Coupling Between the Coastal Ocean and Yaquina Bay, Oregon: Importance of Oceanic Inputs Relative to Other Nitrogen Sources
- First Online:
- Cite this article as:
- Brown, C.A. & Ozretich, R.J. Estuaries and Coasts (2009) 32: 219. doi:10.1007/s12237-008-9128-6
- 925 Downloads
Understanding of the role of oceanic input in nutrient loadings is important for understanding nutrient and phytoplankton dynamics in estuaries adjacent to coastal upwelling regions as well as determining the natural background conditions. We examined the nitrogen sources to Yaquina Estuary (Oregon, USA) as well as the relationships between physical forcing and gross oceanic input of nutrients and phytoplankton. The ocean is the dominant source of dissolved inorganic nitrogen (DIN) and phosphate to the lower portion of Yaquina Bay during the dry season (May through October). During this time interval, high levels of dissolved inorganic nitrogen (primarily in the form of nitrate) and phosphate entering the estuary lag upwelling favorable winds by 2 days. The nitrate and phosphate levels entering the bay associated with coastal upwelling are correlated with the wind stress integrated over times scales of 4–6 days. In addition, there is a significant import of chlorophyll a to the bay from the coastal ocean region, particularly during July and August. Variations in flood-tide chlorophyll a lag upwelling favorable winds by 6 days, suggesting that it takes this amount of time for phytoplankton to utilize the recently upwelled nitrogen and be transported across the shelf into the estuary. Variations in water properties determined by ocean conditions propagate approximately 11–13 km into the estuary. Comparison of nitrogen sources to Yaquina Bay shows that the ocean is the dominant source during the dry season (May to October) and the river is the dominant source during the wet season with watershed nitrogen inputs primarily associated with nitrogen fixation on forest lands.
KeywordsOcean input Upwelling Nutrient sources Yaquina estuary
In most estuaries, the major sources of nitrogen are atmospheric deposition, agricultural nitrogen fixation, fertilizer runoff, and in heavily populated areas point source inputs associated with wastewater treatment facilities (Boyer et al. 2002; Howarth et al. 2002; Driscoll et al. 2003). For many estuaries in the Pacific northwest (PNW) of the United States, there are relatively low population densities in the watersheds and low atmospheric deposition rates. Land use in the watersheds is predominantly forested, resulting in low nitrogen (N) inputs associated with fertilizer and agriculture N fixation. In addition, upwelling provides nutrients to estuaries adjacent to coastal upwelling regions, such as the PNW (e.g., Hickey and Banas 2003). The differences in land use combined with coastal upwelling may result in differences in the dominant N sources to PNW estuaries compared to other regions.
In a recent review, Tappin (2002) found that the N input to temperate and tropical estuaries associated with the ocean is poorly quantified. Previous studies have demonstrated that the oceanic inputs of nutrients and phytoplankton are important for estuaries adjacent to coastal upwelling regions, such as the west coast of the United States (e.g., de Angelis and Gordon 1985; Roegner and Shanks 2001; Roegner et al. 2002; Colbert and McManus 2003). It is important to quantify the contribution of oceanic input to nutrient loading in order to determine reference conditions for estuaries adjacent to upwelling regions and to distinguish natural variability from anthropogenic inputs. In addition, we do not know how susceptible estuaries subjected to large oceanic inputs of nutrients (dissolved inorganic nitrogen and phosphorous) are to future changes in anthropogenic inputs of nutrients. Some studies have suggested that future climate change may lead to changes in the seasonality or intensity of wind-driven upwelling (Snyder et al. 2003), which could modify the nutrient loading to these systems; therefore, it is important to quantify the oceanic input of nutrients to establish a baseline.
Previous studies of the importance of oceanic variability in estuarine water properties consisted of short-term observations of nutrients or chlorophyll a (de Angelis and Gordon 1985; Roegner and Shanks 2001) or examined water temperature or salinity fluctuations (Hickey et al. 2002; Hickey and Banas 2003). De Angelis and Gordon (1985) demonstrated that there was import of oceanic dissolved inorganic nitrogen to Alsea Bay, Oregon; however, they only sampled on six dates during the summer of 1979 and only two of those dates had significant oceanic import of nitrate (NO3−). Roegner and Shanks (2001) demonstrated chlorophyll a was imported into an Oregon estuary from the coastal ocean; however, they had insufficient temporal resolution in their data to examine the coupling between wind stress and chlorophyll a. Hickey et al. (2002) demonstrated that water property fluctuations near the mouth of Willapa Bay, WA, USA, are related to alongshore wind stress and propagate up the estuary, but their study focused on temperature, salinity, and current velocities.
High temporal resolution data are required to demonstrate the coupling between wind stress and water column properties (nutrients and chlorophyll a) and to quantify the gross oceanic loading. In this paper, we quantify the gross oceanic input of nutrients and chlorophyll a using data collected daily during the upwelling season (May–September) for two consecutive years. In addition, we examine the coupling between wind forcing and nutrient and phytoplankton levels entering the estuary and the propagation of these signals into the estuary. We also compare the major N inputs to the estuary (including gross oceanic, riverine, wastewater treatment facility effluent, benthic flux, and atmospheric inputs) to assess the importance of oceanic inputs relative to other sources.
Materials and Methods
During May through October of 2002 and 2003, daily water samples were collected during flood tides approximately 0.5 m below the surface at the Oregon State University Dock (labeled OSU in Fig. 1), which is located inside the bay 4 km from the seaward end of the jetties. The samples were immediately filtered and frozen for storage until analysis. Dissolved inorganic nutrients (NO3− + NO2−, NH4+, PO4−3, and H4SiO4) were analyzed by MSI Analytical Laboratory, University of California-Santa Barbara, CA using Lachat flow injection instrumentation (Zellweger Analytics, Milwaukee WI, USA). One-liter surface water samples were collected daily and analyzed for chlorophyll a. These samples were filtered within 15 min using 47-mm diameter GF/F filters. Chlorophyll a was extracted by sonicating the filters and soaking them overnight in 10 ml of 90% acetone. The next morning the samples were centrifuged and analyzed for chlorophyll a content using a fluorometer (10 AU Fluorometer, Turner Designs, Inc, Sunnyvale, CA, USA). Beginning on July 23, 2002, an in situ fluorometer (SCUFA, Turner Designs, Inc., Sunnyvale, CA, USA) was deployed at the OSU dock providing in situ fluorescence. Commencing on August 28th 2002, an automated water sampler (ISCO®, Model 3700FR, Lincoln, NE, USA) was used to collect water samples for each flood tide and programmed using the predicted time of each high tide. The sampler held the samples in a dark refrigerated compartment and the samples were collected daily, filtered, and frozen for nutrient analyses.
Water temperature, tide height, wind speed, and direction were obtained from South Beach Station (Station: 9435380, latitude 44.625° North, longitude 124.043° West, location presented in Fig. 1) operated by the Center for Operational Oceanographic Products of the National Ocean Service (http://co-ops.nos.noaa.gov). Flood tide water temperatures were extracted from the hourly data using times of predicted high tides. Data from South Beach Station were used because it had minimal data gaps. Water temperature data from the South Beach Station were compared to two YSI, Inc. (Yellow Springs, OH, USA) Multiparameter Monitoring Systems located at Station OSU (Fig. 1), which were deployed near the surface (about 1 m below water surface) and a second at an average depth of 2 m below water surface, as well as data from the bay sampling (see Station 1 in next section). During 1999–2002, there was close agreement between the time series of water temperature at South Beach and other data sources; however, during 2003, the South Beach water temperature was approximately 1°C colder than the OSU Station, so we adjusted (added 1°C) the South Beach time series during this year. Water temperature and salinity were available at four other locations from YSI datasondes (Specht, unpublished data) that were deployed at Riverbend (Station A), Oregon Oyster (Station B), Cragie Point (Station C), and Criteser’s Landing (Station D) (Fig. 1). The datasondes at Riverbend and Cragie Point (Stations A and C) were deployed at an average depth of about 4 m and 2 m below the water surface, respectively, while the datasondes at Oregon Oyster and Criteser’s Landing (Stations B and D) were deployed at a depth of about 1 m below the water surface. All variables were logged every 15 min.
Twelve locations in the estuary were sampled at approximately weekly intervals for dissolved inorganic nutrients at mid-depth and 0.5 m above the bottom (locations shown as Stations 1–12 in Fig. 1). Water samples were collected from depth using a hand-operated pump, filtered (45 μm filter) and frozen until analysis. The samples were analyzed for NO3− + NO2−, NH4+, H4SiO4, and PO4−3. At each station profiles of conductivity, temperature and depth (CTD; SBE 19 SEACAT Profiler, Sea-Bird Electronics, Inc, Bellevue, WA, USA) and in situ fluorescence (WETStar Chlorophyll Fluorometer, WET Labs, Philomath, OR, USA) were measured. The profile measurements were taken at 0.5-s intervals from the water surface to 0.5 m above the bottom and during post-processing the data were binned into 0.25-m intervals. The fluorometer was calibrated by collecting water samples quarterly, filtering them, and analyzing them for chlorophyll a using the same technique used for the oceanic input samples, and developing a relationship between in situ fluorescence and extracted chlorophyll a values (Chlorophyll a (μg l−1) = 0.52 × in situ fluorescence − 1.75, r2 = 0.95, n = 36). In this equation, in situ fluorescence refers to the factory calibration estimate of chlorophyll a. These cruises were conducted during flood tides, tracking the propagation of the tide up the estuary, and were completed in about 3 h. Time series of salinity were examined at Stations A, B, C, and D to confirm that the cruises were tracking the propagation of the tide.
To examine the relationship between shelf upwelling dynamics and nutrient and phytoplankton entering the bay, we performed a cross-correlation analysis between average daily north–south wind stress (at Stations NWP03 and 46050), flood tide water temperature, dissolved inorganic nutrients, chlorophyll a, and in situ fluorescence (at Station OSU) entering the bay. In situ fluorescence data were low-pass filtered using a 3-h Lanczos filter and flood tide values were extracted using times of predicted high tides. Cross-correlation coefficients were calculated using the non-parametric Spearman rank-order correlation coefficient using SigmaStat (version 3.10 Systat Software Inc., Point Richmond, CA, USA). An adjusted sample size (N*) based on the modified Chelton method (Pyper and Peterman 1998) was used in the correlation analysis to adjust for the effect of autocorrelation in the time-series on significance levels. Gaps in the time series were filled with linear interpolation (time step of 1 day), since the modified Chelton method requires no gaps in the time series. These interpolated time series were used only to determine the adjusted sample size for use in determining the significance levels, but were not used in the calculation of the correlation coefficients.
To examine how far into the bay the nutrient and chlorophyll a temporal variability is determined by ocean conditions, we examined the correlation between the NO3− + NO2−, PO4−3, and chlorophyll a concentrations found at Station 1 near the mouth of the estuary (Fig. 1) and those stations further in (Stations 2–12) during the period of May through August. During this period, the mean absolute difference between mid-depth and bottom samples for NO3− + NO2−, NH4+, and PO4−3 was 0.5, 0.3, and 0.1 μM (n = 202), respectively; therefore, we averaged mid-depth and bottom samples for this analysis.
Other Data Used for Comparison of Nutrient Input
The riverine contribution to N inputs to Yaquina Bay was calculated using observations of streamflow and stream nutrient concentrations. The Yaquina River has been gauged by the US Geological Survey and the State of Oregon Water Resources Department at a station near Chitwood, Oregon (USGS Station 1430600), which is 51 km upstream from the mouth of Yaquina Bay. We compared the N sources to Yaquina Bay during the wet and dry seasons. The wet season (November–April) was defined as months when the monthly average discharge of the Yaquina River at Chitwood (computed using data from 1972 to 2002) exceeded the 30-year average discharge of 7.2 m3 s−1, while the dry season (May–October) was defined as months when the monthly average discharge was less than the long-term average.
The wastewater treatment facility input of DIN to the estuary was computed by multiplying the daily volume discharged by the effluent concentration. Data on daily volume discharge and concentration were provided by the City of Toledo, Oregon Wastewater Treatment Facility for 2002–2004. From the analysis of split samples by UCSB the treatment facility NO3− concentrations were found to be biased high and were adjusted (multiplied by 0.78) prior to computing loading. From December 2000 through June 2002 approximately 75% of the discharged nitrogen was NO3−; 15% and 10% as NH4+ and organic N, respectively.
Atmospheric N deposition is monitored by the National Atmospheric Deposition Program (NADP) at a station 40 km away from Yaquina Estuary (Alsea Guard Ranger Station, OR02). The annual atmospheric input at this site averaged over the interval of 1980 to 2002 was used to estimate the atmospheric N input to the estuary (NADP 2003). The atmospheric N input on the watershed was calculated as the product of deposition rate (kg N ha−1 year−1) and watershed area, and the direct input to the estuary was calculated as the product of deposition rate and the estuary area.
The wet season gross oceanic DIN input was estimated using the average wet season (from November 1997–April 2003) surface DIN at an innershelf station off Newport, OR (Wetz et al. 2005) and modeled amount of water entering each flood tide during the wet season of 2002. The wet season average DIN on the inner shelf was 3.3 μM (n = 17) and the average salinity was 32.1 psu. Wet season mixing diagrams (from 1998–2003) from the Yaquina Estuary were used to confirm the wet season average oceanic DIN. Since mixing diagrams generated from estuary data were often influenced by freshwater inflow the mixing diagrams were extrapolated to salinity of 32.1 psu to estimate the oceanic DIN. The average wet season DIN from the innershelf was consistent with extrapolation of wet season mixing diagrams estimate (average = 4.0 μM, n = 32). To estimate the importance of benthic flux on DIN concentrations within the bay, we used published values from Yaquina Bay (De Witt et al. 2004).
Results and Discussion
Flood Tide Input from Ocean
- 1.Flood tide sampling
Dissolved inorganic nutrients
Dissolved inorganic nutrients and chlorophyll a entering Yaquina Bay during flood tides (grab samples from OSU) during May–October of 2002 and 2003 (Mean ± SD)
Mean NO3− + NO2− (μM)
Mean NH4+ (μM)
Mean PO4−3 (μM)
Median N:P ratio
Mean chlorophyll a (μg l−1)
12.8 ± 7.6 (n = 179)
3.8 ± 1.8 (n = 179)
1.7 ± 0.7 (n = 179)
9.7 (n = 179)
6.4 ± 5.6 (n = 120)
10.3 ± 9.4 (n = 284)
3.4 ± 1.6 (n = 284)
1.2 ± 0.7 (n = 284)
11.5 (n = 284)
4.6 ± 2.5 (n = 55)
Potential for nutrient limitation of phytoplankton is often estimated by examining the ratio of dissolved inorganic nutrients relative to the Redfield ratio (16 mol N:1 mol P) and comparing the ambient dissolved inorganic nutrient concentrations to phytoplankton half saturation constants for nutrient uptake (e.g., Eyre 2000). Typically, if the N:P ratio of the water column <10:1 then phytoplankton may be limited by nitrogen and if the ratio >20:1, there is the potential for phosphorous limitation (Boynton et al. 1982). In addition, if the ambient water column concentrations are less than the half saturation constants for nutrient uptake then we assume that the phytoplankton may be nutrient limited. Typical half saturation constants for DIN and DIP are 1.0–2.0 μM and 0.1–0.5 μM, respectively.
Relationship between wind forcing, water temperature, nutrients and chlorophyll a
During the dry seasons of 2002 and 2003, upwelling favorable winds occurred 70% of the time (calculated using daily average wind stress). Even though upwelling favorable winds occurred with similar frequency in 2002 and 2003, there was a difference in the character of the upwelling events. During 2002, upwelling favorable winds were sustained for long time periods (particularly during June through October), while during 2003 upwelling occurred as discrete events. During May and June of 2002, upwelling favorable winds occurred 60% of the time and these periods of upwelling were interrupted by brief periods of downwelling favorable winds. From the end of June through October of 2002, upwelling favorable winds dominated (frequency of occurrence = 74%) with a mean north–south wind stress of 0.26 dyne cm−2, and the NO3− + NO2− entering in flood waters remained elevated. During the interval of May to June 30 of 2002 the mean NO3− + NO2− was 10.6 μM, while during July through October of 2002 the mean NO3− + NO2− was 15.9 μM. During 2003, there were six discrete upwelling events (shown as shaded regions in Fig. 5) that resulted in increases in NO3− + NO2− entering the bay with each event lasting 2 to 3 weeks and peak levels during these events reaching as high as 30 μM. During 2003, the first upwelling event that caused an increase in NO3− + NO2− occurred on May 27th. Between each upwelling event, there were brief periods of downwelling favorable winds or relaxation events, which lasted 1 to 2 weeks, and the NO3− + NO2− levels near the end of these events were as low as 0.3 μM. During some of the upwelling events, there were brief periods (~1 d) of downwelling favorable winds that resulted in brief decreases in NO3− + NO2− (such as that occurring on June 29, of 2003, Fig. 5a, b). For both years, there was a close correspondence between reversals in low-pass filtered north–south wind stress and changes in the flood-tide NO3− + NO2− and PO4−3 levels.
Maximum correlation between average daily north–south wind stress (computed using data from Station NWP03 with no decay coefficient), water temperature, dissolved inorganic nutrients, and chlorophyll a (grab samples and in situ fluorometer)
Time series compared
Sample size (n)
Effective sample size (N*)
Wind stress and flood tide water temperature
Wind stress and flood tide NO3− + NO2−
Wind stress and flood tide PO4−3
Flood tide water temperature and flood tide NO3− + NO2−
Flood tide water temperature and flood tide PO4−3
Wind stress and flood tide chlorophyll a
Wind stress and flood tide chlorophyll a from in situ fluorometer
Flood tide water temperature and flood tide chlorophyll a
Flood tide NO3− + NO2− and flood tide chlorophyll a
Our findings of the close coupling between alongshelf wind stress and water temperature, nutrient, and chlorophyll a and the lags between forcing and response are similar to previous studies. Roegner and Shanks (2001) found similar correlation and lag between wind stress and coastal and estuarine water temperature (r = 0.6, lag 0.5–1.5 days) at Coos Bay, OR, which is located 150 km south of Yaquina Bay. Takesue and van Geen (2002) found that there was an approximately 1.5 days lag between upwelling favorable wind stress and the appearance of nearshore upwelling conditions along the Oregon coast near Coos Bay. They found that the composition of nearshore water responds to local changes in wind forcing, which is similar to our analyses which found stronger correlations between water properties and nearshore wind forcing than offshore wind forcing. Hickey et al. (2002) found similar correlation and lag (r = −0.6, lag 1.25–1.5 days) between wind stress and water temperature and salinity fluctuations near the mouth of Willapa Bay, WA, USA, which is 233 km north of Yaquina Bay. Hickey and Banas (2003) examined variations in temperature, salinity and alongshore winds stress for three estuaries along the Oregon and Washington coasts, spanning 400 km. They demonstrated that there was coherence between estuarine water properties fluctuations (temperature and salinity) among these estuaries during the summer resulting from the large scale patterns in alongshelf wind forcing. However, none of these studies assessed the relationship between wind forcing and chlorophyll a or nutrients. Service et al. (1998) found that off of Monterey Bay, CA, USA wind stress and water temperature were maximally correlated at a lag of 2–3 days and there was a correlation between fluorescence and water temperature and wind stress at lags of 4 days and 6–7 days, respectively, which is similar to our results. Thomas and Strub (2001) performed a cross-correlation analysis between wind forcing (longshore wind stress and wind mixing) and cross-shelf pigment variability. They found that on the shelf off of Washington and northern Oregon (including our study area) the pigment pattern metrics were poorly related to local alongshore winds. However, this is probably due to the temporal resolution of their pigment data being too coarse (10 days) to resolve the relationship between nearshore chlorophyll a and wind stress.
Within the Estuary Patterns During 2002
Data from the cruises were used to examine spatial patterns in nutrients and chlorophyll a within the estuary. A shift in the location of maximum NO3− + NO2− concentrations in the estuary occurred during the transition from spring to summer. During 2002, from January through early June, the maximum NO3− + NO2− occurred at Station 12 suggesting a riverine source for this constituent (with a mean salinity of 4.9 at Station 12 and average riverflow during this time period of 22.6 m3 s−1). From January through mid April of 2002, the NO3− + NO2− for the ocean boundary averaged 5 μM (n = 11), while at Station 12 it averaged 69 μM (n = 10) with peak concentrations of 97 μM. Mixing diagrams of DIN versus salinity (not presented in this paper) revealed conservative transport of DIN during the winter. During late April of 2002, upwelling favorable wind stress resulted in the ocean boundary NO3− + NO2− increasing to about 25 μM.
The median N:P ratio from May through August of 2002 was approximately 13:1, suggesting that nitrogen will be depleted prior to phosphorous for the majority of the estuary. In late April to early May of 2002, there was the potential for phosphorous limitation in the upper portions of the estuary (Stations 11 and 12) with the N:P ratio reaching as high as 176:1. During May through August of 2002, the median DIN concentration was 15 μM, and 92% of the time the DIN was >2 μM (typical half saturation constant for phytoplankton). In only 6% of the estuarine sampling events for the dry season of 2002 was the N:P ratio <10 and DIN <2 μM, and all of these events occurred in late May to early June. In only 7% of the estuarine sampling events for the dry season of 2002 was the N:P ratio >20 and DIP <0.5 μM, suggesting the potential for phosphorous limitation in the upper portions of the estuary (Stations 11 and 12). This suggests that although the N:P ratio often falls below 16:1, the estuary was not usually limited by either nitrogen or phosphorous. This is supported by assimilation ratio data (primary production–chlorophyll a) of Johnson (1981) collected during the dry season near Station 10 (Fig. 1) which found that 77% of the time there were sufficient nutrients for planktonic primary production, 15% of the time there was borderline nutrient deficiency, and only 8% of the time was there evidence of nutrient depletion.
The more rapid decline in the oceanic signal in chlorophyll a compared to nutrients was probably the result of benthic grazing on oceanic phytoplankton. Oyster aquaculture is present in Yaquina Bay in the vicinity of Stations 7–9 and in the lower estuary there are tidal flats that have high densities of burrowing shrimp (DeWitt et al. 2004). Griffen et al. (2004) estimated that the daily filtration rate and density of one species of burrowing shrimp present in Yaquina Bay was sufficient to clear the entire water column of Yaquina Bay on a daily basis.
Comparison of Nitrogen Inputs
Comparison of nitrogen sources during wet and dry seasons for Yaquina Bay, Oregon
Wet season, mol DIN day−1
Dry season, mol DIN day−1
Annual average, mol DIN day−1
2.6 × 105 (±6%)
2.3 × 104 (±6%)
1.6 × 105 (±6%)
8.8 × 104 (±20%)
5.1 × 105 (±4%)
3.0 × 105
3.8 × 105 (±5%)
2.3 × 105
1.8 × 103 (±2%)
1.5 × 103 (±1%)
1.6 × 103 (±1%)
4.3 × 104
2.2 × 102
1.2 × 102
1.7 × 102
1.1 × 104
6.0 × 103
8.5 × 103
There is a ninefold difference in the average daily wet season (13.1 m3 s−1) and dry season (1.5 m3 s−1) riverine discharge at Chitwood. Riverine DIN levels are related to the discharge with wet and dry season DIN levels averaging 99 μM (n = 44) and 40 μM (n = 43), respectively (calculated using observations from Chitwood from 1979–2005). There is an order of magnitude difference in average daily riverine N input to Yaquina Bay during the wet (2.6 × 105 mol N day−1) and dry seasons (2.3 × 104 mol N day−1). In addition, there are considerable interannual differences in riverine N input with wet season riverine input varying from 9.4 × 104 mol N day−1 to 4.7 × 105 mol N day−1 and dry season riverine input ranging from 5.6 × 103 mol N day−1 to 6.7 × 104 mol N day−1 during the interval of 1980 to 2004. During the wet season, riverine input is the largest source of DIN to the estuary, composing approximately 74% of the input, and 92% of the annual riverine N input is delivered during the wet season. Our estimates of riverine N loading are similar to previous published values (Quinn et al. 1991; Sigleo and Frick 2007).
Wastewater treatment facility input
The model estimated volume of water entering Yaquina Bay during each flood tide ranges from about 1.2 × 104 m3 to 2.3 × 107 m3 due to the mixed semidiurnal tides with mean flood tide volume of 1.4 × 107 m3, which compares well to the estimated tidal prism of Shirzad et al. (1989). The volume of oceanic water entering the estuary per day averages 2.71 × 107 m3 day−1.
The gross oceanic input of DIN entering the bay was estimated using Eq. 4 and flood tide samples from OSU during the dry season of 2002 and 2003. During the dry season of 2002, the amount of DIN entering the bay from the ocean during each flood tide varied from 1.3 × 104 mol N to 9.1 × 105 mol N with a mean value of 2.6 × 105 mol N, and the mean daily flood tide input of DIN was 5.1 × 105 mol N day−1. During the 2003 dry season, the mean oceanic input of DIN was 3.8 × 105 mol N day−1 or 25% less than 2002 dry season.
We also calculated the oceanic input of DIN during 2002 and 2003 dry seasons using the modeled water temperature versus NO3− + NO2− relationship (Eq. 5). The oceanic input of DIN estimated using Eqs. 4 and 5 (calculated for each flood tide which was sampled) is 4% higher and 4% lower than estimates calculated using flood tide samples from 2002 and 2003, respectively. This suggests that the error in using Eqs. 4 and 5 to estimate DIN loading is about ±5%. The RMSE in modeled (using Eq. 5) flood tide NO3− + NO2− in 2002 and 2003 was 5.8 μM and 5.0 μM, respectively. Sigleo et al. (2005) calculated the flood tide input of NO3− to Yaquina Bay during August of 2000 to be 13 × 105 mol N day−1, which is about triple our estimate. However, these ocean input numbers were calculated using a constant flood tide NO3− of 30 μM.
Average flood tide water temperature, °C
Average flood tide NO3− + NO2−, μM
Modeled oceanic DIN input, mol N day−1
2.0 × 105
2.7 × 105
3.3 × 105
3.1 × 105
3.9 × 105
4.8 × 105
3.6 × 105
3.3 × 105
Sources of uncertainty in nitrogen loading estimates
Comparison to nitrogen sources for other systems
Population density and forest, agricultural and urban land use in outercoast estuaries in Pacific northwest and northeastern United States
Population density, ind. km−2
Land use (% watershed area)
Coquille River Estuary
Average for Pacific northwest
Hudson River Estuary
Long Island Sound
Average for northeastern US
Comparison of N inputs in Yaquina watershed to average for northeastern catchments (including catchments for Chesapeake Bay, Delaware Bay, Hudson River estuary, Long Island Sound, and Narragansett Bay)
Atmospheric deposition kg N km−2 year−1
N fertilizer usage kg N km−2 year−1
N fixation in forest lands kg N km−2 year−1
Streamflow N export kg N km−2 year−1
The close coupling between oceanic conditions and water column constituents in Yaquina Bay during the dry season is consistent with the high degree of tidal flushing of the estuary (i.e., large tidal prism relative to volume of the estuary and low river inflow). Our results are similar to a study of Boston Harbor (Kelly 1998) that demonstrated that oceanic loading can be a major source of nutrients to coastal embayments. In Yaquina Bay, approximately 60% of the estuary is located in the region where oceanic nutrient inputs dominate.
We found that there was a close coupling between local alongshelf wind stress and flood tide water temperature, NO3−, PO4−3, and chlorophyll a. The maximum cross-correlation between north–south wind stress and flood tide water temperature, NO3−, and PO4−3 occurred at a lag of 2 days (r = 0.5). The maximum correlation between wind stress and chlorophyll a occurred at a lag of 6 days. Numerous other studies have found a close coupling between alongshelf wind stress and coastal and estuarine water properties along the Washington and Oregon coasts (e.g., Service et al. 1998; Roegner and Shanks 2001; Takesue and van Geen 2002; Hickey et al. 2002; Hickey and Banas 2003), which suggests that the results from this study may be extended to other estuaries in this region.
There is considerable interannual variation in oceanic input of nutrients. In determining reference nutrient conditions for estuaries receiving nutrient inputs from coastal upwelling it is important to quantify this interannual variation in oceanic inputs. Measuring flood tide water temperature may be an inexpensive surrogate for estimating this interannual variability. In addition, the strong relationship between integrated alongshore wind stress (Wk) and flood tide nutrients may provide a means to estimate the ocean conditions during the dry season between sampling dates. Further, the seasonal shift in dominant nutrient sources to the estuaries may require establishing nutrient conditions for the wet and dry seasons.
Since all of the bay sampling was conducted during flood tides, we do not have adequate data to compute the N export from the estuary and the net N influx through the tidal inlet. The importance of oceanic input of nutrients to primary production rates within the estuary is dependent upon whether the primary producers are benthic or planktonic. We would expect that most of the oceanic input of nutrients would be exported on the subsequent tidal cycle with little utilization by phytoplankton since the transport time scales are short relative to phytoplankton uptake rates. However, in Yaquina Bay there are intertidal flats which contain benthic primary producers (seagrass, macroalgae, and microalgal mats). These benthic primary producers are inundated with oceanic nutrients twice daily during the dry season. Since these primary producers are located in the intertidal zone they are primarily exposed to flooding ocean water and consequently the gross ocean input may better represent the loading these habitats are exposed to than the net tidally averaged loading. Previous research in Boston Harbor (Kelly 1998) demonstrated that it is important to characterize gross ocean input, not just net ocean input. We suggest this is particularly true for estuaries adjacent to coastal upwelling regions, particularly those with extensive intertidal habitats, such as estuaries in the PNW.
During the dry season, there are seasonal macroalgal blooms on the intertidal flats in the ocean dominated section of the Yaquina Bay (Kentula and DeWitt 2003). The presence of macroalgal blooms is often used as an indicator of anthropogenic eutrophication (e.g., Bricker et al. 2003). However, in Yaquina Bay macroalgal blooms in the lower estuary in the dry season may not be an indicator of cultural eutrophication due to the dominance of oceanic input of nutrients in this area. Natural abundance stable isotope data of the macroalgae in the lower estuary suggests that they are receiving nitrogen from primarily oceanic sources (unpublished data). As the next step in this research, we are using a coupled hydrodynamic and water quality model to examine how much utilization there is of oceanic versus riverine nutrients within different portions of the estuary, the importance of in situ production versus oceanic import on chlorophyll a patterns with in the bay, and the importance of benthic primary producers and grazers on water column properties.
Laura Schumacher and Chris Eide assisted with the sampling and sample analyses. Personnel from Dynamac Corporation conducted field sampling for CTD cruises and provide ISCO sampler support. We would like to acknowledge Pat Wheeler (Oregon State University) and William Peterson (National Marine Fisheries Service) for providing nutrient and water temperature data from the Oregon shelf, Lloyd Van Gordon (Oregon Water Resources Department) for providing river discharge data, and Gary Utiger of the Toledo Wastewater Treatment Facility for providing effluent data. Pat Clinton (EPA) provided GIS support and David Specht (EPA) provided YSI datasonde data. The information in this document has been funded wholly by the USEPA. It has been subjected to review by the National Health and Environmental Effects Research Laboratory’s Western Ecology Division and approved for publication. Approval does not signify that the contents reflect the views of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.