Prepared in cooperation with the New England Interstate Water Pollution Control Commission

Assessment of Total Nitrogen in the Upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts, December 2002–September 2005

Scientific Investigations Report 2006–5144

U.S. Department of the Interior U.S. Geological Survey Cover. Photograph shows the Connecticut River and Mt. Ascutney, Vermont, in the background. Assessment of Total Nitrogen in the Upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts, December 2002–September 2005

By Jeffrey R. Deacon, Thor E. Smith, Craig M. Johnston, Richard B. Moore, Rebecca M. Weidman, and Laura J. Blake

Prepared in cooperation with the New England Interstate Water Pollution Control Commission

Scientific Investigations Report 2006–5144

U.S. Department of the Interior U.S. Geological Survey U.S. Department of the Interior DIRK KEMPTHORNE, Secretary U.S. Geological Survey P. Patrick Leahy, Acting Director

U.S. Geological Survey, Reston, Virginia: 2006

For product and ordering information: World Wide Web: http://www.usgs.gov/pubprod Telephone: 1-888-ASK-USGS For more information on the USGS--the Federal source for science about the Earth, its natural and living resources, natural hazards, and the environment: World Wide Web: http://www.usgs.gov Telephone: 1-888-ASK-USGS

Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Although this report is in the public domain, permission must be secured from the individual copyright owners to reproduce any copyrighted materials contained within this report.

Suggested citation: Deacon, J.R., Smith, T.E., Johnston, C.M., Moore, R.B., Weidman, R.M., and Blake, L.J., 2006, Assessment of total nitrogen in the Upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts, December 2002– September 2005: U.S. Geological Survey Scientific Investigations Report 2006–5144, 89 p. iii

Acknowledgments

The authors acknowledge and appreciate the efforts of their U.S. Geological Survey colleagues provided by Ann Chalmers, Kimberly Campo, Domenic Murino, Joseph Martin, Jason Pollender, Tim Sargent, and Guy Holzer for their assistance in the collection of water-quality samples, and Sarah Flanagan, U.S. Geological Survey, for the formation of computer files for load estimation. Appreciation also is extended to the Connecticut Water Science Center for the collection and use of data from the Connecticut River at Thompsonville, CT. Thanks are extended to Norman Spahr and John Mullaney, U.S. Geological Survey, for their critical review of this report. Appre­ ciation is extended to Kay E. Hedrick for the editorial report review, Ann Marie Squillacci for manuscript and layout, and Tina Cotton for graphics.

v

Contents

Abstract...... 1 Introduction...... 1 Purpose and Scope ...... 2 Previous Studies ...... 2 Description of the Study Area ...... 3 Data Collection and Analysis...... 3 Site Selection...... 3 Sample Collection...... 5 Data Analysis...... 5 Characterization of Total Nitrogen at River Sampling Sites ...... 7 Moose River at Victory, VT, Station 01134500...... 9 Passumpsic River at Passumpsic, VT, Station 01135500 ...... 14 Connecticut River at Wells River, VT, Station 01138500...... 19 Ayers Brook at Randolph, VT, Station 01142500...... 24 White River at West Hartford, VT, Station 01144000...... 29 Sugar River at West Claremont, NH, Station 01152500...... 34 Connecticut River at North Walpole, NH, Station 01154500 ...... 39 Otter River at Otter River, MA, Station 01163200...... 44 North River at Shattuckville, MA, Station 01169000...... 49 South River near Conway, MA, Station 01169900...... 54 Green River near Colrain, MA, Station 01170100...... 59 Mill River at Northhampton, MA, Station 01171500...... 64 Connecticut River at Thompsonville, CT, Station 01184000...... 69 Characterization of Total Nitrogen at Wastewater-Treatment Sampling Sites...... 74 Comparison of Total Nitrogen Concentrations and Yields Among Site Types...... 76 Summary and Conclusions...... 77 References Cited...... 78 Appendix A: Supplemental Data ...... 81 A-1. Annual loads, yields, and confidence intervals for total nitrogen and total phosphorus for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts ...... 82 A-2. Estimates of seasonal total nitrogen load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 85 A-3. Estimates of seasonal total phosphorus load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 87 A-4. Regression equations for estimates of total nitrogen load and total phosphorus load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 89 vi

Figures

1–2. Maps showing— 1. Sampling sites and site-identification numbers in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 4 2. Location, land-use classification, and photograph of sampling site for Moose River at Victory, VT, station 01134500...... 10 3–5. Graphs showing— 3. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Moose River at Victory, VT, station 01134500...... 11 4. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Moose River at Victory, VT, station 01134500 ...... 12 5. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Moose River near Victory, VT, station 01134500, in relation to all other river sampling sites...... 13 6. Map showing location, land-use classification, and photograph of sampling site for Passumpsic River at Passumpsic, VT, station 01135500...... 15 7–9. Graphs showing— 7. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Passumpsic River at Passumpsic, VT, station 01135500 ...... 16 8. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Passumpsic River at Passumpsic, VT, station 01135500...... 17 9. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Passumpsic River at Passumpsic, VT, station 01135500, in relation to all other river sampling sites ...... 18 10. Map showing location, land-use classification, and photograph of sampling site for Connecticut River at Wells River, VT, station 01138500 ...... 20 11–13. Graphs showing— 11. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Connecticut River at Wells River, VT, station 01138500...... 21 12. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Connecticut River at Wells River, VT, station 01138500 ...... 22 13. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Connecticut River at Wells River, VT, station 01138500, in relation to all other river sampling sites ...... 23 14. Map showing location, land-use classification, and photograph of sampling site for Ayers Brook at Randolph, VT, station 01142500 ...... 25 vii

15–17. Graphs showing— 15. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Ayers Brook at Randolph, VT, station 01142500 ...... 26 16. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Ayers Brook at Randolph, VT, station 01142500 ...... 27 17. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Ayers Brook at Randolph, VT, station 01142500, in relation to all other river sampling sites...... 28 18. Map showing location, land-use classification, and photograph of sampling site for White River at West Hartford, VT, station 01144000 ...... 30 19–21. Graphs showing— 19. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for White River at West Hartford, VT, station 01144000...... 31 20. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for White River at West Hartford, VT, station 01144000 ...... 32 21. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for White River at West Hartford, VT, station 01144000, in relation to all other river sampling sites ...... 33 22. Map showing location, land-use classification, and photograph of urban area upstream of sampling site for Sugar River at West Claremont, NH, station 01152500....35 23–25. Graphs showing— 23. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Sugar River at West Claremont, NH, station 01152500...... 36 24. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Sugar River at West Claremont, NH, station 01152500 ...... 37 25. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Sugar River at West Claremont, NH, station 01152500, in relation to all other river sampling sites ...... 38 26. Map showing location, land-use classification, and photograph of sampling site for Connecticut River at North Walpole, NH, station 01154500...... 40 27–29. Graphs showing— 27. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Connecticut River at North Walpole, NH, station 01154500 ...... 41 28. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Connecticut River at North Walpole, NH, station 01154500...... 42 29. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Connecticut River at North Walpole, NH, station 01154500, in relation to all other river sampling sites ...... 43 viii

30. Map showing location, land-use cover classification, and photograph of sampling site for Otter River at Otter River, MA, station 01163200 ...... 45 31–33. Graphs showing— 31. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Otter River at Otter River, MA, station 01163200 ...... 46 32. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Otter River at Otter River, MA, station 01163200 ...... 47 33. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Otter River at Otter River, MA, station 01163200, in relation to all other river sampling sites...... 48 34. Map showing location, land-use classification, and photograph of sampling site for North River at Shattuckville, MA, station 01169000 ...... 50 35–37. Graphs showing— 35. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for North River at Shattuckville, MA, station 01169000...... 51 36. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for North River at Shattuckville, MA, station 01169000...... 52 37. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for North River at Shattuckville, MA, station 01169000, in relation to all other river sampling sites ...... 53 38. Map showing location, land-use classification, and photograph of sampling site for South River near Conway, MA, station 01169900 ...... 55 39–41. Graphs showing— 39. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for South River near Conway, MA, station 01169900...... 56 40. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for South River near Conway, MA, station 01169900...... 57 41. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for South River near Conway, MA, station 01169900, in relation to all other river sampling sites...... 58 42. Map showing location, land-use classification, and photograph of sampling site for Green River near Colrain, MA, station 01170100 ...... 60 43–45. Graphs showing— 43. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Green River near Colrain, MA, station 01170100...... 61 44. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Green River near Colrain, MA, station 01170100...... 62 ix

45. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Green River near Colrain, MA, station 01170100, in relation to all other river sampling sites...... 63 46. Map showing location, land-use classification, and photograph of sampling site for Mill River at Northampton, MA, station 01171500 ...... 65 47–49. Graphs showing— 47. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Mill River at Northampton, MA, station 01171500...... 66 48. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Mill River at Northampton, MA, station 01171500...... 67 49. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Mill River at Northampton, MA, station 01171500, in relation to all other river sampling sites ...... 68 50. Map showing location, land-use classification, and photograph of sampling site for Connecticut River at Thompsonville, CT, station 01184000 ...... 70 51–55. Graphs showing— 51. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Connecticut River at Thompsonville, CT, station 01184000...... 71 52. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phosphorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Connecticut River at Thompsonville, CT, station 01184000...... 72 53. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Connecticut River at Thompsonville, CT, station 01184000, in relation to all other river sampling sites ...... 73 54. A, Mean total nitrogen concentration, and B, Instantaneous mean total nitrogen load for wastewater-treatment sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts ...... 75 55. A, Total nitrogen concentrations among site types in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 76

Tables

1. Selected characteristics for the study sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 6 2. Mean annual loads, yields, confidence intervals, standard error of prediction, and the ratio of the standard error of prediction to the mean annual load for total nitrogen and total phosphorus at river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 8 3. Selected characteristics and summary statistics for concentrations of total nitrogen and instantaneous loads of total nitrogen for the wastewater-treatment sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts...... 74 x

Conversion Factors, Datums, and Abbreviated Water-Quality Units

Multiply By To obtain Length inch (in.) 2.54 centimeter (cm) foot (ft) 0.3048 meter (m) mile (mi) 1.609 kilometer (km) Area acre 0.004047 square kilometer (km2) square foot (ft2) 0.09290 square meter (m2) square mile (mi2) 2.590 square kilometer (km2) Flow rate cubic foot per second (ft3/s) 0.02832 cubic meter per second (m3/s) Weight pound per year (lb/yr) 0.4536 kilogram per year (kg/yr) pound per square mile (lb/mi2) 0.2819 kilogram per square kilometer (kg/km2)

Temperature in degrees Celsius (°C) may be converted to degrees Fahrenheit (°F) as follows: °F=(1.8×°C)+32 Vertical coordinate information is referenced to the North American Vertical Datum of 1988 (NAVD 88). Horizontal coordinate information is referenced to the North American Datum of 1983 (NAD 83). Specific conductance is given in microsiemens per centimeter at 25 degrees Celsius (μS/cm at 25°C). Abbreviated water-quality units used in this report: Chemical concentrations in water are reported in milligrams per liter (mg/L). Milligrams per liter is a unit expressing the concentration of chemical constituents as weight (milligrams) of chemical per unit volume (liter) of water. Assessment of Total Nitrogen in the Upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts, December 2002–September 2005

By Jeffrey R. Deacon1, Thor E. Smith1, Craig M. Johnston1, Richard B. Moore1, Rebecca M. Weidman2, and Laura J. Blake2

Abstract other site types (p<0.05). Concentrations of total nitrogen at agricultural, urban, and main stem sites were not significantly A study of total nitrogen concentrations and loads was different among each other (p>0.05) but were significantly conducted from December 2002 to September 2005 at 13 river greater (p<0.05) than at forested sites and significantly less sites in the upper Connecticut River Basin. Ten sites were than concentrations at wastewater-treatment sites (p<0.05). selected to represent contributions of nitrogen from forested, Total nitrogen concentrations at wastewater-treatment sites agricultural, and urban land. Three sites were distributed were significantly different from all other site types (p<0.05). spatially on the main stem of the Connecticut River to assess Annual yields of total nitrogen ranged from 732 to 2 2 the cumulative total nitrogen loads. To further improve the 1,920 (lb/mi )/yr at forested sites; 1,550 to 2,980 (lb/mi )/yr 2 understanding of the sources and concentrations and loads of at agricultural sites; 1,280 to 1,860 (lb/mi )/yr at urban sites total nitrogen in the upper Connecticut River Basin, ambient that were not directly affected by wastewater effluent; 7,090 to 2 surface water-quality sampling was supplemented with sam­ 7,770 (lb/mi )/yr at an urban site directly affected by wastewa­ 2 pling of effluent from 19 municipal and paper mill wastewater- ter effluent; and 1,300 to 2,390 (lb/mi )/yr at main-stem sites. treatment facilities. In this study, the mean annual load and yield of total Mean concentrations of total nitrogen ranged from 0.19 nitrogen at the Connecticut River at Wells River, VT, was 2 to 2.8 milligrams per liter (mg/L) at river sampling sites. estimated at 4.47 million lb/yr and 1,690 (lb/mi )/yr, respec­ Instantaneous mean loads of total nitrogen ranged from 162 tively. The mean annual load and yield of total nitrogen at the to 58,300 pounds per day (lb/d). Estimated mean annual loads Connecticut River at North Walpole, NH, was estimated at 2 of total nitrogen ranged from 49,100 to 21.6 million pounds 9.60 million lb/yr and 1,750 (lb/mi )/yr, respectively. The per year (lb/yr) with about 30 to 55 percent of the loads being mean annual load and yield of total nitrogen leaving the transported during the spring. The estimated mean annual upper Connecticut River Basin, as estimated at the Connecti­ yields of total nitrogen ranged from 1,190 to 7,300 pounds per cut River at Thompsonville, CT, was 21.6 million lb/yr and 2 square mile per year (lb/mi2)/yr. 2,230 (lb/mi )/yr, respectively. Mean concentrations of total nitrogen ranged from 4.4 to 30 mg/L at wastewater-treatment sampling sites. Instantaneous mean loads of total nitrogen from municipal wastewater-treat­ Introduction ment facilities ranged from 36 to 1,780 lb/d. Instantaneous mean loads of total nitrogen from paper mill wastewater- A long-term study was initiated in 1985 to protect and treatment facilities ranged from 96 to 160 lb/d. restore the health of Long Island Sound (LIS) (U.S. Environ­ The median concentration of total nitrogen was mental Protection Agency, State of Connecticut, State of New 0.24 mg/L at forested sites, 0.48 mg/L at agricultural sites, York, Interstate Sanitation Commission, National Oceanic and 0.54 mg/L at urban sites, 0.48 mg/L at main-stem sites, and Atmospheric Administration, 1985; New York Department of 14 mg/L at wastewater-treatment sites. Concentrations of total Environmental Conservation/Connecticut Department of Envi­ nitrogen at forested sites were significantly less than at all ronmental Protection, 2000). Low concentrations (less than 3.5 mg/L) of dissolved oxygen (hypoxia) as a result of nitro­ 1U.S. Geological Survey, New Hampshire/Vermont Water Science Center, gen enrichment often occur during the summer in the western Pembroke, New Hampshire. part of LIS. Excess nitrogen levels can lead to the enhanced 2New England Interstate Water Pollution Control Commission, Lowell, growth of algae, which eventually die and decay, consuming Massachusetts. oxygen in the process. The ensuing hypoxic conditions have 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005 resulted in the exceedance of water-quality standards in LIS cut River Basin. Estimates of mean annual load and yield of during the summer, with oxygen levels often falling below 1 total nitrogen from river sites and total nitrogen data relative to or 2 mg/L. Nitrogen sources include nonpoint-source runoff land use and land-cover classification also are included. from agricultural and urban areas, atmospheric deposition of Twelve river sites and 19 wastewater-treatment sites were nitrogen, and direct and indirect discharges from wastewater sampled in New Hampshire, Vermont, and Massachusetts, and (Mullaney and others, 2002). one river site was sampled in Connecticut on the Connecticut In response to this problem, Connecticut and New River at Thompsonville, CT. All river water-quality sampling York developed a Total Maximum Daily Load (TMDL) sites were co-located with a USGS streamflow-gaging station that specifies the maximum amount of nitrogen that can (referred to as a “station” in this report). References made be discharged to LIS. According to the TMDL analysis, to land use in the remaining sections of this report include an annual baseline load of about 53,300 tons of nitrogen is land use and land cover. Field measurements of streamflow, delivered to LIS from all sources in Connecticut and New specific conductance, pH, water temperature, and concentra­ York combined. An additional estimated 13,600 tons of tions of dissolved oxygen, and nitrogen, phosphorus, and sus­ nitrogen reach LIS annually from the part of the Connecticut pended-sediment data are presented in the report. The primary River Basin originating north of the State of Connecticut. focus of this report is the assessment of concentrations, loads, Of this total load, an estimated 12,500 tons of nitrogen is and yields of total nitrogen at selected sites; however, concen­ delivered to LIS by way of the Connecticut River. Total trations, loads, and yields of total phosphorus and concentra­ nitrogen loads account for terrestrial, atmospheric, and effluent tions of other water-quality constituents also are presented. from wastewater-treatment facilities. The TMDL specifies a 58.5-percent reduction in human-generated nitrogen from point and nonpoint sources. The TMDL requires Connecticut Previous Studies to remove about 6,300 tons of nitrogen delivered to LIS each year from wastewater dischargers located throughout the State, Several previous studies have provided estimates of as well as another 400 tons of nitrogen resulting from nonpoint nitrogen and phosphorus loads in the Connecticut River Basin sources. The required nitrogen load reduction from wastewater and in the northeastern United States. Trench (1999) provides discharge and other sources in New York is about 17,200 tons a summary of total nitrogen and total phosphorus loads per year. for 25 monitoring stations in the Connecticut River Basin The LIS TMDL also specifies the need for additional through 1995. The estimated load of total nitrogen ranged actions beyond the nitrogen load reductions in Connecticut from 15 million pounds in 1995 to about 41 million pounds in and New York. In particular, nitrogen reduction actions are 1984 at the USGS Connecticut River at Thompsonville, CT, needed for other sources, including sources of nitrogen that water-quality monitoring station. That study also estimates originate in States north of Connecticut (New Hampshire, yields of total nitrogen and total phosphorus for basins Vermont, and Massachusetts) that drain to LIS. representing urban, agricultural, and forested land use. The The New England Interstate Water Pollution Control estimated yields of total nitrogen at the USGS Connecticut Commission (NEIWPCC) is assisting the U.S. Environmental River at Thompsonville, CT, water-quality monitoring Protection Agency (USEPA) and the States of Connecticut, station ranged from 1,550 pounds per square mile in 1995 Massachusetts, New Hampshire, and Vermont to better to 4,210 pounds per square mile in 1984. Mullaney and define nitrogen sources and loads in the upper Connecticut others (2002) estimated annual loads of nonpoint nitrogen River Basin (drainage basin upstream from the Connecticut- for 1988–98 at 28 monitoring stations and 26 unmonitored Massachusetts state line). In order to define these sources, basins that drain to LIS. A multiple-linear regression current data on nitrogen concentrations and loads are needed. equation was developed, on the basis of total nitrogen yields To assess the current nitrogen loads, the U.S. Geological and associated basin characteristics from monitored sites Survey (USGS), in cooperation with the NEIWPCC, designed and applied to unmonitored and other monitored sites, for and operated a surface water-quality monitoring network from estimating nonpoint nitrogen yields. The estimated nonpoint December 2002 to September 2005 to collect water-quality nitrogen load to LIS ranged from 21 million pounds in 1995 and streamflow data to determine the current concentrations to 50 million pounds in 1990 (Mullaney and others, 2002). and loads of nitrogen from selected sites representing different Mullaney (2004) provides a summary of water-quality trends land uses and from sites representing effluent from wastewater in the Connecticut River, including trends in concentrations treatment in the upper Connecticut River Basin. of total nitrogen. Since 1988, downward trends in total nitrogen were observed at the long-term USGS water-quality Purpose and Scope monitoring station at Connecticut River at Thompsonville, CT. The downward trend likely is a result of improved nitrogen This report summarizes concentrations of total nitrogen removal from wastewater-treatment facilities but may relate and other water-quality constituent data collected at 13 river to changes in land use, such as agricultural land changing to sites and 19 wastewater-treatment sites in the upper Connecti­ residential or forested land (Mullaney, 2004). Data Collection and Analysis 

Regional nitrogen modeling studies include application of valley forms a broad lowland just south of the New Hamp­ the USGS New England Spatially Referenced Regression on shire-Vermont-Massachusetts state lines to within about 20 mi Watershed Attributes (SPARROW) model, which uses regres­ of LIS (Grady and Garabedian, 1991). sion equations to relate total nitrogen stream loads to nutrient Average annual temperatures range from less than sources and watershed characteristics. Significant predictor 5 degrees Celsius (oC) in the northern mountainous regions to variables of total nitrogen included (1) nitrogen loadings from about 10oC in the southern regions. Average annual precipita­ permitted municipal wastewater discharge, (2) atmospheric tion ranges from about 34 inches (in.) in the northern areas of deposition, (3) the area of agricultural land, and (4) the area the Connecticut River Basin to more than 65 in. in some of the of urban land (Moore and others, 2004). The estimated total mountainous regions (Garabedian and others, 1998). nitrogen load at the mouth of the Connecticut River for For purposes of this report, the upper Connecticut 1992–93 was 17,800 tons. The estimated total nitrogen load at River Basin is defined as the drainage basin above the USGS the Connecticut River at Thompsonville, CT, was 14,100 tons streamgaging station at Connecticut River at Thompsonville, for 1992–93. Results of simulations made with the SPARROW CT (site 13, fig. 1). The station is near the Connecticut- model also indicated that total nitrogen stream-loss rates were Massachusetts State line and serves as the outlet site to the significant only in streams with average annual flows less than upper basin. Streamflow at the station averages about or equal to 100 cubic feet per second (ft3/s). Seitzinger and 12,700 ft3/s. The upper Connecticut River Basin encompasses others (2002) developed a regression model that predicts the approximately 9,660 mi2 in New Hampshire, Vermont, and amount of nitrogen removed from rivers and its application to Massachusetts (fig. 1). The population is approximately drainages in the northeastern United States. Results from this 1 million people and is distributed amongst densely populated, study indicated that as much as 37 to 76 percent of nitrogen urban areas in the southernmost section of the basin in inputs to rivers is removed during transport and about half of Massachusetts to sparsely populated, rural and agricultural that is removed in first- to fourth-order streams, accounting for regions in the northern areas of the basin in New Hampshire 90 percent of the total stream length in the basins. and Vermont. The agricultural land use in New Hampshire and Mueller and others (1995) compiled historical data on Vermont is predominantly related to dairy farm operations. nutrient concentrations (nitrogen and phosphorus) from 20 The agricultural land use in Massachusetts is predominantly study units of the USGS National Water-Quality Assessment related to orchards, row crops, and some dairy operations. The Program. More than 220,000 surface-water samples were land use and land cover in the upper basin consists of about analyzed on regional and national scales. Results of the study 80 percent forest, 9 percent agriculture, 5 percent urban, and indicated that nutrient concentrations in surface water gener­ 6 percent wetlands and barren. ally were related to land use. Nitrate concentrations were highest in samples from sites downstream from agricultural or urban areas. Results also indicated the elevated concentrations Data Collection and Analysis of nitrate in surface water in the northeastern United States may be related to large amounts of atmospheric deposition. Data were collected for this study to improve the understanding of current total nitrogen concentrations and Description of the Study Area loads at selected sites and to estimate the contributions origi­ nating from different land uses and the effluent from waste­ The Connecticut River Basin, the largest in New England, water-treatment facilities in the basin. Field measurements of encompasses 11,250 square miles (mi2) (fig. 1). The Connecti­ streamflow, specific conductance, pH, water temperature, and cut River traverses 410 miles (mi) from its headwaters near the concentrations of dissolved oxygen, were made, and nitrogen, New Hampshire-Canada border to its mouth in Old Saybrook, phosphorus, and suspended-sediment concentration data were CT, where it drains into Long Island Sound. The river drops collected. Total nitrogen was calculated as the sum of nitrite 2,400 feet (ft) in elevation from its source to its outlet, and plus nitrate and total ammonia plus organic nitrogen. has an estimated average discharge of about 19,200 ft3/s at the outlet (Garabedian and others, 1998). The Connecticut River Site Selection makes up 70 percent of the freshwater flow into LIS . Elevations in the Connecticut River Basin range from sea Approximately 80 stream sites initially were considered level in coastal Connecticut to 6,288 ft above sea level at the as potential sampling sites. The sites were categorized by peak of Mount Washington in the White Mountains of New stream size, percentage of land-use types, and estimated Hampshire. Topography is characterized by rolling hills and total nitrogen loads from the New England SPARROW low, rounded mountains with numerous, generally narrow model (Moore and others, 2004). The final selection of the valleys. The relief is greatest in the northern part of the basin sites was based on topographic map inspections and field where the Green Mountains of Vermont and the White Moun­ reconnaissance, the existing USGS streamgaging network in tains of New Hampshire generally reach elevations between the upper Connecticut River Basin, and whether water-quality 2,000 and 4,000 ft above sea level. The Connecticut River data had previously been collected at the site. The process also  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

73° 72° 71° 70°

45° CANADA U.S.A

14

Burlington 1 15 16 2 17 NEW MAINE Montpelier HAMPSHIRE 18 3 44° EXPLANATION 4 VERMONT Connecticut River Basin boundary 5 19 above Thompsonville, CT 20 21 Connecticut River Basin boundary 22 6 below Thompsonville, CT 23 Concord Rivers 24 7 43° 4 River sampling site and number 25 Manchester 26 15 Wastewater-treatment NEW sampling site and number YORK 11 9 27 8 10 28 MASSACHUSETTS Boston Atlantic Ocean 12 Worcester 29 30 Springfield 31 42° 32 Thompsonville 13 CONNECTICUT Providence Rhode Hartford Island

Waterbury 0 15 30 60 MILES

0 15 30 60 KILOMETERS New Haven Old Saybrook Long Island Sound

Shaded relief from U.S. Geological Survey 30-meter National Eleva ion Dataset, 1999

Figure 1. Sampling sites and site-identification numbers in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts. (Site identification numbers are shown in table 1.) Data Collection and Analysis 5 included an analysis of the New England SPARROW model Water Information System (NWIS) database and are published results that estimated the percentage of land-use classifications in USGS annual Water-Data Reports (Keirstead and others, contributing nitrogen to the stream site. Thirteen river sites 2003, 2004; Kiah and others, 2005; Socolow and others, 2003, were selected for the study (table 1, fig. 1). Ten sites were 2004, 2005). selected to represent contributions of nitrogen from forested, agricultural, and urban land. Forest was the predominant land use at all sites in the basin. Sites that were selected Data Analysis to represent agricultural and urban land use in this study Selected results of the data collection and water-quality generally had agriculture and urban land use as the second analyses for each of the 13 river sampling sites are presented predominant land use in the basin. Three sites were distributed in the following sections. For each river site, water-quality spatially on the main stem of the Connecticut River. The conditions are described, and a map and three pages of graphs Connecticut River at Thompsonville, CT, was sampled as part of selected data are presented. Ranges of constituent values not of the USGS Connecticut Water Science Center programs. plotted in the graphs are statistically summarized in the text. Ambient water-quality sampling at river sites was For nitrite, ammonia, and orthophosphate summary statistics, supplemented with the collection of effluent samples from 19 data were re-censored from the laboratory reporting level wastewater-treatment facilities in New Hampshire, Vermont, (LRL) to the long-term method detection level (LT-MDL). Re- and Massachusetts (table 1, fig. 1). Sixteen sampling sites censoring provides useful information for constituents that are were located at municipal wastewater-treatment facilities and censored at the laboratory reporting level but have qualified three sites were located at paper mill wastewater-treatment values between the LT-MDL and the LRL (Helsel, 2005). For facilities. The facilities were selected on the basis of size, nitrite, the LRL was 0.008 mg/L and the LT-MDL was type, and location in the basin. The samples of effluent from 0.004 mg/L; therefore, nitrite data were re-censored at wastewater-treatment facilities, referred to as wastewater- 0.004 mg/L. For ammonia, the LRL was 0.010 mg/L and treatment sampling sites in this report, will provide an estimate the LT-MDL was 0.005 mg/L; therefore, ammonia was re- of the contribution of total nitrogen from the effluent from censored at 0.005 mg/L. For orthophosphate in the years 2003 these facilities. and 2004, the LRL was 0.02 mg/L and the LT-MDL was 0.01 mg/L; therefore, orthophosphate was re-censored at Sample Collection 0.01 mg/L for data collected in 2003 and 2004. Ortho­ phosphate was analyzed at a lower level during 2005 when Samples were collected monthly from December the LRL was 0.006 mg/L and the LT-MDL was 0.003 mg/L; 2002 through September 2005 at each of the 13 river sites therefore, orthophosphate was re-censored at 0.003 mg/L by using field procedures outlined in Wilde and Radtke for data collected in 2005. Long-term method detection and (1998a,b). Forty to 45 water samples were collected at each laboratory reporting levels for samples analyzed at the USGS river sampling site. Samples were collected in baseflow, National Water Quality Laboratory are described by Childress periodic rainstorm, and snowmelt conditions. About three and others (1999). additional samples were collected during the snowmelt For each river site, the illustrations begin with a map of runoff period each year to characterize the period when the site showing the major stream network and the land-use constituent concentrations might be more variable. Ten to 15 classifications (U.S. Geological Survey, 2000). Information samples were collected at the wastewater-treatment sampling about land use assists in the understanding of natural and sites, generally from April through September, to assist in anthropogenic factors that affect water quality. Two graphs providing current total nitrogen data from the effluent from are presented on the second page of illustrations to describe these facilities. During each site visit, onsite measurements the streamflow and the distribution of water-quality samples were made for specific conductance, pH, water temperature, in relation to streamflow. In the first graph, annual streamflow and concentrations of dissolved oxygen, and water samples is described by use of a bar chart. Streamflow for each water were collected for subsequent laboratory analysis of the year of the study period is represented by one bar in the graph. concentrations of dissolved nitrite plus nitrate, dissolved A water year is the period from October through September ammonia, total ammonia plus organic nitrogen, dissolved and is defined by the year in which it ends. Historical data orthophosphate, total phosphorus, and suspended-sediment. also are provided for the period prior to data collection for Instantaneous streamflow was determined at the time of this study. A comparison of historical data to data collected sample collection using established USGS stage—discharge for this study indicates in general terms whether runoff during relations at streamgaging stations for river sites and from the the data-collection period was above average, below average, facilities gage at wastewater-treatment sites. Water samples or normal. In the second graph, daily mean streamflow for were analyzed at the USGS water-quality laboratory in the data-collection period as well as dates and instantaneous Lakewood, CO. Suspended-sediment concentration analysis streamflow for the water-quality samples are presented. was conducted at the USGS Kentucky sediment laboratory in Because the water-quality samples are displayed with the Louisville, KY. All data are available from the USGS National corresponding instantaneous streamflow, these points do not  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Table 1. Selected characteristics for the study sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts.

[USGS, U.S. Geological Survey; mi2, square miles; WY, water year; na, not applicable; WWTF, wastewater-treatment facility; --, no data]

Mean Mean Site annual annual Drainage num­ USGS station Site type in stream- stream- Station name Latitude Longitude area ber number this study flow flow (mi2) (fig. 1) (WY (WY 1980–2002) 200–05) (in thousands of River sampling sites acre-feet) 1 01134500 Moose River at Victory, VT Forest 44°30'42" 71°50'13" 75.2 111 107 2 01135500 Passumpsic River at Agriculture 44°21'56" 72°02'23" 436 572 554 Passumpsic, VT 3 01138500 Connecticut River at Main stem 44°09'13" 72°02'34" 2,644 3,750 3,560 Wells River, VT 4 01142500 Ayers Brook at Randolph, VT Agriculture 43°56'04" 72°39'30" 30.5 37.3 38.2 5 01144000 White River at West Hartford, VT Agriculture 43°42'51" 72°25'07" 690 885 959 6 01152500 Sugar River at West Urban 43°23'15" 72°21'45" 269 304 341 Claremont, NH 7 01154500 Connecticut River at North Main stem 43°07'34" 72°26'14" 5,493 7,160 7,070 Walpole, NH 8 01163200 Otter River at Otter River, MA Urban 42°35'18" 72°02'29" 34.1 45.3 53.4 9 01169000 North River at Shattuckville, MA Agriculture 42°38'18" 72°43'32" 89.0 138 167 10 01169900 South River near Conway, MA Agriculture 42°32'31" 72°41'39" 24.1 37.1 48.1 11 01170100 Green River near Colrain, MA Forest 42°42'12" 72°40'16" 41.4 61.4 74.4 12 01171500 Mill River at Northhampton, MA Urban 42°19'08" 72°39'56" 54.0 73.0 88.4 13a 01184000 Connecticut River at Main stem 41°59'14" 72°36'21" 9,660 12,700 13,000 Thompsonville, CT Point-source sampling sites (in millions of gallons) 14 443558071303001 Wasau Paper WWTF, NH Point source 44°35'58" 71°30'30" na na 1,670 15 442906071354401 Lancaster WWTF, NH Point source 44°29'06" 71°35'44" na na -- 16 442436072010001 St. Johnsbury WWTF, VT Point source 44°24'36" 72°01'00" na na 407 17 441827071473701 Littleton WWTF, NH Point source 44°18'27" 71°47'37" na na 303 18 440835072022501 Woodsville WWTF, NH Point source 44°08'35" 72°02'25" na na 63 19 434145072175601 Hanover WWTF, NH Point source 43°41'45" 72°17'56" na na 531 20 433844072185501 Hartford/White River Junction Point source 43°38'44" 72°18'55" na na 365 WWTF, VT 21 433813072192001 Lebanon WWTF, NH Point source 43°38'13" 72°19'20" na na 659 22 432359072234001 Claremont WWTF, NH Point source 43°23'59" 72°23'40" na na 578 23 431648072280801 Springfield WWTF, VT Point source 43°16'48" 72°28'08" na na -- 24 430745072263701 Bellows Falls WWTF, VT Point source 43°07'45" 72°26'37" na na 206 25 425322072323701 Fibermark Products WWTF, VT Point source 42°53'22" 72°32'37" na na 518 26 425029072325901 Brattleboro WWTF, VT Point source 42°50'29" 72°32'59" na na 575 27 423542072031001 Seaman Paper WWTF, MA Point source 42°35'42" 72°03'10" na na 329 28 423413072011201 Gardner WWTF, MA Point source 42°34'13" 72°01'12" na na 1,300 29 421132072365001 Holyoke WWTF, MA Point source 42°11'32" 72°36'50" na na 3,270 30 420908072373301 Chicopee WWTF, MA Point source 42°09'08" 72°37'33" na na 3,590 31 420702072435601 Westfield WWTF, MA Point source 42°07'02" 72°43'56" na na 1,400 32 420508072351401 Springfield WWTF, MA Point source 42°05'08" 72°35'14" na na 15,900 a Sampled by the USGS Connecticut Water Science Center. Characterization of Total Nitrogen at River Sampling Sites  always fall on the line representing the daily mean discharge. blank samples indicated that concentrations of constituents The distribution of water-quality samples with respect to time discussed in this report were less than the LRL. Replicate also is represented. samples provide information on the variability of analytical The third page of illustrations presents the distribution results caused by sample collection, processing, and analysis of selected water-quality constituents at the various sites. (Spahr and Boulger, 1997). Differences in concentrations of Concentrations of selected constituents are shown in relation the constituents of interest in environmental and replicate to streamflow and instantaneous loads of total nitrogen are samples at river sampling sites were within 0.01 mg/L and shown in relation to time. Box plots in the right margins of the 0.05 mg/L for nitrite plus nitrate and total ammonia plus graphs summarize the distribution of concentrations or loads organic nitrogen, respectively. This resulted in concentration for each constituent. differences in environmental and replicate samples for total Two bar graphs are presented on the fourth page of illus­ nitrogen being within 0.1 mg/L. Differences in concentrations trations to describe the total nitrogen loads and mean annual in environmental and replicate samples at wastewater- yields. The first graph is the estimated total nitrogen load by sampling sites, where concentrations generally were greater, water year for the river site, and the second graph presents the were within 10 percent for nitrite plus nitrate and total mean annual nitrogen yield (2003–05) compared to all other ammonia plus organic nitrogen. This resulted in differences river sampling sites. in environmental and replicate samples for total nitrogen Selected results of the analysis of total nitrogen in also being within 10 percent. Differences in concentrations samples collected at the wastewater-treatment sampling sites in environmental and replicate samples for other constituents are presented as summary statistics in the text and as bar discussed in this report were low. It was concluded that sample graphs in the illustrations. The concentrations of total nitrogen processing and analysis did not introduce enough variation in among and in relation to site types are summarized in box the environmental data to affect interpretation of results. plots. The Kruskal-Wallis test, a nonparametric analysis of variance test (ANOVA), was used to determine significant statistical differences among groups of data (Helsel and Hirsch, 1992). The level of significance for ANOVA was Characterization of Total Nitrogen at set at alpha equals 0.05. When the nonparametric ANOVA River Sampling Sites indicated significant statistical differences, ranked transformed data were used for the Tukey’s multiple comparison test to Mean concentrations of total nitrogen at river sampling determine which groups were significantly different (Helsel sites ranged from 0.19 to 2.8 mg/L and mean instantaneous and Hirsch, 1992). All statistical analyses were performed loads of total nitrogen ranged from 162 to 58,300 pounds using Statview and SAS statistical software (SAS Institute, per day (lb/d). Estimated mean annual loads of total nitrogen Inc., 1998, 2000). ranged from 49,100 to 21.6 million pounds per year (lb/yr) The LOAD ESTimator (LOADEST) computer with about 30 to 55 percent of the load being transported program (Runkel and others, 2004) was used to estimate during the spring. The estimated mean annual yields of total total nitrogen and total phosphorus loads at the study sites. nitrogen ranged from 1,190 to 7,300 (lb/mi2)/yr. LOADEST is based on a regression equation in which time In the following sections, selected constituent series of streamflow and constituent concentrations are used concentrations and loads of total nitrogen are summarized as calibration data for the estimation of constituent loads. in the text and are plotted in graphs to examine relations to Explanatory variables within the regression model include streamflow and time at each site. Ranges of field measurement various functions of streamflow and time (seasonality). For values and other water-quality constituent values are presented total nitrogen or total phosphorus at each site, models were in the text. Estimated loads and yields of total nitrogen and fit using all possible combinations of these variables and the total phosphorus at river sampling sites are presented in best model was selected based on the Akaike Information table 2. The 95-percent confidence intervals of the mean load, Criteria (Akaike, 1981). The specific method used for model the standard error of prediction, and the ratio of the standard calibration and load estimation is described by Runkel and error of prediction to the mean load also are presented in others (2004). Data on annual loads, yields, confidence table 2. The ratio of the standard error of prediction to the intervals, seasonal load estimates, and the regression equations mean load standardizes the model error and provides a used in LOADEST for total nitrogen and total phosphorus comparison among sites having large differences in load estimates are in appendix tables A-1 through A-4. estimates. The prediction error ranged from 2.2 to 10 percent Quality-control procedures included analyses of of the mean loads for total nitrogen estimates and from field blank and replicate samples. Field blanks provide 9.7 to 54 percent of the mean loads for total phosphorus information on bias or the potential for contamination of estimates. This indicates that the models had low error in the analytical results by sample collection, processing, and estimates of total nitrogen and more error in the estimates of analysis (Spahr and Boulger, 1997). Analyses of the field total phosphorus. 8 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005 5.3 4.2 2.2 3.5 2.7 3.9 4.1 3.9 7.0 4.4 3.0 3.0 9.7 10 20 16 10 44 33 11 10 10 34 54 28 11 (percent) to mean load Ratio of standard error of prediction 679 325 6,470 2,660 9,790 2,350 5,070 2,860 1,650 2,790 1,520 7,180 2,580 35,000 97,500 34,300 14,600 13,600 11,600 24,200 17,600 64,200 error (lb/yr) 400,000 650,000 116,000 189,000 Standard of prediction 3,990 4,020 82,300 59,800 61,800 12,000 83,900 32,100 18,100 38,000 11,400 138,000 901,000 413,000 270,000 229,000 103,000 100,000 285,000 626,000 765,000 (lb/yr) Upper 4,690,000 1,360,000 2,190,000 10,500,000 22,900,000 ermont, and Massachusetts. interval 5,530 1,580 1,360 2,740 otal nitrogen 67,900 48,000 38,500 88,800 54,600 16,300 21,100 12,200 10,300 T otal phosphorus tandard error of prediction to the mean annual load for 106,000 749,000 335,000 230,000 163,000 191,000 177,000 512,000 (lb/yr) T Lower 95-percent confidence 4,250,000 1,170,000 8,820,000 1,450,000 20,300,000 )/yr 2 89 97 58 62 110 172 507 114 438 235 200 185 ield Y 1,610 1,890 1,690 2,450 1,830 1,380 1,750 7,300 2,180 2,230 1,190 1,770 2,230 1,320 , pounds per square mile year] (lb/mi )/yr 2 8,300 4,810 2,420 3,340 74,900 53,700 49,100 95,400 74,800 40,300 26,200 15,000 20,900 121,000 822,000 372,000 249,000 194,000 234,000 350,000 626,000 (lb/yr) 4,470,000 1,260,000 9,600,000 1,790,000 Mean load 21,600,000 7 9 23 72 41 57 13 onnecticut River Basin in New Hampshire, V 331 205 682 531 147 134 262 205 642 111 959 , pounds per year; (lb/mi 2,250 3,460 1,020 1,720 4,900 (lb/d) 12,200 26,300 59,000 Mean load y; lb/yr e , VT , VT ville, CT ville, CT er er alpole, NH alpole, NH v v , MA , MA , MA , MA er er ay ay ells Ri ells Ri , VT , VT v v assumpsic, VT assumpsic, VT ictory ictory est Hartford, VT est Hartford, VT est Claremont, NH est Claremont, NH er at W er at North W er at Thompson er at W er at North W er at Thompson er at P er at P v v v v v v v v Station name er at V er at V er at W er at W er near Colrain, MA er near Colrain, MA er near Conw er near Conw er at Shattuckville, MA er at Shattuckville, MA er at W er at W v v er at Otter Ri er at Otter Ri v v v v v v v v v v er at Northhampton, MA er at Northhampton, MA v v v v ar Ri ar Ri yers Brook at Randolph, VT yers Brook at Randolph, VT assumpsic Ri assumpsic Ri Moose Ri P Connecticut Ri A White Ri Sug Connecticut Ri Otter Ri North Ri South Ri Green Ri Mill Ri Connecticut Ri Moose Ri P Connecticut Ri A White Ri Sug Connecticut Ri Otter Ri North Ri South Ri Green Ri Mill Ri Connecticut Ri USGS station number e been independently rounded; USGS, U.S. Geological Surv v 01134500 01135500 01138500 01142500 01144000 01152500 01154500 01163200 01169000 01169900 01170100 01171500 01184000 01134500 01135500 01138500 01142500 01144000 01152500 01154500 01163200 01169000 01169900 01170100 01171500 01184000 Mean annual loads, yields, confidence intervals, standard error of prediction, and the ratio s 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 11 12 13 10 11 12 13 Site fig. 1) number able 2. (table 1; T total nitrogen and phosphorus at river sampling sites in the upper C [Numbers ha Characterization of Total Nitrogen at River Sampling Sites 9

Moose River at Victory, VT, Station 011500 year 2004 (fig. 4D). Instantaneous loads of total nitrogen gen­ erally were representative of the instantaneous streamflow at Moose River at Victory, VT, represents a forested site in the time of sample collection and ranged from 22 to 2,700 lb/d the study area with minimal effects on water quality from land (fig. 4E). In other words, instantaneous loads of total nitrogen use. The contributing drainage basin upstream from the site were greater in samples collected during high streamflows and encompasses 75 mi2 and is about 90 percent forested, 7 percent less in samples collected during base-flow or low-flow condi­ wetlands, and 3 percent barren (fig. 2). Annual streamflow tions. Ranges of concentrations or values for water-quality for 1980–2002 averaged 111,000 acre-feet, with considerable constituents not shown in figure 4 are listed below. year-to-year variation (fig. 3A). Annual streamflow for the data-collection period for water years 2003–05 averaged <, less than; —, not calculated 107,000 acre-feet (table 1). Streamflow during the period was Water-quality Mini- Maxi­ less than the long-term mean for the first year of the study constituenta Mean Median mum mum period, greater than the long-term mean for the second year, Specific conductance 18 32 31 49 and about equal to the long-term mean for the third year (fig. 3A). pH 5.7 6.5 6.4 7.4 Daily mean streamflow and time distribution for 42 Water temperature .0 6.8 3.0 21.8 water-quality samples are shown in figure 3B. The minimum Dissolved oxygen 7.0 10.8 11.2 14.1 streamflow at which the water samples were collected was Dissolved nitrite < .004 — <.004 .005 16 ft3/s, and the maximum streamflow at which samples were Dissolved ammonia .005 .017 .013 .074 collected was 721 ft3/s. Dissolved orthophos­ < .01 — <.01 < .01 The minimum concentration of dissolved oxygen at this phate (WY 2003–04) site (7.0 mg/L) equaled the State of Vermont water-quality Dissolved orthophos­ <.003 — <.003 .003 criterion for Class B waters for dissolved oxygen of 7.0 mg/L phate (WY 2005) for cold water fish habitat (Vermont Water Resources Board, Suspended sediment 1 13 3 165 2000). Concentrations of selected water-quality constituents aAll constituents are reported as milligrams per liter except for specific were plotted relative to streamflow (fig. 4). Concentrations of conductance, which is reported as microsiemens per centimeter at 25 dissolved nitrite plus nitrate ranged from an estimated 0.016 degrees Celsius; pH, which is reported as standard units; and water tem­ to 0.373 mg/L, and showed no apparent relation to streamflow perature, which is reported as degrees Celsius; WY, water year. < less than; —, not calculated. (fig. 4A). Total ammonia plus organic nitrogen ranged from 0.13 to 0.81 mg/L and also showed no apparent relation to The estimated load of total nitrogen varied among years streamflow (fig. 4B). The maximum concentration of total during the study (appendix A-1; fig. 5A). The mean annual ammonia plus organic nitrogen (0.81 mg/L) was in a sample load of total nitrogen was 121,000 lb/yr with a ratio of the collected at a streamflow of 570 ft3/s, just prior to the maxi­ mum daily mean streamflow, which occurred during snowmelt standard error of prediction to the mean load of 5.3 percent in water year 2004 (fig. 4B). Concentrations of total nitrogen (table 2, fig. 5A). An estimated 47 percent of the total nitrogen (dissolved nitrite plus nitrate plus total ammonia plus organic load was transported during the spring (appendix A-2). 2 nitrogen) ranged from 0.20 to 0.90 mg/L (fig. 4C), with the The mean total nitrogen yield (1,610 (lb/mi )/yr) generally maximum concentration observed during spring snowmelt in was less than at sites representing other land uses (fig. 5B). water year 2004. Concentrations of total phosphorus generally Trench (1999) reported similar results in that forested basins were low and ranged from 0.005 to 0.124 mg/L and showed a generally had lower yields of total nitrogen than agricultural slight increase with an increase in streamflow (fig. 4D). The and urban basins of similar size. In general, water quality at maximum concentration of total phosphorus (0.124 mg/L) this site typifies background conditions for this area of the also was collected at a streamflow of 570 ft3/s, just prior to the basin. Concentrations of total nitrogen and other water-quality maximum daily mean streamflow during snowmelt in water constituents generally were low. 10 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Moose River at Victory, VT, Station 01134500

EXPLANATION 71°50' Land-Use 44°40' Classification

Water Urban Barren

71°45' Forest Shrubland Ja M m o es o Agriculture s B e r 44°35' o Wetlands o

k k

Granby o o Basin boundary r

R B i y v nb e ra Streams W r G e ir Mill Broo rook Sampling site k Rogers B

44°30'

0 1 2 3 4 MILES 1 Land-use data from U.S. Geological Survey, 1992 2 NEW National Land Cover Dataset, 2000 HAMPSHIRE 0 1 2 3 4 KILOMETERS 3

4 VERMONT 5 Lebanon

r e 6 v i

R

t

u

c

i t 7 c

e n n o C Keene

11 9 8 10 MASSACHUSETTS 12 Springfield

13

Photograph by Thor Smith, U.S. Geological Survey

Figure 2. Location, land-use classification, and photograph of sampling site for Moose River at Victory, VT, station 01134500. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 11

A 160 Historical record 140 1980-2002 Study period Mean annual

120

100

80 , IN THOUSANDS OF ACRE FEET

60

40

ANNUAL STREAMFLOW 20

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 2,000

Daily mean streamflow

Water-quality sample

1,500

, IN CUBIC FEET PER SECOND 1,000

500 Y MEAN STREAMFLOW DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure . A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Moose River at Victory, VT, station 01134500. 12 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 0.38 0.9 0.36 0.34 0.8 0.32 0.30 0.28 0.7 0.26 0.24 0.6 0.22 0.20 0.5 0.18 0.16 0.14 0.4 0.12 0.10 0.3 0.08 DISSOLVED NITRITE PLUS NITRATE 0.06 0.2

0.04 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.02 0.00 0.1 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 C D 0.9 0.13

TION, IN MILLIGRAMS PER LITER 0.12 0.8 0.11 0.10 0.7 0.09 CONCENTRA 0.08 0.6 0.07

NITROGEN 0.06 L PHOSPHORUS A

0.5 L T A

O 0.05 T T O T 0.04 0.4 0.03

0.3 0.02 0.01 0.2 0.00 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 3,000 800

700 EXPLANATION , WATER-QUALITY SAMPLE 600 OUTLYING SAMPLE 2,000 500 95th PERCENTILE 75th PERCENTILE MEDIAN 400 25th PERCENTILE 5th PERCENTILE 300 ANEOUS STREAMFLOW

1,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND

200 INST

INSTANTANEOUS TOTAL NITROGEN LOAD, 100

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure . Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Moose River at Victory, VT, station 01134500. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 1

A 160,000

140,000 2003–05 Mean annual 120,000

100,000

80,000

NITROGEN LOAD, 60,000 L A T O IN POUNDS PER YEAR 40,000 T 20,000 0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Moose River at Victory, VT All other river sampling sites

Figure 5. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Moose River near Victory, VT, station 01134500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) 1 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Passumpsic River at Passumpsic, VT, 1,600 ft3/s, when streamflow was rising as a result of snowmelt Station 0115500 (fig. 8B). Concentrations of total nitrogen ranged from 0.25 to 1.2 mg/L (fig. 8C). Concentrations of total phosphorus ranged Passumpsic River at Passumpsic, VT, is classified as from 0.008 to 0.155 mg/L and increased with an increase in an agricultural site in the study area and represents potential streamflow (fig. 8D). Instantaneous loads of total nitrogen effects on water quality from agricultural land use. The con­ ranged from 210 to 23,800 lb/d (fig. 8E). Ranges of concen­ tributing drainage basin upstream from the site encompasses trations or values for water-quality constituents not shown in 436 mi2 and is about 80 percent forested, 11 percent agricul­ figure 8 are listed below. ture, 2 percent urban, and about 7 percent wetlands and barren (fig. 6). Annual streamflow for 1980–2002 averaged <, less than; —, not calculated 572,000 acre-feet, with considerable year-to-year variation Water-quality Mini- Maxi­ (fig. 7A). Annual streamflow for the data-collection period constituenta Mean Median mum mum for water years 2003–05 averaged 554,000 acre-feet (table 1). Specific conductance 74 182 197 302 Streamflow was less than the long-term mean for the first year pH 6.3 7.7 7.6 9.0 of the study, greater than the long-term mean for the second year, and slightly less than the long-term mean for the third Water temperature .0 9.1 6.2 24.4 year (fig. 7A). Dissolved oxygen 8.0 11.3 11.9 14.5 Daily mean streamflow and time distribution for 43 Dissolved nitrite <.008 — <.008 <.008 water-quality samples are shown in figure 7B. The minimum Dissolved ammonia .005 .032 .019 .208 streamflow at which the water samples were collected was Dissolved orthophos­ <.01 — <.01 .02 152 ft3/s, and the maximum streamflow at which samples were phate (WY 2003–04) collected was 3,820 ft3/s. Dissolved orthophos­ <.003 — .005 .022 The minimum concentration of dissolved oxygen at phate (WY 2005) this site (8.0 mg/L) was above the State of Vermont water- Suspended sediment 1 24 4 358 quality criterion for dissolved oxygen (7.0 mg/L; Vermont aAll constituents are reported as milligrams per liter except for specific Water Resources Board, 2000). Concentrations of selected conductance, which is reported as microsiemens per centimeter at 25 water-quality constituents were plotted relative to streamflow degrees Celsius; pH, which is reported as standard units; and water tem­ perature, which is reported as degrees Celsius; WY, water year. (fig. 8). Concentrations of dissolved nitrite plus nitrate ranged from 0.064 to 0.515 mg/L, and showed no apparent relation to streamflow (fig. 8A). Total ammonia plus organic nitrogen The estimated load of total nitrogen varied among years ranged from 0.13 to 0.67 mg/L and generally showed no during the study (appendix A-1; fig. 9A). The mean annual relation to streamflow for lower flow samples; however, total nitrogen load was 822,000 lb/yr, with a ratio of the concentrations increased at streamflows greater than standard error of prediction to the mean load of 4.2 percent 2,500 ft3/s (fig. 8B). One exception was the maximum con­ (table 2, fig. 9A). An estimated 48 percent of the total nitrogen centration of total ammonia plus organic nitrogen (0.67 mg/L) load was transported during the spring (appendix A-2). The that was in a sample collected at a streamflow of about mean yield of total nitrogen was 1,890 (lb/mi2)/yr (fig. 9B). Characterization of Total Nitrogen at River Sampling Sites 15

Passumpsic River at Passumpsic, VT, Station 01135500

71°56'15" EXPLANATION Land-Use 44°42'45" Classification

r 72°06' e v i Water R c 71°46'30" i s Urban p

m

u s Barren s

a P ch Forest n ra B M st Shrubland ille a r R E un Agriculture

r e Granby v i Wetlands 44°30' R

c

i Basin boundary

s Lyndon

p

m

u Streams S s l s e a e Sampling site p P

e

r

s

R iv er

St Johnsbury

44°23'15"

Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000 0 5 10 MILES

1 0 5 10 KILOMETERS 2 NEW HAMPSHIRE 3

4 VERMONT 5 Lebanon r e v i 6

R

t

u

c i t 7 c

e

n n o C Keene 11 9 8 10 MASSACHUSETTS 12 Springfield

13

Photograph by Thor Smith, U.S. Geological Survey

Figure . Location, land-use classification, and photograph of sampling site for Passumpsic River at Passumpsic, VT, station 01135500. (Refer to table 1 and figure 1 for additional site information.) 1 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 800 Historical record 1980-2002 700 Mean annual Study period

600

500

400 , IN THOUSANDS OF ACRE FEET

300

200

ANNUAL STREAMFLOW 100

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 7,000

Daily mean streamflow 6,000 Water-quality sample

5,000

4,000 , IN CUBIC FEET PER SECOND

3,000

2,000 Y MEAN STREAMFLOW 1,000 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure . A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Passumpsic River at Passumpsic, VT, station 01135500. Characterization of Total Nitrogen at River Sampling Sites 1

Box Box A plot B plot 0.52 0.7 0.50 0.48 0.46 0.44 0.6 0.42 0.40 0.38 0.36 0.5 0.34 0.32 0.30 0.28 0.4 0.26 0.24 0.22 0.3 0.20 0.18 0.16

DISSOLVED NITRITE PLUS NITRATE 0.14 0.2 0.12 0.10 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.08 0.06 0.1 0 1,000 2,000 3,000 4,000 0 1,000 2,000 3,000 4,000 C D 1.2 0.16 0.15 TION, IN MILLIGRAMS PER LITER 1.1 0.14 1.0 0.13 0.12 0.9 0.11 CONCENTRA 0.10 0.8 0.09 0.7 0.08 NITROGEN L

PHOSPHORUS 0.07 A L T

0.6 A O T 0.06 T O 0.5 T 0.05 0.04 0.4 0.03 0.02 0.3 0.01 0.2 0.00 0 1,000 2,000 3,000 4,000 0 1,000 2,000 3,000 4,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 30,000 4,000

EXPLANATION , WATER-QUALITY SAMPLE 3,000 OUTLYING SAMPLE 20,000 95th PERCENTILE 75th PERCENTILE MEDIAN 2,000 25th PERCENTILE 5th PERCENTILE ANEOUS STREAMFLOW

10,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND

1,000 INST INSTANTANEOUS TOTAL NITROGEN LOAD,

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 8. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Passumpsic River at Passumpsic, VT, station 01135500. (Refer to table 1 and figure 1 for station location.) 18 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 1,200,000

1,000,000 2003–05 Mean annual 800,000

600,000 NITROGEN LOAD, L 400,000 A T O IN POUNDS PER YEAR T 200,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Passumpsic River at Passumpsic, VT All other river sampling sites

Figure 9. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Passumpsic River at Passumpsic, VT, station 01135500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) Characterization of Total Nitrogen at River Sampling Sites 19

Connecticut River at Wells River, VT, with an increase in streamflow (fig. 12D). The sample that 3 Station 0118500 was collected at about 4,500 ft /s when streamflow was rising during snowmelt also had the second highest concentration of Connecticut River at Wells River, VT, represents a main total phosphorus (0.110 mg/L; fig. 12D). Instantaneous loads stem site in the study area and is located on the Connecticut of total nitrogen ranged from 2,300 to 71,900 lb/d (fig. 12E). River where there are potential effects on water quality from Ranges of concentrations or values for water-quality constitu­ a mix of land uses. The contributing drainage basin upstream ents not shown in figure 12 are listed below. from this site encompasses 2,644 mi2 and is about 84 percent forested, 6 percent agriculture, 2 percent urban, and 8 percent <, less than; —, not calculated Water-quality wetlands and barren (fig. 10). Annual streamflow for 1980– Mini- Maxi­ constituenta Mean Median 2002 averaged 3,750,000 acre-feet, with considerable year- mum mum to-year variation (fig. 11A). Annual streamflow for the data- Specific conductance 58 87 87 119 collection period for water years 2003–05 averaged pH 6.4 7.1 7.1 7.7 3,560,000 acre-feet (table 1). Streamflow was less than the Water temperature .0 8.3 5.5 23.2 long-term mean for the first year of the study, greater than the long-term mean for the second year, and slightly less than the Dissolved oxygen 7.6 11.3 11.6 15.2 long-term mean for the third year (fig. 11A). Dissolved nitrite <.004 — <.004 .005 Daily mean streamflow and the time distribution for Dissolved ammonia .007 .021 .017 .080 Dissolved orthophos­ 43 water-quality samples are shown in figure 11B. The <.01 — <.01 <.01 minimum streamflow at which water samples were collected phate (WY 2003–04) 3 Dissolved orthophos­ was 1,310 ft /s, and the maximum streamflow at which <.003 — <.003 .006 samples were collected was 22,600 ft3/s. phate (WY 2005) The minimum concentration of dissolved oxygen at Suspended sediment 1 24 5 337 this site (7.6 mg/L) was above the State of Vermont water- aAll constituents are reported as milligrams per liter except for specific quality criterion for dissolved oxygen (7.0 mg/L; Vermont conductance, which is reported as microsiemens per centimeter at 25 Water Resources Board, 2000). Concentrations of selected degrees Celsius; pH, which is reported as standard units; and water tem- perature, which is reported as degrees Celsius; WY, water year. water-quality constituents were plotted relative to streamflow (fig. 12). Concentrations of dissolved nitrite plus nitrate The estimated load of total nitrogen varied among years ranged from 0.110 to 0.370 mg/L and showed scatter or a during the study (appendix A-1; fig. 13A). The mean annual poorly defined relation to streamflow (fig. 12A). Total ammo­ load of total nitrogen was 4,470,000 lb/yr, with a ratio of the nia plus organic nitrogen ranged from 0.15 to 0.46 mg/L and generally showed an increase with an increase in streamflow standard error of prediction to the mean load of 2.2 percent (fig. 12B). However, a sample that was collected at a (table 2, fig. 13A). An estimated 45 percent of the total nitro­ streamflow of about 4,500 ft3/s during snowmelt had the gen load was transported during the spring (appendix A-2). 2 second highest concentration of total ammonia plus organic The mean yield of total nitrogen (1,690 (lb/mi )/yr) was simi­ nitrogen (0.40 mg/L; fig. 12B). Concentrations of total lar to the yield of total nitrogen at Connecticut River at North nitrogen ranged from 0.30 to 0.74 mg/L (fig. 12C). Concentra­ Walpole, NH (site 7), but less than the yield of total nitrogen at tions of total phosphorus ranged from 0.007 to 0.112 mg/L the Connecticut River at Thompsonville, CT (site 13), the site and generally showed some scatter but also a slight increase representing the outlet to the upper basin (fig. 13B). 20 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Connecticut River at Wells River, VT, Station 01138500

72°30' EXPLANATION Land-Use Classification

QUEBEC Water 45° Urban Barren Colebrook Forest 72° NEW Shrubland HAMPSHIRE VERMONT Agriculture r Amm r e on e pp o Wetlands v o i U s Lyndon R u ic c s Groveton R Basin boundary p i v m r e e r u v Lancaster 44°30' s i

s I Streams R s a t r P cu a ti e St Johnsbury c l Sampling site ne R n i Co ver mon Am oo su Littleton c River

Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000 0 5 10 15 20 MILES

1 0 5 10 15 20 KILOMETERS 2 NEW HAMPSHIRE 3

4 VERMONT 5 Lebanon r e

v i 6 R

t

u

c i

t

c 7

e n n o C Keene 11 9 8 10 MASSACHUSETTS 12 Springfield

13

Photograph by Thor Smith, U.S. Geological Survey

Figure 10. Location, land-use classification, and photograph of sampling site for Connecticut River at Wells River, VT, station 01138500. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 21

A 6,000 Historical record Study period

5,000 1980-2002 Mean annual

4,000

3,000 , IN THOUSANDS OF ACRE FEET

2,000

1,000 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 30,000

Daily mean streamflow

25,000 Water-quality sample

20,000

, IN CUBIC FEET PER SECOND 15,000

10,000

Y MEAN STREAMFLOW 5,000 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 11. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Connecticut River at Wells River, VT, station 01138500. 22 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 0.37 0.47 0.36 0.35 0.45 0.34 0.43 0.33 0.32 0.41 0.31 0.30 0.39 0.29 0.37 0.28 0.27 0.35 0.26 0.33 0.25 0.24 0.31 0.23 0.22 0.29 0.21 0.27 0.20 0.19 0.25 0.18 0.17 0.23 0.16 DISSOLVED NITRITE PLUS NITRATE 0.21 0.15 0.14 0.19 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.13 0.17 0.12 0.11 0.15 0 5,000 10,000 15,000 20,000 25,000 0 5,000 10,000 15,000 20,000 25,000 C D 0.74 0.12 0.72

TION, IN MILLIGRAMS PER LITER 0.70 0.11 0.68 0.66 0.10 0.64 0.62 0.09 0.60

CONCENTRA 0.08 0.58 0.56 0.07 0.54 0.52 0.06 NITROGEN

L 0.50 PHOSPHORUS A L 0.05 T 0.48 A O T

T 0.46 O

0.44 T 0.04 0.42 0.40 0.03 0.38 0.02 0.36 0.32 0.01 0.30 0.28 0.00 0 5,000 10,000 15,000 20,000 25,000 0 5,000 10,000 15,000 20,000 25,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 80,000 30,000

70,000 EXPLANATION , WATER-QUALITY SAMPLE 60,000 OUTLYING SAMPLE 20,000 50,000 95th PERCENTILE 75th PERCENTILE MEDIAN 40,000 25th PERCENTILE 5th PERCENTILE 30,000 ANEOUS STREAMFLOW

10,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND

20,000 INST

INSTANTANEOUS TOTAL NITROGEN LOAD, 10,000

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 12. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Connecticut River at Wells River, VT, station 01138500. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 2

A 6,000,000

5,000,000 2003–05 Mean annual 4,000,000

3,000,000 NITROGEN LOAD, L 2,000,000 A T O IN POUNDS PER YEAR T 1,000,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Connecticut River at Wells River, VT All other river sampling sites

Figure 1. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Connecticut River at Wells River, VT, station 01138500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Ayers Brook at Randolph, VT, Station 0112500 (fig. 16D). Instantaneous loads of total nitrogen ranged from 27 to 2,760 lb/d (fig. 16E). Ranges of concentrations or values Ayers Brook at Randolph, VT, represents an agricul­ for water-quality constituents not shown in figure 16 are listed tural site in the study area. The contributing drainage basin below. upstream from the site encompasses 31 mi2 and is about 73 percent forested, 21 percent agriculture, 1 percent urban, <, less than; —, not calculated and 5 percent wetlands and barren (fig. 14). Annual stream- Water-quality Mini- Maxi­ flow for 1980–2002 averaged 37,300 acre-feet, with consider­ constituenta Mean Median mum mum able year-to-year variation (fig. 15A). Annual streamflow for Specific conductance 127 210 209 293 the data-collection period for water years 2003–05 averaged pH 6.3 7.5 7.6 8.2 38,200 acre-feet (table 1). Streamflow was less than the long-term mean for the first year of the study, greater than the Water temperature .0 7.9 6.5 22 long-term mean for the second year, and slightly greater than Dissolved oxygen 8.2 11.6 11.7 15.0 the long-term mean for the third year (fig. 15A). Dissolved nitrite <.004 — <.004 .006 Daily mean streamflow and time distribution for 43 Dissolved ammonia .005 .016 .011 .111 Dissolved orthophos­ water-quality samples are shown in figure 15B. The mini­ <.01 — <.01 <.01 mum streamflow at which water samples were collected was phate (WY 2003–04) 3 Dissolved orthophos­ 10 ft /s, and the maximum streamflow at which samples were <.003 — <.003 .004 collected was 310 ft3/s. phate (WY 2005) The minimum concentration of dissolved oxygen at this Suspended sediment 1 73 5 776 site (8.2 mg/L) was above the State of Vermont water-quality aAll constituents are reported as milligrams per liter except for specific criterion for dissolved oxygen (7.0 mg/L; Vermont Water conductance, which is reported as microsiemens per centimeter at 25 Resources Board, 2000). Concentrations of selected water- degrees Celsius; pH, which is reported as standard units; and water tem- perature, which is reported as degrees Celsius; WY, water year. quality constituents were plotted relative to streamflow (fig. 16). Concentrations of dissolved nitrite plus nitrate ranged The estimated load of total nitrogen varied among years from 0.291 to 0.733 mg/L and showed scatter or a poorly during the study (appendix A-1; fig. 17A). The mean annual defined relation to streamflow (fig. 16A). Total ammonia plus organic nitrogen ranged from an estimated 0.06 to 1.3 mg/L load of total nitrogen was 74,900 lb/yr, with a ratio of the and showed a slight variation with an increase in streamflow standard error of prediction to the mean load of 3.5 percent (fig. 16B). Concentrations of total nitrogen ranged from (table 2, fig. 17A). An estimated 51 percent of the load of 0.39 to 1.8 mg/L and also showed a slight variation with total nitrogen was transported during the spring (appendix streamflow (fig. 16C). Concentrations of total phosphorus A-2). The mean yield of total nitrogen (2,450 (lb/mi2)/yr) was ranged from an estimated concentration of 0.003 to greater than the yield of total nitrogen at all other sites, with 0.708 mg/L and increased with an increase in streamflow, the exception of one urban site that was directly affected by especially at streamflows greater than about 100 ft3/s wastewater (site 8; fig. 17B). Characterization of Total Nitrogen at River Sampling Sites 25

Ayers Brook at Randolph, VT, Station 01142500

EXPLANATION Land-Use

72°42'30" Classification

44°02'30" 72°37'30" Water Urban Barren Forest Shrubland

k Agriculture

o

o r

B Wetlands

s r

e y Basin boundary

A Streams Sampling site

43°57'30"

Land-use data from U.S. Geological Survey, 1992 1 2 0 1 2 3 4 MILES National Land Cover Dataset, 2000 NEW HAMPSHIRE 3 0 1 2 3 4 KILOMETERS 4 VERMONT 5 Lebanon

r e 6 v i

R

t

u

c

i t 7

c e n n o C Keene

11 9 8 10 MASSACHUSETTS 12 Springfield

13

Photograph by Thor Smith, U.S. Geological Survey

Figure 1. Location, land-use classification, and photograph of sampling site for Ayers Brook at Randolph, VT, station 01142500. (Refer to table 1 and figure 1 for additional site information.) 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 60 Historical record Study period 50 1980-2002 Mean annual

40

30 , IN THOUSANDS OF ACRE FEET

20

10 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 800

Daily mean streamflow 700 Water-quality sample

600

500

, IN CUBIC FEET PER SECOND 400

300

200 Y MEAN STREAMFLOW

DAIL 100

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 15. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Ayers Brook at Randolph, VT, station 01142500. Characterization of Total Nitrogen at River Sampling Sites 2

Box Box A plot B plot 0.74 1.4 0.72 0.70 0.68 1.2 0.66 0.64 0.62 0.60 1.0 0.58 0.56 0.54 0.8 0.52 0.50 0.48 0.6 0.46 0.44 0.42 0.4 0.40 0.38

DISSOLVED NITRITE PLUS NITRATE 0.36 0.34 0.2 0.32 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.30 0.28 0.0 0 50 100 150 200 250 300 0 50 100 150 200 250 300 C D 1.9 0.8 1.8 TION, IN MILLIGRAMS PER LITER 1.7 0.7 1.6 1.5 0.6 1.4 CONCENTRA 1.3 0.5 1.2 1.1 0.4 NITROGEN L

1.0 PHOSPHORUS A L T A O 0.9 T 0.3 T O 0.8 T 0.7 0.2 0.6 0.5 0.1 0.4 0.3 0.0 0 50 100 150 200 250 300 0 50 100 150 200 250 300

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 3,000 400

EXPLANATION

, WATER-QUALITY SAMPLE 300 OUTLYING SAMPLE 2,000 95th PERCENTILE 75th PERCENTILE MEDIAN 200 25th PERCENTILE 5th PERCENTILE ANEOUS STREAMFLOW 1,000 IN POUNDS PER DAY

ANT Instantaneous streamflow 100 IN CUBIC FEET PER SECOND INST INSTANTANEOUS TOTAL NITROGEN LOAD,

0 00 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 1. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Ayers Brook at Randolph, VT, station 01142500. (Refer to table 1 and figure 1 for station location.) 28 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 100,000

2003–05 80,000 Mean annual

60,000

40,000 NITROGEN LOAD, L A T O IN POUNDS PER YEAR T 20,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Ayers Brook at Randolph, VT All other river sampling sites

Figure 1. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Ayers Brook at Randolph, VT, station 01142500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) Characterization of Total Nitrogen at River Sampling Sites 29

White River at West Hartford, VT, streamflow, especially at streamflows greater than about 3 Station 011000 2,000 ft /s (fig. 20D). Instantaneous loads of total nitrogen ranged from 351 to 52,700 lb/d (fig. 20E). Ranges of White River at West Hartford, VT, represents an concentrations or values for water-quality constituents not agricultural site in the study area. The contributing drainage shown in figure 20 are listed below. basin upstream from this site encompasses 690 mi2 and is about 84 percent forested, 13 percent agriculture, 1 percent <, less than; —, not calculated Water-quality urban, and 2 percent wetlands and barren (fig. 18). Annual Mini- Maxi­ constituenta Mean Median streamflow for 1980–2002 averaged 885,000 acre-feet, mum mum with considerable year-to-year variation (fig. 19A). Annual Specific conductance 59 150 150 217 streamflow for the data-collection period for water years pH 6.5 7.6 7.6 8.3 2003–05 averaged 959,000 acre-feet (table 1). Streamflow was Water temperature .0 8.6 6.5 27 less than the long-term mean for the first year of this study, Dissolved oxygen 8.0 11.4 11.8 14.4 greater than the long-term mean for the second year, and about Dissolved nitrite <.004 — <.004 .007 equal to the long-term mean for the third year (fig. 19A). Dissolved ammonia .005 .014 .010 .077 Daily mean streamflows and time distribution for 45 Dissolved orthophos­ <.01 — <.01 <.01 water-quality samples are shown in figure 19B. The minimum phate (WY 2003–04) streamflow at which water samples were collected was Dissolved orthophos­ 3 <.003 — <.003 .008 216 ft /s, and the maximum streamflow at which samples phate (WY 2005) were collected was 11,600 ft3/s. Suspended sediment 1 72 2 752 The minimum concentration of dissolved oxygen at this aAll constituents are reported as milligrams per liter except for specific site (8.0 mg/L) was above the State of Vermont water-quality conductance, which is reported as microsiemens per centimeter at 25 criterion for dissolved oxygen (7.0 mg/L; Vermont Water degrees Celsius; pH, which is reported as standard units; and water tem- Resources Board, 2000). Concentrations of selected water- perature, which is reported as degrees Celsius; WY, water year. quality constituents were plotted relative to streamflow (fig. 20). Concentrations of dissolved nitrite plus nitrate The estimated load of total nitrogen varied among years ranged from 0.138 to 0.535 mg/L and showed scatter or during the study (appendix A-1; fig. 21A). The mean annual a poorly defined relation to streamflow (fig. 20A). Total load of total nitrogen was 1,260,000 lb/yr, with a ratio of the ammonia plus organic nitrogen ranged from an estimated 0.06 standard error of prediction to the mean load of 2.7 percent to 0.81 mg/L and generally increased as streamflow increased (table 2, fig. 21A). An estimated 55 percent of the total (fig. 20B). Concentrations of total nitrogen ranged from 0.24 nitrogen load was transported during the spring (appendix to 1.1 mg/L and increased as streamflow increased (fig. 20C). A-2). The mean yield of total nitrogen (1,830 (lb/mi2)/yr) was Concentrations of total phosphorus ranged from an estimated similar to the yield of total nitrogen at most other sites in this 0.002 to 0.604 mg/L and also increased with an increase in study (fig. 21B). 0 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

White River at West Hartford, VT, Station 01144000

72°45' 72°30' EXPLANATION Land-Use Classification

Water

44° Urban

k

o

o r

Barren

B s

k r

o e k Forest o y o r Randolph A o

r B h B

s c Shrubland

r n s

e a r y r e

y A B

d A Agriculture h c n h

o c n

c n a e a r Wetlands r S B B

d t

r s

i r W i Basin boundary

h

F

T h

i t e e Whit Streams R Royalton iv Sampling site er

R

i 43°45' v e r

Land-use data from U.S. Geological Survey, 1992 0 5 10 MILES National Land Cover Dataset, 2000

1 0 5 10 KILOMETERS NEW 2 HAMPSHIRE 3

4

5 VERMONT Lebanon

r 6

e

v i

R

t

u

c i 7 t

c e n n o C Keene 11 9 8 10 MASSACHUSETTS 12 Springfield

13

Photograph by Thor Smith, U.S. Geological Survey

Figure 18. Location, land-use classification, and photograph of sampling site for White River at West Hartford, VT, station 01144000. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 1

A 1,400 Historical record Study period 1,200 1980-2002 Mean annual

1,000

800 , IN THOUSANDS OF ACRE FEET 600

400

ANNUAL STREAMFLOW 200

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 20,000

Daily mean streamflow

Water-quality sample

15,000

, IN CUBIC FEET PER SECOND 10,000

5,000 Y MEAN STREAMFLOW DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 19. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for White River at West Hartford, VT, station 01144000. 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 0.54 0.9 0.52 0.50 0.8 0.48 0.46 0.44 0.7 0.42 0.40 0.6 0.38 0.36 0.5 0.34 0.32 0.30 0.4 0.28 0.26 0.3 0.24 0.22 0.2

DISSOLVED NITRITE PLUS NITRATE 0.20 0.18 0.1 0.16 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.14 0.12 0.0 0 2,000 4,000 6,000 8,000 10,000 12,000 0 2,000 4,000 6,000 8,000 10,000 12,000 C D 1.1 0.7

TION, IN MILLIGRAMS PER LITER 1.0 0.6 0.9 0.5

CONCENTRA 0.8

0.7 0.4 NITROGEN

0.6 PHOSPHORUS L

L 0.3 A A T T O O T

0.5 T 0.2 0.4 0.1 0.3

0.2 0.0 0 2,000 4,000 6,000 8,000 10,000 12,000 0 2,000 4,000 6,000 8,000 10,000 12,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 60,000 12,000

EXPLANATION

50,000 10,000 , WATER-QUALITY SAMPLE

OUTLYING SAMPLE 40,000 8,000 95th PERCENTILE 75th PERCENTILE MEDIAN 30,000 6,000 25th PERCENTILE 5th PERCENTILE ANEOUS STREAMFLOW

20,000 4,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND INST 10,000 2,000 INSTANTANEOUS TOTAL NITROGEN LOAD,

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 20. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for White River at West Hartford, VT, station 01144000. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 

A 1,600,000

1,400,000 2003–05 Mean annual 1,200,000

1,000,000

800,000

NITROGEN LOAD, 600,000 L A T O IN POUNDS PER YEAR 400,000 T

200,000 0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION White River at West Hartford, VT All other river sampling sites

Figure 21. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for White River at West Hartford, VT, station 01144000, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Sugar River at West Claremont, NH, Concentrations of total phosphorus ranged from 0.011 to Station 01152500 0.275 mg/L (fig. 24D). Maximum concentrations of total nitrogen and total phosphorus also were observed at the Sugar River at West Claremont, NH, represents an urban maximum streamflow at which samples were collected. site in the study area. A urban site represents potential effects Instantaneous loads of total nitrogen ranged from 187 to on water quality from urban land use. The contributing drain­ 22,400 lb/d (fig. 24E). Ranges of concentrations or values age basin upstream from the site encompasses 269 mi2 and for water-quality constituents not shown in figure 24 are is about 78 percent forested, 9 percent agriculture, 6 percent listed below. urban, and 7 percent wetlands and barren (fig. 22). Although the second predominant percentage of land use in the upstream <, less than; —, not calculated drainage basin is agriculture, this site was selected to represent Water-quality Mini- Maxi­ constituenta Mean Median urban land use because the site is downstream from and near mum mum an urban area (fig. 22). The annual streamflow for 1980–2002 Specific conductance 49 120 121 191 averaged 304,000 acre-feet, with considerable year-to-year pH 6.1 7.0 6.8 8.7 variation (fig. 23A). The annual streamflow for the data-col­ Water temperature .0 9.5 7.5 26 lection period for water years 2003–05 averaged 341,000 acre- feet (table 1). Streamflow was less than the long-term mean Dissolved oxygen 7.4 11.8 11.8 14.4 for the first year of this study and greater than the long-term Dissolved nitrite <.004 — <.004 .005 mean for the second and third years (fig. 23A). Dissolved ammonia .005 .038 .021 .156 Dissolved orthophos­ Daily mean streamflow and time distribution for 43 <.01 — <.01 .01 water-quality samples are shown in figure 23B. The mini­ phate (WY 2003–04) Dissolved orthophos­ mum streamflow at which water samples were collected was <.003 — <.003 .005 73 ft3/s, and the maximum streamflow at which samples were phate (WY 2005) collected was 4,380 ft3/s. Suspended sediment 1 14 2 267 The minimum concentration of dissolved oxygen was aAll constituents are reported as milligrams per liter except for specific 7.4 mg/L and was within the water-quality criterion conductance, which is reported as microsiemens per centimeter at 25 for dissolved oxygen (New Hampshire Department of degrees Celsius; pH, which is reported as standard units; and water tem­ perature, which is reported as degrees Celsius; WY, water year. Environmental Services, 2004). Concentrations of selected water-quality constituents were plotted relative to streamflow The estimated load of total nitrogen varied among years (fig. 24). Concentrations of dissolved nitrite plus nitrate ranged from 0.059 to 0.440 mg/L and showed some scatter but also during the study (appendix A-1; fig. 25A). The mean annual a slight dilution effect relative to streamflow (fig. 24A). Total load of total nitrogen was 372,000 lb/yr, with a ratio of the ammonia plus organic nitrogen ranged from an estimated 0.16 standard error of prediction to the mean load of 3.9 percent to 0.82 mg/L and generally showed no relation to streamflow; (table 2, fig. 25A). An estimated 49 percent of the total nitro­ however, the maximum concentration of total ammonia plus gen load was transported during the spring (appendix A-2). organic nitrogen was observed at the maximum streamflow The mean yield of total nitrogen (1,380 (lb/mi2)/yr) was less at which samples were collected (fig. 24B). Concentrations than the mean yield of total nitrogen from most other sites in of total nitrogen ranged from 0.24 to 0.95 mg/L (fig. 24C). the study (fig. 25B). Characterization of Total Nitrogen at River Sampling Sites 5

Sugar River at West Claremont, NH, Station 01152500

EXPLANATION Land-Use Classification

Water Urban

Barren 72°10' Forest 1 2 NEW HAMPSHIRE Shrubland 3 Agriculture 72° 4 VERMONT Wetlands 5 43°30' er Lebanon Basin boundary iv r R e r v 6 a i g

R Streams u

t S

u h c i Sampling site c t 7 72°20' n c a e r n B n o th r C Keene o N 11 9 8 10 ar R g i MASSACHUSETTS u v Sunapee S e Claremont r Lake 12 Springfield Newport S

o

u

t 13 h

B

43°20' r a n c h S u g

a

r

R

i

v

e

r

Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000

0 5 10 MILES

0 5 10 KILOMETERS

Photograph by Sanborn Ward, U.S. Geological Survey

Figure 22. Location, land-use classification, and photograph of urban area upstream of sampling site for Sugar River at West Claremont, NH, station 01152500. (Refer to table 1 and figure 1 for additional site information.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 500 Historical record Study period

400 1980-2002 Mean annual

300 , IN THOUSANDS OF ACRE FEET 200

100 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 6,000

Daily mean streamflow

Water-quality sample 5,000

4,000

, IN CUBIC FEET PER SECOND 3,000

2,000 Y MEAN STREAMFLOW 1,000 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 2. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Sugar River at West Claremont, NH, station 01152500. Characterization of Total Nitrogen at River Sampling Sites 

Box Box A plot B plot 0.44 0.9 0.42 0.40 0.38 0.8 0.36 0.34 0.7 0.32 0.30 0.6 0.28 0.26 0.24 0.5 0.22 0.20 0.4 0.18 0.16 0.14 0.3 0.12 DISSOLVED NITRITE PLUS NITRATE 0.10 0.2

0.08 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.06 0.04 0.1 0 1,000 2,000 3,000 4,000 0 1,000 2,000 3,000 4,000 C D 1.0 0.28 0.27

TION, IN MILLIGRAMS PER LITER 0.26 0.9 0.25 0.24 0.23 0.22 0.8 0.21 0.20

CONCENTRA 0.19 0.7 0.18 0.17 0.16 0.15 0.6 0.14 NITROGEN

PHOSPHORUS 0.13 L L

A 0.12 A T 0.5 T 0.11 O O T

T 0.10 0.09 0.4 0.08 0.07 0.06 0.05 0.3 0.04 0.03 0.02 0.2 0.01 0 1,000 2,000 3,000 4,000 0 1,000 2,000 3,000 4,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 30,000 5,000

EXPLANATION , 4,000 WATER-QUALITY SAMPLE

OUTLYING SAMPLE 20,000 95th PERCENTILE 3,000 75th PERCENTILE MEDIAN 25th PERCENTILE 5th PERCENTILE

2,000 ANEOUS STREAMFLOW

10,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND INST 1,000 INSTANTANEOUS TOTAL NITROGEN LOAD,

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 2. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Sugar River at West Claremont, NH, station 01152500. (Refer to table 1 and figure 1 for station location.) 8 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 450,000 2003–05 400,000 Mean annual 350,000 300,000 250,000 200,000 NITROGEN LOAD, L 150,000 A T O IN POUNDS PER YEAR

T 100,000 50,000 0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Sugar River at West Claremont, NH All other river sampling sites

Figure 25. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Sugar River at West Claremont, NH, station 01152500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) Characterization of Total Nitrogen at River Sampling Sites 9

Connecticut River at North Walpole, NH, from 2,460 to 186,000 lb/d (fig. 28E). Ranges of concentra­ Station 0115500 tions or values for water-quality constituents not shown in figure 28 are listed below. Connecticut River at North Walpole, NH, represents a main-stem site for the study area. The contributing drainage <, less than; —, not calculated Water-quality basin upstream from the site encompasses 5,493 mi2 and is Mini- Maxi­ constituenta Mean Median about 83 percent forested, 9 percent agriculture, 2 percent mum mum urban, and 6 percent wetlands and barren (fig. 26). Annual Specific conductance 64 117 114 172 streamflow for 1980–2002 averaged 7,160,000 acre-feet, with pH 6.2 7.3 7.3 7.8 considerable year-to-year variation (fig. 27A). Annual stream- Water temperature .0 9.5 7.2 27 flow for the data-collection period for water years 2003–05 Dissolved oxygen 6.9 11.2 11.9 14.8 averaged 7,070,000 acre-feet (table 1). Streamflow was less Dissolved nitrite <.004 — <.004 .006 than the long-term mean for the first year of the study, greater Dissolved ammonia .005 .021 .017 .055 than the long-term mean for the second year, and about equal Dissolved orthophos­ <.01 — <.01 .01 to the long-term mean for the third year (fig. 27A). phate (WY 2003–04) Daily mean streamflow and time distribution for 44 Dissolved orthophos­ <.003 — <.003 .003 water-quality samples are shown in figure 27B. The mini­ phate (WY 2005) mum streamflow at which water samples were collected was Suspended sediment 1 13 2 169 1,290 ft3/s, and the maximum streamflow at which samples aAll constituents are reported as milligrams per liter except for specific 3 were collected was 46,700 ft /s. conductance, which is reported as microsiemens per centimeter at 25 The minimum concentration of dissolved oxygen was degrees Celsius; pH, which is reported as standard units; and water tem- 6.9 mg/L and was within the water-quality criterion for dis­ perature, which is reported as degrees Celsius; WY, water year. solved oxygen (New Hampshire Department of Environmental Services, 2004). Concentrations of selected water-quality The estimated load of total nitrogen varied among years constituents were plotted relative to streamflow (fig. 28). during the study (appendix A-1; fig. 29A). The mean annual Concentrations of dissolved nitrite plus nitrate ranged from load of total nitrogen was 9,600,000 lb/yr, with a ratio of the 0.136 to 0.406 mg/L and showed a poorly defined relation to standard error of prediction to the mean load of 4.1 percent streamflow (fig. 28A). Total ammonia plus organic nitrogen (table 2, fig. 29A). An estimated 47 percent of the total ranged from an estimated concentration of 0.15 to 0.46 mg/L nitrogen load was transported during the spring (appendix and showed a slight increase with an increase in streamflow A-2). The mean yield of total nitrogen (1,750 (lb/mi2)/yr) was (fig. 28B). Concentrations of total nitrogen ranged from 0.28 similar to the yield of total nitrogen at the Connecticut River to 0.83 mg/L (fig. 28C) and also showed a slight variation with at Wells River, VT (site 3), but less than the yield of total streamflow. Concentrations of total phosphorus ranged from nitrogen at the Connecticut River at Thompsonville, CT 0.005 to 0.155 mg/L and increased with an increase in stream- (site 13), the site representing the outlet to the upper basin flow (fig. 28D). Instantaneous loads of total nitrogen ranged (fig. 29B). 0 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Connecticut River at North Walpole, NH, Station 01154500 72°15'

EXPLANATION Land-Use Classification QUEBEC 45° Water Urban 72° Colebrook Barren Forest 1 2 NEW HAMPSHIRE Shrubland NEW 3 St Johnsbury Agriculture HAMPSHIRE 4 osuc R o ive Wetlands on r m 5 43°15' m Lebanon Basin boundary A 72°45' r e 6 v i

Streams R

t

u c Sampling site i 7 t c e VERMONT n White n R o Keene i C v e 11 r 9 8 VERMONT 10 Lebanon MASSACHUSETTS 12 Springfield 43°30' Claremont 13

Springfield

Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000

0 10 20 30 40 MILES

0 10 20 30 40 KILOMETERS

Photograph by Thor Smith, U.S. Geological Survey

Figure 2. Location, land-use classification, and photograph of sampling site for Connecticut River at North Walpole, NH, station 01154500. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 1

A 12,000 Historical record Study period 10,000 1980-2002 Mean annual

8,000

6,000 , IN THOUSANDS OF ACRE FEET

4,000

2,000 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 70,000

Daily mean streamflow 60,000 Water-quality sample

50,000

40,000 , IN CUBIC FEET PER SECOND

30,000

20,000 Y MEAN STREAMFLOW 10,000 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 2. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Connecticut River at North Walpole, NH, station 01154500. 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 0.41 0.6 0.39 0.37 0.35 0.5 0.33 0.31 0.4 0.29 0.27 0.25 0.3 0.23 0.21 0.19 0.2 DISSOLVED NITRITE PLUS NITRATE 0.17 0.15 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.13 0.1 0 10,000 20,000 30,000 40,000 50,000 0 10,000 20,000 30,000 40,000 50,000 C D 1.0 0.16 0.15 TION, IN MILLIGRAMS PER LITER 0.9 0.14 0.13 0.8 0.12 0.11 CONCENTRA 0.7 0.10 0.09 0.6 0.08 NITROGEN PHOSPHORUS 0.07 L L A A T 0.5 T 0.06 O O T T 0.05 0.4 0.04 0.03 0.3 0.02 0.01 0.2 0.00 0 10,000 20,000 30,000 40,000 50,000 0 10,000 20,000 30,000 40,000 50,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 300,000 50,000

EXPLANATION , 40,000 WATER-QUALITY SAMPLE

OUTLYING SAMPLE 200,000 95th PERCENTILE 30,000 75th PERCENTILE MEDIAN 25th PERCENTILE 5th PERCENTILE

20,000 ANEOUS STREAMFLOW

100,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND INST 10,000 INSTANTANEOUS TOTAL NITROGEN LOAD,

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 28. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Connecticut River at North Walpole, NH, station 01154500. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 

A 14,000,000

12,000,000 2003–05 10,000,000 Mean annual

8,000,000

6,000,000 NITROGEN LOAD, L A

T 4,000,000 O IN POUNDS PER YEAR T 2,000,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Connecticut River at Walpole, NH All other river sampling sites

Figure 29. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Connecticut River at North Walpole, NH, station 01154500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Otter River at Otter River, MA, Station 011200 0.337 mg/L (fig. 32D). Instantaneous loads of total nitrogen ranged from 268 to 2,140 lb/d (fig. 32E). Ranges of Otter River at Otter River, MA, represents a urban site concentrations or values for water-quality constituents not in the study area. The contributing drainage basin upstream shown in figure 32 are listed below. from the site encompasses about 34 mi2 and is about 58 per­ cent forested, 6 percent agriculture, 22 percent urban, and <, less than; —, not calculated 14 percent wetlands and barren (fig. 30). The annual stream- Water-quality Mini- Maxi­ constituenta Mean Median flow for 1980–2002 averaged 45,300 acre-feet, with consider­ mum mum able year-to-year variation (fig. 31A). Annual streamflow for Specific conductance 209 378 362 765 the data-collection period for water years 2003–05 averaged 53,400 acre-feet (table 1). Streamflow was greater than the pH 5.7 6.3 6.3 7.0 long-term mean for all 3 years of the study (fig. 31A). Water temperature .0 10.1 8.6 24.9 Daily mean streamflow and time distribution for 42 water Dissolved oxygen 6.6 9.8 9.3 14.7 samples are shown in figure 31B. The minimum streamflow at Dissolved nitrite .004 .024 .016 .214 which samples were collected was 6.4 ft3/s, and the maximum streamflow was 348 ft3/s. Dissolved ammonia <.020 .165 .104 .710 The minimum concentration of dissolved oxygen was Dissolved orthophos- .01 .06 .03 .21 6.6 mg/L and was within the Massachusetts water-quality phate (WY 2003–05) criterion for dissolved oxygen (Massachusetts Department of Suspended sediment 2 6 6 16 Environmental Protection, 2005). Concentrations of selected aAll constituents are reported as milligrams per liter except for specific water-quality constituents were plotted relative to streamflow conductance, which is reported as microsiemens per centimeter at 25 (fig. 32). Concentrations of dissolved nitrite plus nitrate degrees Celsius; pH, which is reported as standard units; and water tem- ranged from 0.226 to 9.66 mg/L and showed a typical dilution perature, which is reported as degrees Celsius; WY, water year. curve relative to streamflow (fig. 32A). Total ammonia plus organic nitrogen ranged from an estimated 0.33 to 1.1 mg/L The estimated load of total nitrogen was similar during and showed a decrease in concentrations with an increase in the first and second year of the study and slightly greater dur­ streamflow with the exception of the sample collected at a ing the third year (appendix A-1; fig. 33A). The mean annual streamflow of 348 ft3/s. This sample was collected just prior load of total nitrogen was 249,000 lb/yr, with a ratio of the to the first peak in maximum daily mean streamflow during standard error of prediction to the mean load of 3.9 percent snowmelt in water year 2005 (fig. 32B). Concentrations of (table 2, fig. 33A). An estimated 29 and 22 percent of the total nitrogen ranged from 0.59 to 10 mg/L and also showed total nitrogen load was transported in the spring and summer, a well-defined relation with streamflow (fig. 32C). The water respectively. The load of total nitrogen at this site was simi­ quality at this site is dominated by effluent and most of the lar among seasons (appendix A-2). The mean yield of total total nitrogen is in the form of dissolved nitrite plus nitrate. nitrogen (7,300 (lb/mi2)/yr) was greater than at all other sites This relation is typical of dilution curves found at sites (fig. 33B). Trench (1999) reported similar results in that basins that may be dominated by wastewater-treatment effluent. receiving effluent from wastewater-treatment facilities gener­ Concentrations of total phosphorus ranged from 0.023 to ally are greater in yields of total nitrogen than other basins. Characterization of Total Nitrogen at River Sampling Sites 5

Otter River at Otter River, MA, Station 01163200

EXPLANATION Land-Use Classification

Water Urban Barren Forest 1 2 NEW HAMPSHIRE Shrubland 71°58' 3 72°01'30" Agriculture 4 42°35'30" VERMONT Wetlands 5 Lebanon Basin boundary r e Otte v 6 r i

R R Streams i Gardner t v u e c r i Sampling site t 7 c

e

n n o C Keene 11 9 10 8 MASSACHUSETTS 12 Springfield 42°32'

13

Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000

0 1 2 3 4 MILES

0 1 2 3 4 KILOMETERS

Photograph by Tim Driskell, U.S. Geological Survey

Figure 0. Location, land-use cover classification, and photograph of sampling site for Otter River at Otter River, MA, station 01163200. (Refer to table 1 and figure 1 for additional site information.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 70 Historical record Study period 60 1980-2002 Mean annual

50

40 , IN THOUSANDS OF ACRE FEET 30

20

ANNUAL STREAMFLOW 10

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 800

Daily mean streamflow 700 Water-quality sample

600

500

, IN CUBIC FEET PER SECOND 400

300

200 Y MEAN STREAMFLOW

DAIL 100

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 1. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Otter River at Otter River, MA, station 01163200. Characterization of Total Nitrogen at River Sampling Sites 

Box Box A plot B plot 10 1.1

9 1.0 8 0.9 7 0.8 6

5 0.7

4 0.6 3 0.5 2 DISSOLVED NITRITE PLUS NITRATE 0.4

1 TOTAL AMMONIA PLUS ORGANIC NITROGEN

0 0.3 0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350 C D 11 0.34 0.32 TION, IN MILLIGRAMS PER LITER 10 0.30 9 0.28 8 0.26 0.24 CONCENTRA 7 0.22 6 0.20 0.18 5 NITROGEN PHOSPHORUS 0.16 L L A A T 4 T 0.14 O O T T 0.12 3 0.10 2 0.08 0.06 1 0.04 0 0.02 0 50 100 150 200 250 300 350 0 50 100 150 200 250 300 350

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 2,200 400

2,000 EXPLANATION , 1,800 WATER-QUALITY SAMPLE 300 1,600 OUTLYING SAMPLE 95th PERCENTILE 1,400 75th PERCENTILE MEDIAN 1,200 200 25th PERCENTILE 5th PERCENTILE

1,000 ANEOUS STREAMFLOW ANT IN POUNDS PER DAY Instantaneous streamflow

800 IN CUBIC FEET PER SECOND

100 INST 600

INSTANTANEOUS TOTAL NITROGEN LOAD, 400

200 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 2. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Otter River at Otter River, MA, station 01163200. (Refer to table 1 and figure 1 for station location.) 8 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 300,000 2003–05 Mean annual 250,000

200,000

150,000 NITROGEN LOAD, L 100,000 A T O IN POUNDS PER YEAR T 50,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Otter River at Otter River, MA All other river sampling sites

Figure . A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Otter River at Otter River, MA, station 01163200, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) Characterization of Total Nitrogen at River Sampling Sites 9

North River at Shattuckville, MA, generally followed a similar pattern as concentrations of total Station 0119000 ammonia plus organic nitrogen and likely were affected by an increase in concentrations of suspended sediment at greater North River at Shattuckville, MA, represents an streamflows. Instantaneous loads of total nitrogen ranged agricultural site in the study area. The contributing drainage from 47 to 6,590 lb/d (fig. 36E). Ranges of concentrations or basin upstream from the site encompasses about 89 mi2 and values for water-quality constituents not shown in figure 36 is about 84 percent forested, 9 percent agriculture, 2 percent are listed below. urban, and 5 percent wetlands and barren (fig. 34). Annual streamflow for 1980–2002 averaged 138,000 acre-feet, <, less than; —, not calculated with considerable year-to-year variation (fig. 35A). Annual Water-quality Mini- Maxi­ constituenta Mean Median streamflow for the data-collection period for water years mum mum 2003–05 averaged 167,000 acre-feet (table 1). Streamflow Specific conductance 47 107 105 212 was greater than the long-term mean for all 3 years of the pH 6.6 7.5 7.4 8.7 study (fig. 35A). Water temperature .0 9.4 7.7 22.5 Daily mean streamflows and the time distribution for 41 water samples are shown in figure 35B. The minimum stream- Dissolved oxygen 8.0 11.7 11.6 15.8 flow at which water samples were collected was 22 ft3/s, and Dissolved nitrite <.004 — <.004 .036 the maximum streamflow at which samples were collected Dissolved ammonia <.005 — <.005 .092 3 Dissolved orthophos­ was 2,030 ft /s. <.01 — <.01 .13 The minimum concentrations of dissolved oxygen was phate (WY 2003–04) Dissolved orthophos­ 8.0 mg/L and was within the Massachusetts water-quality <.003 — .007 .065 criterion for dissolved oxygen (Massachusetts Department of phate (WY 2005) Environmental Protection, 2005). Concentrations of selected Suspended sediment 1 10 2 184 water-quality constituents were plotted relative to streamflow aAll constituents are reported as milligrams per liter except for specific (fig. 36). Concentrations of dissolved nitrite plus nitrate conductance, which is reported as microsiemens per centimeter at 25 ranged from 0.082 to 1.35 mg/L and generally showed a slight degrees Celsius; pH, which is reported as standard units; and water tem- perature, which is reported as degrees Celsius; WY, water year. variation (dilution) in relation to streamflow (fig. 36A). Total ammonia plus organic nitrogen ranged from an estimated The estimated load of total nitrogen varied among 0.07 to 0.47 mg/L and generally showed at slight dilution years during the study period (appendix A-1; fig. 37A). with an increase in streamflow; however, an increase in total The mean annual load of total nitrogen was 194,000 lb/yr, with ammonia plus organic nitrogen was observed in samples collected at streamflows greater than about 500 ft3/s (fig. 36B). a ratio of the standard error of prediction to the mean load of An increase in concentration of suspended sediment relative 7.0 percent (table 2, fig. 37A). An estimated 43 percent of the to the other samples also was observed for the three samples total nitrogen load was transported during the spring (appendix collected at streamflows greater than 500 ft3/s. Concentrations A-2). The mean yield of total nitrogen (2,180 (lb/mi2)/yr) was of total nitrogen ranged from 0.25 to 1.7 mg/L (fig. 36C). greater than yield of total nitrogen at two of the agricultural Concentrations of total phosphorus ranged from 0.007 to sites in Vermont and less than the yield of total nitrogen at the 0.199 mg/L (fig. 36D). Concentrations of total phosphorus other agricultural site in Massachusetts (fig. 37B). 50 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

North River at Shattuckville, MA, Station 01169000

72°49'15"

EXPLANATION Land-Use Classification

Water 42°47'45" Urban Barren E a s Forest t B ranch N or th Shrubland Ri ve VERMONT r Agriculture 72°42' Wetlands

Basin boundary MASSACHUSETTS W Streams es ch No t Bran rth Ri ver Sampling site

Land-use data from U.S. Geological Survey, 1992 42°40'20" National Land Cover Dataset, 2000

NColrain o r t

h

R

i

0 1 2 3 4 MILES v

e

r

1 0 1 2 3 4 KILOMETERS 2 NEW HAMPSHIRE 3

4 VERMONT 5 Lebanon r e v i 6

R

t

u

c i t 7 c

e

n n o C Keene

11 9 8 10 MASSACHUSETTS 12 Springfield

13

Photograph courtesy of Roy Socolow, U.S. Geological Survey

Figure . Location, land-use classification, and photograph of sampling site for North River at Shattuckville, MA, station 01169000. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 51

A 250 Historical record Study period

200 1980-2002 Mean annual

150 , IN THOUSANDS OF ACRE FEET 100

50 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 5,000

Daily mean streamflow

Water-quality sample 4,000

3,000 , IN CUBIC FEET PER SECOND

2,000

Y MEAN STREAMFLOW 1,000 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 5. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for North River at Shattuckville, MA, station 01169000. 52 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 1.4 0.48 1.3 0.46 0.44 1.2 0.42 1.1 0.40 0.38 1.0 0.36 0.9 0.34 0.32 0.8 0.30 0.7 0.28 0.26 0.6 0.24 0.22 0.5 0.20 0.4 0.18 0.16 0.3

DISSOLVED NITRITE PLUS NITRATE 0.14 0.2 0.12

TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.10 0.1 0.08 0 0.06 0 500 1,000 1,500 2,000 0 500 1,000 1,500 2,000 C D 1.7 0.20 1.6 TION, IN MILLIGRAMS PER LITER 0.18 1.5 1.4 0.16 1.3 0.14

CONCENTRA 1.2 1.1 0.12 1.0 0.10 0.9 NITROGEN PHOSPHORUS L L A

0.8 A 0.08 T T O O T

0.7 T 0.06 0.6 0.5 0.04 0.4 0.02 0.3 0.2 0.00 0 500 1,000 1,500 2,000 0 500 1,000 1,500 2,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 7,000 3,000

6,000 EXPLANATION , WATER-QUALITY SAMPLE

5,000 OUTLYING SAMPLE 2,000 95th PERCENTILE 4,000 75th PERCENTILE MEDIAN 25th PERCENTILE 3,000 5th PERCENTILE ANEOUS STREAMFLOW

1,000 ANT IN POUNDS PER DAY Instantaneous streamflow 2,000 IN CUBIC FEET PER SECOND INST

1,000 INSTANTANEOUS TOTAL NITROGEN LOAD,

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure . Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for North River at Shattuckville, MA, station 01169000. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 5

A 250,000 2003–05 200,000 Mean annual

150,000

100,000 NITROGEN LOAD, L A T O IN POUNDS PER YEAR T 50,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION North River at Shattuckville, MA All other river sampling sites

Figure . A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for North River at Shattuckville, MA, station 01169000, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) 5 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

South River near Conway, MA, Station 0119900 estimated 0.002 to 0.050 mg/L (fig. 40D). Instantaneous loads of total nitrogen ranged from 11 to 525 lb/d (fig. 40E). Ranges South River near Conway, MA, represents an agricul­ of concentrations or values for water-quality constituents not tural site in the study area. The contributing drainage basin shown in figure 40 are listed below. upstream from the site encompasses about 24 mi2 and is about 81 percent forested, 10 percent agriculture, 4 percent urban, <, less than; —, not calculated and 5 percent wetlands and barren (fig. 38). Annual stream- Water-quality Mini- Maxi­ flow for 1980–2002 averaged 37,100 acre-feet, with consider­ constituenta Mean Median mum mum able year-to-year variation (fig. 39A). Annual streamflow for Specific conductance 103 161 157 218 the data-collection period for water years 2003–05 averaged 48,100 acre-feet (table 1). Streamflow was greater than the pH 6.7 7.4 7.4 8.5 long-term mean for all 3 years of the study (fig. 39A). Water temperature .0 9.2 7.4 24 Daily mean streamflows and time distribution for 41 Dissolved oxygen 7.9 11.2 11.3 14.4 water-quality samples are shown in figure 39B. The minimum Dissolved nitrite <.004 — <.004 .123 streamflow at which water samples were collected was 5 ft3/s, Dissolved ammonia <.005 — <.005 .020 and the maximum streamflow at which samples were collected Dissolved orthophos­ <.01 — <.01 <.01 was 197 ft3/s. phate (WY 2003–04) The minimum concentration of dissolved oxygen was Dissolved orthophos­ <.003 — <.003 <.003 7.9 mg/L and was within the Massachusetts water-quality phate (WY 2005) criterion for dissolved oxygen (Massachusetts Department of Suspended sediment 1 5 2 49 Environmental Protection, 2005). Concentrations of selected aAll constituents are reported as milligrams per liter except for specific water-quality constituents were plotted relative to streamflow conductance, which is reported as microsiemens per centimeter at 25 (fig. 40). Concentrations of dissolved nitrite plus nitrate degrees Celsius; pH, which is reported as standard units; and water tem- ranged from 0.175 to 0.635 mg/L and showed no relation to perature, which is reported as degrees Celsius; WY, water year. streamflow (fig. 40A). Total ammonia plus organic nitrogen ranged from an estimated 0.05 to 0.31 mg/L and showed no The estimated load of total nitrogen varied among years relation to streamflow (fig. 40B). The maximum concentration during the study (appendix A-1; fig. 41A). The mean annual of 0.31 mg/L was in a sample collected at a streamflow of load of total nitrogen was 53,700 lb/yr, with a ratio of the about 70 ft3/s during a small increase in streamflow that standard error of prediction to the mean load of 4.4 percent resulted from rain in August 2003. The concentration of (table 2, fig. 41A). An estimated 31 and 39 percent of the total suspended sediment in this sample was one of the highest nitrogen load was transported during the winter and spring, concentrations relative to other samples collected at the site. respectively (appendix A-2). The mean yield of total nitrogen Concentrations of total nitrogen ranged from 0.26 to 1.0 mg/L (2,230 (lb/mi2)/yr) was greater than at most of the other agri­ (fig. 40C). Concentrations of total phosphorus ranged from an cultural sites (fig. 41B). Characterization of Total Nitrogen at River Sampling Sites 55

South River near Conway, MA, Station 01169900

EXPLANATION Land-Use Classification

Water Urban 1 2 NEW Barren HAMPSHIRE 3 Forest 4 Shrubland VERMONT 5 Agriculture Lebanon r e v 6 Wetlands i R

t

u

c i Basin boundary t 7 c

e

n n Streams o C Keene Sampling site 11 9 Photograph by Tim Driskell, U.S. Geological Survey 8 10 MASSACHUSETTS 12 72°48' Springfield

13

42°32' 72°44'

Ashfield R th iv u e Land-use data from U.S. Geological Survey, 1992 o r S National Land Cover Dataset, 2000 Conway

42°28' 0 1 2 3 4 MILES

0 1 2 3 4 KILOMETERS

Figure 8. Location, land-use classification, and photograph of sampling site for South River near Conway, MA, station 01169900. (Refer to table 1 and figure 1 for additional site information.) 5 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 70 Historical record Study period 60

1980-2002 50 Mean annual

40 , IN THOUSANDS OF ACRE FEET 30

20

ANNUAL STREAMFLOW 10

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 1,400

Daily mean streamflow 1,200 Water-quality sample

1,000

800 , IN CUBIC FEET PER SECOND

600

400 Y MEAN STREAMFLOW 200 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 9. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for South River near Conway, MA, station 01169900. Characterization of Total Nitrogen at River Sampling Sites 5

Box Box A plot B plot 0.64 0.31 0.60 0.29 0.56 0.27 0.52 0.25 0.23 0.48 0.21 0.44 0.19 0.40 0.17 0.36 0.15 0.32 0.13 0.28 0.11

DISSOLVED NITRITE PLUS NITRATE 0.24 0.09

0.20 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.07 0.16 0.05 0 50 100 150 200 0 50 100 150 200 C D 1.1 0.050

TION, IN MILLIGRAMS PER LITER 0.046 1.0 0.042 0.9 0.038

CONCENTRA 0.8 0.034 0.030 0.7 0.026 NITROGEN

0.6 PHOSPHORUS L L 0.022 A A T T O O T

0.5 T 0.018 0.014 0.4 0.010 0.3 0.006 0.2 0.002 0 50 100 150 200 0 50 100 150 200

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 600 200

180 EXPLANATION

500 , 160 WATER-QUALITY SAMPLE

140 OUTLYING SAMPLE 400 95th PERCENTILE 120 75th PERCENTILE MEDIAN 300 100 25th PERCENTILE 5th PERCENTILE

80 ANEOUS STREAMFLOW

200 ANT IN POUNDS PER DAY Instantaneous streamflow

60 IN CUBIC FEET PER SECOND INST 40 100

INSTANTANEOUS TOTAL NITROGEN LOAD, 20

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 0. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for South River near Conway, MA, station 01169900. (Refer to table 1 and figure 1 for station location.) 58 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 70,000 2003–05 60,000 Mean annual 50,000

40,000

30,000 NITROGEN LOAD, L A

T 20,000 O IN POUNDS PER YEAR T 10,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION South River near Conway, MA All other river sampling sites

Figure 1. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for South River near Conway, MA, station 01169900, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) Characterization of Total Nitrogen at River Sampling Sites 59

Green River near Colrain, MA, Station 0110100 values for water-quality constituents not shown in figure 44 are listed below. Green River near Colrain, MA, represents a forested site in the study area. The contributing drainage basin <, less than; —, not calculated 2 Water-quality upstream from the site encompasses about 41 mi and is Mini- Maxi­ constituenta Mean Median about 92 percent forested, 5 percent agriculture, 1 percent mum mum urban, and about 2 percent wetlands and barren (fig. 42). Specific conductance 51 84 88 128 Annual streamflow for 1980–2002 averaged 61,400 acre-feet, pH 6.6 7.4 7.5 8.4 with considerable year-to-year variation (fig. 43A). Annual Water temperature .0 8.8 7.7 21.2 streamflow for the data-collection period for water years Dissolved oxygen 8.4 11.6 11.3 15.7 2003–05 averaged 74,400 acre-feet (table 1). Streamflow Dissolved nitrite <.004 — <.004 .006 was greater than the long-term mean for all 3 years of the Dissolved ammonia <.005 — <.005 .006 study (fig. 43A). Dissolved orthophos­ Daily mean streamflow and time distribution for 41 <.01 — <.01 <.01 phate (WY 2003–04) water-quality samples are shown in figure 43B. The mini­ Dissolved orthophos­ mum streamflow at which water samples were collected was <.003 — <.003 <.003 phate (WY 2005) 10 ft3/s, and the maximum streamflow at which samples were Suspended sediment 1 4 1 61 collected was 602 ft3/s. a The minimum concentration of dissolved oxygen was All constituents are reported as milligrams per liter except for specific conductance, which is reported as microsiemens per centimeter at 25 8.4 mg/L and was within the Massachusetts water-quality degrees Celsius; pH, which is reported as standard units; and water tem- criterion for dissolved oxygen (Massachusetts Department of perature, which is reported as degrees Celsius; WY, water year. Environmental Protection, 2005). Concentrations of selected water-quality constituents were plotted relative to streamflow The estimated load of total nitrogen varied among years (fig. 44). Concentrations of dissolved nitrite plus nitrate ranged and was considerably less during the third year of the study from less than 0.016 to 0.201 mg/L and showed no relation to (fig. 45A). The mean annual flow also was less during the streamflow (fig. 44A). Total ammonia plus organic nitrogen third year of the study (appendix A-1). The mean annual load ranged from an estimated 0.05 to 0.32 mg/L and also showed of total nitrogen was 49,100 lb/yr, with a ratio of the standard no relation to streamflow (fig. 44B). Concentrations of total error of prediction to the mean load of 10 percent (table 2, nitrogen ranged from 0.06 to 0.46 mg/L (fig. 44C). Concentra­ fig. 45A). This site had the greatest percentage of error relative tions of total phosphorus ranged from an estimated 0.002 to to other sites. An estimated 47 percent of the total nitrogen 0.067 mg/L (fig. 44D). The maximum concentration of total load was transported during the spring (appendix A-2). The phosphorus was observed at the second greatest streamflow at mean yield of total nitrogen (1,190 (lb/mi2)/yr) was less than which samples were collected; this sample had the maximum all of the other study sites (fig. 45B). In general, water quality concentration of suspended sediment (61 mg/L) relative to the at this site typifies reference conditions for this area of the other samples. Instantaneous loads of total nitrogen ranged basin. Concentrations of total nitrogen and other water-quality from 5.5 to 1,240 lb/d (fig. 44E). Ranges of concentrations or constituents generally were low. 0 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Green River near Colrain, MA, Station 01170100

EXPLANATION Land-Use Classification

Water Urban Barren Forest Shrubland Agriculture Photograph by Tim Driskell, U.S. Geological Survey Wetlands

Basin boundary 72°45' Streams Sampling site

Harrisville Brook

1 42°50' 2 NEW HAMPSHIRE 3 72°40' 4 VERMONT Green Riv 5 er Lebanon r e v 6 i

R

t

u

c i t 7 c

e

n VERMONT n o Roaring Brook C Keene 11 9 8 42°45' 10 MASSACHUSETTS 12 Springfield

Thorne Brook 13 MASSACHUSETTS 0 1 2 3 4 MILES

Land-use data from U.S. Geological Survey, 1992 0 1 2 3 4 KILOMETERS National Land Cover Dataset, 2000

Figure 2. Location, land-use classification, and photograph of sampling site for Green River near Colrain, MA, station 01170100. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 1

A 100 Historical record Study period 1980-2002 80 Mean annual

60 , IN THOUSANDS OF ACRE FEET 40

20 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 2,000

Daily mean streamflow

Water-quality sample

1,500

, IN CUBIC FEET PER SECOND 1,000

500 Y MEAN STREAMFLOW DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure . A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Green River near Colrain, MA, station 01170100. 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 0.21 0.32 0.31 0.30 0.19 0.29 0.28 0.17 0.27 0.26 0.25 0.15 0.24 0.23 0.22 0.13 0.21 0.20 0.19 0.11 0.18 0.17 0.09 0.16 0.15 0.14 0.07 0.13 0.12 0.05 0.11

DISSOLVED NITRITE PLUS NITRATE 0.10 0.09 0.08 0.03 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.07 0.06 0.01 0.05 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700 C D 0.46 0.07

TION, IN MILLIGRAMS PER LITER 0.42 0.06 0.38

0.34 0.05 CONCENTRA 0.30 0.04 0.26 NITROGEN PHOSPHORUS L

L 0.03 A

0.22 A T T O O T T 0.18 0.02 0.14 0.01 0.10

0.06 0.00 0 100 200 300 400 500 600 700 0 100 200 300 400 500 600 700

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 1,300 700 1,200 1,100 600 EXPLANATION , WATER-QUALITY SAMPLE 1,000 900 500 OUTLYING SAMPLE 95th PERCENTILE 800 75th PERCENTILE 400 700 MEDIAN 600 25th PERCENTILE 300 5th PERCENTILE 500 ANEOUS STREAMFLOW ANT IN POUNDS PER DAY 400 Instantaneous streamflow 200 IN CUBIC FEET PER SECOND

300 INST 200 100 INSTANTANEOUS TOTAL NITROGEN LOAD, 100 0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure . Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Green River near Colrain, MA, station 01170100. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 

A 70,000

60,000 2003–05 50,000 Mean annual

40,000

30,000 NITROGEN LOAD, L A

T 20,000 O IN POUNDS PER YEAR T 10,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Green River near Colrain, MA All other river sampling sites

Figure 5. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Green River near Colrain, MA, station 01170100, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Mill River at Northhampton, MA, snowmelt. This sample also had the greatest concentration of Station 0111500 suspended sediment (17 mg/L) relative to the other samples. Instantaneous loads of total nitrogen ranged from 39 to Mill River at Northhampton, MA, represents an urban 958 lb/d (fig. 48E). Ranges of concentrations or values site in the study area. The contributing drainage basin for water-quality constituents not shown in figure 48 are upstream from this site encompasses about 54 mi2 and is listed below. about 77 percent forested, 6 percent agriculture, 10 percent urban, and about 7 percent wetlands and barren (fig. 46). <, less than; —, not calculated Water-quality Annual streamflow for 1980–2002 averaged 73,000 acre-feet, Mini- Maxi­ constituenta Mean Median with considerable year-to-year variation (fig. 47A). Annual mum mum streamflow for the data-collection period for water years Specific conductance 65 108 107 160 2003–05 averaged 88,400 acre-feet (table 1). Streamflow pH 6.3 7.2 7.2 8.0 was greater than the long-term mean for all 3 years of the Water temperature .0 10.1 8.3 22.6 study (fig. 47A). Daily mean streamflow and time distribution for 43 water Dissolved oxygen 7.7 11.2 11.2 15.1 samples are shown in figure 47B. The minimum streamflow Dissolved nitrite <.004 — <.004 .087 at which water samples were collected was 12 ft3/s, and the Dissolved ammonia <.005 <.005 .022 Dissolved orthophos­ maximum streamflow at which samples were collected was <.01 — <.01 <.01 512 ft3/s. phate (WY 2003–04) Dissolved orthophos­ The minimum concentration of dissolved oxygen was <.003 — .003 .007 7.7 mg/L and was within the Massachusetts water-quality phate (WY 2005) criterion for dissolved oxygen (Massachusetts Department of Suspended sediment 1 3 2 17 Environmental Protection, 2005). Concentrations of selected aAll constituents are reported as milligrams per liter except for specific water-quality constituents were plotted relative to streamflow conductance, which is reported as microsiemens per centimeter at 25 (fig. 48). Concentrations of dissolved nitrite plus nitrate ranged degrees Celsius; pH, which is reported as standard units; and water tem- perature, which is reported as degrees Celsius; WY, water year. from less than 0.160 to 0.570 mg/L and, although some scatter was shown, there also was a slight dilution effect relative to The estimated load of total nitrogen was similar during streamflow (fig. 48A). Total ammonia plus organic nitrogen the first and second year of the study and less during the third ranged from an estimated 0.07 to 0.31 mg/L and showed no relation to streamflow (fig. 48B). Concentrations of total year of the study (appendix A-1; fig. 49A). The mean annual nitrogen ranged from 0.29 to 0.68 mg/L and showed a pattern load of total nitrogen was 95,400 lb/yr, with a ratio of the similar to dissolved nitrite plus nitrate (fig. 48C). Most of the standard error of prediction to the mean load of 3.0 percent total nitrogen at this site was in the form of dissolved nitrite (table 2, fig. 49A). An estimated 31 and 37 percent of the total plus nitrate. Concentrations of total phosphorus ranged from nitrogen load was transported during the winter and spring, an estimated 0.006 to 0.026 mg/L (fig. 48D). The maximum respectively (appendix A-2). The mean yield of total nitrogen concentration of total phosphorus was observed at the greatest (1,770 (lb/mi2)/yr) was similar to the yield of total nitrogen at streamflow at which samples were collected which was during other sites (fig. 49B). Characterization of Total Nitrogen at River Sampling Sites 5

Mill River at Northampton, MA, Station 01171500

EXPLANATION Land-Use Classification

Water Urban Barren Forest Shrubland Agriculture Wetlands Photograph by Tim Driskell, U.S. Geological Survey Basin boundary 72°45' Streams Sampling site

1 2 NEW HAMPSHIRE 3 72°40' 4 42°25' VERMONT 5 Lebanon r e v 6 i

R

t

Williamsburg u

c i t 7 c M e n il n l o R C Keene i v

e

r 11 9 10 8

R MASSACHUSETTS o b 12 e Springfield rt s M ea k d oo ow Br 13 42°20'

Northampton 0 1 2 3 4 MILES Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000 0 1 2 3 4 KILOMETERS

Figure . Location, land-use classification, and photograph of sampling site for Mill River at Northampton, MA, station 01171500. (Refer to table 1 and figure 1 for additional site information.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 120 Historical record Study period 100 1980-2002 Mean annual

80

60 , IN THOUSANDS OF ACRE FEET

40

20 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 2,500

Daily mean streamflow

Water-quality sample 2,000

1,500 , IN CUBIC FEET PER SECOND

1,000

Y MEAN STREAMFLOW 500 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure . A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Mill River at Northampton, MA, station 01171500. Characterization of Total Nitrogen at River Sampling Sites 

Box Box A plot B plot 0.58 0.32 0.56 0.31 0.54 0.30 0.29 0.52 0.28 0.50 0.27 0.48 0.26 0.46 0.25 0.44 0.24 0.23 0.42 0.22 0.40 0.21 0.38 0.20 0.36 0.19 0.34 0.18 0.32 0.17 0.16 0.30 0.15 0.28 0.14 0.26 0.13 0.12

DISSOLVED NITRITE PLUS NITRATE 0.24 0.22 0.11 0.10 0.20 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.09 0.18 0.08 0.16 0.07 0 100 200 300 400 500 0 100 200 300 400 500 C D 0.70 0.026 0.68 TION, IN MILLIGRAMS PER LITER 0.66 0.024 0.64 0.62 0.022 0.60 0.58 0.020

CONCENTRA 0.56 0.54 0.018 0.52 0.50 0.48 0.016 NITROGEN PHOSPHORUS L

0.46 L A

A 0.014 T

0.44 T O O T

0.42 T 0.40 0.012 0.38 0.36 0.010 0.34 0.32 0.008 0.30 0.28 0.006 0 100 200 300 400 500 0 100 200 300 400 500

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 1,000 600

900 EXPLANATION

500 , 800 WATER-QUALITY SAMPLE

700 OUTLYING SAMPLE 400 95th PERCENTILE 600 75th PERCENTILE MEDIAN 500 300 25th PERCENTILE 5th PERCENTILE

400 ANEOUS STREAMFLOW

200 ANT IN POUNDS PER DAY Instantaneous streamflow

300 IN CUBIC FEET PER SECOND INST 200 100

INSTANTANEOUS TOTAL NITROGEN LOAD, 100

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 8. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Mill River at Northampton, MA, station 01171500. (Refer to table 1 and figure 1 for station location.) 8 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

A 120,000 2003–05 100,000 Mean annual

80,000

60,000 NITROGEN LOAD, L 40,000 A T O IN POUNDS PER YEAR T 20,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Mill River at Northampton, MA All other river sampling sites

Figure 9. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Mill River at Northampton, MA, station 01171500, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.) Characterization of Total Nitrogen at River Sampling Sites 9

Connecticut River at Thompsonville, CT, at the maximum streamflow at which samples were collected. Station 0118000 This sample also had the greatest concentration of suspended sediment (135 mg/L) relative to the other samples. Instanta­ Connecticut River at Thompsonville, CT, represents a neous loads of total nitrogen ranged from 14,300 to main-stem site in the study area and serves as the outlet site to 288,000 lb/d (fig. 52E). Ranges of concentrations or values the upper Connecticut River Basin. The contributing drainage for water-quality constituents not shown in figure 52 are basin upstream from the site encompasses about 9,660 mi2 listed below. and is about 80 percent forested, 9 percent agriculture, 4 percent urban, and about 6 percent wetlands and barren <, less than; —, not calculated Water-quality (fig. 50). Annual streamflow for 1980–2002 averages Mini- Maxi­ constituenta Mean Median 12,700,000 acre-feet, with considerable year-to-year variation mum mum (fig. 51A). Annual streamflow for the data-collection period Specific conductance 85 142 141 230 for water years 2003–05 averaged 13,000,000 acre-feet pH 6.8 7.4 7.4 7.7 (table 1). Streamflow was less than the long-term mean for Water temperature .0 13.3 14.5 27 the first year and greater than the long-term mean for the Dissolved oxygen 6.7 10.7 10.4 14.6 second year and about average for the third year of the Dissolved nitrite <.004 — .007 .031 study (fig. 51A). Dissolved ammonia .020 .057 .041 .227 Daily mean streamflows and the time distribution of Dissolved orthophos­ <.01 — <.01 .094 the collection of 41 water-quality samples are shown in phate (WY 2003–04) figure 51B. The minimum streamflow at which water samples 1 12 3 135 3 Suspended sediment were collected was 4,930 ft /s, and the maximum streamflow a 3 All constituents are reported as milligrams per liter except for specific at which samples were collected was 77,200 ft /s. conductance, which is reported as microsiemens per centimeter at 25 The minimum concentration of dissolved oxygen was degrees Celsius; pH, which is reported as standard units; and water tem- 6.7 mg/L and was within the Connecticut water-quality perature, which is reported as degrees Celsius; WY, water year. criterion for dissolved oxygen (Connecticut Department of Environmental Protection, 2002). Concentrations of selected The estimated load of total nitrogen varied among years water-quality constituents were plotted relative to stream- during the study period (appendix A-1; fig. 53A). The mean flow (fig. 52). Concentrations of dissolved nitrite plus nitrate annual load of total nitrogen was 21,600,000 lb/yr with a ranged from less than 0.126 to 0.632 mg/L and showed a ratio of the standard error of prediction to the mean load slight dilution relative to streamflow except at extreme high of 3.0 percent (table 2, fig. 53A). An estimated 41 percent streamflows (fig. 52A). Total ammonia plus organic nitrogen of the total nitrogen load was transported during the spring ranged from 0.22 to 0.46 mg/L and showed no apparent rela­ (appendix A-2). The mean load estimate at this site represents tion to streamflow (fig. 52B). Concentrations of total nitrogen the cumulative load of total nitrogen leaving the upper ranged from 0.44 to 1.0 mg/L (fig. 52C). Concentrations of Connecticut River Basin. The mean yield of total nitrogen total phosphorus ranged from 0.021 to 0.122 mg/L (fig. 52D). (2,230 lb/mi2/yr) was greater than the yield at other main-stem The maximum concentration of total phosphorus was observed sites and at most agricultural and urban sites (fig. 53B). 0 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Connecticut River at Thompsonville, CT, Station 01184000

71°15'

EXPLANATION Land-Use 45° Classification

Water 72° Urban Barren Forest 1 St Johnsbury Shrubland 2 NEW HAMPSHIRE Agriculture 44°15' Wetlands 71°45' 3 Basin boundary Streams 4 W Sampling site hi te R i v 5 er VERMONT Lebanon

r 43°30' e v i Claremont R

t 0 10 20 30 40 MILES

u

c i 6 t

c e 0 10 20 30 40 KILOMETERS n W n es o

t C 7 R iv e r

Keene BratD tleboro

e

e

r

f i 42°45' e ld r 9 ve Rive i r Millers R 10 8 11 Gardner MASSACHUSETTS Amherst W Quabbin

s 12

t Reservoir f

i

e

l d R iv er Springfield

42° 13 Land-use data from U.S. Geological Survey, 1992 National Land Cover Dataset, 2000 Photograph courtesy of John Mullaney, U.S. Geological Survey

Figure 50. Location, land-use classification, and photograph of sampling site for Connecticut River at Thompsonville, CT, station 01184000. (Refer to table 1 and figure 1 for additional site information.) Characterization of Total Nitrogen at River Sampling Sites 1

A 20,000 Historical record Study period 1980-2002 Mean annual 15,000

10,000 , IN THOUSANDS OF ACRE FEET

5,000 ANNUAL STREAMFLOW

0 1980 1985 1990 1995 2000 2005 WATER YEAR

B 120,000

Daily mean streamflow

100,000 Water-quality sample

80,000

, IN CUBIC FEET PER SECOND 60,000

40,000 Y MEAN STREAMFLOW 20,000 DAIL

0 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 Apr 1 Oct 1 2003 2004 2005

WATER YEAR

Figure 51. A, Historical and study-period annual streamflow, and B, Daily mean streamflow and time distribution of water-quality samples for Connecticut River at Thompsonville, CT, station 01184000. 2 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Box Box A plot B plot 0.7 0.46 0.45 0.44 0.43 0.6 0.42 0.41 0.40 0.39 0.5 0.38 0.37 0.36 0.35 0.34 0.4 0.33 0.32 0.31 0.30 0.3 0.29 0.28 0.27 0.26 DISSOLVED NITRITE PLUS NITRATE 0.2 0.25 0.24 TOTAL AMMONIA PLUS ORGANIC NITROGEN 0.23 0.22 0.1 0.21 0 15,000 30,000 45,000 60,000 75,000 0 15,000 30,000 45,000 60,000 75,000 C D 1.0 0.13

TION, IN MILLIGRAMS PER LITER 0.12

0.9 0.11

0.10

CONCENTRA 0.8 0.09

0.08 0.7

NITROGEN 0.07 PHOSPHORUS L L A A T T

O 0.06 O T

0.6 T 0.05

0.5 0.04 0.03

0.4 0.02 0 15,000 30,000 45,000 60,000 75,000 0 15,000 30,000 45,000 60,000 75,000

INSTANTANEOUS STREAMFLOW, IN CUBIC FEET PER SECOND

E 300,000 80,000

70,000 EXPLANATION , WATER-QUALITY SAMPLE 60,000 OUTLYING SAMPLE 200,000 50,000 95th PERCENTILE 75th PERCENTILE MEDIAN 40,000 25th PERCENTILE 5th PERCENTILE 30,000 ANEOUS STREAMFLOW

100,000 ANT IN POUNDS PER DAY Instantaneous streamflow IN CUBIC FEET PER SECOND

20,000 INST

INSTANTANEOUS TOTAL NITROGEN LOAD, 10,000

0 0 October April October April October April October 2002 2003 2003 2004 2004 2005 2005

Figure 52. Distribution of A, Dissolved nitrite plus nitrate, B, Total ammonia plus organic nitrogen, C, Total nitrogen, D, Total phos­ phorus concentrations relative to streamflow, and E, Instantaneous total nitrogen load relative to time for Connecticut River at Thompsonville, CT, station 01184000. (Refer to table 1 and figure 1 for station location.) Characterization of Total Nitrogen at River Sampling Sites 

A 25,000,000 2003–05 Mean annual 20,000,000

15,000,000

10,000,000 NITROGEN LOAD, L A T O IN POUNDS PER YEAR T 5,000,000

0 2003 2004 2005 WATER YEAR

B 8,000

7,000

6,000

5,000

4,000 NITROGEN YIELD, L A T

O 3,000 T N A

E 2,000 M

IN POUNDS PER SQUARE MILE YEAR 1,000

0 1 11 2 4 5 9 10 6 8 12 3 7 13 Forest Agricultural Urban Main stem

SAMPLING SITE NUMBER AND CLASSIFICATION

EXPLANATION Connecticut River at Thompsonville, CT All other river sampling sites

Figure 5. A, Total nitrogen load, by year, and B, Mean annual total nitrogen yield (2003–05) for Connecticut River at Thompsonville, CT, station 01184000, in relation to all other river sampling sites. (Refer to table 1 and figure 1 for site names and locations.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Characterization of Total Nitrogen at which are summed to calculate total nitrogen, varied among wastewater-treatment sampling sites (fig. 54A). Ranges in Wastewater-Treatment Sampling Sites mean concentrations of total nitrogen at wastewater-treatment sampling sites (4.4 to 30 mg/L) were greater than the ranges Data collected for this study were used to characterize in mean concentrations of total nitrogen at river sampling sites the spatial distribution and variability of total nitrogen among (0.19 to 2.8 mg/L). 19 wastewater-treatment sampling sites. Sixteen sampling Instantaneous mean loads of total nitrogen from sites were located at municipal wastewater-treatment facilities. municipal wastewater-treatment sampling sites ranged from Three sampling sites were located at paper mill wastewater- 36 to 1,780 lb/d (table 3; fig. 54B). Instantaneous mean loads treatment facilities. Summary statistics concentrations and of total nitrogen at the three paper mill wastewater-treatment instantaneous loads of total nitrogen are listed in table 3. sampling sites ranged from 96 to 160 lb/d (table 3; fig. 54B). Mean concentrations of total nitrogen ranged from 4.4 Based on the sites and data for this study, the paper mill to 30 mg/L at wastewater-treatment sampling sites (table 3; wastewater-treatment facilities were contributing less total fig. 54A). The concentrations of dissolved nitrite plus nitrate nitrogen load than the municipal wastewater-treatment and total ammonia plus organic nitrogen, the constituents, facilities.

Table . Selected characteristics and summary statistics for concentrations of total nitrogen and instantaneous loads of total nitrogen for the wastewater-treatment sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts.

[USGS, U.S. Geological Survey; ft3/s, cubic feet per second; mg/L, milligrams per day; lb/d, pounds per day; Min, minimum; Max, maximum; WWTF, wastewater-treatment facility]

Mean Instantaneous total Total nitrogen Site Number instanta­ nitrogen load USGS (mg/L) number Station name of neous (lb/d) station number (fig. 1) samples flow (ft/s) Min Mean Max Min Mean Max 14a 443558071303001 Wasau Paper WWTF, NH 12 7.1 1.1 4.4 16 44 160 521 15 442906071354401 Lancaster WWTF, NH 10 2.1 4.7 8.8 13 38 106 221 16 442436072010001 St, Johnsbury WWTF, VT 19 2.3 4.5 12 21 54 146 318 17 441827071473701 Littleton WWTF, NH 10 1.7 3.3 10 18 31 88 131 18 440835072022501 Woodsville WWTF, NH 13 0.4 4.6 16 59 4 36 119 19 434145072175601 Hanover WWTF, NH 17 3.1 19 30 38 201 498 697 20 433844072185501 Hartford/White River 10 1.7 23 30 35 130 267 376 Junction WWTF, VT 21 433813072192001 Lebanon WWTF, NH 17 3.4 12 19 29 194 346 658 22 432359072234001 Claremont WWTF, NH 14 2.5 4.1 14 22 55 185 330 23 431648072280801 Springfield WWTF, VT 14 2.1 8.4 12 21 57 145 293 24 430745072263701 Bellows Falls WWTF, VT 10 1.4 12 21 32 73 147 262 25a 425322072323701 Fibermark Products 10 2.5 2.3 7.3 11 31 96 150 WWTF, VT 26 425029072325901 Brattleboro WWTF, VT 12 3.2 13 20 35 200 330 636 27a 423542072031001 Seaman Paper WWTF, MA 11 1.4 3.3 14 50 21 97 262 28 423413072011201 Gardner WWTF, MA 10 6.2 9.5 22 28 478 702 914 29 421132072365001 Holyoke WWTF, MA 11 12.8 1.9 5.6 12 150 339 621 30 420908072373301 Chicopee WWTF, MA 15 17.3 13 18 24 1,080 1,640 5,330 31 420702072435601 Westfield WWTF, MA 11 6.4 13 19 26 500 654 822 32 420508072351401 Springfield WWTF, MA 14 66.5 3.2 5.0 6.3 1,290 1,780 2,150 a Paper mill. Characterization of Total Nitrogen at Wastewater-Treatment Sampling Sites 5

A 35

30

25 EXPLANATION 20 Dissolved nitrite plus nitrate

15 Total ammonia plus organic nitrogen NITROGEN CONCENTRATION, L A

T 10 IN MILLIGRAMS PER LITER O T N A

E 5 M

0 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

B 2,000

1,800

1,600

1,400 EXPLANATION NITROGEN LOAD, L 1,200 A

T Municipal wastewater- O T 1,000 treatment facility N A E Paper mill wastewater-

M 800

S treatment facility U IN POUNDS PER DAY

O 600 E N A

T 400 N A T

S 200 N I 0 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

WASTEWATER-TREATMENT SAMPLING SITE

Figure 5. A, Mean total nitrogen concentration, and B, Instantaneous mean total nitrogen load for wastewater- treatment sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts. (Refer to table 1 and figure 1 for site names and locations.)  Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Comparison of Total Nitrogen nitrogen at forested sites were significantly less than at all other site types (p<0.05). Concentrations of total nitrogen at Concentrations and Yields Among Site agricultural, urban, and main-stem sites were not significantly different among each other (p>0.05), but were significantly Types greater (p<0.05) than at forested sites and significantly less than concentrations at wastewater-treatment sites (p<0.05). The sampling sites were grouped to compare concentra­ tions and yields of total nitrogen among site types. The five Concentrations of total nitrogen at wastewater-treatment site types were forested, agricultural, urban, main stem, and sites were significantly different from those at all other site wastewater treatment. Forest was the predominant land use types (p<0.05). at all sites in the basin. Sites that were selected to represent Nitrogen yields at forested sites generally were less agricultural and urban land use in this study generally had than yields at agricultural and urban sites; however, nitrogen agriculture and urban land use as the second predominant land yields at agricultural sites generally were greater than yields use in the basin. at urban sites. Annual yields of total nitrogen ranged from 732 The median concentration of total nitrogen was to 1,920 (lb/mi2)/yr at forested sites; 1,550 to 2,980 (lb/mi2)/yr 0.24 mg/L at forested sites, 0.48 mg/L at agricultural sites, at agricultural sites; 1,280 to 1,860 (lb/mi2)/yr at urban sites 0.54 mg/L at urban sites, 0.48 mg/L at main-stem sites, and that were not directly affected by wastewater effluent; 7,090 14 mg/L at wastewater-treatment sites (fig. 55). Comparison to 7,700 (lb/mi2)/yr at a urban site directly affected by waste­ of concentrations of total nitrogen among the site types water effluent; and 1,300 to 2,390 (lb/mi2)/yr at main-stem indicated significant differences. Concentrations of total sites (appendix A-1).

A B B B D 100

EXPLANATION

90th percentile 10 75th percentile 50th percentile (median) 25th percentile 10th percentile 1 (19, 240) Number of sites, number of samples TOTAL NITROGEN, IN MILLIGRAMS PER LITER 0.1 Forest Agricultural Urban Main stem WWTF (2, 84) (5, 213) (3, 128) (3, 128) (19, 240)

Figure 55. Total nitrogen concentrations among site types in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts. (WWTF, wastewater-treatment facility; results of Tukey’s multiple comparison test [Helsel and Hirsch, 1992] among sites are presented as letters, and concentrations with at least one letter in common do not differ significantly.) Summary and Conclusions 

Summary and Conclusions Instantaneous mean loads of total nitrogen ranged from 162 to 58,300 pounds per day (lb/d). Estimated mean annual loads Low concentrations (less than 3.5 milligrams per liter of total nitrogen at river sampling sites ranged from 49,100 (mg/L)) of dissolved oxygen (hypoxia) caused by nitrogen to 21.6 million pounds per year (lb/yr) with about 30 to 55 enrichment often occur during the summer in the western percent being transported during the spring. The estimated part of Long Island Sound (LIS). Excess nitrogen levels mean annual yields of total nitrogen ranged from 1,190 to lead to the enhanced growth of algae, which eventually die 7,300 pounds per square mile per year (lb/mi2)/yr. and decay, consuming oxygen in the process. The ensuing Mean concentrations of total nitrogen ranged from 4.4 to hypoxic conditions have resulted in the exceedance of 30 mg/L at wastewater-treatment sampling sites. Instantaneous dissolved-oxygen water-quality standards in Long Island mean loads of total nitrogen from municipal wastewater-treat­ Sound (LIS) during the summer. As a result, Connecticut and ment facilities ranged from 36 to 1,780 lb/d. Instantaneous New York developed a Total Maximum Daily Load (TMDL) mean loads of total nitrogen from three paper mill wastewater- for dissolved oxygen that specifies the maximum amount treatment facilities ranged from 96 to 160 lb/d. of nitrogen that can be discharged to LIS without exceeding Sampling sites were grouped to compare concentrations water-quality standards for dissolved oxygen. According to the and yields of total nitrogen among forested, agricultural, TMDL analysis, an annual baseline load of about 53,300 tons urban, main stem, and wastewater-treatment site types. The of nitrogen is delivered to LIS from all sources in Connecticut median concentration of total nitrogen was 0.24 mg/L at and New York combined. An additional estimated 13,600 forested sites, 0.48 mg/L at agricultural sites, 0.54 mg/L at tons of nitrogen reach LIS annually from the portion of the urban sites, 0.48 mg/L at main-stem sites, and 14 mg/L at watershed originating north of Connecticut. Of this total load, wastewater-treatment sites. Concentrations of total nitrogen an estimated 12,500 tons of nitrogen are delivered to LIS by at forested sites were significantly less than at all other way of the Connecticut River. site types (p<0.05). Concentrations of total nitrogen at To better define loads in the upper Connecticut River agricultural, urban, and main-stem sites were not significantly Basin, the U.S. Geological Survey (USGS), in cooperation different among each other (p>0.05) but were significantly with the New England Interstate Water Pollution Control greater (p<0.05) than at forested sites and significantly less Commission, designed and operated a surface-water quality than concentrations at wastewater-treatment sites (p<0.05). monitoring network from December 2002 to September 2005 Concentrations of total nitrogen at wastewater-treatment sites to collect water-quality and streamflow data and to use these were significantly different from all other site types (p<0.05). data to determine the current concentrations, loads, and yields Nitrogen yields at forested sites generally were less of nitrogen from selected sites in the upper Connecticut than yields at agricultural and urban sites; however, nitrogen River Basin. yields at agricultural sites generally were greater than yields A study was conducted at 13 river sites in the upper at urban sites. Annual yields of total nitrogen ranged from 732 Connecticut River Basin. Ten sites were selected to represent to 1,920 (lb/mi2)/yr at forested sites; 1,550 to 2,980 (lb/mi2)/yr contributions of nitrogen from forested, agricultural, and 2 urban land. Forest was the predominant land use at all sites at agricultural sites; 1,280 to 1,860 (lb/mi )/yr at urban sites in the basin. Sites that were selected to represent agricultural that were not directly affected by wastewater effluent; 7,090 to 2 and urban land use in this study generally had agriculture 7,700 (lb/mi )/yr at a urban site directly affected by wastewater 2 and urban land use as the second predominant land use in the effluent; and 1,300 to 2,390 (lb/mi )/yr at main-stem sites. basin. Three sites were distributed spatially on the main stem In this study, the mean annual load and yield of total of Connecticut River to assess the cumulative loads of total nitrogen at the Connecticut River at Wells River, VT, was nitrogen. To further improve the understanding the sources estimated at 4.47 million lb/yr and 1,690 (lb/mi2)/yr, respec­ and concentrations and loads of total nitrogen in the upper tively. The mean annual load and yield of total nitrogen at Connecticut River Basin, ambient water-quality sampling at the Connecticut River at North Walpole, NH, was estimated river sites was supplemented with sampling and analysis of at 9.60 million lb/yr and 1,750 (lb/mi2)/yr, respectively. The effluent from wastewater treatment at 19 municipal and paper mean annual load and yield of total nitrogen leaving the upper mill wastewater-treatment facilities. Connecticut River Basin, as estimated at the Connecticut Mean concentrations of total nitrogen ranged from 0.19 River at Thompsonville, CT, was 21.6 million lb/yr and to 2.8 milligrams per liter (mg/L) at river sampling sites. 2,230 (lb/mi2)/yr, respectively. 8 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

References Cited Moore, R.B., Johnston, C.M., Robinson, K.W., and Deacon, J.R., 2004, Estimation of total nitrogen and total phosphorus in New England streams using spatially referenced regres­ Akaike, H., 1981, Likelihood of a model and information sion models: U.S. Geological Survey Scientific Investiga­ criterion: Journal of Econometrics, v. 16, p. 13–14. tions Report 2004–5012, 42 p. Connecticut Department of Environmental Protection, 2002, Mueller, D.K., Hamilton, P.A., Helsel, D.R., Hitt, K.J., and Water-quality standards: Hartford, CT, Connecticut Depart­ ment of Environmental Protection, variously paged. Ruddy, B.C., 1995, Nutrients in ground water and surface water of the United States—An analysis of data through Childress, C.J.O., Foreman, W.T., Connor, B.F., and Malo­ 1992: U.S. Geological Survey Water Resources Investiga­ ney, T.J., 1999, New reporting procedures based on long- tions Report 95–4031, 74 p. term method detection levels and some considerations for interpretations of water-quality data provided by the U.S. Mullaney, J.R., 2004, Summary of water quality trends in the Geological Survey National Water Quality Laboratory: Connecticut River, 1968–1998: American Fisheries Society U.S. Geological Survey Open-File Report 99–193, 19 p. Monograph, v. 9, p. 273–286.

Garabedian, S.P., Coles, J.F., Grady, S.J., Trench, E.C.T., and Mullaney, J.R., Schwarz, G.E., and Trench, E.C.T., 2002, Zimmerman, M.J., 1998, Water quality in the Connecticut, Estimation of nitrogen yields and loads from basins draining Housatonic, and Thames River Basins, Connecticut, Mas­ to Long Island Sound, 1998–98: U.S. Geological Survey sachusetts, New Hampshire, and Vermont, 1992–95: U.S. Water-Resources Investigations Report 02–4044, 84 p. Geological Survey Circular 1155, 32 p. New Hampshire Department of Environmental Services, 2004, Grady, S.J., and Garabedian, S.P., 1991, National Water-Qual­ Surface water-quality assessments, sections 305(b) and ity Assessment Program—The Connecticut River and Long 303(d), Consolidated assessment and listing methodology: Island Sound Coastal Rivers: U.S. Geological Survey Concord, NH, Watershed Management Bureau, variously Open-File Report 91–159, 2 p. paged. Helsel, D.R., 2005, Insider censoring: Distortion of data with New York State Department of Environmental Conservation/ nondetects: Human and Ecological Risk Assessment, v. 11, Connecticut Department of Environmental Protection, 2000, p. 1127–1137. A total maximum daily load analysis to achieve water-qual­ Helsel, D.R., and Hirsch, R.M., 1992, Statistical methods in ity standards for dissolved oxygen in Long Island Sound: water resources: New York, Elsevier Science Publishers, Albany, NY, and Hartford, CT, New York State Department 522 p. of Environmental Conservation/Connecticut Department of Environmental Protection, 73 p. Keirstead, Chandlee, Kiah, R.G., Brown, R.O., and Ward, S.L., 2003, Water resources data, New Hampshire and Ver­ Runkel, R.L., Crawford, C.G., and Cohn, T.A., 2004, Load mont, water year 2003: U.S. Geological Survey Water-Data estimator (LOADEST)—A FORTRAN program for esti­ Report NH-VT-03-1, 323 p. mating constituent loads in streams and rivers: U.S. Geo­ Keirstead, Chandlee, Kiah, R.G., Ward, S.L., and Hilgendorf, logical Survey Techniques and Methods, book 4, chap. A5, G.S., 2004, Water resources data, New Hampshire and Ver­ 69 p., available online at http://pubs.water.usgs.gov/tm4A5. mont, water year 2004: U.S. Geological Survey Water-Data SAS Institute, Inc., 1998, Statview user’s guide, version 5: Report NH-VT-04-1, 335 p. Cary, NC, SAS Institute, Inc. Kiah, R.G., Keirstead, Chandlee, Brown, R.O., and Hilgen­ dorf, G.S., 2005, Water resources data, New Hampshire and SAS Institute, Inc., 2000, SAS/STAT user’s guide, version 8: Vermont, water year 2005: U.S. Geological Survey Water- Cary, NC, SAS Institute, Inc. Data Report NH-VT-05-1, 309 p. Seitzinger, S.P., Styles, R.V., Boyer, E.W., Alexander, R.B., Massachusetts Department of Environmental Protection, 2005, Billen, Gilles, Howarth, R.W., Mayer, Bernhard, and Massachusetts surface water quality standards, Division of Breemen, N.V., 2002, Nitrogen retention in rivers—Model Water Pollution Control, accessed December 2005, at development and application to watersheds in the northeast­ http://www.mass.gov/dep/water/laws/regulati.htm#wqual. ern U.S.A.: Biogeochemistry, v. 57/58, p. 199–237. References Cited 9

Socolow, R.S., Zanca, J.L., Driskell, T.R., and Ramsbey, L.R., U.S. Environmental Protection Agency, State of Connecti­ 2003, Water resources data, Massachusetts and Rhode cut, State of New York, Interstate Sanitation Commission, Island, water year 2003: U.S. Geological Survey Water- National Oceanic and Atmospheric Administration, 1985, Data Report MA-RI-03-1, 368 p. Long Island Sound estuarine study, water quality manage­ ment strategy: 10 p. Socolow, R.S., Comeau, L.Y., Murino, Domenic, Jr., 2004, Water resources data, Massachusetts and Rhode Island, U.S. Geological Survey, 2000, National land cover data water year 2004: U.S. Geological Survey Water-Data (NLCD) circa 1992, completed nationwide September 2000, Report MA-RI-04-1, 310 p. accessed November 2004, at http://seamless.usgs.gov/.

Socolow, R.S., Leighton, C.R., Whitley, J.F., and Ramsbey, Vermont Water Resources Board, 2000, Vermont Water L.R., 2005, Water resources data, Massachusetts and Rhode Quality Standards: Montpelier, VT, 61 p. Island, water year 2005: U.S. Geological Survey Water- Data Report MA-RI-05-1, 348 p. Wilde, F.D., and Radtke, D.B., 1998a, Collection of water samples: U.S. Geological Survey Techniques of Water- Spahr, N.E., and Boulger, R.W., 1997, Interim results of quality-control sampling of surface-water for the Upper Resources Investigations, Handbooks for Water-Resources Colorado River National Water-Quality Assessment Study Investigations, National field manual for the collection of Unit, water years 1995–96: U.S. Geological Survey Water- water-quality data, book 9, chap. A4, variously paged. Resources Investigations Report 97–4227, 34 p. Wilde, F.D., and Radtke, D.B., 1998b, Field measurements: Trench, E.C.T., 1999, Nutrient sources and loads in the U.S. Geological Survey Techniques of Water-Resources Connecticut, Housatonic, and Thames River Basins: Investigations, Handbooks for Water-Resources Investiga­ U.S. Geological Survey Water-Resources Investigations tions, National field manual for the collection of water- Report 99–4236, 66 p. quality data, book 9, chap. A6, variously paged.

Appendix A: Supplemental Data 82 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005 9,380 10,200 14,100 13,000 65,800 33,600 (lb/yr) Upper 124,000 125,000 212,000 342,000 325,000 218,000 277,000 333,000 1,410,000 interval in 3,910 6,310 5,690 3,800 30,900 66,500 57,600 11,600 32,100 96,600 (lb/yr) 137,000 217,000 202,000 153,000 247,000 Lower 95-percent confidence Basin River )/yr 2 86 65 98 otal phosphorus 128 117 105 211 199 104 201 672 246 334 943 T ield 3,100 Y (lb/mi Connecticut 6,500 9,640 8,790 6,130 Load 45,900 92,200 86,600 20,500 94,500 (lb/yr) 172,000 274,000 258,000 170,000 231,000 651,000 upper the in sites 61,900 99,100 85,900 (lb/yr) 122,000 162,000 130,000 733,000 880,000 Upper 1,090,000 3,610,000 5,680,000 4,780,000 1,250,000 1,480,000 1,350,000 sampling interval river 91,300 96,800 51,600 83,100 69,100 (lb/yr) Lower 129,000 625,000 881,000 740,000 95-percent confidence for 3,290,000 5,100,000 4,340,000 1,080,000 1,310,000 1,140,000 , pounds per square mile year] otal nitrogen )/yr 2 phorus )/yr T 2 ield Y 1,410 1,920 1,500 1,550 2,250 1,850 1,300 2,040 1,720 1,860 2,980 2,530 1,680 2,020 1,800 phos (lb/mi total and 56,600 90,800 77,200 Load (lb/yr) 106,000 145,000 113,000 677,000 981,000 808,000 3,400,000 5,400,000 4,600,000 1,160,000 1,390,000 1,240,000 , pounds per year; (lb/mi nitrogen - y; lb/yr e 87 30 46 38 total 126 109 453 672 538 837 904 feet) Mean sands 2,769 4,326 3,594 1,135 for of acre- (in thou streamflow intervals er at er at er at er at er at er at v v v v v v , VT , VT , VT er er er v v v er at er at er at er at er at er at , VT , VT , VT v v v v v v confidence Station name ells Ri ells Ri ells Ri est Hartford, VT est Hartford, VT est Hartford, VT ictory ictory ictory assumpsic, VT assumpsic, VT assumpsic, VT V V V P P P W W W Randolph, VT Randolph, VT Randolph, VT W W W and yers Brook at yers Brook at yers Brook at assumpsic Ri assumpsic Ri assumpsic Ri Moose Ri Moose Ri Moose Ri P P P Connecticut Ri Connecticut Ri Connecticut Ri A A A White Ri White Ri White Ri Massachusetts. yields, and ater year 2003 2004 2005 2003 2004 2005 2003 2004 2005 2003 2004 2005 2003 2004 2005 W loads, ermont, V USGS station number Annual e been independently rounded; USGS, U.S. Geological Surv 01134500 01134500 01134500 01135500 01135500 01135500 01138500 01138500 01138500 01142500 01142500 01142500 01144000 01144000 01144000 v Hampshire, 1 1 1 2 2 2 3 3 3 4 4 4 5 5 5 Site fig. 1) number (table 1; able A-1. T New [Numbers ha Appendix A 83 4,850 17,400 17,700 19,700 16,800 51,100 53,700 13,200 16,700 31,000 34,700 34,900 (lb/yr) Upper 539,000 921,000 881,000 interval in 8,690 9,380 1,430 1,730 1,600 11,700 11,700 12,800 11,100 18,300 22,800 19,000 (lb/yr) Lower 349,000 604,000 535,000 95-percent confidence Basin River )/yr 2 otal phosphorus 90 97 79 422 425 469 138 289 278 115 227 257 T 105 137 126 ield Y (lb/mi Connecticut 2,770 5,480 6,190 14,400 14,500 16,000 12,300 25,700 24,800 Load 24,100 28,300 26,100 (lb/yr) 437,000 751,000 692,000 upper the in sites 64,000 53,700 61,800 261,000 263,000 285,000 221,000 258,000 207,000 (lb/yr) 384,000 435,000 419,000 Upper 8,340,000 12,600,000 10,600,000 sampling interval river 52,100 41,700 50,300 224,000 224,000 241,000 161,000 187,000 142,000 (lb/yr) Lower 307,000 362,000 335,000 95-percent confidence for 7,200,000 8,960,000 10,200,000 , pounds per square mile year] otal nitrogen )/yr 2 phorus )/yr T 2 ield 7,090 7,130 7,700 2,120 2,480 1,940 2,320 2,400 1,970 Y 1,280 1,480 1,400 1,410 2,070 1,780 phos (lb/mi total and 55,800 57,800 47,400 Load 242,000 243,000 262,000 189,000 221,000 173,000 (lb/yr) 344,000 397,000 376,000 7,800,000 9,760,000 11,400,000 , pounds per year; (lb/mi nitrogen - y; lb/yr e 51 50 59 43 53 48 total 145 190 167 285 380 356 feet) Mean sands 5,756 8,475 6,990 for of acre- (in thou streamflow intervals er at er at er at v v v , MA , MA , MA er er er alpole, NH alpole, NH alpole, NH , MA , MA , MA v v v er near er near er near er at er at er at er at er at er at er at er at er at v v v v v v ay ay ay v v v v v v confidence Station name est Claremont, NH est Claremont, NH est Claremont, NH ar Ri ar Ri ar Ri Otter Ri Otter Ri Otter Ri Shattuckville, MA Shattuckville, MA Shattuckville, MA Conw Conw Conw W W W North W North W North W and Otter Ri Otter Ri Otter Ri North Ri North Ri North Ri South Ri South Ri South Ri Sug Sug Sug Connecticut Ri Connecticut Ri Connecticut Ri Massachusetts.—Continued yields, and ater 2003 2004 2005 2003 2004 2005 2003 2004 2005 year 2003 2004 2005 2003 2004 2005 W loads, ermont, V USGS station number Annual e been independently rounded; USGS, U.S. Geological Surv 01163200 01163200 01163200 01169000 01169000 01169000 01169900 01169900 01169900 01152500 01152500 01152500 01154500 01154500 01154500 v Hampshire, 8 8 8 9 9 9 6 6 6 7 7 7 10 10 10 Site fig. 1) number (table 1; able A-1. T New [Numbers ha 84 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005 2,940 7,140 2,370 3,360 4,200 4,710 (lb/yr) Upper 1,840,000 2,430,000 2,350,000 interval in 876 1,280 1,750 2,370 2,860 2,850 (lb/yr) Lower 95-percent confidence 1,280,000 1,580,000 1,450,000 Basin River )/yr 2 48 91 36 52 64 68 otal phosphorus 159 205 192 T ield Y (lb/mi Connecticut 1,980 3,770 1,500 2,810 3,480 3,650 Load (lb/yr) upper 1,540,000 1,980,000 1,860,000 the in sites 67,000 80,900 37,500 95,300 108,000 105,000 (lb/yr) Upper 21,200,000 24,700,000 23,000,000 sampling interval river 43,300 47,800 24,300 93,400 92,200 80,700 (lb/yr) Lower 95-percent confidence for 18,800,000 21,600,000 20,300,000 , pounds per square mile year] otal nitrogen )/yr 2 phorus )/yr T 2 732 ield 1,310 1,520 1,860 1,820 1,620 2,070 2,390 2,240 Y phos (lb/mi total and 54,200 62,800 30,300 98,500 87,600 Load 100,000 (lb/yr) 20,000,000 23,100,000 21,600,000 , pounds per year; (lb/mi nitrogen - y; lb/yr e 70 89 65 82 92 91 total feet) Mean sands for of acre- (in thou 11,653 14,398 12,805 streamflow intervals er at er at er at v v v ville, CT ville, CT ville, CT er near er near er near v v v er at er at er at v v v confidence Station name Colrain, MA Colrain, MA Colrain, MA Northhampton, MA Northhampton, MA Northhampton, MA Thompson Thompson Thompson and Green Ri Green Ri Green Ri Mill Ri Mill Ri Mill Ri Connecticut Ri Connecticut Ri Connecticut Ri Massachusetts.—Continued yields, and ater 2003 2004 2005 2003 2004 2005 2003 2004 2005 year W loads, ermont, V USGS station number Annual e been independently rounded; USGS, U.S. Geological Surv 01170100 01170100 01170100 01171500 01171500 01171500 01184000 01184000 01184000 v Hampshire, 11 11 11 12 12 12 13 13 13 Site fig. 1) number (table 1; able A-1. T New [Numbers ha Appendix A 85

Table A-2. Estimates of seasonal total nitrogen load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts.

[Numbers have been independently rounded; USGS, U.S. Geological Survey; lb/season, pounds per season; (lb/mi2)/season, pounds per square mile per sea­ son; Winter, December–February; Spring, March–May; Summer, June–August; Fall, September–November]

Site Total Total USGS number nitrogen Percent nitrogen Lower 95th percent Upper 95th percent Standard error station (table 1, Season load of load yield confidence interval confidence interval of prediction number fig. 1) (lb/season) (lb/mi2)/season Moose River at Victory, VT 1 01134500 Winter 74,300 15 988 67,300 81,800 3,700 1 01134500 Spring 228,000 47 3,040 201,000 258,000 14,700 1 01134500 Summer 70,100 15 932 62,800 77,900 3,850 1 01134500 Fall 110,000 23 1,460 97,200 124,000 6,880 Passumpsic River at Passumpsic, VT 2 01135500 Winter 758,000 23 1,740 669,000 857,000 47,800 2 01135500 Spring 1,580,000 48 3,620 1,420,000 1,740,000 82,500 2 01135500 Summer 385,000 12 882 342,000 431,000 22,600 2 01135500 Fall 561,000 17 1,290 468,000 668,000 51,100 Connecticut River at Wells River, VT 3 01138500 Winter 4,130,000 23 1,560 3,840,000 4,430,000 150,000 3 01138500 Spring 7,950,000 45 3,000 7,460,000 8,440,000 249,000 3 01138500 Summer 2,550,000 14 959 2,370,000 2,710,000 8,100 3 01138500 Fall 3,230,000 18 1,220 2,950,000 3,500,000 141,000 Ayers Brook at Randolph, VT 4 01142500 Winter 68,600 23 2,250 62,800 74,800 3,060 4 01142500 Spring 151,000 51 4,970 138,000 166,000 7,060 4 01142500 Summer 24,100 8 789 22,100 26,100 1,030 4 01142500 Fall 54,700 18 1,790 47,900 62,300 3,680 White River at West Hartford, VT 5 01144000 Winter 1,100,000 22 1,600 1,020,000 1,190,000 43,400 5 01144000 Spring 2,760,000 55 3,990 2,560,000 2,970,000 104,000 5 01144000 Summer 435,000 9 631 401,000 472,000 18,300 5 01144000 Fall 756,000 15 1,100 682,000 835,000 39,100 Sugar River at West Claremont, NH 6 01152500 Winter 356,000 24 1,320 316,000 399,000 21,200 6 01152500 Spring 727,000 49 2,700 654,000 805,000 38,300 6 01152500 Summer 179,000 12 665 159,000 200,000 10,500 6 01152500 Fall 224,000 15 834 193,000 260,000 17,200 Connecticut River at North Walpole, NH 7 01154500 Winter 8,960,000 24 1,630 7,990,000 10,000,000 533,000 7 01154500 Spring 18,500,000 47 3,370 16,700,000 20,400,000 946,000 7 01154500 Summer 4,520,000 12 820 4,100,000 4,900,000 232,000 7 01154500 Fall 6,540,000 18 1,190 5,550,000 7,650,000 536,000 86 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Table A-2. Estimates of seasonal total nitrogen load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts.—Continued

[Numbers have been independently rounded; USGS, U.S. Geological Survey; lb/season, pounds per season; (lb/mi2)/season, pounds per square mile per season; Winter, December–February; Spring, March–May; Summer, June–August; Fall, September–November]

Site Total Total USGS number nitrogen Percent nitrogen Lower 95th percent Upper 95th percent Standard error station (table 1, Season load of load yield confidence interval confidence interval of prediction number fig. 1) (lb/season) (lb/mi2)/season Otter River at Otter River, MA 8 01163200 Winter 257,000 26 7,530 226,000 290,000 16,300 8 01163200 Spring 293,000 29 8,600 263,000 326,000 16,200 8 01163200 Summer 212,000 22 6,210 189,000 236,000 12,100 8 01163200 Fall 234,000 23 6,860 202,000 269,000 17,100 North River at Shattuckville, MA 9 01169000 Winter 169,000 22 1,900 151,000 190,000 9,890 9 01169000 Spring 331,000 43 3,720 267,000 404,000 35,000 9 01169000 Summer 93,000 12 1,050 83,700 103,000 4,970 9 01169000 Fall 182,000 23 2,050 159,000 209,000 12,800 South River near Conway, MA 10 01169900 Winter 66,700 31 2,770 59,800 74,100 3,620 10 01169900 Spring 84,200 39 3,490 74,500 94,700 5,160 10 01169900 Summer 20,600 10 856 18,500 22,900 1,120 10 01169900 Fall 43,200 20 1,790 37,200 49,900 3,230 Green River near Colrain, MA 11 01170100 Winter 39,000 20 943 33,600 45,100 2,930 11 01170100 Spring 92,200 47 2,230 68,300 122,000 13,700 11 01170100 Summer 17,000 9 410 14,900 19,200 1,090 11 01170100 Fall 47,800 24 1,150 38,800 58,100 4,940 Mill River at Northhampton, MA 12 01171500 Winter 120,000 31 2,220 111,000 130,000 4,710 12 01171500 Spring 141,000 37 2,610 130,000 152,000 5,520 12 01171500 Summer 43,700 11 810 40,600 47,000 1,630 12 01171500 Fall 78,300 20 1,450 70,500 86,800 4,170 Connecticut River at Thompsonville, CT 13 01184000 Winter 22,900,000 27 2,370 20,900,000 25,000,000 1,040,000 13 01184000 Spring 35,100,000 41 3,640 32,200,000 38,300,000 1,580,000 13 01184000 Summer 12,600,000 15 1,300 11,700,000 13,500,000 448,000 13 01184000 Fall 15,600,000 18 1,610 14,200,000 17,000,000 726,000 Appendix A 87

Table A-3. Estimates of seasonal total phosphorus load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts.

[Numbers have been independently rounded; USGS, U.S. Geological Survey; lb/season, pounds per season; (lb/mi2)/season, pounds per square mile per sea­ son; Winter, December–February; Spring, March–May; Summer, June–August; Fall, September–November]

Site Total Total USGS number phosphorus Percent phosphorus Lower 95th percent Upper 95th percent Standard error station (table 1, Season load of load yield confidence interval confidence interval of prediction number fig. 1) (lb/season) (lb/mi2)/season Moose River at Victory, VT 1 01134500 Winter 3,300 10 44 2,510 4,260 445 1 01134500 Spring 17,700 53 235 11,100 26,900 4,070 1 01134500 Summer 3,730 11 50 2,710 5,010 588 1 01134500 Fall 8,390 25 112 4,740 13,800 2,330 Passumpsic River at Passumpsic, VT 2 01135500 Winter 46,500 16 107 34,300 61,600 7,000 2 01135500 Spring 161,000 54 369 110,000 227,000 30,000 2 01135500 Summer 27,900 9 64 22,400 34,300 3,060 2 01135500 Fall 63,500 21 146 41,900 92,200 12,900 Connecticut River at Wells River, VT 3 01138500 Winter 168,000 18 64 126,000 220,000 23,900 3 01138500 Spring 534,000 57 202 408,000 688,000 71,500 3 01138500 Summer 102,000 11 39 76,600 134,000 14,600 3 01138500 Fall 130,000 14 49 88,100 186,000 25,100 Ayers Brook at Randolph, VT 4 01142500 Winter 6,630 4 217 4,050 10,200 1,590 4 01142500 Spring 135,000 84 4,410 48,900 299,000 65,700 4 01142500 Summer 770 1 25 610 960 91 4 01142500 Fall 18,600 11 610 8,960 34,400 6,580 White River at West Hartford, VT 5 01144000 Winter 114,000 8 166 60,300 198,000 35,500 5 01144000 Spring 1,120,000 80 1,620 499,000 2,170,000 433,000 5 01144000 Summer 22,200 2 32 16,800 28,700 3,040 5 01144000 Fall 141,000 10 204 86,200 218,000 33,900 Sugar River at West Claremont, NH 6 01152500 Winter 19,300 18 72 15,900 23,200 1,860 6 01152500 Spring 58,800 56 219 44,100 76,900 8,410 6 01152500 Summer 9,730 9 36 8,350 11,300 741 6 01152500 Fall 16,700 16 62 13,400 20,600 1,850 Connecticut River at North Walpole, NH 7 01154500 Winter 414,000 17 75 342,000 500,000 40,200 7 01154500 Spring 1,470,000 59 267 1,160,000 1,860,000 181,000 7 01154500 Summer 181,000 7 33 157,000 208,000 13,100 7 01154500 Fall 435,000 17 79 343,000 548,000 52,200 88 Assessment of Total Nitrogen in the Upper Connecticut River Basin in NH, VT, and MA, December 2002-September 2005

Table A-3. Estimates of seasonal total phosphorus load for river sampling sites in the upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts.—Continued

[Numbers have been independently rounded; USGS, U.S. Geological Survey; lb/season, pounds per season; (lb/mi2)/season, pounds per square mile per season; Winter, December–February; Spring, March–May; Summer, June–August; Fall, September–November]

Site Total Total USGS number phosphorus Percent phosphorus Lower 95th percent Upper 95th percent Standard error station (table 1, Season load of load yield confidence interval confidence interval of prediction number fig. 1) (lb/season) (lb/mi2)/season Otter River at Otter River, MA 8 01163200 Winter 21,700 36 636 16,000 28,800 3,280 8 01163200 Spring 18,300 31 536 14,000 23,400 2,410 8 01163200 Summer 7,780 13 228 5,750 10,300 1,160 8 01163200 Fall 12,100 20 356 8,360 17,100 2,230 North River at Shattuckville, MA 9 01169000 Winter 8,340 10 94 5,540 12,100 1,670 9 01169000 Spring 54,100 65 608 20,500 117,000 25,300 9 01169000 Summer 6,770 8 76 4,820 9,260 1,140 9 01169000 Fall 14,100 17 158 7,260 24,800 4,520 South River near Conway, MA 10 01169900 Winter 2,460 13 102 1,330 4,170 734 10 01169900 Spring 12,500 65 518 3,270 33,500 8,100 10 01169900 Summer 832 4 35 533 1,240 183 10 01169900 Fall 3,420 18 142 1,220 7,670 1,690 Green River near Colrain, MA 11 01170100 Winter 1,120 12 27 745 1,640 230 11 01170100 Spring 5,290 55 128 2,380 10,200 2,040 11 01170100 Summer 628 7 15 456 840 99 11 01170100 Fall 2,600 27 63 1,370 4,480 803 Mill River at Northhampton, MA 12 01171500 Winter 3,020 23 56 2,530 3,570 266 12 01171500 Spring 5,500 41 102 4,060 7,300 829 12 01171500 Summer 1,710 13 32 1,470 1,970 128 12 01171500 Fall 3,100 23 58 2,420 3,920 383 Connecticut River at Thompsonville, CT 13 01184000 Winter 1,520,000 21 157 1,280,000 1,780,000 126,000 13 01184000 Spring 3,300,000 46 342 2,390,000 4,450,000 526,000 13 01184000 Summer 944,000 13 98 826,000 1,070,000 63,100 13 01184000 Fall 1,390,000 19 144 1,160,000 1,650,000 125,000 Appendix A 89 ­ ) 2 .97 .98 .99 .99 .96 .98 .65 .90 .96 .92 .97 .94 .92 .89 .97 .97 .62 .92 .95 .78 .88 .89 .91 .77 (R 0.96 0.89 nation of determi Coefficient Hampshire, 2 New in Basin River Connecticut dtime) dtime) dtime) dtime) dtime) – 0.07 (dtime) dtime) – 0.06 (dtime) dtime) – 0.06 (dtime) dtime) dtime) – 0.08 (dtime) 0.06 dtime) + 0.08 (dtime) dtime) π π π π π π π π π π π upper the in dtime) dtime) dtime) dtime) – 0.14 (dtime) dtime) – 0.11 (dtime) sites π π π π π dtime) + 0.03 cos (2 dtime) – 0.15 cos (2 , pi; dtime, centered decimal time] dtime) + 0.08 cos (2 dtime) + 0.19 cos (2 dtime) + 0.25 cos (2 dtime) + 0.19 cos (2 dtime) + 0.10 cos (2 dtime) + 0.31 cos (2 dtime) – 0.05 cos (2 dtime) + 0.14 cos (2 dtime) – 0.08 cos (2 π π π π π π π π π π π π sampling river otal nitrogen regression equations T for otal phosphorus regression equations T dtime) – 0.14 cos (2 dtime) + 0.21 cos (2 + 0.26 sin (2 + 0.40 sin (2 dtime) + 0.13 cos (2 dtime) + 0.09 cos (2 dtime) + 0.02 cos (2 2 2 – 0.27 sin (2 – 0.31 sin (2 – 0.17 sin (2 – 0.17 sin (2 – 0.13 (dtime) – 0.22 (dtime) + 0.16 (dtime) – 0.13 sin (2 + 0.13 (dtime) – 0.14 (dtime) + 0.30 sin (2 – 0.37 sin (2 – 0.27 sin (2 – 0.34 sin (2 π π π π π 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 load phosphorus total and , in cubic feet per second; sin, sine; cos, cosine; load w 7.13 + 1.11 ln(Q) 0.15 4.64 + 1.12 ln(Q) – 0.09 (dtime) 6.69 + 1.02 ln(Q) 0.09 8.60 + 1.07 ln(Q) – 0.12 sin (2 4.24 + 1.10 ln(Q) 0.19 5.99 + 0.88 ln(Q) 0.12 8.99 + 1.04 ln(Q) 0.05 5.64 + 0.40 ln(Q) 0.07 sin (2 5.17 + 0.83 ln(Q) 0.09 3.62 + 0.84 ln(Q) – 0.20 sin (2 3.40 + 1.17 ln(Q) 0.09 4.37 + 0.75 ln(Q) – 0.13 sin (2 10.1 + 0.77 ln(Q) 0.03 sin (2 1.06 + 1.26 ln(Q) 0.27 3.39 + 1.49 ln(Q) 0.46 5.17 + 1.42 ln(Q) 0.19 0.44 + 2.33 ln(Q) 0.59 3.63 + 2.27 ln(Q) 0.48 2.81 + 1.24 ln(Q) 0.32 2.00 + 1.02 ln(Q) 0.29 5.34 + 1.56 ln(Q) 0.38 2.70 + 0.56 ln(Q) – 0.14 -0.68 + 1.44 ln(Q) 0.22 -0.40 + 1.61 ln(Q) 0.20 0.66 + 1.06 ln(Q) 0.07 7.18 + 0.93 ln(Q) 0.44 nitrogen , VT , VT ville, CT ville, CT er er total alpole, NH alpole, NH v v of , MA , MA , MA , MA er er ay ay ells Ri ells Ri , VT , VT v v assumpsic, VT assumpsic, VT arithm; Q, centered streamflo ictory ictory est Hartford, VT est Hartford, VT est Claremont, NH est Claremont, NH estimates er at W er at North W er at Thompson er at W er at North W er at Thompson er at P er at P v v v v v v v v Station name for er at V er at V er at W er at W er near Colrain, MA er near Colrain, MA er near Conw er near Conw er at Shattuckville, MA er at Shattuckville, MA er at W er at W v v er at Otter Ri er at Otter Ri v v v v v v v v v v er at Northhampton, MA er at Northhampton, MA v v v v ar Ri ar Ri y; ln, natural log e equations yers Brook at Randolph, VT yers Brook at Randolph, VT assumpsic Ri assumpsic Ri White Ri Moose Ri P Connecticut Ri A Sug Connecticut Ri Otter Ri North Ri South Ri Green Ri Mill Ri Connecticut Ri Moose Ri P Connecticut Ri A White Ri Sug North Ri Connecticut Ri Otter Ri South Ri Green Ri Mill Ri Connecticut Ri Massachusetts. USGS Regression station number 01144000 01134500 01135500 01138500 01142500 01152500 01154500 01163200 01169000 01169900 01170100 01171500 01184000 01134500 01135500 01138500 01142500 01144000 01152500 01169000 01154500 01163200 01169900 01170100 01171500 01184000 and 5 1 2 3 4 6 7 8 9 1 2 3 4 5 6 9 7 8 10 11 12 13 10 11 12 13 Site fig. 1) ermont, number able A-4. (table 1, T V [USGS, U.S. Geological Surv Prepared by Publications Service Center 1

For more information concerning the research in this report, contact: Keith W. Robinson, Director U.S. Geological Survey New Hampshire-Vermont Water Science Center 361 Commerce Way Pembroke, NH 03275 or visit our Web site at: http://nh.water.usgs.gov Deacon, J.R., and others—Assessment of Total Nitrogen in the Upper Connecticut River Basin in New Hampshire, Vermont, and Massachusetts, December 2002–September 2005—USGS Scientific Investigations Report 2006–5144