Modeling daily variation of trihalomethane compounds in drinking water system, Houston, Texas.

Total trihalomethanes (TTHM) concentrations vary widely and periodically between 70 and 130 ppb. Data from the National Environmental Services Laboratory, Houston, Texas indicate that pH and free residual chlorine contribute minimally to the wide variability of TTHM levels. Temperature variation in drinking fluctuates from 11 to 27 degrees C. The objective of this research is to formulate a model that delineates more clearly the daily variations of the most prevalent volatile trihalomethane by-products: chloroform (CHCl3), bromodichloromethane (CHBr2Cl), and bromoform (CHBr3) levels from drinking water. This model simulates the daily fluctuation of THM at a single location and at any time during the day as a function of the water temperature and the average concentration of TTHM, which can be estimated. The hypothesis of this study is that observed daily fluctuations of TTHM, CHCl3, CHCl2Br, CHClBr2, and CHBr3 are periodic. This hypothesis is tested using autocorrelation functions and it is shown that for the series of pH the correlation coefficient is maximal at zero lags, rapidly decreases to zero, and increases again between 4- and 6-h period. Such pattern suggests random fluctuation unrelated to time. However, the series of free residual chlorine, temperature, TTHM, CHCl3, CHCl2Br, CHClBr2, and CHBr3 suggest a different pattern. The correlation coefficient increases when the time-shift approaches 24 h. These repetitions in fluctuation of content over a 24-h period are statistically significant. The model formulated in this study provides insights in TTHM variation and is a necessary tool to reduce the error when estimating potential risk from exposure to trihalomethane compounds in drinking water system. In general, calculation of potential risk by using a value measured early morning or late afternoon concentrations were found minimal lead to an underestimation of the population risk.

Modeling brominated trihalomethane compounds in drinking water at a treatment plant in Beaumont, Texas.

The premise of this study is that the presence of bromide has a substantial effect on both the speciation and total formation of trihalomethane (THM). Consequently, models of water containing substantial bromide concentrations require refinement because they are only calibrated with raw water with high humic acid content. This study investigates and reports efforts on such refinement. The objectives of work reported in this paper are to formulate and validate a new correlative model that is based on physical principles and incorporates high levels of bromide that affect THM formation using raw water rich in both humic and fulvic acid. Two types of THM precursors are considered in the model discussed in this paper: (1) activated aromatic groups, which are more reactive with chlorine than with bromine, and (2) aliphatic groups, which are more reactive with bromine than with chlorine. Aliphatic and aromatic carbons are incorporated in the model by the inclusion of pertinent variables such as C/N and Br/Cl2 in the algorithm. The model also includes NH4+. This variable affects chlorine consumption that, in turn, potentially affects THM formation. For the first time ever, organic carbon to organic nitrogen and bromide to chlorine ratios is also used as variables potentially affecting THM formation when treating drinking water. The THM model formulated in this paper is an empirical model based on scientific principles and not simply a regression equation. The formulated model is pronounced valid based on a validation effort that employs a portion of an EPA database that was not used for model formulation.

Modeling spatial variation of brominates trihalomethane in a water distribution system of Ontario, Canada.

Conventional approaches to characterize and model the formation of trihalomethanes (THM) species in the distribution system use either residence time or water temperature. A significant deviation of THM levels were observed at the beginning and the end of a selected distribution system in Ontario, which may be because the consumption rate of residual chlorine is not constant in the distribution system. The approach developed in this study incorporates water temperature and proceeds with a trend and decomposition modeling method to incorporate the traveled distance and to explain the seasonal THM variation in the distribution system. The model has been tested and verified using a database from the Bettravia distribution system in Ontario, Canada. The deviations at the extremes of the distribution system were minimized due to the modeling technique used to develop the model and by including more factors that affect THM formation in the distribution system. The agreement between predicted and measured THM values at the beginning and the end of the distribution system is pronounced. The model presented in this paper is a robust tool that may be used by SDWAA to evaluate regulatory options and justify potential regulations regarding THM levels of the drinking water distribution system.