<emphasis role="strong">Methods</emphasis>
Study system
We conducted our study in the northern Delta, a freshwater tidal portion of the SFE, a large and highly altered estuary on the Pacific coast of North America. Our study sites were located in a terminal channel formed by the upper portions of the Sacramento River Deep Water Ship Channel (DWSC), which connects the West Sacramento Port to the Sacramento River and San Francisco Bay/Pacific Ocean. The 69 km long channel has a width of ~150 m, depth of ~10 m, and has one set of non-operational ships locks at its northern terminus.
We sampled at ten fixed sites along the DWSC, identified by channel marker (CM) number, to characterize water quality parameters and plankton community composition and biomass (depths 9.8-12.8 m). Study sites were located in three distinct hydrodynamic zones: the no-exchange zone (NE), where waters are trapped in the upper, landward section of the DWSC (above CM 70), a zone of mixing and high turbidity in the mid-zone (CM62 - CM66; low-exchange = LE), and the seaward zone that experiences tidal exchange each day (below CM62; high exchange zone = HE; P. R. Stumpner et al. 2020).
Sample Collection and Processing
We collected water chemistry and plankton samples at each station approximately monthly from April 2012 until September 2019 (74 sampling dates). Two YSI 6600 probes were used to collect field measurements of temperature and specific conductance. Turbidity was measured in the field on samples collected from 1 m depth using a Hach 2100P turbidimeter.
Two 1-L water chemistry samples were collected at 1 m depth using a submerged water pumping system and stored in acid-washed HDPE bottles for laboratory analyses. A 500 mL water sample for phytoplankton identification and biovolume quantification was collected at 1-m depth and preserved in 3% Lugol's solution (final concentration). Zooplankton samples were collected by vertical tow using a 150-µm mesh zooplankton net with a retrieval rate of ~0.33 m s -1
. The 150-µm mesh size was selected to avoid complications from high suspended sediment concentrations that frequently occur at some sampling sites, but as a result, smaller life stages (nauplii, early copepodites) and microzooplankton were not sampled effectively (Kayfetz et al., 2020). Zooplankton samples were preserved in a 2% final concentration of Lugol’s solution.
Sample processing was initiated within 24 hours of collection time. A subsample was filtered through a pre-rinsed 0.2 µm polycarbonate membrane (Millipore) for quantification of soluble reactive phosphorus (SR-PO 4
), nitrate-N (NO 3
-N) + nitrite-N (NO 2
-N), and ammonium (NH 4
-N). SR-PO 4
was determined using the ammonium molybdate spectrophotometric method (limit of detection (LOD) ~0.005 mg LP-1; Clesceri et al., 1998). 3
+NO 2
-N (LOD =0.01 mg L -1
; 3
-N constituted >95% of the combined NO 3
+NO 2
-N concentration, we report the NO 3
+NO 2
-N concentration as NO 3
-N in this study. NH 4
-N was determined spectroscopically with the Berthelot reaction, using a salicylate analog of indophenol blue (LOD ~ 0.010 mg L-1; Forster 1995). Dissolved Si was determined using the molybdate-reactive spectroscopic method (SM4500-SiO2 C; MRL=0.5 mg L-1; Clesceri et al., 1998).
Chlorophyll-a concentrations were determined from duplicate samples collected on Whatman GF/F filters, following methods in Clesceri et al. (1998). Samples were filtered in the field using low vacuum and stored on ice until storage in a -20ºC freezer. Samples were extracted in 90% ethanol, and filters were freeze dried but not ground (Sartory and Grobselaar, 1984). Samples were analyzed by fluorometric determination with the limit of detection dependent on the volume of water filtered (200 - 1000 mL, generally 0.5 µg L -1
).
Phytoplankton identification and enumeration were performed using standard membrane filtration (McNabb, 1960). A Leica DMLB compound microscope was used for random field counts of at least 300 natural units and taxa were identified to the lowest possible taxonomic level. Cell biovolumes were quantified on a per milliliter basis (Hillebrand et al., 1999). Zooplankton abundance and identification were measured on three 1-ml aliquots using a Wilovert inverted microscope at 100x with a target tally of 200-400 specimens. Biomass estimates were based on established length/width relationships (Dumont et al., 1975; McCauley, 1984; Lawrence et al., 1987).
Species-level phytoplankton biovolume (μm 3
L -1
) or mesozooplankton biomass (μg dry weight L -1
) were summed according to the following taxonomic divisions, which represent the dominant taxa in our study system: Bacillariophyta, Chlorophyta, and Cryptophyta (phytoplankton), and Copepoda and Cladocera (zooplankton).
Quantifying hydrodynamic conditions
Since hydrodynamic processes can affect constituents over short (hours to days) to long (weeks – months) timescales we used several metrics that vary over longer timescales to more closely align with the time scale (monthly) of sample collection. Tidally averaged discharge (hourly; cubic feet per second) was obtained from two USGS monitoring stations, one near Cache Slough in the HE zone (CM 45, USGS 11455350) and one within the DWSC (CM 54; USGS 11455335; U.S. Geological Survey, 2021). Tidally-averaged discharge at CM 45 was used as a metric of seasonal hydrology for sites in the HE zone—when high flow events through the Yolo Bypass were excluded, it was highly correlated (R=0.81) with Sacramento River discharge at Freeport (USGS 11447650). In the landward LE and NE zones, we used discharge at CM 54 to capture variation in flow conditions (no gauges exist within the NE zone).
Normalized tidal amplitude was computed as a measure of tidal strength that varies from spring-neap (~14 days) to yearly timescales. During periods of higher tidal amplitude (normalized values greater than 1) water parcels are subject to stronger mixing and longer transport, decreasing water residence time. During periods of lower tidal amplitude (normalized values less than 1) water parcels will have less transport and mixing, with commensurately longer water residence time. Normalized tidal amplitude was estimated for each discrete sampling event. Discharge (15 min frequency) at CM 45 and CM 54 was used for the normalized tidal amplitude calculation. Because the discharge time series at CM45 ended prior to the end of the study, we substituted discharge at CM 41 (USGS 11455385) for the four missing dates (May - August 2019). For each discharge signal (Q) the tidally filtered discharge (<Q>) was computed using a Godin filter (Godin, 1972), and the tidal discharge (Q’) was computed as Q’ = Q - <Q>. The Q’ signal was then used to compute the tidal amplitude by finding the outer envelopes (or tidal maxima) using a 30-hr. moving window. The tidal amplitude is the difference between the upper and lower envelope, which was then normalized by median tidal amplitude over the length of the record.
We used tidal excursion length to assign fixed sampling stations to each of the three hydrodynamic zones (HE, LE, or NE) on each sampling date, and designated final site groupings based on a combination of hydrodynamic and environmental characteristics. Exchange zones were defined based on estimates of tidal excursion from the mouth of the DWSC using methods developed in P.R. Stumpner et al. (2020) and Young et al. (2021), and are a proxy for water residence time. The tidal excursion from the mouth of the DWSC was estimated using water velocity data at CM 54, which was then corrected by a scaling factor of 1.54 based on comparison to drifters that travelled up the channel over a course of a single flood tide. Corrected water velocity was then integrated over each tidal period (~25 hours) continuously over the length of the record to provide an estimate of the distance travelled along the DWSC. The distance travelled along the channel defines the upper boundary of the HE zone. The range of tidal excursions, over a month, was used to bound the LE zone. As a result, the boundaries of the HE and LE zones varied based on the tides, and the boundary of the LE zone also varied on monthly timescales due to the variation in the range of tidal excursions across spring-neap cycles.
For stations that switched zones between sampling dates (CM 56, CM 62), group assignment was based on both the proportion of sampling dates the site fell within each zone, as well as by comparing average water quality parameters with nearby stations. CM 56 fell within the HE zone on 57% of sampling dates, within the LE zone on 39% of sampling dates, and the NE zone on 4% of dates, and it was therefore grouped with stations in the HE zone. CM 62 fell within the LE zone on 39% of sampling dates and within the NE on 56% of sampling dates, but was assigned to the LE group based on its high mean turbidity. Although CM 66 technically fell within the NE zone, we assigned it to the LE zone based on its high turbidity.
References
Clesceri, L.S., A.E. Greenberg, and A.D. Eaton (eds.). (1998). Standard Methods for the
Doane, T. A., Horwath, W. R. (2003). Spectrophotometric determination of nitrate with a single
reagent. Analytical Letters,
36:2713-2722.
Dumont, H.J., I. Van de Velde & S. Dumont. (1975). The dry weight estimate of biomass in a
Feyrer, F., Sommer, T., & Slater, S. B. (2009). Old School vs. New school: Status of Threadfin Shad (Dorosoma petenense) Five Decades After its Introduction to the Sacramento–San Joaquin Delta. San Francisco Estuary and Watershed Science
, 7
(1). https://doi.org/10.15447/sfews.2009v7iss1art3
Feyrer, F., Slater, S. B., Portz, D. E., Odom, D., Morgan-King, T., & Brown, L. R. (2017). Pelagic Nekton Abundance and Distribution in the Northern Sacramento–San Joaquin Delta, California. Transactions of the American Fisheries Society
, 146
(1), 128–135.
Godin, G. (1972). The analysis of tides
. University of Toronto Press.
Hillenbrand, H., C.D. Dürselen, D. Kirschtel, U. Pollingher, and T. Zohary. (1999). Biovolume
calculation for pelagic and benthic microalgae. Journal of Phycology
35:403-424.
Kayfetz, K., Bashevkin, S. M., Thomas, M., Hartmann, R., Burdi, C. E., Hennessy, A., et al. (2020). Zooplankton Integrated Dataset Metadata Report
. IEP Technical Report 93. California Department of Water Resources, Sacramento, CA.
Lawrence, S. G., D.F. Malley, W.J. Findlay, M.A. MacIver & I.L. Delbaere. (1987). Method for
McCauley, E. (1984). The estimation of the abundance and biomass of zooplankton in samples.
pp. 228-265, In: J.A. Downing & F.H. Rigler (eds.) A Manual for the Assessment of Secondary Productivity in Fresh Waters. Blackwell Scientific Publishers.
McNabb, C.D. (1960). Enumeration of freshwater phytoplankton concentrated on the membrane
filter. Limnology and Oceanography
5:57-61.
Sartory, D.P., and J.U. Grobbelaar. (1984). Extraction of chlorophyll-a from freshwater
phytoplankton for spectrophotometric analysis. Hydrobiologia
. 114:177-187.
Stumpner, P. R., Burau, J. R., & Forrest, A. L. (2020). A Lagrangian-to-Eulerian Metric to Identify Estuarine Pelagic Habitats. Estuaries and Coasts
, 44
(5), 1231–1249. https://doi.org/10.1007/s12237-020-00861-7
United States Geological Survey. (2021). USGS water data for the Nation: U.S. Geological Survey National Water Information System database. . https://doi.org/http://dx.doi.org/10.5066/F7P55KJN
Young, M. J., Feyrer, F., Stumpner, P. R., Larwood, V., Patton, O., & Brown, L. R. (2021).
