Sweetwater Authority (SWA) owns and operates the Robert A. Perdue Water Treatment
Plant (WTP), a 30 mgd conventional filtration plant located in San Diego County, CA.
After testing ozone, enhanced coagulation, chlorine dioxide (ClO2), MIEX and
submerged microfiltration/ultrafiltration (MF/UF), SWA has, determined that chlorine
dioxide and membrane filtration should be in the Authority's future. This paper
documents the results of SWA's five-month pilot testing of two submerged MF/UF
membrane systems, presents an overview of SWA's decision-making process, and
highlights some important lessons learned in the preliminary cost estimating and design
phases of this project.
The Perdue WTP has the ability to treat two source waters, local surface water from
Sweetwater Reservoir (SWR) (average total organic carbon = 6.5 mg/L, bromide up to
0.5 mg/L) and San Diego County Water Authority (CWA) aqueduct water (average TOC
= 2.5 mg/L, low bromide). SWA currently applies free chlorine and potassium
permanganate to the raw SWR water and uses chloramines in the distribution system.
However, excessive trihalomethanes (THMs) formation is a potential concern for future
compliance with the Stage 2 Disinfection Byproducts (DBP) Rule using the Locational
Running Annual Average. When treating water from the local reservoir during periods of
high runoff, THM formation could potentially exceed 120 ppb in the distribution system,
with typical distribution system values in the 60 to 80 ppb range.
The testing investigated various combinations of ClO2, dissolved air flotation (DAF), and
conventional flocculation/sedimentation (floc/sed) as pretreatment processes to the
MF/UF membranes. In addition to monitoring the performance of the DAF, floc/sed
process, and the membrane systems; simulated distribution system (SDS) testing was
performed on the membrane filtrate on a weekly basis to estimate disinfection by-product
(DBP) formation.
With DAF as pretreatment, both of the submerged membrane systems tested provided
reliable water quality at fluxes above 35 U.S. gallons per square foot per day (gfd), with
water recoveries around 95%, for at least 30 days before the need for membrane clean-inplace
(CIP). Based on these initial results, the flux rates were increased to above 40 gfd
and higher with conventional flocculation/sedimentation pretreatment. Membrane
performance under the higher flux conditions was mixed but provided valuable design information. SWA also held a series of peer review meetings to analyze the results of this and previous
testing. The consensus was that chlorine dioxide, DAF or floc/sed, and submerged
MF/UF membranes represent the best process combination for SWA and potentially for
other agencies with similar water quality concerns. SWA has begun design of a 40 mgd
upgrade to ClO2 and submerged MF/UF membranes. Includes 3 references, figures.