Visualization of Productivity Zones Based on Nitrogen Mass Balance Model in Narragansett Bay, Rhode Island

Jongsun Kim; Myung Hwangbo; Patricia S. Thibodeau; Georgia Rhodes; Emma Hogarth; Stewart Copeland

doi:10.3791/65728

JoVE Journal > Environment

Umwelt

Visualization of Productivity Zones Based on Nitrogen Mass Balance Model in Narragansett Bay, Rhode Island

Published: July 14, 2023

doi:

10.3791/65728

Jongsun Kim¹, Myung Hwangbo¹, Patricia S. Thibodeau², Georgia Rhodes³, Emma Hogarth⁴, Stewart Copeland^3,4

¹School of Earth, Environmental and Marine Sciences,University of Texas – Rio Grande Valley, ²Graduate School of Oceanography,University of Rhode Island, ³Rhode Island School of Design, ⁴Department of Art,University of New Mexico

Summary

Here, we aim to visualize the zonation of biological productivity in Narragansett Bay, Rhode Island, based on the nitrogen mass balance model. The results will inform nutrient management in the coastal regions to reduce hypoxia and eutrophication.

Abstract

Primary productivity in the coastal regions, linked to eutrophication and hypoxia, provides a critical understanding of ecosystem function. Although primary productivity largely depends on riverine nutrient inputs, estimation of the extent of riverine nutrient influences in the coastal regions is challenging. A nitrogen mass balance model is a practical tool to evaluate coastal ocean productivity to understand biological mechanisms beyond data observations. This study visualizes the biological production zones in Narragansett Bay, Rhode Island, USA, where hypoxia frequently occurs, by applying a nitrogen mass balance model. The Bay is divided into three zones – brown, green, and blue zones – based on primary productivity, which are defined by the mass balance model results. Brown, green, and blue zones represent a high physical process, a high biological process, and a low biological process zone, depending on river flow, nutrient concentrations, and mixing rates. The results of this study can better inform nutrient management in the coastal ocean in response to hypoxia and eutrophication.

Introduction

Primary productivity, the production of organic compounds by phytoplankton, fuels ecosystem food webs, and is important for understanding the system's function in response to environmental changes¹^,². Estuarine primary productivity is also closely linked to eutrophication which is defined as excessive nutrients in the ecosystem¹, causing several harmful consequences in the coastal regions, such as an overgrowth of phytoplankton leading to large algal blooms and subsequent hypoxia³^,⁴. Importantly, primary productivity in estuaries is highly dependent on the riverine nutrient loading, particularly nitrogen concentrations, which are the typical limiting nutrient in most temperate ocean ecosystems⁵^,⁶. However, an estimation of the extent of riverine nitrogen impacts in coastal areas remains challenging.

To estimate the estuarine primary productivity, a nitrogen (N) mass balance model is a useful tool to calculate nitrogen fluxes². The N-mass balance model also provides an understanding of biological mechanisms beyond data observations, revealing information at the edges of different primary productivity zones⁷. Three different zones⁸, defined as brown, green, and blue zones, are particularly useful for predicting the impact of nutrient loading in hypoxic regions. The brown zone, defined as the nearest region of a river mouth, represents a high physical process, the green zone has high biological productivity, and the blue zone represents low biological process. The boundary of each zone depends on river flow, nutrient concentrations, and mixing rates⁸.

Narragansett Bay (NB) is a coastal, temperate estuary in Rhode Island, USA, supporting economic and ecological services and goods⁹^,¹⁰^,¹¹, in which hypoxia has been consistently occurring. These hypoxic events, defined as the period of low dissolved oxygen (i.e., less than 2-3 mg of oxygen per liter), are particularly prevalent in July and August and are heavily impacted by riverine nitrogen loading during these months¹². With an increase in primary production and hypoxia due to anthropogenic emissions of nutrients¹³, understanding the nitrogen inputs into NB is critical to managing and addressing coastal issues such as eutrophication and hypoxia. Thus, in this study, the rate of primary production in NB is calculated from the N-mass balance model using historically observed nutrient data, especially dissolved inorganic nitrogen (DIN). Based on the results of the N-mass balance model by converting to carbon units using the Redfield ratio, three different primary productivity zones were identified to visualize the extent of nitrogen influence from the river in NB. The model was then recreated into a 3D representation to better visualize the different zones. The products produced from this study can better inform nutrient management in NB in response to hypoxia and eutrophication. Further, results from this study are applicable to other coastal regions to visualize the effects of riverine transport on nutrients and primary productivity.

Protocol

1. Applying the N-mass balance model

Download the dissolved inorganic nitrogen (DIN) data from the US Environmental Protection Agency (USEPA) for 166 stations in Narragansett Bay from 1990 to 2015.
NOTE: In this study, the sum of ammonium (NH₄⁺), nitrite (NO₂^–), and nitrate (NO₃^–) concentrations were considered as the DIN concentration.
Split the Narragansett Bay into fifteen boxes along its axis modified from the previous study¹⁴ using Adobe Illustrator to divide the Bay in the map (Figure 1).
Apply the N-mass balance model to calculate the mean concentration of DIN at each box.
NOTE: In this study, the N-mass balance model, consisting of DIN input and output terms, was modified from previous studies²^,¹⁵ and applied to each box (1-15) of the Narragansett Bay as Equation 1.
Eq. (1)
Table 1 shows the definitions of each term and unit used in this model of Narragansett Bay. The model calculates the mean DIN concentration by determining the difference in each box of Narragansett Bay, representing the net DIN removal by biological production. Detailed information on the N-mass balance model is shown in the previous studies²^,¹⁵. The detailed values used in the model of this study were derived from the previous studies¹⁴.
Calculate the potential primary production (PPP) rate based on the N-mass balance model results by converting the net DIN removal to carbon units using the Redfield ratio (C: N = 106: 16, molar ratio) in a spreadsheet file.

2. Visualizing three zones in the map of Narragansett Bay

Plot the identified three zones in the map of Narragansett Bay as a contour plot using the Ocean Data View software.
1. Save the PPP rate data of each box as a text file (.txt) from the spreadsheet file.
  NOTE: The .txt file also includes the location of each box number as latitude and longitude. Put the longitude as a negative value. The PPP rate data is labeled as PPP [gC·m^-2·day^-1].
2. Load the PPP rate data into the Ocean Data View software.
  1. Go to open in the File menu.
  2. Click Associate Variables Box, Latitude, Longitude with Station, latitude [degrees_north], and Longitude [degrees_east], in the Metadata Variable Association window, then click the OK button.
  3. Click the OK button in the Import window.
3. Draw the contour plot to show the PPP ranges in the map of Narragansett Bay.
  1. Right-click on the map, click Zoom, drag the red box to zoom into the data area of the map, and then click on Enter.
  2. Click the 1 SCATTER window of the Layout Templates in the View menu.
  3. Right-click in the Sample panel and select Derived Variables.
  4. Click the Add button after selecting Latitude under Metadata from the list of Choices panel. Do the same thing for Longitude and then click the OK button.
  5. Select drvd: Longitude [degrees_East] as X-Variable by right-clicking on the scatter window.
  6. Select drvd: Latitude [degrees_North] as Y-Variable by right-clicking on the scatter window.
  7. Select PPP [gC·m^-2·day ^-1] as Z-Variable by right-clicking on the scatter window.
  8. Select Properties by right-clicking on the scatter window and go to the Display Style option.
    1. Select the Gridded field.
    2. Go to the Contours option and click the << button to make 0, 0.1, and 2 values only remain in the Already Defined Panes of the left.
    3. Click the OK button.
Based on the contour plot from the Ocean Data View Software, define the edge of the brown, green, and blue zones in Narraganset Bay, and visualize the zones using Adobe Illustrator to plot three zones in the map.
NOTE: Following the previous study¹⁵, the PPP rate of the brown zone was over 2 gC·m^-2·day^-1, the green zone was between 0.1-2 gC·m^-2·day^-1, and the blue zone was less than 0.1 gC·m^-2·day^-1, respectively.

3. Converting the contour plot of three zones into the three-dimensional (3D) frame with LED light

Etch three acrylic panels as 5.5'' x 8'' with a laser cutter to show the boundary of each zone.
Stack three acrylic panels in an illuminated frame.Overlap each acrylic panel showing the blue, green, and brown zones. Place a panel showing green zones on top of the blue zones panel and a brown zones panel on top of that.
For the second physical model, etch four acrylic sheets as 5.5'' x 8'' with a laser cutter, with the UV printed three boundaries of zones and one panel to represent the entire Narragansett Bay (as per steps 3.1-3.2).
Change the color of each zone into brown, green, and blue using the LEDs placed at the bottom of the frame.

Representative Results

Three theoretical zones of Narragansett Bay based on the N-mass balance model
The three theoretical zones in Narragansett Bay (NB) were defined based on the N-mass balance model results, in which the DIN data were applied to fifteen boxes of NB, and then the mean DIN in each box was converted to the PPP rates for the summer period. As shown in Figure 2, based on the mean summer (June to September) PPP rates of each box, three (brown, green, and blue) zones in NB were identified as following the criteria of the PPP rates of each zone from the previous study¹⁵. During the summer period, boxes 1, 2, 5, 6, 7, and 10, mostly located near the river mouth, were defined as brown zones with high PPP rates greater than 2 gC·m^-2·day^-1, indicating there was a strong physical process and biological process with high turbidity and light limitation. Boxes 3, 4, 8, 9, and 11 were classified as green zones, having PPP ranges from 0.1-2 gC·m^-2·day^-1, where a strong biological process occurred, showing nutrient limitation and high primary production. Due to high turbidity in the brown zone, the penetration of light was limited, which was a significant difference from the green zone. In contrast, blue zones, with low PPP rates of less than 0.1 gC·m^-2·day^-1, were identified in boxes 12, 13, 14, and 15 and were the furthest offshore, representing low biological productivity.

Visualization of three zones of Narragansett Bay using physical frameworks
To visually implement the boundaries of three theoretical zones in NB, a 3D representation was created in which layered acrylic panels were used and etched, creating two physical frameworks as described in section 3. As shown in Figure 3, three acrylic panels were used with the LED lights on the bottom of the frame, which can be changed to show a better representation of the characteristics of each boundary. In addition, dot-matrix patterns were etched with a different degree to represent the amount of sediment turbidity in each zone. Figure 4 shows the second physical framework with four acrylic sheets containing three boundaries of each zone, UV printed, and one layer etched to show the entire NB. The images of the development stage of the second framework are shown in Figure 4A, with three sheets representing each zone and one additional sheet showing the entire three zones. In Figure 4B, the second physical framework was illuminated by the LED lights and showed the overlaps of the boundaries for each zone.

Figure 1: Map of Narragansett Bay. The numbered segments show the 15 boxes along the axis, which is modified from a previous study¹⁴. Please click here to view a larger version of this figure.

Figure 2: Extent of the three theoretical zones in Narragansett Bay. Zones were defined based on the N-mass balance model results. Each zone is divided by the mean summer (June to September) potential primary production (PPP) rates, which are converted to the N-mass balance model results defined from the previous study¹⁵. The mean summer PPP rate of brown zones is over 2 gC·m^-2·day^-1, the green zones is between 0.1-2 gC·m^-2·day^-1, and the blue zones is less than 0.1 gC·m^-2·day^-1. Please click here to view a larger version of this figure.

Figure 3: The first physical framework of the three zones in Narragansett Bay. The physical framework uses three acrylic panels and dot-matrix patterns to represent the amount of sediment turbidity in each zone. Please click here to view a larger version of this figure.

Figure 4: The second physical framework of three theoretical zones in Narragansett Bay. (A) The schematic images of the entire three zones in the Bay for UV printing and stacking of the second physical framework. (B) The framework created using four acrylic sheets to show overlaps of zones' boundaries. Please click here to view a larger version of this figure.

Unit	Definitions
	DIN flux from each river discharge
	Diffusive flux from atmospheric deposition
	Benthic flux from the bottom sediments
	Denitrification in the water column
	An advection term which calculated from the current velocity
	Removal by biological production

Table 1: Definitions of each term in N mass balance model. The detailed values used in the model were derived from previous studies¹⁴^,¹⁶^,¹⁷.

Discussion

This study estimated the extent of nutrient impacts from riverine inputs in Narraganset Bay (NB) based on the N-mass balance model by defining the three theoretical zones. Historically, hypoxic zones appeared near the Providence River, the western side of Greenwich Bay, and Mount Hope Bay during the summer period¹⁸, which were defined as brown zones in this study. Moreover, the zonation of NB is comparable to the results of a previous study¹⁹, which examined nutrient concentration and primary production in NB. Both highlight the importance of nutrient reduction efforts. In addition, the boundaries of each zone in this study were similar to results from a previous study¹⁹, indicating that hypoxia in the upper bay of NB may be controlled by the advection of organic matter from the Providence River, yielding elevated respiration with high productivity over 2.6 gC·m^-2·day^-1. These results in the upper bay of NB were represented by the brown zone in this study. In addition, productivity continued to decrease towards the ocean, indicated by the green and blue zones.

In contrast, during the summer season, Mt. Hope Bay (box 10) was defined as the brown zone in this study, showing higher primary productivity over 2 gC·m^-2·day^-1 compared to the previous study¹⁹. This increased productivity indicates that other nutrient input sources, in addition to the riverine nitrogen input, might affect this region and should be considered as another DIN input term in N-mass balance models. The zonation in this study is expected to inform better management efforts for nutrients in NB aimed at reducing riverine nitrogen discharge as well as atmospheric nitrogen deposition, which has been highlighted in other estuarine systems, including the Chesapeake Bay²^,²⁰. Oviatt et al. (2002) found that mixing rate and light penetration influenced PPP²¹, but future work is needed to better quantify these factors attributed to high PPP in the brown zones.

Finally, by representing the three theoretical zones of NB as two physical frameworks, an improved understanding of the extent of riverine or other nutrient inputs to the coastal area is visually achieved. While frameworks can have fixed boundaries for each zone, in our framework, flexibility is additionally shown to inform that the three theoretical zones can change from month to month according to the nutrient concentrations of freshwater, mixing rate, and river flow, as pointed out from the previous applications of the N-mass balance model²^,¹⁵. For example, several boxes in Figure 3 and Figure 4 were represented as mixed zones because they were categorized as different zones monthly during summer periods based on N-mass balance model results. The frameworks show the effect of riverine nutrients in NB by providing an integrated visualization of scientific biogeochemical data through an art form, which is useful for nutrient management in the coastal area and for scientific communication.

Offenlegungen

The authors have nothing to disclose.

Acknowledgements

This study was supported by the National Science Foundation (OIA-1655221, OCE-1655686) and Rhode Island Sea Grant (NA22-OAR4170123, RISG22-R/2223-95-5-U). We also would like to thank the Rhode Island School of Design for developing the Vis-A-Thon project and this visualization.

Materials

Adobe Illustrator	Adobe	version 27.6.1	https://www.adobe.com/products/illustrator.html
Ampersand Gessobord Uncradled 1/8" Profile 8" x 8"	Risdstore	70731053088	https://www.risdstore.com/ampersand-gessobord-8×8-flat-1-8-profile.html
Ocean Data View software			https://odv.awi.de/en/software/download/
W-Series (Wide) Flexible LED Strip Light – Ultra Bright (18 LEDs/foot)	aspectLED	SKU AL-SL-W-U	https://www.aspectled.com/products/w-wide-5050-ultra-bright?gclid=CjwKCAjwm4ukBhAuEiwA0z QxkyqisRPqBcHvXEW8KcJE-bK0d2cvGtqlOxXWJI_ E2rd6DzttPR0FLRoCgfkQAvD_BwE

Referenzen

Nixon, S. W. Coastal marine eutrophication: A definition, social causes, and future concerns. Ophelia. 41, 199-219 (1995).
Kim, J. S., Brush, M. J., Song, B., Anderson, I. C. Reconstructing primary production in a changing estuary: A mass balance modeling approach. Limnology and Oceanography. 66 (6), 2535-2546 (2021).
Kemp, W. M., et al. Eutrophication of Chesapeake Bay: historical trends and ecological interactions. Marine Ecology Progress Series. 303, 1-29 (2005).
Brush, M. J., Malone, T. C., Malej, A., Faganeli, F., et al. . Coastal Ecosystems in Transition: A Comparative Analysis of the Northern Adriatic and Chesapeake Bay. Chapter 5, (2021).
Howarth, R. W., Marino, R. Nitrogen as the limiting nutrient for eutrophication in coastal marine ecosystems: Evolving views over three decades. Limnology and Oceanography. 51 (1 part 2), 364-376 (2006).
Paerl, H. W. Controlling eutrophication along the freshwater-marine continuum: Dual nutrient (N and P) reductions are essential. Estuaries and Coasts. 32, 593-601 (2009).
Kim, J. S., Chapman, P., Rowe, G., DiMarco, S. F. Categorizing zonal productivity on the continental shelf with nutrient-salinity ratios. Journal of Marine Systems. 206, 103336 (2020).
Rowe, G. T., Chapman, P. Continental shelf hypoxia: Some nagging questions. Gulf of Mexico Science. 20 (2), 153-160 (2002).
Nixon, S. W. Eutrophication and the macroscope. Hydrobiologia. 629, 5-19 (2009).
Barbier, E. B., et al. The value of estuarine and coastal ecosystem services. Ecological Monographs. 81 (2), 169-193 (2011).
Cloern, J. E., Foster, S. Q., Kleckner, A. E. Phytoplankton primary production in the world’s estuarinecoastal ecosystem. Biogeosciences. 11 (9), 2477-2501 (2014).
Codiga, D. L., Stoffel, H. E., Oviatt, C. A., Schmidt, C. E. Managed nitrogen load decrease reduces chlorophyll and hypoxia in warming temperate urban estuary. Frontiers in Marine Science. 9, 930347 (2022).
Sigman, D. M., Hain, M. P. The biological productivity of the ocean. Nature Education Knowledge. 3 (10), 21 (2012).
Kremer, J. N., et al. Simulating property exchange in estuarine ecosystem models at ecologically appropriate scales. Ecological Modelling. 221 (7), 1080-1088 (2010).
Kim, J. S., Chapman, P., Rowe, G., DiMarco, S. F., Thornton, D. C. O. Implications of different nitrogen input sources for potential production and carbon flux estimates in the coastal Gulf of Mexico (GOM) and Korean Peninsula coastal waters. Ocean Science. 16, 45-63 (2020).
Lake, S. J., Brush, M. J. The contribution of microphytobenthos to total productivity in upper Narragansett Bay, Rhode Island. Estuarine, Coastal and Shelf Science. 95 (2-3), 289-297 (2011).
Brush, M. J., Nixon, S. W. Modeling the role of macroalgae in a shallow sub-estuary of Narragansett Bay, RI (USA). Ecological Modelling. 221 (7), 1065-1079 (2010).
Deacutis, C. F., Murray, D., Prell, W., Saarman, E., Korhun, L. Hypoxia in the upper half of Narragansett Bay, RI, during August 2001 and 2002. Northeastern Naturalist. 13 (Special Issue 4), 173-198 (2006).
Oviatt, C., et al. Managed nutrient reduction impacts on nutrient concentrations, water clarity, primary production, and hypoxia in a north temperate estuary. Estuarine, Coastal and Shelf Science. 199, 25-34 (2017).
Boesch, D. F. Barriers and bridges in abating coastal eutrophication. Frontiers in Marine Science. 6, 123 (2019).
Oviatt, C. A., Keller, A. A., Reed, L. Annual primary production in Narragansett Bay with no bay-wide winter-spring phytoplankton bloom. Estuarine, Coastal and Shelf Science. 54, 1013-1026 (2002).

Play Video

PDF

DOI

DOWNLOAD MATERIALS LIST

Diesen Artikel zitieren

Kim, J., Hwangbo, M., Thibodeau, P. S., Rhodes, G., Hogarth, E., Copeland, S. Visualization of Productivity Zones Based on Nitrogen Mass Balance Model in Narragansett Bay, Rhode Island. J. Vis. Exp. (197), e65728, doi:10.3791/65728 (2023).