Graduation Year


Document Type




Degree Name

MS in Environmental Engr. (M.S.E.V.)

Degree Granting Department

Civil and Environmental Engineering

Major Professor

Katherine Alfredo, Ph.D.

Committee Member

James Mihelcic, Ph.D.

Committee Member

Jeffrey Cunningham, Ph.D.


Contamination, Drinking water, Flushing, Self-supply, Stagnation


Previous sampling campaigns in Tamatave, Madagascar, conducted by the University of South Florida (USF) have identified lead contamination in drinking water sourced from self-supply pitcher pumps. The pitcher pump piston and foot valves were identified as the primary source of lead contamination since they are often manufactured from recycled lead acid batteries. The previous field sampling technique consisted of collecting an aliquot of a 10-15-liter composite sample from the pitcher pump for dissolved lead analysis; however, this technique (1) neglects the contribution of particulate lead, (2) does not properly account for the effects of water stagnation inside the pitcher pump chamber, and (3) dilutes the sample which stagnates in the pump chamber. This sampling technique is referred to as the “First-draw composite (FDC)” throughout this research. A true first draw sample represents the water which stagnates overnight within the pump chamber and predicts the maximum possible dissolved lead concentration exposure.

To determine the extent to which this FDC sampling technique underestimates the true first draw sample, this research utilizes laboratory batch stagnation and column flushing studies on harvested pitcher pump foot valves. For the eight batch stagnation trials completed through this research, three different solutions with varying water quality were evaluated. A total of six column flushing trials were performed using two different solutions. Both total and dissolved lead samples (n=548) were collected for analysis using Atomic Absorption Spectroscopy. Stagnation of the foot valves demonstrated that aqueous lead concentrations increase with time as 24 h total lead concentrations were higher than 0.5 h total lead concentrations in 87% of the trials. The average particulate lead concentrations observed in the batch study for stagnation periods 0.5, 8, and 24 h were 100 μg/L, 172 μg/L, and 165 μg/L respectively and the particulate lead contribution was significantly greater than the World Health Organization guideline of 10 μg/L across all stagnation trials. Particulate lead concentrations also varied throughout the column flushing study and were typically the highest during the first 100 mL flushed as column flow was likely not great enough to dislodge any additional particulates through continued flushing. Using the obtained first draw and composite samples from the column flushing studies, this research estimates that the previous field sampling technique diluted first draw samples by 3-50 times dilution. This research shows that the prior field events likely underestimated the maximum total lead concentrations of first draw samples by neglecting the particulate lead contribution and by underestimating stagnation effects. This has important health implications for human lead exposure as water collectors who used pitcher pumps after prolonged stagnation periods may be exposed to higher water lead levels.