The Lake Saltonstall Water Treatment Plant (LSWTP), owned and operated by theRegional Water Authority (Authority) in New Haven, Connecticut, is a 12-milliongallons per day (mgd) conventional treatment plant. The LSWTP consists of rapidmixing, flocculation, sedimentation, filtration, and disinfection with chlorine.Currently, the LSWTP meets all existing regulations. However, it was determinedthat the LSWTP would have difficulty meeting anticipated regulations regardingdisinfection byproduct formation, increased Cryptosporidium inactivation orremoval, and difficulty continuing to meet customers' expectations regarding thecontrol of taste and odor-causing compounds. A review of the available watertreatment processes was performed to determine the appropriate combinations ofthese processes to meet the water quality goals. The treatment trains consistedof combinations of high-rate sedimentation, filtration, alternative disinfection,microfiltration, ultrafiltration, and nanofiltration processes. The treatmenttrains were evaluated and rated based on their ability to meet water-qualitygoals. A cost analysis was performed for the five treatment trains that receivedthe highest ratings. A benefit-to-cost ratio was then calculated for thesetrains. The use of ozone followed by biologically activated carbon (BAC) filterswas found to have the highest benefit-to-cost ratio. Ozone is a powerfuldisinfectant that has the ability to inactivate Cryptosporidium. The combinationof ozone and BAC filters has also been used for DBP removal and taste and odorcontrol. However, retrofits to the existing facility to accommodate theseprocesses would be difficult because of the plant's layout. Additionally, newstudies show that ozone is not as effective at inactivating Cryptosporidium atlow temperatures. Winter conditions require much higher contact times and ozonedosage. Therefore, this process may not be as cost effective as originallythought. The treatment train with the next highest benefit-to-cost ratioconsisted of immersed ultrafiltration (UF) with coagulant and powdered activatedcarbon (PAC) addition. An immersed UF system has several advantages over theozone/BAC process. However, with this treatment train, many of the advantageswere still unproven hypotheses. Therefore, a pilot-testing program was designedto answer the following questions: can coagulant and PAC be used in this reactormode to provide sufficient DBP precursor removal to meet anticipated DBPrequirements; can PAC be used in a reactor mode to address taste and odorsrelated to algae; what impacts will using coagulant and PAC have on the fluxrate, solids waste rate, membrane performance, and cleaning frequency; and, isthis all-in-one membrane approach really a cost-effective solution for addressingthese issues. Includes figures.