Estimating Nutrient Concentrations in
Wetlands and Biofilters Using Proxy Data
Norma Galaviz, Emily Parker, and Cameron Patel
August 1, 2013
NSF PIRE
Powerpoint Templates
Estimating Nutrient Concentrations in
Wetlands and Biofilters Using Proxy Data
Norma Galaviz, Emily Parker, and Cameron Patel
August 1, 2013
NSF PIRE
Powerpoint Templates
Estimating Nutrient Concentrations in
Wetlands and Biofilters Using Proxy Data
Norma Galaviz, Emily Parker, and Cameron Patel
August 1, 2013
NSF PIRE
Powerpoint Templates
Estimating Nutrient Concentrations in
Wetlands and Biofilters Using Proxy Data
Norma Galaviz, Emily Parker, and Cameron Patel
August 1, 2013
NSF PIRE
Powerpoint Templates
Outline
•
•
•
•
•
•
•
Background
Introduction
Hypothesis
Methods
Data Analysis
Results
Discussion
Powerpoint Templates
Powerpoint Templates
Background
• The Millennium Drought threatened water security in
Melbourne, Australia
• Stormwater capture, treatment, and harvesting provide
a new source of water for Melbourne, and reduce
nitrogen input to Port Phillip Bay
Powerpoint Templates
Introduction
•
•
•
Two treatment approaches are evaluated in this study -biofilters and constructed wetlands
Stormwater treatment is affected by flow path,
vegetation, and filter media
The fate of nitrate and phosphate in wetlands and
biofilters is evaluated using multiple linear regression
(MLR)
Powerpoint Templates
Hypothesis
Phosphate and nitrate in constructed wetlands and
biofilters can be estimated using sensor measurements
and other proxy data.
Methods
Sampling took place at a total
of six sites in Melbourne,
Victoria, Australia.
• Two constructed wetlands: Royal
Gardens and Hampton Park
Powerpoint Templates
Methods
Sampling took place at a total
of six sites in Melbourne,
Victoria, Australia.
• Two constructed wetlands: Royal
Gardens and Hampton Park
• Two biofilters: Wicks Reserve and
Hereford Road
Powerpoint Templates
Methods
Sampling took place at a total
of six sites in Melbourne,
Victoria, Australia.
• Two constructed wetlands: Royal
Gardens and Hampton Park
• Two biofilters: Wicks Reserve and
Hereford Road
• One bioretention/ constructed
wetland system: Lynbrook Estates
Powerpoint Templates
Methods
Sampling took place at a total
of six sites in Melbourne,
Victoria, Australia.
• Two constructed wetlands: Royal
Gardens and Hampton Park
• Two biofilters: Wicks Reserve and
Hereford Road
• One bioretention/ constructed
wetland system: Lynbrook Estates
• One natural wetland system:
Edithvale-Seaford Wetlands
Powerpoint Templates
Methods
• Measurements and water samples were collected
at multiple locations within each site (including
outflow and inflow locations)
Powerpoint Templates
Methods
• Measurements and water samples were collected
at multiple locations within each site (including
outflow and inflow locations)
• Temperature (T), dissolved oxygen (DO), and pH
measurements were recorded ten times per
location using portable instruments
Powerpoint Templates
Methods
• Measurements and water samples were collected
at multiple locations within each site (including
outflow and inflow locations)
• Temperature (T), dissolved oxygen (DO), and pH
measurements were recorded ten times per
location using portable instruments
• Water samples taken in triplicate were analyzed
for chlorophyll (CHL), pheophytin (PHEO),
nitrate, and phosphate (spectrophotometry), and
total suspended solids (TSS; weight
measurements) as outlined in Standard Methods
for the Examination of Water and Wastewater
Powerpoint Templates
Data Analysis
MLR models were developed
using Virtual Beach 2.3 with:
• nitrate or phosphate as the
dependent variable
• T, DO, pH, CHL, PHEO,
and TSS as independent
variables
Powerpoint Templates
4 groups were evaluated:
Powerpoint Templates
Results
• MLR models created for
each group, ranked by
AICc
• The R2 values for
phosphate models are
greater than for nitrate
models
• Nitrate and phosphate
were most highly
correlated with TSS and
DO
Powerpoint Templates
Powerpoint Templates
Discussion
Phosphate
• DO and TSS were the
strongest correlates
• Consistent with the fact
that phosphate binds to
suspended solids
• Highlights the
importance of solids
removal as a wetland
function
Powerpoint Templates
Powerpoint Templates
Discussion
Nitrate
• Nitrate models explained less variance than phosphate
models
Powerpoint Templates
Discussion
Nitrate
• Nitrate models explained less variance than phosphate
models
• DO, TSS, and PHEO were explanatory variables
Powerpoint Templates
Discussion
Nitrate
• Nitrate models explained less variance than phosphate
models
• DO, TSS, and PHEO were explanatory variables
• Nitrate/inverse PHEO relationship suggests nutrient
uptake by phytoplankton
Phytoplankton
Chlorophyll
Powerpoint Templates
Pheophytin
Discussion
Nitrate
• Nitrate models explained less variance than phosphate
models
• DO, TSS, and PHEO were explanatory variables
• Nitrate/inverse PHEO relationship suggests nutrient
uptake by phytoplankton
Phytoplankton
Chlorophyll
Phytoplankton
Pheophytin
NO3-
Powerpoint Templates
Discussion
Nitrate
• Nitrate models explained less variance than phosphate
models
• DO, TSS, and PHEO were explanatory variables
• Nitrate/inverse PHEO relationship suggests nutrient
uptake by phytoplankton
• Note: It is unclear why DO is a correlate. We noticed
that DO tended to be low in shallower areas that had
higher TSS.
Powerpoint Templates
Conclusion
• Phosphate can be
reasonably estimated
using proxy data
• Nitrate analysis reveals
weak relationship
• Further study needed, or
perhaps nitrate cannot be
reasonably predicted
using proxy data.
Powerpoint Templates
References
1. Andrieux-Loyer, F.; Aminot, A. Phosphorus forms related to sediment grain
size and geochemical characteristics in French coastalA areas. Estuarine,
Coastal, Shelf Sci. 2001, 52, 617−629.
2. Eaton, A. D.; Franson, M. A. H.; American Public Health Association (Eds.).
(2005) Standard Methods for the Examination of Water and Wastewater.
(21st ed.) American Public Health Association.
3. Koike, I.; Furuya, K.; Otobe, H.; Nakai, T.; Menoto, T.; Hattori, A.
Horizontal distributions of surface chlorophyll a and nitrogenous nutrients
near Bering Strait and Unimak Pass. Deep-Sea Res., Part A 1982, 29,
149−155.
4. Rippy, M.; Franks, P.; Feddersen, F.; Guza, R.; Warrick, J. Beach
Nourishment Impacts on Bacteriological Water Quality and Phytoplankton
Bloom Dynamics. Environ. Sci. Technol. 2013, 47, 6146-6154.
5. Rocha, R. R.; Thomaz, S. M.; Carvalho, P.; Gomez, L. C. Modeling
chlorophyll a and dissolved oxygen concentration in tropical floodplain lakes
(Parana River Brazil). Braz. J. Biol. 2009, 69, 491−500.
Powerpoint Templates
Acknowledgements
This project was funded through a Partnerships for International Research Education grant from the
National Science Foundation. Special thanks to the University of Melbourne Engineering
Department for lab and equipment accommodations, and to Melbourne Water for site access and
cooperation. Thanks also to Dr. Stanley Grant, Dr. Sunny Jiang, Dr. Andrew Mehring, Dr. Meg
Rippy, and Alexander McCluskey for their assistance and advice.