22
geosmin. The presence of certain culprit algae may not necessarily mean that MIB
and/or geosmin is being produced. Several of the MIB producers appear to be able to
turn on/off production in the laboratory. Production of T&O compounds should be
verified by GC/MS analysis.
MIB and geosmin. Analyze by FPA, or preferably by GC/MS. Rapid turnaround is
paramount for making decisions on water supply management, canal treatment, or in-
plant process options. Analytical results reported within 72 hours of sampling is a
reasonable goal.
Flavor Profile Analysis (FPA) panels should be trained by experts and chlorine should be
quenched with hydrogen peroxide in order to specifically detect MIB or geosmin.
MIB and geosmin can be measured using Solid-Phase Microextraction/Gas
Chromatography/Mass Spectroscopy (SPME-GC/MS) (Watson et al., 2000; Lloyd et al.,
1998). Details of the method used in the recent study are given in Appendix C.
4.3 MONITORING IN PRESSURIZED WATER DISTRIBUTION SYSTEMS
Algae do not pass through sand filtration systems at the WTPs nor grow in the absence
of light, and therefore typically are not present in pressurized pipe water distribution
systems carrying finished drinking water from the WTPs. However, fungi, bacteria (e.g.,
actinomycetes) and other organisms can grow in water distribution systems and also
impart earthy, musty, moldy odors to water. Therefore, customer complaints about tap
water may be caused from either: (1) T&O compounds in finished water leaving the WTP
or (2) T&O compounds produced within the pressurized water distribution system. If a
water utility suspects the latter due to numerous complaints from a localized region of the
water distribution system, then samples from the distribution system should be collected.
Preferred sampling would include a responsive program where samples from the service
tap and customer faucet are collected and analyzed for chlorine residual (field) and
MIB/geosmin (lab). Many complaints are actually from chlorine residual. If a routine
monitoring program is desired, then points within the water distribution system should
represent points of historic microbial concern (e.g., high plate counts, low chlorine
residuals).
4.4 PREDICTION OF T&O PROBLEMS
4.4.1 Rationale for Prediction
The ability to predict the occurrence of T&O episodes is important for water resources
planning and for water treatment plant staff for three reasons:
1.
Some T&O episodes may be prevented. For example,
canal treatments can be
used to prevent the growth of T&O culprit algae. To utilize canal treatments
effectively, it is desirable to know where and when T&O episodes are likely to
23
occur, because it is too expensive to treat the entire canal throughout the T&O
season.
2.
Some T&O episodes may be avoided. In many cases, it may not be possible to
prevent the production of T&O compounds upstream, but water resources
managers could manage the water supply system to prevent poor-tasting water
from reaching consumers. This can be done through reservoir management,
blending of source waters, or source switching (reducing production at a WTP
receiving poor-tasting water while simultaneously increasing production at a WTP
receiving better quality water). These management procedures will be discussed
later.
3.
Water treatment plants can remove T&O compounds,
but need advance warning to
do this effectively. T&O compounds can generally be reduced below threshold
levels using powdered activated carbon (PAC) within the WTPs. However,
treatment plant operators need to know when T&O episodes will occur so that they
can order PAC and prepare their WTPs for an episode.
4.4.2 Reservoirs
MIB was produced in all three of the water storage reservoirs above the canal
distribution system (Section 2). Prediction of when MIB is likely to be released from each
reservoir, and its impact on Phoenix’s water supply, is now possible.
As noted above, MIB is produced by certain blue-green algae, which grow primarily
either in shallow near shore areas or suspended in the epilimnion. MIB concentrations in
the epilimnia of all three reservoirs commonly exceed 20 ng/L during the summer
stratification period. Concentrations of MIB in the outlets from the reservoirs during
summer stratification are nearly always much lower than found in the epilimnion,
because water is released from the hypolimnion. Figures 4-2 through 4-4 show MIB
concentrations in the reservoirs and their outlets.
Lake Pleasant
0
10
20
30
40
50
Aug-99
Oct-99
Dec-99
Feb-00
Apr-00
Jun-00
Aug-00
Oct-00
Dec-00
Feb-01
Apr-01
Jun-01
Aug-01
Oct-01
Dec-01
Feb-02
Apr-02
MIB, ng/L
Epilimnion
Hypolimnion
CAP@7th St.
Figure 4-2. MIB in the epilminion, hypolimnion, and outflow from Lake Pleasant.