Patterns in Anthropogenic Nitrogen and Water Quality Leading to Phytoplankton Blooms in Urban Estuaries

Abstract

The San Francisco Estuary (SFE) ecosystem receives anthropogenic ammonium (NH₄) from agricultural runoff and sewage treatment plants and has low chlorophyll levels. As observed in other aquatic systems, NH₄ at concentrations < 4 µmol/L inhibits nitrate (NO₃) uptake by phytoplankton and can reduce the frequency with which phytoplankton assimilate all available inorganic nitrogen (i.e., NO₃ and NH₄); paradoxically, elevated NH₄ can reduce the chances of phytoplankton blooms in some high NH₄ ecosystems. For blooms to occur, NH₄ must first be reduced to non-repressive levels, then NO₃ uptake can occur and is accompanied by more rapid carbon (C) uptake and chlorophyll accumulation. The consequence of this sequence is that when NO₃ uptake, C uptake, or chlorophyll concentrations are plotted against ambient NH₄, a rectangular hyperbola results. Here, these relationships are statistically described for a variety of SFE field data, and their management applications are discussed. These relationships enable ambient NH₄ to be used to predict both the likelihood of blooms and to investigate historical changes in productivity when no rate measurements were made. We apply the statistical relationship to a 40-year time series from the SFE during which there was an ecosystem-scale change in the estuarine foodweb with a drastic decline in pelagic fishes (the pelagic organism decline) and suggest that this period aligned with the lowest annual primary production and highest NH₄. The relationship may be generalizable to other high-nitrogen, low-growth systems and aid nutrient management decisions, assuming potential limitations are considered.

1. Introduction

The San Francisco Estuary (SFE) is an urbanized estuary with altered freshwater flows from a changing climate and municipal and agricultural water demand, nutrient input from wastewater and agriculture, and introduced species, primarily through ballast water introductions. Beginning in the mid-1980s the northern SFE experienced a decline in phytoplankton biomass, despite high nutrient concentrations [1], and was described as a high-nutrient, low-chlorophyll system [2], like the high-nutrient low-growth (HNLG) in the Delaware Estuary [3]. Summer phytoplankton blooms, dominated by diatoms, were observed in the northernmost embayment, Suisun Bay, during the 1970s, supporting a rich pelagic food web [4]. Blooms of the magnitude once seen occur only rarely today [5,6,7,8]. Not only has overall phytoplankton biomass decreased, but also the composition of the phytoplankton community has shifted away from diatoms with an increasing contribution by picocyanobacteria [9,10].

Following the decline in phytoplankton biomass, there was a sharp decline in the abundances of some endangered species of pelagic fish, including Delta Smelt (Hypomesus transpacificus) and Longfin Smelt (Spirinchus thaleichthys) beginning in the 2000s, raising concerns from natural resource agencies of a “Pelagic Organism Decline” (POD) [11]. Hypotheses for the POD include mortality from entrainment in California’s water conveyance system and poor food resources due to declines in phytoplankton biomass (and presumably primary productivity). Several factors likely contribute to the HNLG characteristics in the SFE, including low light availability due to high suspended sediments (e.g., [12,13]), the introduction of a voracious invasive clam [14], and nutrient effects resulting from both the chemical forms, ratios, and concentrations, all altered because of increasing human populations and differences in wastewater treatment practices [1,15,16].

Wastewater effluent often has elevated NH₄ that has been shown to be associated with depressed primary productivity in the Delaware Estuary [17], Wascana Creek, Canada [18], the southern California coast [19], and the Sacramento River [20]. This apparent paradox of low productivity associated with high NH₄ is related to decreased phytoplankton NO₃ uptake, as shown for Hong Kong waters [21] and for the SFE [22].

More than fifteen years ago, it was shown that in the SFE, NH₄ concentrations, even at low levels (~1 µmol/L) repress NO₃ uptake by phytoplankton and reduce the chances that all the available nitrogen can be assimilated to support phytoplankton growth; this has been called the NH₄ paradox [22,23,24]. The NH₄ paradox was described using plots of NO₃ uptake or chlorophyll concentration versus NH₄ concentration that showed an inverse hyperbolic relationship fit with a power exponential function [20,22,23] like that observed by other investigators working with cultures [25]. This same inverse hyperbolic function was also shown to fit carbon uptake versus NH₄ but not NH₄ uptake versus NH₄ [20] data from the SFE.

Enclosure experiments conducted using water collected from throughout the SFE showed a consistent sequence of nitrogen drawdown [23] that was confirmed in the field [26]. The sequence of events to occur before high chlorophyll concentrations were observed [22,23] starts with NH₄ being taken up by phytoplankton, and when they have reduced NH₄ to low non-repressive levels, NO₃ uptake begins, accompanied by a rapid chlorophyll increase, and concludes with all inorganic nutrients being consumed [23].

The ecological consequences of the NH₄ paradox may contribute to the low chlorophyll situation and low primary productivity in this anthropogenically modified estuary. The phenomenon of repressed growth and NO₃ utilization by NH₄ is well known to plant physiology as the “NH₄ syndrome” [27,28] and is no surprise for the SFE in view of large inputs of NH₄ from local sewage plants in the Bay Area and especially from the Sacramento Regional sewage facility upgrades in the 1980s in response to the U.S. Clean Water Act. It was shown that high chlorophyll concentrations were only observed at very low NH₄ concentrations (e.g., [5,22]).

The consistent, reproducible relationships between NH₄ concentration and NO₃ uptake, carbon (C) uptake and chlorophyll concentration should allow the prediction of phytoplankton blooms (as chlorophyll) or rates of primary production to be obtained from ambient NH₄ concentrations if the relationships are quantified. Our hypothesis is that higher NH₄ concentrations result in lower values of NO₃ uptake, C uptake, and chlorophyll that can be described statistically. The research objective is to test the statistical relationships of NO₃ uptake, C uptake, and chlorophyll versus NH₄ using a variety of datasets from SFE that vary in space and time. Here, a model is developed using an iterative, nonlinear least squares regression of chlorophyll, NO₃, and C uptake versus NH₄ data from several sampling programs carried out throughout the full spatial extent of the SFE over the past 13 years. Model performance and potential management applications are discussed.

2. Materials and Methods

2.1. Data Sources

Data sources used here are shown in Table 1 and include data collected from a variety of field studies made in the SFE (Figure 1). A series of datasets from a variety of projects conducted in the SFE were used to investigate the ubiquity of the inverse hyperbola, “hockey stick” shaped relationship and how such relationships might be used in a predictive manner. All these datasets, collected by the same laboratory group using the same methods summarized in Section 2.2 and [26] contain surface nutrient and chlorophyll-a concentrations and measurements of NO₃ and C uptake obtained using incubations made at 50% of surface light with trace additions of ¹⁵N-labeled NO₃ and ¹³C-labeled bicarbonate. Some projects covered many years of data (e.g., Nutrient Ratio project—[16,29]), while other projects covered a single cruise in May 2016 but over many different sites, i.e., the entire SFE from the Sacramento River to South Bay (HICO project). Another dataset was focused on one location, D7, sampled by several of our projects from 2011 to 2016. In 2011 and 2012 [26] (SWAMP project), there were also uptake measurements made at different light levels and depths such that trapezoidal depth-integrated values could be calculated, and the NH₄ relationship was evaluated for that in addition to surface data.

One location was selected from the ~40-year historical NH₄ dataset collected by the California Department of Water Resources Environmental Monitoring Program (DWR-EMP) (https://emp-dwr.github.io/emp-website/data-links.html, accessed 22 July 2024). This was D7 in Grizzly Bay, a shoal habitat located in the Suisun Bay sub embayment of northern SFE (Figure 1). This station has been used previously in analyses of nutrients and chlorophyll (e.g., [9,15,30], and depth-integrated production estimates exist for the site [26], allowing published regressions [31] to evaluate fish yield. The fish abundance of Delta Smelt and Longfin Smelt was obtained from a fish monitoring survey conducted by the California Department of Fish and Wildlife since 1967 called the Fall Midwater Trawl Survey (FMWT) (https://wildlife.ca.gov/Conservation/Delta/Fall-Midwater-Trawl, accessed on 2 August 2024) that provides FMWT indices for these fishes.

2.2. Analytical Protocols

Nitrate concentrations were measured in twenty-ml water samples filtered through pre-combusted (450 °C, 4 h) 25 mm Whatman GF/F filters using a Bran and Luebbe AutoAnalyzer II with MT-19 manifold chemistry module according to Whitledge et al. and Bran and Luebbe [32,33]. Separate 25 mL GF/F-filtered samples were analyzed for NH₄ following Solorzano [34] using a 10 cm pathlength cell. Samples for chlorophyll-a (termed chlorophyll throughout) were prepared by filtering 50–100 mL of sample water onto 25 mm diameter GF/F filters. Chlorophyll on the filters was determined by in vitro fluorometry after extraction in 90% acetone using a Turner 10 AU fluorometer [35] calibrated with commercially available chlorophyll-a (Turner Designs, San Jose, CA, USA) and corrected for phaeophytin [36]. Dissolved inorganic carbon (DIC) concentrations, required for calculating the C uptake rates, were measured in 20 mL samples using a Monterey Bay Research Institute-clone DIC analyzer (custom-built by the University of Delaware, Lewes, DE, USA) with acid-sparging and non-dispersive infrared analysis [37,38] following preservation with 200 μL of 5% w/v HgCl₂. Preserved samples were held for up to a month before DIC analysis.

Carbon and NO₃ uptake rates were measured using dual-labeled ¹³C/¹⁵N stable isotope tracer incubations [39,40]. Trace additions of NaH¹³CO₃ (20 µmoles) and K¹⁵NO₃ (typically 0.2 to 0.6 µmoles), both 99 atom %, were added to 150 mL samples to approximately 10% of the ambient concentration. For the Heterosigma study, ambient NO₃ concentrations were unknown at the time of inoculation, so saturated isotope additions of 1.5 µmoles of ¹⁵NO₃ were made. Samples were incubated in 160 mL polycarbonate bottles for 24 h, held in baywater-cooled incubator tables screened to 50% (and 25%, 10%, and 5% for some D7 samples) of surface photosynthetically active radiation (PAR). Incubations were terminated by gentle vacuum filtration onto pre-combusted (450 °C for four hours) 25 mm Whatman GF/F filters. Filters were frozen and stored at −20 °C for up to a year, until analysis for ¹³C and ¹⁵N enrichment and particulate organic carbon and nitrogen concentration with a Europa 20/20 isotope ratio-mass spectrometer system. Near-surface (50% of surface PAR) carbon and nitrogen uptake rates (ρ, µmol C or N L⁻¹ h⁻¹) were calculated according to [41,42]. Depth-integrated C and N uptake rates were calculated at D7 during 2011 and 2012 by trapezoidal integration of rates measured at 4 different irradiances (50%, 25%, 10%, 5%) of surface, based on an empirical conversion factor of light attenuation from Secchi depth determined previously for the low salinity zone in the northern SFE [6].

2.3. Fitting Data

For each of the datasets, chlorophyll, NO₃ uptake, and C uptake were plotted versus ambient NH₄ concentration to determine if there was a “hockey stick” pattern, and then the strength of any inverse rectangular hyperbola relationship was determined. The relationship takes the form of an inverse function, Y = 1/X, where Y is the chlorophyll or uptake rate and X is NH₄ concentration. Data were then fit to this equation in the Solver add-in using Microsoft Excel that uses iterative, nonlinear least squares regression to adjust two “tuning” parameters, C and A, to calculate fitted values of chlorophyll or uptake rate and determine the parameters of NH₄ suppression of NO₃ uptake, carbon uptake, and chlorophyll production. The first of these tuning parameters, term C, was added to account for baseline values for Y. The second parameter, term A, is an “ammonium influence factor” that accounts for weakening or strengthening the effect of NH₄. For NO₃ uptake (ρNO₃), the equation is:

ρNO₃ computed = ρNO₃ observed/(NH₄ * A) + C(1)

Carbon uptake (ρC) and chlorophyll were treated likewise. The sum of squares of the observed and calculated chlorophyll or uptake is used to determine the coefficients of determinations (r²). This least squares analysis offers a statistical fit to quantify the relationships.

Increased or higher values of A result in lower C uptake, NO₃ uptake, and chlorophyll for the same NH₄ concentration, indicating stronger NH₄ suppression of uptake and chlorophyll production. Values of A < 1 will be favorable for blooms, with little impact of NH₄ on NO₃ uptake and chlorophyll, while A > 1 indicates NH₄ suppression of phytoplankton NO₃ uptake and blooms. The term C describes a background or baseline level of uptake or chlorophyll that occurs independent of the NH₄ concentration. The values of A and C, along with the coefficients of determination for the different datasets, are given in Table 2.

3. Results

The Nutrient Ratio project conducted eight seasonal spring and fall surveys in the Sacramento River through to Suisun Bay from 2011 to 2014 (Figure 1). A large, regional wastewater treatment facility (Regional San) discharges municipal wastewater into the river south of Sacramento, resulting in a wide range in ambient nutrient concentrations along the transect as well as varied chlorophyll. NH₄ ranged from 0.5 to 70 µmol/L and reached a maximum in March 2014 near the Regional San discharge point. Chlorophyll ranged from 0.5 to 31.6 µg/L, with a maximal value recorded in March 2013 near the confluence of the Sacramento River with Suisun Bay. When plotted against NH₄ concentrations, NO₃ uptake, carbon uptake and chlorophyll concentrations all showed the same inverse relationship (Figure 2) as observed in the earlier studies. When the least squared best fit was applied using Equation (1), the calculated values closely matched the observed, and all relationships show a coefficient of determination > 0.6 and values of A near or close to 1, indicating unmodified NH₄ repression of NO₃ uptake (Table 2).

The HICO cruise made in May 2016 sampled the Sacramento River from above Sacramento, CA, and continued through all northern, central, and south San Francisco Bay (Figure 1) and observed a large bloom with chlorophyll of almost 80 µg/L near the confluence of the Sacramento River with Suisun Bay. This was likely the same prolonged bloom shown to be a diatom bloom dominated by Aulacoseira [8]. The highest NH₄ concentration was 32 µmol/L just downstream from Regional San WWTP. Plots of the NO₃ uptake, C uptake, and chlorophyll concentration (Figure 3) show the “hockey stick” relationship for all with least squares best-fit regressions of 0.9 and values for A, the NH₄ influence factor, of 0.3 to 0.4 (Table 2).

A third dataset that is available shows how NH₄ monitoring could aid in the prediction of the likelihood of phytoplankton blooms—in this case, a harmful algal bloom (HAB) or red tide of Heterosigma akashiwo that was observed in the SFE in 2022. In Central San Francisco Bay, our research group has maintained a time series over the last 20 years at the seawall adjacent to our laboratory in Tiburon (Figure 1), where surface water is sampled for nutrients and chlorophyll analyses once or twice a week. The available chlorophyll and NH₄ data collected in 2021 and 2022 are plotted in Figure 4a and show that typically NH₄ ranges from 3 to 10 µmol/L and chlorophyll rarely exceeds 5 µg/L. However, in 2022, an inverse rectangular hyperbola with high chlorophyll reaching 77 µg/L was observed at low NH₄ levels (<1 µmol/L). When the relationship was quantified, the least squares regression r² was 0.97 with parameters A = 0.4 and C = 4.4.

This bloom occurred in the summer of 2022, peaking at the seawall on 22 August, and its timing matched well the observations seen in the East Bay for the initiation of a Heterosigma bloom a few weeks earlier. Closer observation showed that before the peak in chlorophyll, the sequence of nutrient drawdown described by [22,23] occurred, with first NH₄ being drawn down to near zero and then NO₃ with an accompanying increase in chlorophyll (Figure 4b).

During this time there was also a monitoring cruise in SFE (green line on Figure 1) by the United States Geological Survey (https://sfbay.wr.usgs.gov/water-quality-database/, accessed on 14 November 2023) on 19 August 2022 that captured the bloom as it progressed from the Central Bay (station 18) to the South Bay (station 30) (Figure 5a) and along the transect a similar sequence with high chlorophyll (110 µg/L) associated with low NH₄ and low NO₃. These data were part of monthly surveys carried out each year by USGS of the entire SFE (Figure 5c, yellow dots). When all data for SFE from all the monthly 2022 transects were plotted, the chlorophyll versus NH₄ shows an inverse hyperbola (Figure 5b) with a least squares fit of r² = 0.92 with parameters A = 1.3 and C = 5.7 (Table 2).

To ensure that all the physiological steps (i.e., NO₃ and C uptake) leading to the hockey stick relationship for chlorophyll vs. NH₄ were met, another 2022 dataset that included the Heterosigma bloom was used to plot NO₃ uptake and C uptake versus NH₄ measured near the decline of the bloom at locations throughout the SFE to see if they showed the same rectangular hyperbolae as the other datasets described above. Figure 6 shows that they did, although the least squares fit was less than the other projects (Table 2). These plots provide support for the use of NH₄ to project the possibility of bloom occurrence—in this case, a HAB.

Finally, a fourth dataset was examined to evaluate how historical NH₄ datasets might be used to estimate historical primary productivity. The historical dataset considered was at station D7 in Suisun Bay, where a 43-year time series of surface nutrients has been used for SFE analyses—e.g., [9,15,30], but there are few phytoplankton NO₃ and C uptake rate data.

NO₃ and C uptake rates were measured at D7 during four separate projects from 2011 to 2016 and plotted versus NH₄ concentration in Figure 7. The ambient NH₄ range was less than other datasets (Figure 2 and Figure 3), with a maximum of 8 µmol/L, while the highest chlorophyll concentration observed was 60 µg/L in May 2012. There is a “hockey stick” relationship with least squares regression (r²) of 0.8 for NO₃ and for C uptake and 0.9 for chlorophyll, with values of A greater than 1 (Table 2), indicating a strong influence of NH₄ on the system.

A subset of data at D7 during 2011 and 2012 included estimates of depth-integrated C uptake, which was reported previously [26]. This dataset allowed for further exploration of the “hockey stick” relationship, this time using depth-integrated production versus NH₄ concentration. Most published relationships between primary productivity and fish abundance or yield use depth-integrated annual primary productivity (e.g., [31,43,44,45]) allowing further exploration between NH₄ concentration and primary production and the decline of fish in the SFE. When integrated C production was plotted versus NH₄ (Figure 8), it resulted in a good quantitative relationship (r² = 0.95) and similar A (A = 3.4) to the surface data (Figure 7b and Table 2).

The tuning parameters A and C were applied to the 43-year dataset of NH₄ concentrations at D7, using the maximum integrated carbon uptake rate (maximum ΣρC observed = 441 gC/m²/yr) at A and C values from 2011 and 2012 (Figure 8), according to

ΣρC computed = maximum ΣρC observed/(NH₄ * A) + C(2)

The computed productivity at D7 is shown in Figure 9a plotted versus NH₄ along with the time-series of mean annual depth-integrated primary production (estimated from monthly DWR-EMP NH₄ values) from 1979 to 2022 (Figure 9b, linear regression (r² = 0.4)). There is a decline in annual average productivity, with the lowest value at the highest annual average NH₄ (Figure 9a) near 2001 and just before the decrease in fish abundance (POD) of two endemic fish, the Delta Smelt and the Longfin Smelt (Figure 10b). This coincidental relationship is confounded by the extreme temporal variability of these fish populations associated with hydrology, including declines during dry years [46], but offers a provocative way to extend our results to help explain the large ecosystem scale change shown by the POD.

In addition, the higher annual productivity predicted from lower ambient NH₄ at the start of the time series in the late 1970s (Figure 9b) coincides with the timing of the productive summer diatom blooms (e.g., 70 µg/L [4]). This offers promise for using ammonium to help assess historical productivity. The next step in such a process could be to use productivity estimates obtained like this to estimate fish yield using published relationships (e.g., [31,43,44,45]).

4. Discussion

Repeatable inverse rectangular hyperbolae (hockey stick relationships) were observed for NO₃ uptake, C uptake, and chlorophyll vs. NH₄ applied across several datasets from the SFE, as have been described for numerous studies, including oceanic HNLC regions [47] and in phytoplankton culture studies (e.g., [25,48]). The observation that NH₄ is strongly related to NO₃ uptake, C uptake, and chlorophyll concentrations is consistent with the known sequence of bloom development observed in SFE [23]. These results reveal a cascade of events that are all controlled by NH₄, beginning with the alleviation of NH₄ repression of NO₃ uptake by phytoplankton and enzyme syntheses [49] for C uptake and the synthesis of chlorophyll.

These repeatable relationships allow for quantification of the repression of NO₃ accessibility by an NH₄ influence factor, A, and a baseline parameter, C, assuming there is plentiful NO₃ to fuel productivity. This is the case for the SFE [50,51]. The progression of blooms in the SFE has been shown to be influenced by ambient nitrogen and requires several conditions to be met. An important step is that phytoplankton need to take the ambient NH₄ down to levels that allow access to NO₃ and use of all the available DIN (both NH₄ and NO₃). It also assumes that flow is optimal such that washout is minimized (e.g., [7], and extended residence times occur to enable chlorophyll accumulation as described by Ball and Arthur [4] and Cloern [52], who reported optimal flows that supported the regular summer phytoplankton blooms in the 1970s in SFE. This is in addition to optimal light and abundant NO₃ to serve nitrogen demand.

The least squares regression analysis of the equation output, model parameters, and the coefficient of determination, r², are summarized in Table 2. The coefficients of determination offer a measure of statistical fit to quantify the relationships and range between 0.62 and 0.97. Eleven out of the 15 datasets have r² values of ≥0.8, indicating a high degree of confidence in the model. For 8 out of the 15 datasets, the ammonium influence factor A was less than or equal to 1, indicating a low influence of NH₄ on the initiation of the phytoplankton bloom sequence, i.e., bloom favorable. The D7 datasets show higher values of A, suggesting that NH₄ is more likely to suppress blooms there. All datasets except one show a baseline factor, C. Without this factor C, the statistical least squares fits are weakened considerably. When all the surface data are pooled, there is still a good statistical relationship, with r² of 0.7 (Table 2), even with the larger sample sizes of ~200 for rates and 432 for chlorophyll. The A values are high, ranging from 2 to 3, reflecting the high A values of the D7 dataset, and the high C value for the chlorophyll relationship is likely due to the contribution of the D7 data also. The compiled dataset relationships support the application of this inverse rectangular model.

This quantitative analysis offers potential applications of NH₄ data to indicate the possibility of future blooms, including HABs, as shown in the Heterosigma dataset described above and to look at historical blooms or reduced primary productivity and outcomes, as illustrated by the forty-year dataset from D7. Interestingly, NH₄ favors many estuarine HAB species [49,53,54] and could possibly trigger HABs while suppressing the faster phytoplankton bloomers, although most HABs observed in the SFE have shown complete depletion of all available DIN, including NO₃, once uptake is enabled. Systems monitoring NH₄ may be able to offer projections of possible HAB occurrences and possibly fish yield or abundance in a system, and to evaluate the impact of anthropogenic NH₄, once there are a few datasets available from the location with NH₄, NO₃ uptake, C uptake, and chlorophyll to generate the least squares parameters.

Another aspect of the fitted hyperbolas for the different areas of the SFE is that the parameters can be used for both management and prediction purposes. One example might be to provide information for nutrient reduction policies being considered for aquatic systems with increased urbanization and associated NH₄ from point sources, such as wastewater treatment systems [8]. The ecosystem response to wastewater treatment upgrades (e.g., [16]) that have major societal costs could be investigated. Another example might be to evaluate how to improve productivity in HNLG estuaries like the Delaware Estuary which shows the same inverse relationship of primary productivity versus NH₄ [17]. Acknowledging potential limitations such as dominant phytoplankton species, light availability, etc., knowing the hyperbola relationship for a given system and the NH₄ influence factor, A, could provide managers insight to determine the NH₄ reduction needed to realize enhanced productivity and chlorophyll.

To illustrate how the hyperbolic relationship could be used in a management context, one can explore how much present-day NH₄ concentrations should be reduced to increase potential productivity and chlorophyll in Suisun Bay to support a restoration of its historic production and benefit fish populations contributing to the POD. The range of NH₄ values for 2022 in the DWR-EMP monthly dataset is 4 to 10 µmol/L (mean of 7 µmol/L) with associated average chlorophyll of 2.9 µg/L, equivalent to 4% of the maximum previously observed for D7 (70 µg/L, Figure 7).

To determine what reduction in NH₄ is needed to reach the maximal chlorophyll values, first the hyperbolic plot from Equation (1) is reorganized and the maximal percent chlorophyll concentration (y-axis) plotted against NH₄ concentrations with different values of NH₄ impact factor A (0.5 to 3) from Table 2 (Figure 11). The current situation with 4% of maximal chlorophyll and an average concentration of NH₄ of 7 µmol/L is shown as a black circle. It happens to lie on the A = 3 plot. Knowing the value of A for the system, the NH₄ required to reach the maximum chlorophyll (i.e., 100% on the Y axis) can be inferred from these curves (Figure 11). From data for D7 provided in Table 2, A for chlorophyll = 3.3. The point on the A = 3 curve for 100% of the potential maximum chlorophyll is shown as a black circle in the upper left corner, equivalent to when NH₄ is 0.3 µmol/L. For productivity, if A = 2 is used, the corresponding NH₄ is 0.5 µmol/L to reach maximum values. A reduction from 7 to 0.3 µmol/L NH₄ should increase the potential chlorophyll to maximal.

Ball and Arthur [4] observed peak chlorophyll concentrations in 1979 at D7 of 70 µg/L with an NH₄ concentration of 2 µmol/L. When their dataset was plotted (Figure 12), the parameters from the rectangular hyperbolic relationship were A = 0.5, C = 6.8 with r² of 0.99. The observed value of NH₄ of 2 µmol/L for peak chlorophyll values would have been predicted when using A = 0.5 by Figure 11, supporting this approach.

These results, with high statistical confidence and agreement with plant physiology and ecological studies in other estuaries, confirm that anthropogenic inputs of NH₄ can negatively impact the primary productivity and nutrient processes in the SFE ecosystem. All the rectangular hyperbolae relationships from the SFE (Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8) and the hypothetical (Figure 11) show the importance of relatively low NH₄ concentrations (<2 µmol/L) in determining whether NO₃ uptake and C uptake will reach near 100% maximum capacity, i.e., nitrogen exhaustion. Previously, 4 µmol/L was introduced as a critical minimum threshold for NO₃ uptake and bloom formation in the SFE [7,20,22,24]. One of the outcomes of this study is that the original value of 4 µmol/L is likely too high, and the real range is closer to 1–2 µmol/L, like algal culture studies [25,48] and observations made in the Delaware Estuary [55] and suggested earlier for SFE [22].

This quantitative investigation into the effect of NH₄ concentrations on estuarine production processes should be helpful to regulators and ecologists. The relationships provide a tool to determine reductions in NH₄ to ensure maximal phytoplankton NO₃ uptake to ameliorate low productivity and chlorophyll situations in HNLG systems or to assess the likelihood of imminent blooms. The ability to predict primary productivity (carbon uptake) from NH₄ concentrations provides a way to investigate historical changes in primary productivity from long-term nutrient datasets and can be extended to link with higher trophic levels. Continuous NH₄ measurements, either on board vessels carrying out high-speed mapping (e.g., [56]) or installed at fixed locations, are being implemented in estuaries, including in the SFE. Thus, real-time estimates of nutrient uptake, chlorophyll synthesis, primary production, and fishery yield are near-term outgrowths of this research.

Author Contributions

Conceptualization, R.C.D. and A.E.P.; methodology, R.C.D. and A.E.P.; formal analysis, R.C.D. and A.E.P.; investigation, R.C.D. and A.E.P.; data curation, F.P.W., A.E.P., and R.C.D.; writing—original draft preparation, F.P.W.; writing—review and editing, F.P.W., R.C.D., and A.E.P.; project administration, F.P.W.; funding acquisition, R.C.D., F.P.W., and A.E.P. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

Data are being deposited in the Environmental Data Initiative (EDI) https://environmentaldatainitiative.org/, accessed on 5 August 2024.

Acknowledgments

We thank S. Blaser, E. Antell, A. Pimenta, J. Lee, N. Travis, T. Lee, S. Strong (Estuary and Ocean Science Center, San Francisco State University), P. Glibert, J. Alexander (University of Maryland Center for Environmental Sciences), and S. Murasko (Florida Fish and Wildlife Research Institute) for participation in sample collection on cruises and lab analyses.

Conflicts of Interest

The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures and Tables

Figure 1. Map of project sites for the datasets used in the study.

Figure 2. (a) NO3 uptake (µmol/L/h); (b) carbon uptake (µmol/L/h); and (c) chlorophyll (µg/L) vs. NH4 (µmol/L) for all nutrient ratio cruises (observed data, black circles) combined with least squares best fit and Solver calculated values (blue, green, and red squares, respectively).

Figure 3. (a) NO3 uptake (µmol/L/h); (b) carbon uptake (µmol/L/h); and (c) chlorophyll (µg/L) vs. NH4 (µmol/L) for HICO16-2 cruise (observed data, black circles) combined with least squares best fit and solver-calculated values (blue, green, and red squares, respectively).

Figure 4. Data collected from the EOS Seawall. (a) Chlorophyll (µg/L) versus NH4 (µmol/L) measured from January to December in 2021 (cyan circles) and 2022 (black circles) with least squares hyperbolic curve fitted using Equation (1) and calculated chlorophyll for 2022; (b) time series in August 2022 of concentrations of chlorophyll (green circles), NH4 (red circles), NO3 (blue circles), and phosphate (purple circles).

Figure 5. Data collected by USGS. (a) Concentrations of NH4 (µmol/L, red), NO3 (µmol/L, blue), and chlorophyll (µg/L, green) measured on 19 August 2022 at stations 18 to 36 along a north–south transect in South San Francisco Bay shown as red circled location in the map (c); (b) chlorophyll (µg/L) versus NH4 (µmol/L) over the entire 2022 that includes 12 monthly transects of the entire SFE (stations are represented by the yellow dots in (c)), measured by USGS (black circles) and solver calculated values (red squares) and curve (red line) fitted using Equation (1); (c) map to show transect and stations 18–36.

Figure 6. (a) NO3 uptake (µmol/L/h); (b) carbon uptake (µmol/L/h); and (c) chlorophyll (µg/L) vs. NH4 (µmol/L) for four Heterosigma cruises from 21 August to 13 October 2022 (observed data, black circles) combined with least squares best fit and solver calculated values (blue, green, and red squares, respectively).

Figure 7. (a) NO3 uptake (µmol/L/h); (b) carbon uptake (µmol/L/h); and (c) chlorophyll (µg/L) vs. NH4 (µmol/L) for D7 from projects conducted 2011–2016 (observed data, black circles) combined with least squares best fit and solver calculated values (blue, green, and red squares, respectively).

Figure 8. Observed depth-integrated ρC (gC/m2/yr) vs. NH4 (µmol/L) for D7 from 20 cruises in 2011 and 2012 (SWAMP project [26]) (black circles), combined with least squares best fit solver calculated values (green squares) using Equation (1).

Figure 9. Depth-integrated primary productivity (gC/m2/yr) at D7 calculated from historical DWR-EMP NH4 data from 1979 to 2022 using tuning parameters A and C for D7 versus (a) surface NH4 (µmol/L); or (b) year. The arrow shows the start of the POD.

Figure 10. (a) Annual average surface ammonium (µmol/L) at D7 from 1979 to 2022 with linear regression (r2 = 0.4); (b) annual fish indices of Delta Smelt (purple triangles) and Longfin Smelt (green) plotted against the year of sampling. Arrows show the start of the POD.

Figure 11. Calculated maximal chlorophyll percentage vs. NH4 (µmol/L) using varying A—NH4 impact factor. Black circles indicate the current situation at D7 and the NH4 required for the maximum chlorophyll when A = 3.

Figure 12. Observed chlorophyll (µg/L) vs. NH4 (µmol/L) for D7 (Grizzly Bay) in 1979 from [4] (black circles), combined with solver calculated values (red squares).

Table 1

San Francisco Estuary Delta datasets used in the study.

Study Name	Location	Date	Data
Nutrient Ratio (NUR) project	Transects from above Sacramento to Carquinez Straits	March or Sept. 2011, 2012, 2013, and 2014	ρNO3, ρC, chl, NH4
HICO	Transect from I80 to bottom of South Bay (Sta US36)	May 2016	ρNO3, ρC, chl, NH4
EOS Seawall	Tiburon	2021 and 2022	chl, NH4
USGS data	Golden Gate to Lower South Bay	2022	chl, NH4
Heterosigma	San Pablo Bay to South Bay	Aug to Oct 2022	ρNO3, ρC, chl, NH4
D7 surface from SWAMP, HICO, NUR	D7	Spring 2011, 2012, 2013, 2014, 2015, and 2016	ρNO3, ρC, chl, NH4
D7 integrated from SWAMP Project	D7	2011 and 2012	Int ρC
DWR-EMP D7 40-year time series	D7	1980–2022	NH4
Ball and Arthur	D7	1979	chl, NH4

Table 2

Least mean square parameters r², tuning parameters A and C, and number of samples (n) for different SFE datasets and compiled surface datasets.

Study Name and Data	r2	A	C	n
Nutrient Ratio—ρNO3	0.76	1.0	0.01	100
—ρC	0.68	1.5	0.44	105
—chlorophyll	0.62	1.0	2.7	106
HICO—ρNO3	0.90	0.3	0	33
—ρC	0.90	0.4	0.43	33
—chlorophyll	0.95	0.3	3.9	33
Seawall 2022—chlorophyll	0.97	0.4	4.4	36
USGS 2022—chlorophyll	0.92	1.3	5.7	187
Heterosigma—ρNO3	0.76	0.2	0.04	35
—ρC	0.80	3.9	0.77	35
—chlorophyll	0.67	0.3	5.0	35
D7 (2014-16)—ρNO3	0.82	3.1	0.02	34
—ρC	0.78	1.9	0.45	32
—chlorophyll	0.90	3.3	6.0	34
D7 (2011, 2012)—integrated ρC	0.95	3.4	42.0	24
Compiled dataset—ρNO3	0.67	3.3	0.02	202
—ρC	0.72	3.4	0.62	206
—chlorophyll	0.70	2.3	6.0	432

References

1. Jassby, A. Phytoplankton in the upper San Francisco Estuary: Recent biomass trends, their causes and their trophic significance. San Franc. Estuary Watershed Sci.; 2008; 6, pp. 1-26.

2. Cloern, J.E. Our evolving conceptual model of the coastal eutrophication problem. Mar. Ecol. Prog. Ser.; 2001; 210, pp. 223-253. [DOI: https://dx.doi.org/10.3354/meps210223]

3. Sharp, J.H. Marine and aquatic communities, stress from eutrophication. Encyclopedia of Biodiversity; Levine, S. Academic Press: Amsterdam, The Netherlands, 2001; Volume 4, pp. 1-11.

4. Ball, M.D.; Arthur, J.F. Planktonic chlorophyll dynamics in the Northern San Francisco Bay and Delta. San Francisco Bay: The Urbanized Estuary; Conomos, T.J. Pacific Division of the American Association for the Advancement of Science: San Francisco, CA, USA, 1979; pp. 265-286.

5. Wilkerson, F.P.; Dugdale, R.C.; Hogue, V.E.; Marchi, A. Phytoplankton blooms and nitrogen productivity in San Francisco Bay. Estuaries Coasts; 2006; 29, pp. 401-416. [DOI: https://dx.doi.org/10.1007/BF02784989]

6. Kimmerer, W.J.; Parker, A.E.; Lidström, U.E.; Carpenter, E.J. Short-term and interannual variability in primary production in the low-salinity zone of the San Francisco Estuary. Estuaries Coasts; 2012; 35, pp. 913-929. [DOI: https://dx.doi.org/10.1007/s12237-012-9482-2]

7. Dugdale, R.; Wilkerson, F.; Parker, A.E.; Marchi, A.; Taberski, K. River flow and ammonium discharge determine spring phytoplankton blooms in an urbanized estuary. Estuar. Coast. Shelf Sci.; 2012; 115, pp. 187-199. [DOI: https://dx.doi.org/10.1016/j.ecss.2012.08.025]

8. Jungbluth, M.; Lee, C.; Patel, C.; Ignoffo, T.; Bergamaschi, B.; Kimmerer, W. Production of the copepod Pseudodiaptomus forbesi is not enhanced by ingestion of the diatom Aulacoseira granulata during a bloom. Estuaries Coasts; 2021; 44, pp. 1083-1099. [DOI: https://dx.doi.org/10.1007/s12237-020-00843-9]

9. Glibert, P.M. Long-term changes in nutrient loading and stoichiometry and their relationships with changes in the food web and dominant pelagic fish species in the San Francisco Estuary, California. Rev. Fish. Sci.; 2010; 18, pp. 211-232. [DOI: https://dx.doi.org/10.1080/10641262.2010.492059]

10. Lehman, P. The increase of cyanobacteria and benthic diatoms over 43 years in upper San Francisco Estuary, California. Estuar. Coast. Shelf Sci.; 2022; 275, 107988. [DOI: https://dx.doi.org/10.1016/j.ecss.2022.107988]

11. Sommer, T.R.; Armor, C.; Baxter, R.; Breuer, R.; Brown, L.; Chotkowski, M.; Culberson, S.; Feyrer, F.; Gingas, M.; Herbold, B. et al. The collapse of pelagic fishes in the upper San Francisco Estuary. Fisheries; 2007; 32, pp. 270-277. [DOI: https://dx.doi.org/10.1577/1548-8446(2007)32[270:TCOPFI]2.0.CO;2]

12. Jassby, A.D.; Cloern, J.E.; Cole, B.E. Annual primary production: Patterns and mechanisms of change in a nutrient-rich tidal ecosystem. Limnol. Oceanogr.; 2002; 47, pp. 698-712. [DOI: https://dx.doi.org/10.4319/lo.2002.47.3.0698]

13. Cloern, J.E. Turbidity as a control on phytoplankton biomass and productivity in estuaries. Cont. Shelf Res.; 1987; 7, pp. 1367-1381. [DOI: https://dx.doi.org/10.1016/0278-4343(87)90042-2]

14. Carlton, J.T.; Thompson, J.K.; Schemel, L.E.; Nichols, F.H. Remarkable invasion of San Francisco Bay (California, USA) by the Asian clam, Potamocorbula amurensis. 1. Invasion and dispersal. Mar. Ecol. Prog. Ser.; 1990; 66, pp. 81-94. [DOI: https://dx.doi.org/10.3354/meps066081]

15. Glibert, P.M.; Fullerton, D.; Burkholder, J.M.; Cornwell, J.; Kana, T.M. Ecological stoichiometry, biogeochemical cycling, invasive species, and aquatic food webs: San Francisco Estuary and comparative systems. Rev. Fish. Sci.; 2011; 19, pp. 358-417. [DOI: https://dx.doi.org/10.1080/10641262.2011.611916]

16. Glibert, P.M.; Wilkerson, F.P.; Dugdale, R.C.; Parker, A.E. Ecosystem recovery in progress? Initial nutrient and phytoplankton response to nitrogen reduction from sewage treatment upgrade in the San Francisco Bay Delta. Nitrogen; 2022; 3, pp. 569-591. [DOI: https://dx.doi.org/10.3390/nitrogen3040037]

17. Yoshiyama, K.; Sharp, J.H. Phytoplankton response to nutrient enrichment in an urbanized estuary: Apparent inhibition of primary production by overeutrophication. Limnol. Oceanogr.; 2006; 51, pp. 424-434. [DOI: https://dx.doi.org/10.4319/lo.2006.51.1_part_2.0424]

18. Waiser, M.J.; Tumber, V.; Holm, J. Effluent-dominated streams. Part 1: Presence and effects of excess nitrogen and phosphorus in Wascana Creek, Saskatchewan, Canada. Environ. Toxicol. Chem.; 2010; 30, pp. 496-507. [DOI: https://dx.doi.org/10.1002/etc.399]

19. MacIsaac, J.J.; Dugdale, R.C.; Huntsman, S.A.; Conway, H.L. The effect of sewage on uptake of inorganic nitrogen and carbon by natural populations of marine phytoplankton. J. Mar. Sci.; 1979; 37, pp. 51-66.

20. Parker, A.E.; Dugdale, R.C.; Wilkerson, F.P. Elevated ammonium concentrations from wastewater discharge depress primary productivity in the Sacramento River and the northern San Francisco Estuary. Mar. Pollut. Bull.; 2012; 64, pp. 574-586. [DOI: https://dx.doi.org/10.1016/j.marpolbul.2011.12.016]

21. Xu, J.; Glibert, P.M.; Liu, H.; Yin, K.; Yuan, X.; Chen, M.; Harrison, P.J. Nitrogen sources and rates of phytoplankton uptake in different regions of Hong Kong waters in summer. Estuaries Coasts; 2012; 35, pp. 559-571. [DOI: https://dx.doi.org/10.1007/s12237-011-9456-9]

22. Dugdale, R.C.; Wilkerson, F.P.; Hogue, V.E.; Marchi, A. The role of ammonium and nitrate in spring bloom development in San Francisco Bay. Estuar. Coast. Shelf Sci.; 2007; 73, pp. 17-29. [DOI: https://dx.doi.org/10.1016/j.ecss.2006.12.008]

23. Parker, A.E.; Hogue, V.E.; Wilkerson, F.P.; Dugdale, R.C. Inorganic nitrogen speciation and phytoplankton growth in the high nutrient, low chlorophyll San Francisco Estuary. Estuar. Coast. Shelf Sci.; 2012; 104–105, pp. 91-101. [DOI: https://dx.doi.org/10.1016/j.ecss.2012.04.001]

24. Wilkerson, F.P.; Dugdale, R.C. The ammonium paradox of a high nutrient low chlorophyll estuary. Aquatic Microbial Ecology and Biogeochemistry: A Dual Perspective; Glibert, P.M.; Kana, T.M. Springer: Berlin/Heidelberg, Germany, 2016; pp. 117-126.

25. Cochlan, W.P.; Harrison, P.J. Inhibition of nitrate uptake by ammonium and urea in the eucaryotic picoflagellate Micromonas pusilla (Butcher) Manton et Parke. J. Exp. Mar. Biol. Ecol.; 1991; 153, pp. 143-152. [DOI: https://dx.doi.org/10.1016/0022-0981(91)90221-H]

26. Wilkerson, F.P.; Dugdale, R.C.; Parker, A.E.; Blaser, S.B.; Pimenta, A. Nutrient uptake and primary productivity in an urban estuary: Using rate measurements to evaluate phytoplankton response to different hydrological and nutrient conditions. Aquat. Ecol.; 2015; 49, pp. 211-233. [DOI: https://dx.doi.org/10.1007/s10452-015-9516-5]

27. Gerendas, J.; Zhu, Z.; Bendixen, R.; Ratcliffe, R.G.; Sattelmacher, B. Physiological and bio- chemical processes related to ammonium toxicity in higher plants. Z. Pflanzenernährung Bodenkd.; 1997; 160, pp. 239-251. [DOI: https://dx.doi.org/10.1002/jpln.19971600218]

28. Britto, D.T.; Kronzucker, H.J. Ecological significance and complexity of N-source preference in plants. Ann. Bot.; 2013; 112, pp. 957-963. [DOI: https://dx.doi.org/10.1093/aob/mct157]

29. Glibert, P.M.; Dugdale, R.C.; Wilkerson, F.; Parker, A.E.; Alexander, J.; Antell, E.; Blaser, S.; Johnson, A.; Lee, J.; Lee, T. et al. Major—But rare—Spring blooms in San Francisco Bay Delta, California, a result of the long-term drought, increased residence times, and altered nutrient loads and forms. J. Exp. Mar. Biol. Ecol.; 2014; 460, pp. 8-18. [DOI: https://dx.doi.org/10.1016/j.jembe.2014.06.001]

30. Cloern, J.E.; Jassby, A.D. Drivers of change in estuarine–coastal ecosystems: Discoveries from four decades of study in San Francisco Bay. Rev. Geophys.; 2012; 50, pp. 1-33. [DOI: https://dx.doi.org/10.1029/2012RG000397]

31. Nixon, S.W. Physical energy inputs and the comparative ecology of lake and marine ecosystems. Limnol. Oceanogr.; 1988; 33, pp. 1005-1025. [DOI: https://dx.doi.org/10.4319/lo.1988.33.4_part_2.1005]

32. Whitledge, T.E.; Malloy, S.C.; Patton, C.J.; Winck, C.D. Automated Nutrient Analyses in Seawater; Department of Energy and Environment, Brookhaven National Laboratory: Upton, NY, USA, 1981.

33. Bran Luebbe, Inc. Bran Luebbe AutoAnalyzer Applications; AutoAnalyzer Method No. G-172-96 Nitrate and Nitrite in Water and Seawater Bran Luebbe, Inc.: Buffalo Grove, IL, USA, 1999.

34. Solorzano, L. Determination of ammonia in natural waters by the phenolhypochlorite method. Limnol. Oceanogr.; 1969; 14, pp. 799-801.

35. Arar, E.; Collins, G.B. In vitro determination of chlorophyll a and phaeophytin in marine and freshwater phytoplankton by fluorescence. USEPA Methods for Determination of Chemical Substances in Marine and Estuarine Environmental Samples; USEPA: Cincinnati, OH, USA, 1992.

36. Holm-Hansen, O.; Lorenzen, C.J.; Holmes, R.W.; Strickland, J.D.H. Fluorometric determination of chlorophyll. J. Cons. Perm. Int. l’Exploration Mer; 1965; 30, pp. 3-5. [DOI: https://dx.doi.org/10.1093/icesjms/30.1.3]

37. Friederich, G.E.; Walz, P.M.; Burczynski, M.G.; Chavez, F.P. Inorganic carbon in the central California upwelling system during the 1997–1999 El Niño-La Niña event. Prog. Oceanogr.; 2002; 54, pp. 185-203. [DOI: https://dx.doi.org/10.1016/S0079-6611(02)00049-6]

38. Parker, A.; Fuller, J.; Dugdale, R.C. Estimating dissolved inorganic carbon concentrations from salinity in San Francisco Bay for use in 14C-primary production studies. Interag. Ecol. Program San Franc. Estuary Newsl.; 2006; 19, pp. 17-22.

39. Slawyk, G.; Collos, Y.; Auclair, J.C. The use of the 13C and 15N isotopes for the simultaneous measurement of carbon and nitrogen turnover rates in marine phytoplankton. Limnol. Oceanogr.; 1977; 22, pp. 925-932. [DOI: https://dx.doi.org/10.4319/lo.1977.22.5.0925]

40. Parker, A.E. Differential supply of autochthonous organic carbon and nitrogen to the microbial loop of the Delaware Estuary. Estuaries; 2005; 286, pp. 856-867. [DOI: https://dx.doi.org/10.1007/BF02696015]

41. Dugdale, R.C.; Wilkerson, F.P. The use of 15N to measure nitrogen uptake in eutrophic oceans: Experimental considerations. Limnol. Oceanogr.; 1986; 31, pp. 673-689. [DOI: https://dx.doi.org/10.4319/lo.1986.31.4.0673]

42. Legendre, L.; Gosselin, M. Estimation of N or C uptake rates by phytoplankton using 15N or 13C: Revisiting the usual computation formulae. J. Plankton Res.; 1996; 19, pp. 263-271. [DOI: https://dx.doi.org/10.1093/plankt/19.2.263]

43. Houde, E.D.; Rutherford, E.S. Recent trends in estuarine fisheries: Predictions of fish production and yield. Estuaries; 1993; 16, pp. 161-176. [DOI: https://dx.doi.org/10.2307/1352488]

44. Downing, J.A.; Plante, C.; Lalonde, S. Fish production correlated with primary productivity, not the morphoedaphic index. Can. J. Fish. Aquat. Sci.; 1990; 47, pp. 1929-1936. [DOI: https://dx.doi.org/10.1139/f90-217]

45. Iverson, R.L. Control of marine fish production. Limnol. Oceanogr.; 1990; 35, pp. 1593-1604. [DOI: https://dx.doi.org/10.4319/lo.1990.35.7.1593]

46. Sommer, T.; Mejia, F. A place to call home: A synthesis of Delta Smelt habitat in the upper San Francisco Estuary. San Franc. Estuary Watershed Sci.; 2013; 11, pp. 1-25. [DOI: https://dx.doi.org/10.15447/sfews.2013v11iss2art4]

47. Wheeler, P.A.; Kokkinakis, S.A. Ammonium recycling limits nitrate use in the oceanic subarctic Pacific. Limnol. Oceanogr.; 1990; 35, pp. 1267-1278. [DOI: https://dx.doi.org/10.4319/lo.1990.35.6.1267]

48. Varela, D.E.; Harrison, P.J. Effect of ammonium on nitrate utilization by Emiliania huxleyi, a coccolithophore from the oceanic northeastern Pacific. Mar. Ecol. Prog. Ser.; 1999; 186, pp. 67-74. [DOI: https://dx.doi.org/10.3354/meps186067]

49. Glibert, P.M.; Wilkerson, F.P.; Dugdale, R.C.; Raven, J.A.; Dupont, C.L.; Leavitt, P.R.; Parker, A.E.; Burkholder, J.M.; Kana, T.M. Pluses and minuses of ammonium and nitrate uptake and assimilation by phytoplankton and implications for productivity and community composition, with emphasis on nitrogen-enriched conditions. Limnol. Oceanogr.; 2016; 61, pp. 165-197. [DOI: https://dx.doi.org/10.1002/lno.10203]

50. Cloern, J.E.; Schraga, T.S.; Nejad, E.; Martin, C. Nutrient status of San Francisco Bay and its management implications. Estuaries Coasts; 2020; 43, pp. 1299-1317. [DOI: https://dx.doi.org/10.1007/s12237-020-00737-w]

51. Sutula, M.; Kudela, R.K.; Hagy, J.D., III; Harding, L.W., Jr.; Senn, D.; Cloern, J.E.; Bricker, S.; Berg, G.M. Novel analyses of long-term data provide a scientific basis for chlorophyll-a thresholds in San Francisco Bay. Estuar. Coast. Shelf Sci.; 2017; 197, pp. 107-118. [DOI: https://dx.doi.org/10.1016/j.ecss.2017.07.009]

52. Cloern, J.E.; Alpine, A.E.; Cole, B.E.; Wong, R.L.; Arthur, J.F.; Ball, M.D. River discharge controls phytoplankton dynamics in the northern San Francisco Bay estuary. Estuar. Coast. Shelf Sci.; 1983; 16, pp. 415-429. [DOI: https://dx.doi.org/10.1016/0272-7714(83)90103-8]

53. Lee, J.; Parker, A.E.; Wilkerson, F.P.; Dugdale, R.C. Uptake and inhibition kinetics of nitrogen in Microcystis aeruginosa: Results from cultures and field assemblages collected in the San Francisco Bay Delta, CA. Harmful Algae; 2015; 47, pp. 126-140. [DOI: https://dx.doi.org/10.1016/j.hal.2015.06.002]

54. Herndon, J.; Cochlan, W.P. Nitrogen utilization by the raphidophyte Heterosigma akashiwo: Growth and uptake kinetics in laboratory cultures. Harmful Algae; 2007; 6, pp. 260-270. [DOI: https://dx.doi.org/10.1016/j.hal.2006.08.006]

55. Pennock, J.R. Temporal and spatial variability in phytoplankton ammonium and nitrate uptake in the Delaware Estuary. Estuar. Coast. Shelf Sci.; 1987; 24, pp. 841-857. [DOI: https://dx.doi.org/10.1016/0272-7714(87)90156-9]

56. Richardson, E.T.; Hansen, A.M.; Kraus, T.E.; Downing, B.D.; Forsberg, D.; Stillian, J.; O’Donnell, K.; Sturgeon, C.L.; Bergamaschi, B.A. A novel boat-based field application of a high-frequency conductometric ammonium analyzer to characterize spatial variation in aquatic ecosystems. Limnol. Oceanogr. Methods; 2023; 21, pp. 761-774. [DOI: https://dx.doi.org/10.1002/lom3.10579]