mesh opening has remained consistent, the denier size has varied between 210/6 to 210/15. In 2015, nets were constructed using 210/6 twine, which is a thinner yarn than historically used. Database changes also occurred. The Access database was first linked to the Tier-3 Server located in Sacramento. All data is now stored on the server and backed up locally (Stockton office). Fish collection is modified again, when UCD begins to preserve fish using liquid nitrogen to address the Fall Low Salinity Habitat (FLASH) studies. Delta Smelt and other pelagic fish are preserved in liquid nitrogen and transferred to UCD. CDFW continues to preserve fish in (formalin) and (ethanol) for in-house diet analysis.

2017- Depth loggers, SENSU loggers, were regularly attached to the townet to record actual depth of the fish net during the season. Analysis has shown the actual depth closely matches targeted depth based on cable length and angle.

2020- Station 721 could not be sampled this year due to large amounts of invasive aquatic vegetation.

2021- Station 721 could not be sampled again due to large amounts of invasive aquatic vegetation. Station 721 is replaced with station 722, located two kilometers downstream from 721. It is the same location used by the Spring Kodiak Trawl (SKT 716).

6) Timeline of organism identifications added to the STN database. Tows prior to these dates will have NA to indicate that the organism (or other identification, i.e. damaged fish) was either not yet present in the Estuary or not included in STN identifications.

1997- First detection of Shokihaze Goby, Tridentiger barbatus in Estuary by CDFW Bay Study.

2001- Tridentiger spp. added as a category due to difficulty identifying larvae and small juveniles to species among Shimofuri and Shokihaze gobies.

2007-Invertebratre species: Crangon spp., Palaemon macrodactylus, Exopalaemon modestus, Maeotias marginata, Aurelia spp., and Aequorea spp.

2009- Polyorchis penicillatus and Dungeness Crab

2012- Pleurobrachia bachei and Blackfordia virginica added to species list

2019- Species Code Name for Inland Silversides (INLDSLVR) was changed to reflect Mississippi Silversides (MSSIL). In addition, a UNID category was added for the genus’s Pomoxis (POM, 191) & Lepomis (LEP, 190).

2021- UNID code updated to reflect unidentified fish. Codes for damaged UNIDs, unidentified invertebrates, and larval Striped Bass/larval Tridentiger spp. too undeveloped to be distinguished from one another.

7.) QA/QC

Description of the full Quality Assurance and Quality Control (QA/QC) procedure for STN has been divided into 4 components. First, the data used to inform management decisions are identified and labeled as ‘critical data’. The next two sections describe the QA/QC procedures for: field, laboratory and database.

Field QA/QC procedures

STN field staff conduct a field QA/QC procedure at 5% of the minimum number of tows per year (24 tows). These QA/QC stations will be spread out among the 6 surveys, such that each survey contains at least 4 QA/QC tows. QA/QC tows are selected at random using R statistics software or a similar random number generator. In 2021, the number of QA/QC tows was increased to one per boat day, or 6-8 tows per survey. This helped to match QA/QC procedures of other CDFW surveys such as the Fall Midwater Trawl. QA/QC tows are marked on a Station Checklist that is provided to the Boat Operator at the beginning of the survey. The boat operator notifies the science crew that a quality assurance check is required after the initial sample processing is complete.

After each tow the net codend is emptied into a sorting tub for processing and the net is checked for any remaining fish. Once a quality check is revealed assigned to a station, the net is checked up to the fyke, a second check, to ensure that all fish have been removed. The deck is examined to ensure that all fish made it into the sorting tub. Any debris in the net or tub is inspected and then placed in a bucket of water. The water in this bucket, as well as that in the sorting tub is strained through a net to check for missed fish. Any fish or invertebrates located during the second check are recorded on the field data sheet and noted on a QA/QC datasheet. All Age-0 Striped Bass are re-identified and their fork lengths are re-measured on the QA/QC datasheet

Species counts are split between species not preserved and preserved. All species (fish and invertebrate) are recounted that have not been preserved and the recount is recorded on the QA/QC datasheet. The species to be preserved are also confirmed using a second count and this is recorded on the QA/QC datasheet. A representative of each species is collected, other than those already brought to the lab, to be re-identified in the lab.

The final flowmeter and CB flowmeter counts are re-checked. If the end values for the flowmeter counts are different, the datasheet is corrected, and a note made in the QA/QC comments. The starting and final flowmeter counts (from the field datasheet) for both CB and net flowmeters are recorded in the QA/QC datasheet, and the flowmeter total values are calculated for the tow.

Finally, GPS coordinates on datasheet are compared with those listed in the STN protocol and those saved in the GPS unit. Any significant discrepancies are noted on the QA/QC datasheet. The towing path should cross within 1 quarter mile of the STN location. GPS locations are also checked electronically, following datasheet entry, comparing the start and end locations with expected STN station location.

Lab QA/QC Procedures

Any fish from the field with a fork length ≤20 mm, and any fish that cannot be identified to species, are preserved and brought to the lab to have their identity confirmed, 1 tow each day (or 5% of all tows returned to the lab) go through a second ID by a second individual. Any discrepancies are noted in the QA/QC datasheet and discussed between the two ID’ers. If no consensus is reached in fish ID, the fish in question is ID’d by a third individual. Any corrections are made in the field datasheet in colored pencil and noted on the QA/QC datasheet.

Database Procedures

After all UNID fish are identified the field datasheets are entered into an Access database. Each station entry is checked again in a line-by-line procedure. Line-by-line consists of a comparing of field data sheets and entered data for each value entered. Following this, a series of R-scripts and plotting is used to identify outliers and potential errors in the database. These outliers are checked with the original datasheets and patterns with nearby stations. Comments are used to describe any missing information. In some cases, such as flowmeter counts, average values are used to replace missing information or errors in the datasheet. Use of averaging is identified in the database, and on datasheets. These comments are not available in this version of the catch matrix, but can be found in the Access Database located here: https://filelib.wildlife.ca.gov/Public/TownetFallMidwaterTrawl/TNS%20MS%20Access%20Data/TNS%20data/