Presentation is loading. Please wait.

Presentation is loading. Please wait.

201 529 5151 www.hydroqual.com Use of Mechanistic Modeling to Enhance Derivation of Great Bay TN Criteria and Inform Restoration Strategy Thomas W. Gallagher,

Similar presentations


Presentation on theme: "201 529 5151 www.hydroqual.com Use of Mechanistic Modeling to Enhance Derivation of Great Bay TN Criteria and Inform Restoration Strategy Thomas W. Gallagher,"— Presentation transcript:

1 Use of Mechanistic Modeling to Enhance Derivation of Great Bay TN Criteria and Inform Restoration Strategy Thomas W. Gallagher, P.E. Cristhian A. Mancilla, EIT Southeast Watershed Alliance Symposium May 11, 2011

2 Stressor-Response Regression Analysis  Regresses response variable (high chla, low DO, eelgrass decline) to stressor (nutrients). MAJOR ASSUMPTION: Other factors not significantly influencing “dose:response”  Empirical relationship (Vollenweider and others) of lake chla to lake phosphorus have been successfully applied for over 40 years.  Stressor-response relationships for rivers and estuaries have been more problematic.  Primary difficulties have been the proper consideration of other factors that influence the stressor-response relationship and the covariance across some stressors.

3 Mechanistic Modeling  Relates response variables to stressor with consideration of site specific physical, chemical, and biological factors.  Example: Low DO due to algal photosynthesis and respiration, BOD oxidation, nitrification, sediment oxygen demand and atmospheric reaeration.  Requires a considerable amount of data in comparison to the stressor-response approach.  More expensive than stressor-response approach but yields more scientifically defensible results; however, frequently beyond the budget of many regulatory agencies.

4 EPA Science Advisory Board Stressor Response Observations “In order to be scientifically defensible, empirical methods must take into consideration the influence of other variables…. The statistical methods in the Guidance require careful consideration of confounding variables before being used as predictive tools. … Without such information, nutrient criteria developed using bivariate methods may be highly inaccurate.” SAB Stressor Response Review – April 27, 2010

5 Proposed Numeric Nitrogen Criteria for Great Bay Estuary

6 Trend Monitoring Stations for Water Quality in the Great Bay Estuary (New Hampshire DES, 2009)

7 Relationship between Light Attenuation Coefficient and TN at Trend Stations (New Hampshire DES, 2009) 0.75

8 Relationship between Turbidity and TN at Datasonde Stations (New Hampshire DES, 2009)

9 Contributions to Kd (PAR) measured at the Great Bay Buoy (From Morrison et al, 2008)

10 Measured Chla and Secchi Disk at Adams Point ( )

11 Relationship between Minimum DO and Chlorophyll-a at Trend Stations (New Hampshire DES, 2009) Coastal Bay Tidal River  Other factors:  - Residence time  - Reaeration  - SOD  - BOD oxidation  -Stratification

12 DO at the Squamscott River Datasonde July 2008 DO (%) DO (mg/L) Depth Salinity

13 DO at the Squamscott River Datasonde July-October 2009 (New Hampshire DES, 2011)

14 Factors Influencing Water Column Dissolved Oxygen BOD5DO Reaeration Algal photosynthesis and respiration SOD NH4 BOD5 Water column stratification NitrogenPhosphorus Residence Time

15 Recommendations  Continue to look for other factors responsible for eelgrass decline other than changes in water column light transparency.  To better understand the factors responsible for low DO in tributaries to Great Bay, perform a data collection effort to support mechanistic modeling.  Start with Squamscott River DO model to address Exeter permit issues.

16 STOP

17 Relationship between TN and Chlorophyll-a at Trend Stations (New Hampshire DES, 2009)

18 COASTAL MARINE LABORATORY GREAT BAY DATASONDE LAMPREY RIVER DATASONDE 90 th chla = 9.3 ug/L Median TN = 0.39 mg/L 90 th chla = 7.5 ug/L Median TN = 0.45 mg/L SALMON FALLS RIVER DATASONDE OYSTER RIVER DATASONDE SQUAMSCOTT RIVER DATASONDE 90 th chla = 14.3 ug/L Median TN = 0.57 mg/L 90 th chla = 12.1 ug/L Median TN = 0.74 mg/L 90 th chla = 13.7 ug/L Median TN = 0.52 mg/L Daily Minimum DO, Jun-Sep (New Hampshire DES, 2009) 90 th chla = 3.3 ug/L Median TN = 0.30 mg/L

19 Key Transparency Issues to Resolve Did Bay transparency significantly change over time? Degree of a 1-2 ug/l Chlorophyll a change on long term average transparency? Degree of chlorophyll a reduction achievable with TN control considering bay hydrodynamics? Does pattern of eelgrass loss fit transparency theory? Are other factors at play in eelgrass decline?

20 Summary statistics for DO and chlorophyll-a for grab samples collected from grouped by assessment zone (NH DES Figure 26) Predicted threshold for DO violations: Chl-a (90 th %ile) > 7 ug/L


Download ppt "201 529 5151 www.hydroqual.com Use of Mechanistic Modeling to Enhance Derivation of Great Bay TN Criteria and Inform Restoration Strategy Thomas W. Gallagher,"

Similar presentations


Ads by Google