Impact of Simulated Solar Irradiation on Disinfection Byproduct Precursors ALEX T. CHOW, † DINA M. LEECH,* ,‡ TREAVOR H. BOYER, ‡ AND PHILIP C. SINGER ‡ College of Environmental Sciences and Engineering, South China University of Technology Guangzhou, 510641, P.R. China, Department of Environmental Sciences and Engineering, University of North Carolina-Chapel Hill, Chapel Hill, North Carolina, North Carolina 27599 Received February 14, 2008. Revised manuscript received April 25, 2008. Accepted May 5, 2008. The Sacramento-San Joaquin Delta is the major drinking water source for 23 million California residents. Consequently, many studies have examined disinfection byproduct (DBP) formation in relation to Delta dissolved organic carbon (DOC) concentration. However, DOC characteristics within the Delta are not the same as those entering downstream water treatment facilities. As water is transferred to Southern California through the California Aqueduct, a 714.5 km-open channel, it is exposed to sunlight, potentially altering DBP precursors. We collected water from three sites within the Delta and one near the California Aqueduct, representing different DOC sources, and irradiated them in a solar simulator at a dose equivalent to that received during four days conveyance in the aqueduct. Photolytic changes in DOC were assessed by measuring CO 2 and organic acid production, fluorescence, and ultraviolet absorbance over time. Trihalomethane (THM) and haloacetic acid (HAA) formation potentials, as well as the distribution of hydrophobic, transphilic, and hydrophilic acid fractions were determined at exposures equivalent to one and four days. Solar irradiation significantly decreased ultraviolet absorbance and fluorescence intensity, produced organic acids, and increased the hydrophilic fraction of waters. These changes in DOC caused a shift in bromine incorporation among the THM and HAA species. Our results are the first to demonstrate the importance of sunlight in altering DOC with respect to DBP formation. 1. Introduction The Sacramento-San Joaquin Delta (hereafter referred to as the Delta) is the major drinking water source for 23 million people in California. Water quality in the region is of great concern because of the significant increases in organic carbon concentration as water passes through the Delta and into downstream water treatment plants (1–3). During chlorina- tion, dissolved organic carbon (DOC) acts as a major precursor in the formation of carcinogenic disinfection byproducts (DBPs), such as trihalomethanes (THMs) and haloacetic acids (HAAs) (4). Many studies have evaluated the reactivity of DOC in forming DBPs by collecting water samples from upstream rivers feeding the Delta and from within the central Delta (2, 5–8). However, the majority of Delta waters used for municipal water supply purposes are delivered 714.5 km downstream to southern California through the concrete open channel California Aqueduct. Given the potential for photolytic alterations to DOC during conveyance, DBP precursors found in upland Delta sources could be quite different from those at downstream water treatment plant intakes. The Delta is a complex ecosystem, with a variety of carbon sources contributing DOC (2, 3). A mass balance calculation indicates that the two major carbon sources are tributary- borne loads and in situ phytoplankton production. These two sources contribute an annual average of 270 and 47 Mg per day of total organic carbon, equivalent to 69 and 12% of the annual average organic carbon load, respectively (3). Peat soil is also considered a major organic carbon contributor to Delta waterways, through agricultural drainage returns and wetland outflows (5, 8). Drainage from about 1011 km 2 of peat soils was estimated to contribute an average of 36 Mg of total organic carbon per day (3). Large areas of Delta peat soil are being considered for conversion to wetland habitats, which could substantially increase organic carbon releases (9). Urban runoff within the Delta contributes less than 1% of the organic carbon in Delta waters, and the effect of this carbon source on DBP formation is relatively minimal (2, 3). Studies have demonstrated that organic carbon in river water, wetland outflow, and agricultural drainage are reactive precursors for DBP formation during drinking water treat- ment (2, 5, 8, 9). However, before entering water treatment facilities, organic materials in the water are exposed to solar radiation for an average of 3-4 days in the aqueduct. Many studies have shown that humic substances and DOC can be broken down into smaller, more labile organic carbon moieties by solar irradiation (10, 11). Despite this, solar effects on the reactivity of DOC with respect to DBP formation have not been studied. Our objectives were to evaluate the effects of sunlight on the properties and reactivity of DOC from different Delta sources, particularly with respect to their impact on DBP formation. 2. Materials and Methods 2.1. Sample Collection and Preparation. Grab samples were collected in five 2 L polyethylene bottles on January 4-5, 2006 at (1) the Sacramento River at Freeport; (2) a drainage ditch on Twitchell Island; (3) the outflow of a United States Geological Survey experimental wetland on Twitchell Island; and (4) the California Aqueduct at Banks Pumping Station (Figure 1). Water collected at Freeport represented upstream inputs to the Delta, drainage water represented typical agricultural runoff, the outflow from the experimental wet- land represented waters from rehabilitated wetlands, and water collected at the Banks Pumping Station represented outflow from the Delta to the Aqueduct. All water samples were filtered through 0.22 μm polycarbonate filters (Millipore) upon return to the laboratory. Filtered samples were stored in sterilized polyethylene bottles at 4 °C. Within two weeks of collection, samples were transported to the University of North Carolina at Chapel Hill for the solar exposure experi- ments and DBP analysis. To ensure sterility, all samples were again sterile-filtered before the exposure to simulated sunlight, and 500 mL aliquots were transferred into sterile 1 L quartz tubes (5 cm diameter, 33 cm length). Filtrates were analyzed for DOC, ultraviolet absorbance at 254 nm * Corresponding author phone: 252-726-6841; e-mail: dmleech@ email.unc.edu. † South China University of Technology Guangzhou. ‡ University of North Carolina-Chapel Hill. Environ. Sci. Technol. 2008, 42, 5586–5593 5586 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 15, 2008 10.1021/es800206h CCC: $40.75  2008 American Chemical Society Published on Web 06/28/2008