Coliform and E. coli Testing (3rd of 3 articles)
Tim Loftus
The first article in this series emphasized that the ways of validating
bacteriological testing differ from those methods used to validate chemical or
physical analyses. The second article was an overview of federally required
techniques and procedures needed to produce reportable results. This article will cover the requirements for reading the
results and for calculating the geometric mean for NPDES reporting.
Reading the results:
After incubating, count the colonies under 10X to 15X magnification. It’s important to report only the results that meet certain requirements. Often background colonies will grow on the membrane filter (will be a different color than the target organism). The total of the fecal coliform colonies and the background should not exceed 200 colonies per filter. If it does, consider that dilution to be invalid. Note that the E.coli test methods are not specific on this point.
Report only the fecal coliform results from membrane filters that have between 20 and 60 colonies or the E. coli colonies that have between 20 to 80 colonies per filter. Disregard the other dilutions. Don’t forget to factor all dilutions back up to a 100 mL sample volume. The following example shows how the result of a fecal coliform analysis is to be determined:
sample
colony count background
dilution
dilution
per filter colonies
factor
calculation
result
25 mL to 100 mL 15 67 4 4x15 60 col/100 mL
50 mL to 100 mL 34 110 2 2x34 68 col/100 mL
no dilution 100 mL 55 190 1 1x55 55 col/100 mL
Report on your NPDES DMR a fecal coliform value of 68 col/100 mL. Here’s the reason: The first dilution (25 mL to 100 mL) does not produce a fecal colony count of 20 to 60 colonies per membrane filter (it produces only 15). Therefore it doesn’t meet the counting requirements of the test. The third sample (no dilution) does produce enough colonies on the filter (55), but the total of fecal and background colonies is too high (55 + 190 >200). Likewise, this dilution does not meet the reporting requirements. Only the second dilution (50 mL to 100 mL) produces a reportable result.
There are a few exceptions to this. If your effluent disinfection system is working well, you may not have any, or very few, fecal coliform or E. coli bacteria in the sample. No matter what dilution you use, you will get a very low result. Choose the dilution closest to the requirements and report that result as an estimate.
Conversely, if your disinfection
system isn’t working properly, even a highly diluted sample will exceed the
reporting requirement limits. While the actual number of colonies may be
inaccurate, it is obviously an NPDES discharge violation. The results should be
reported as TNTC (Too Numerous To Count).
Calculating the Geometric Mean:
The geometric mean is a calculation to determine an average when the set of numbers covers a wide range. Results of bacteriological testing often cover such a large range. The easiest way to calculate the geometric mean is to use a scientific calculator. The first step in the calculation involves converting all your daily coliform results to log values. Add up these log values and divide by the number of samples. For example: The daily coliform results during the week were 25 col/100 mL, 285, 15, and 460.
(log 25 + log 285 + log 15 + log 460)/4 = (1.3979 + 2.4584 + 1.1760 + 2.6627)/4 = 1.9238.
The second step
is to take this value and find the antilog. This will be the 10x button
on your calculator. Press the button for 10x then type in 1.9238. Press “enter” and the display will
show 84. Report 84 colonies/100 mL as the weekly geometric mean. Monthly
geometric means are done the same way, except you will be using many more data
points.
The information in this article is based on general test methods that are used throughout the New England area for NPDES monitoring of the fecal coliform group and for E. coli. As usual, check your federal, state, and local regulations. You may have additional regulations or reporting requirements that you must meet.