PIRE-presentation_final - UCI Water-PIRE

Ava Moussavi
Jessica Satterlee
Garfield Kwan
Started in the late 1990s and lasted more than a decade
Bureau of Meteorology, 2011
Sparked widespread use of alternate water sources
◦ Recycled water
◦ Rainwater harvesting
Grant et al. 2012
Western Treatment
Wastewater and
stormwater recycling can
be a potential risk to
human and ecosystem
health if methods for water
treatment do not perform
Larval stage of midges
Thrive in anoxic
Feed on organic matter
Associated with degraded
wetland conditions
 The
objective of this project was to
assess the relationship between
chironomid abundance and overall
water quality.
Water quality parameters were
measured at 2 biofilters and 3
constructed wetlands in
Melbourne, Australia
 Chironomids
 Chlorophyll concentrations
 Dissolved oxygen and
 Conductivity, Turbidity, ORP,
and pH
Virtual Beach 2.3 was used to perform multiple linear
Identified correlations between chironomid abundance
and water quality parameters:
Chlorophyll Content
Dissolved Oxygen (DO)
Oxidation Reduction Potential (ORP)
Chironomidae = B0 – B1Temp-1 + B2Turb-1
B0 = 170.14
B1 = 1948.40
B2 = 2315.22
p-value (Turb-1): 0.02
p-value (Temp-1): 0.03
Chironomidae = B0 – B1 poly(pH) + B2Turb-1
B0 = -34.56
B1 = 1.30
B2 = 1505.51
• Chironomid abundance can be predicted from
temperature and turbidity (top ranked model) or pH
and turbidity (second model)
• Turbidity is the most credible explanatory variable
because it appears in both top-ranked models, and
was identified as an important correlate in a
preliminary Classification Tree analysis (data not
• Data set is small and more advanced analytical
techniques for categorical data would need to be
Our study has identified temperature, pH and
turbidity as possible indicators of chironomid
abundance, but our data/methods are insufficient for
us to conclude that these water quality parameters
can be used to predict chironomid abundance.
Increase sampling size and sampling intensity
Survey alternative variables i.e. wetland birds
Use advanced statistical tools (Generalized Linear
Models, Classification Tree analysis) that permit
evaluation of categorical variables
Functional role of chironomidae
We want to thank Stanley Grant, Sunny Jiang, Megan
Rippy, Andrew Mehring, Alex McCluskey, Laura
Weiden, Nicole Patterson, and Leyla Riley, the faculty
of University of California - Irvine, and the staff of
University of Melbourne for contributing and
facilitating our research. We also want to thank NSF
for funding this research.

similar documents