Graduation Year


Document Type




Degree Granting Department

Civil and Environmental Engineering

Major Professor

Sarina J. Ergas, Ph.D.

Co-Major Professor

James R. Mihelcic, Ph.D.

Committee Member

Valerie J. Harwood, Ph.D.


coliforms, groundwater, health, microbial test kits, Millennium Development Goals


It has been reported that globally we have achieved Millennium Development Goal (MDG) Target 7C, to halve the proportion of the population without access to safe drinking water; however, there is a major flaw with this statement. While Target 7C calls for access to `safe' drinking water, what is actually being measured and reported is access to an `improved' water source. The World Health Organization (WHO) maintains that they must use this proxy measure because the methods for water quality testing are too expensive and logistically complicated, but by doing so, they may be over reporting safe water coverage.

This was shown to be true in Tamatave, Madagascar, where thermotolerant coliforms were detected in water from a type of `improved' source, the Pitcher Pump system. This research looked at several parameters - Pitcher Pump system depth, sampling neighborhood, requirement of pump priming, frequency that the system was repaired, distance from on-site sanitation, and number of users - to see if they were influencing water quality. Of all the parameters tested, only priming was found to be significantly associated with the levels of thermotolerant coliforms detected (Fisher exact test p = 0.03). Using a Mann-Whitney U test, it was shown that the median thermotolerant coliform concentration was significantly higher in primed wells (41.3 cfu/100 ml) than unprimed wells (3.5) (p = 0.01 cfu/100 ml).

A pilot study was conducted to look at only the effect of depth and to determine if a depth could be identified that could provide safe drinking water. The result of the pilot study showed that, while thermotolerant coliform concentration did decrease with increasing depth, even at the deepest well of 9.4 m, levels were still above 100 cfu/100 ml.

Additional research was conducted to investigate the performance and cost of three test kits for both total coliform and Escherichia coli quantification for water quality analysis in developing countries. IDEXX Colilert Quanti-trays[reg] (Colilert), Micrology Laboratories Coliscan[reg] Membrane Filtration tests (Coliscan MF) and a modified method for 3-M PetrifilmTM Coliform/E. coli plates (modified 3-M) were compared with standard membrane filtration (standard MF) methods under a range of incubation temperature conditions (22.0, 35.0 and 44.5[deg]C). Each test method was also performed by inexperienced volunteers, with the results compared to those of an experienced technician. At non-standard temperatures, Coliscan MF proved to be the most accurate when compared to standard methods, with a significant difference with only total coliforms at 44.5[deg]C. Modified 3-M had the poorest correlation with standard MF over the range of temperatures tested, with significant differences noted for all the temperatures except for E. coli at 44.5[deg]C. Inexperienced university volunteers found Colilert easiest to use, but Coliscan MF produced E. coli results that were most similar to the experts. Coliscan MF was found to have the overall best performance and lowest cost in this study; however, it did produce high numbers of false positive results.