Surface-water-quality data from selected monitoring stations in Connecticut were analyzed for trend, using the Seasonal Kendall test, for water years 1969-88, 1975-88, and 1981-88. The number of constituents and stations evaluated varied with the different time periods. The 39 monitoring stations included 26 freshwater streams with associated discharge data, 7 tidally affected streams, 4 harbor stations, and 2 surface impoundments. Flow-adjustment procedures were used where possible to minimize the effects of stream- flow variability on trend results.

The drainage area of the monitoring stations includes approximately 5,000 mi² covering the State of Connecticut and about 11,000 mi² in upstream drainage areas outside of the State. Drainage basin size for the freshwater streams ranges from 4.1 mi² to 9,660 mi². Land uses in the drainage basins range from undeveloped forested areas to highly urbanized metropolitan areas. During the period covered by the trend study, the State's population has grown, suburban development has increased, agricultural land use has decreased, and wastewater-treatment practices have improved.

Increases in specific conductance and in the concentrations of calcium, magnesium, chloride, sulfate, dissolved solids, and total solids were geographically widespread and numerous during water years 1975-88 and indicate a general increase statewide in dissolved constituents in streamflow, both in urbanized and less developed areas. The effects of increasing urbanization, including municipal and industrial wastewater, septic system leachate, nonpoint runoff, and atmospheric deposition of contaminants, are possible causes for these increases.

Decreases in turbidity and in the concentrations of total phosphorus, total organic carbon, and fecal coliform bacteria were geographically widespread and numerous during 1975-88. This general decrease in suspended material and bacteria may be attributable to basic improvements in the treatment of municipal and industrial wastewater during the period of record. Decreasing concentrations of total phosphorus may also be related to decreases in agricultural land use and to a decline in the use of detergents containing phosphorus. Detected decreases in total organic carbon and turbidity may have been caused, in part, by changes in sampling or analytical methods.

Increases in total nitrogen, total organic nitrogen, and total nitrite-plus- nitrate were geographically widespread and numerous during 1975-88 and appear to indicate effects from both point sources in urbanized basins and nonpoint sources in less developed basins. The number of stations with increasing concentrations of nitrogen constituents was much smaller during 1981-88 than during 1975-88. Decreases in total ammonia nitrogen were detected at 11 stations during 1981-88. Decreases in total ammonia, sometimes paired with increases in total nitrite-plus-nitrate, may result from improvements in wastewater treatment.

Increases in the concentration of dissolved oxygen, or dissolved oxygen as a percent of saturation, were geographically widespread and numerous during 1969-88 and 1975-88. Increases were less common during 1981-88. Increases in dissolved oxygen in urbanized basins may be related to major improvements in wastewater treatment during the 1970's and 1980's. The magnitude of the trends detected during 1969-88 may have been affected in part by a change, around 1974, in the model of the instrument used to measure dissolved oxygen in the field.

Statewide increases in pH were detected during 1969-88, 1975-88, and 1981-88, in both urbanized and less developed basins. The widespread increases in pH were unexpected, given the relatively acidic quality of precipitation in the region during the study period. Only two decreases in pH were detected, both in relatively undeveloped basins. Increases in pH in urbanized areas may be related to decreasing concentrations of ammonia and to requirements for neutralization of municipal and industrial wastewater.