AWWA WQTC58908

Quabbin Reservoir, located in central Massachusetts, supplies drinking water to the Boston area through Wachusett Reservoir via the Quabbin Aqueduct. The reservoir is under the joint control of the Metropolitan District Commission (MDC) and the Massachusetts Water Resources Authority (MWRA). The drinking water supplied from both the Quabbin and Wachusett source reservoirs is unfiltered, therefore the quality of water delivered to the consumers depends primarily on watershed protection and disinfection. Based on previous studies of natural organic matter (NOM) in tributaries and the reservoir, and hydrodynamic data collected by the MDC, Quabbin Reservoir NOM was modeled using CE-QUAL-W2, a hydrodynamic and water quality model. The objective of this study was to assess the applicability of the model for simulating observed temporal and spatial trends in total and dissolved organic carbon (TOC, DOC), algae, and UV absorbance (UV-254). The effect of nutrient loadings on reservoir TOC, DOC, and algae was investigated as the long-term goal of NOM modeling is to provide support for watershed management. CE-QUAL-W2 is a two-dimensional, longitudinal/vertical (assuming lateral homogeneity), hydrodynamic and water quality model. Based on specified flow inputs and outputs, meteorological data and reservoir bathymetry, the model predicts water surface elevations, velocities, and temperatures. The water quality algorithms of the model incorporate 21 constituents in addition to temperature, including several related to NOM and nutrient/phytoplankton/dissolved oxygen interactions. The model requires specification of initial conditions and constituent concentrations in all inflows and outflows. The water quality constituents of interest for this study are those that affect the fate and transport of NOM; algae, detritus, labile and refractory dissolved organic matter (LDOM, RDOM), dissolved oxygen, ammonia-nitrogen, nitrate-nitrogen, orthophosphorus, and UV-254. The model requires specification of values for at least 39 model parameters related to processes affecting the transformation (decay, growth) of constituents. Much of this study was dedicated to estimating and calibrating many of the parameter values. The two-year period of model simulation was from January 1998 through December 1999. NOM levels in the reservoir are relatively stable over the two-year study period (and longer) and NOM levels in tributaries exceed levels observed in the reservoir. Decay or loss of organic matter within the reservoir is thus significant and is simulated in the model. In addition, NOM increases due to phytoplankton growth are modeled. In general, the model simulated the relatively constant average levels of reservoir outlet UV-254, TOC and DOC with some seasonal variability. For example, average TOC (which is 95% DOC) levels of approximately 2.2 mg/L are simulated by assuming that 20% of the DOC is labile DOM and use of temperature dependent labile and refractory DOM decay rates of 0.003 per day and 0.0003 per day (at 20 C), respectively. Algae levels are low (0.05 to 0.4 mg/L as carbon), and estimates of algal growth and respiration rates of 3.5 day-1 and 0.2 day-1 respectively, resulted in algal levels consistent with observed concentrations and that account for seasonal variability in TOC concentrations. Additionally, modeling results revealed that nutrient inputs to the reservoir via precipitation are a very important source of nitrogen and phosphorus relative to tributary inputs. In-reservoir processes are necessary to maintain the low levels of NOM in Quabbin Reservoir and thus its excellent water quality. The calibrated water quality model can be used to examine potential impacts of watershed management practices, such as increases in nutrient loadings from tributaries or precipitation, on water quality and required treatment practices. Includes 13 references, table, figures.