Methods
I. Field Collection Methods
Water Quality
Water quality parameters are recorded at the start of each sampling event. Water temperature (degrees Celsius), electrical conductivity (microSiemens/cm), specific conductance (microSiemens/cm), pH, dissolved oxygen (mg/L), and turbidity (FNU) are sampled with a YSI ProDSS handheld meter. Turbidity values are averaged over three readings. Secchi depth is measured in the shade. Light attenuation (subsurface irradiance) is measured with a light meter (LI-COR LI-250A) at the surface, and at 75%, 50%, 25%, and 1% of the surface reference value in the water. The corresponding depth and measured micromoles are noted. Tide, condition of sampling (condition code), Microcystis level, and weather are also recorded with water quality parameters.
Zooplankton Collection
Fixed zooplankton nets are used to collect zooplankton samples from the Yolo Bypass at the base of the Toe Drain and the Sacramento River at Sherwood Harbor. Samples are collected on an ebb tide, either on a biweekly or weekly basis (during floodplain inundation). In the Yolo Bypass, nets are deployed year-round from the rotary screw trap, which is anchored in the middle of the Toe Drain channel. At the Sacramento River site, nets are deployed dockside during periods of higher flows (typically January-June), and from a boat at 2-3 mph when downstream flows are insufficient (e.g. less than 2 fps, typically July-December). The 150 micron net is fished just below the surface for 5 minutes, and the 50 micron net is fished just below the surface for 2 minutes, though sampling times may be shortened when high levels of debris interfere with sample collection. Tow times are recorded with each sampling event. Samples are washed down with deionized water and preserved in 10% Formalin with Rose Bengal dye to aid in separating organisms from detritus and algae.
Dimensions of the plankton nets are as follows: 1) 150 micron net, 0.50 m diameter mouth, 2 m in length with polyethylene cod-end jar screened with 150 micron mesh. 2) 50 micron net, 0.50m diameter mouth, 2m in length with polyethylene cod-end jar screened with 50 micron mesh. Water volumes are recorded using General Oceanics flow meters (Model 2030R) that are mounted inside each plankton net.
II. Sample Processing and Tracking
Sample Tracking
After being stored for a minimum of two weeks in 10% formalin, zooplankton samples are rinsed through a 106 micron (150 micron net) or 45 micron (50-micron net) sieve. Large debris (leaves, sticks, etc.) are carefully rinsed and removed. The remaining sample is retained for identification, and transferred to 8% Lugol’s Iodine solution. Samples are securely packaged to prevent leakage and breakage. Samples are then shipped to contractors for taxonomic identification. The contractors return the samples to DWR after completing analyses.
Sample transfer
Zooplankton samples are transferred to 8% Lugol’s Iodine solution after being stored for a minimum of two weeks in formalin. Samples are securely packaged to prevent leakage and breakage. Samples are then shipped to contractors for taxonomic identification. The contractors return the samples to DWR after completing analyses.
Contractor
Since July 2015 (see historical changes for more information)
BSA Environmental Services
23400 Mercantile Rd. #8, Beachwood, OH 44122
216-765-0582
https://www.bsaenv.com/
Subsampling
Contractors rinse and filter each sample through appropriately-sized sieves. Samples filtered through a 43 micron sieve are retained for analysis. For mesozooplankton, sample volumes are adjusted to achieve 200-250 organisms per sample. For microzooplankton, sample volumes are adjusted to achieve 300 organisms per sample.
Identification and Enumeration
Samples are stirred to distribute organisms homogeneously. Subsamples are extracted with a Hensen Stempel subsampling pipette, dispensed into a Ward zooplankton counting wheel, and examined under a compound microscope at a minimum of 100x magnification. Zooplankton are identified to the lowest taxon possible. Mesozooplankton are enumerated by differentiating both species and life stages. Calanoid copepods, Acanthocyclops vernalis, Oithona, and Limnothoina are identified to species and separated into juveniles and adults. Harpacticoids are identified to order, with juveniles and adults combined. Cladocera are identified to genus, with juveniles and adults combined. Microzooplankton are enumerated. If the sample has any suspended sediments, the sediment volume will be recorded separately.
Sample Archive
Samples are stored by the contractor for up to 90 days, then returned to DWR for storage.
III. Quality Assurance & Control
Calibrations
YSIs are calibrated for pH, turbidity, dissolved oxygen (DO), and electrical conductivity (EC) monthly. Percent dissolved oxygen is also calibrated daily to local barometric pressure before use in the field. pH is checked two weeks after calibration to see if drift in readings is occurring. If drift has occurred, pH is recalibrated.
Sample Identification Quality Control
Contractors re-identify 10% of samples to ensure 90% similarity.
Data Quality Control
Four levels of quality control are conducted on data:
Field data are checked by someone other than the data recorder prior to leaving each field site,
Datasheets are checked while being entered into the Microsoft Access database, which has customized error-checking and data validation checks,
A separate DWR staff member compares data from original field sheets to data entered into the database,
Data are examined and visualized in R to look for outliers by station, year, and month. Values that are out of range are flagged, and select values are modified and/or re-calculated (see table below for more information). Water quality data are overlaid on real-time sonde data from Lisbon (data obtained from the California Data Exchange Center or CDEC) to ensure values are within range.
PQC: Physical data outliers/errors are flagged, and incorrectly entered values are modified.
QC1: Flowmeter difference values are filtered and flagged for exceedingly high and low values (see table for more details). Select values are re-calculated.
QC2: Samples with comments indicating poor sample collection are flagged (see table for more details)
QC3: CPUE values are filtered and flagged for exceedingly high values (see table for more details). Select values are re-calculated.
See metadata pdf for more information.
Notes on Data Quality
While some level of QC was conducted on monitoring data, we recognize that there are more QC tests that could be conducted in the future to further clean up the data.
Flowmeter: There are known issues with the flowmeter data, and flowmeters generally. Flowmeter values are key to calculating CPUE, so QC efforts were conducted to try to account for some of these inaccuracies.
Historically when flowmeters were not working, readings from another net’s flowmeter (e.g. egg & larval) were used instead. Because these nets have a different mesh and diameter, values from other nets cannot be reliably used. While samples noted using values from other nets have been flagged, these comments may not have always been recorded.
Before 2012, low-flow flowmeters were not consistently used under low flows, sampling at Sherwood Harbor often was done from shore even under low flows, and sampling was not always at a consistent tide. At flows that are too low, or during a switching tide, flowmeter values are not accurate. Samples noted to have had very low flows, or that were taken during a high tide, were flagged, but these comments may not have always been recorded.
50-micron net: During high flow/high turbidity events, it is difficult to get a good sample, as silt and debris block the sample. Tow times are shortened to mitigate this problem to an extent.
Taxonomist changes: There have been changes in taxonomists throughout the years, which may influence the direct comparability of data from different contractors. Additionally, some samples were taken during years, but sent to contractors much later, and thus samples may have been in poorer condition.
Notes on how to use data
For un-edited CPUE values, use the column “CPUE.”
For CPUE values subjected to described QC, used the column “CPUE_ed.”
To further remove other flagged data, filter out data with Flag_QC1, Flag_QC2 and/or Flag_QC3 = 5 (highly suspect), or Flag_QC1, Flag_QC2 and/or Flag_QC3 = 3 (suspect).
At the bottom of the .Rmd file, there is also code to combine all the flags into one column so that any flagged data can be filtered out.
IV. Data Storage
Data Management and Archiving
Field data are collected and recorded on paper datasheets by DWR personnel, then entered into a Microsoft Access database. The monitoring program is currently transitioning to a new database called WISKI, which will replace Microsoft Access by the end of 2022. Paper datasheets are archived in binders that are stored at the West Sacramento DWR office, and electronic copies are archived on DWR/AES Network drives.
Taxonomic results are received from the contractor via email in an Excel spreadsheet. Data are printed and entered into the Access database by DWR personnel. Hard copies of the data are printed and stored in binders at the West Sacramento DWR office. Electronic copies of results for taxonomic analyses are archived on DWR/AES Network drives.
V. Calculations and Analysis
CPUE Calculations
Calculations for CPUE are conducted in R. Separate calculations are conducted by net mesh size, flowmeter type, and mesozooplankton (calanoids, cyclopoids, harpacticoids, cladocerans) & macrozooplankton (mysids, clams, snails, etc.) versus microzooplankton and nauplii (rotifers, barnacles, copepod nauplii, cladocera nauplii, ostracods), as these categories have different associated values for the equation below:
CPUE = (C / (Vsub/Vsamp)) / V
C: Organisms counted in sample
Vsub [SubMeso * Subsample or SubMicro * Subsample in data]:
Total subsample volume = Subsample volume x Number of subsamples
Vsamp [VolMeso or VolMicro in data]:
Total sample volume (send to contractors)
V [VolNet in data]:
Volume of water sampled through net =
(((Flowmeter count end-Flowmeter count start) x R)/999999*NV)
R: Rotor constant (Regular flow = 26873, Low flow = 57560)
NV: Net volume = (pi * (net diameter)^2))/4
VI. Historical Changes
Equipment
Physical Water Quality: Prior to 2011, YSI 63, and possibly others. Between 2011-October 2016, a YSI 556 - was used. Starting October 2016, a YSI ProDSS was used.
Turbidity: February 2012-October 2016: Glass vials of water were collected and analyzed with Hach 2100Q portable turbidimeter. Starting in late October 2016, the YSI ProDSS was used. Thus, units shift from NTU to FNU. Although these units are not exactly the same and cannot be standardized, they are very close to each other (Memo, DWR, 2020), so we have decided to keep all turbidity readings in one column for ease of analyzing data.
Nets: (150 micron and 50 micron): Sea-Gear Corporation, 50 micron net started in 2015.
Methods
Field Sampling
2002: Conductivity starts being collected
May 2008: Specific Conductivity, pH, Dissolved Oxygen start being collected
2011 – Present: Zooplankton starts being sampled year-round (originally Jan-June)
February 2012: Turbidity starts being collected
2012-Present: During low flows (usually June-January) Sherwood samples are collected from a boat (rather than from shore) to ensure sufficient flow for adequate sample collection.
2015-Present: The 50 micron net was added to the zooplankton sampling procedure to capture smaller zooplankton that may be important prey for larval and juvenile fish.
Late October 2016-present: Turbidity readings taken by YSI instead of Hach turbidimeter.
Lab
1999-2004: Additional 1 millileter subsamples (cells) examined until at least 200 organisms have been counted
2007-2010 (Tenera): Differentiate life stages of holoplankton and identify to species (including all lifestages) a subsample of species in samples. Water is to be added to the retained sample to achieve a target concentration of 200 organisms per mL. A 1mL sub-sample is to be extracted with an automatic pipette, dispensed into a 1mL Sedgwick-Rafter counting slide, and examined under a compound microscope. Holoplankton will be identified to the lowest taxon possible and enumerated by life stage and species. Additional 1mL sub-samples will be used until at least 6mL were processed.
2004-2006; 2011-2015 (EcoAnalysts): The contractor shall be responsible for differentiating the life stages of all holoplankton and identify to genus or species a sub-sample of species in each sample…Water will be added to the remaining sample to achieve a target concentration of 200 organisms per mL… A 1 mL (target volume) subsample will be extracted with an automatic pipette, dispensed into a 1mL Sedgwick-Rafter counting slide, and examined under a compound microscope at a minimum of a 100x magnification. A total of at least 200 organisms, excluding microzooplankton will be identified over at least five subsamples. Microzooplankton will be identified in the same subsample as the mesozooplankton of there are at least 100 organisms per sub-sample. If not, the dilution volume will be adjusted to achieve 100 organisms per mL and the second dilution volume will be recorded.
2015-2018 (BSA): A sub-sample will be extracted with a Hensen Stempel subsampling pipette, dispensed into a Ward zooplankton counting wheel, and examined under a compound microscope at a minimum of 100x magnification. For mesozooplankton, water will be added to the remaining sample to achieve a target 40-50 organisms and subsample each sample 5 times to accumulate a total of 200-250 organisms per sample. For microzooplankton, the sample total volume will be readjusted to target 100 organisms per mL and 3 subsamples will be taken to accumulate 300 organisms total per sample.
2018-Present (BSA): Total counts and volumes are summed, and considered as 1 subsample.
Taxonomists
See Metadata PDF for contractor history.
