To fulfill the goals of the NAWQA program, the USGS plans to examine 60 areas (study units) across the United States during full implementation of the program. In 1991, the NAWQA program went into full implementation with the intensive investigation of 20 of these study units; one of these study units is the Western Lake Michigan Drainages (WMIC) study unit.
The Western Lake Michigan Drainages Study Unit
The WMIC study unit drains 51,500 km2 (square kilometers) of Wisconsin and part of the Upper Peninsula of Michigan (Upper Michigan) emptying into Lake Michigan and Green Bay. The study unit is bounded on the south by the Illinois State line and extends north through Wisconsin and Upper Michigan to approximately 50 km (kilometers) north of Escanaba, Mich. Drainage basins of nine major rivers and numerous smaller tributaries are contained in the WMIC study unit. The northern half of this area is dominated by forests, and the southern half is dominated by agriculture, primarily dairy-farm operations.
Staff from each of the NAWQA study units, during the early years of investigation, examine, assemble, and summarize available water-quality and ancillary information in retrospective reports. The purposes of this report are to (I) provide information that will assist in designing the sampling components of the intensive phase of the WMIC study unit; (2) propose preliminary conclusions and hypotheses that can be further tested using data collected during the intensive phase; (3) describe the spatial and temporal distribution and the availability of nutrient and suspended-sediment data collected within the study unit (and the gaps in the available data); (4) describe the natural and anthropogenic factors affecting the spatial patterns in concentrations and loads within the study unit; (5) describe long term trends (or lack of trends) in water quality and pro vide data for future examinations for trends; and (6) provide data to the NAWQA national synthesis team, who will assemble data from the individual study units and interpret this information from a national perspective.
Amount and Geographical Distribution of Data
This retrospective report summarizes the nutrient (phosphorus and nitrogen species) and suspended sediment data collected in the surface water and ground water of the WMIC study unit during the 1971-90 water yearsl. Surface-water samples were collected at 712 sites; total phosphorus was the constituent sampled at the most sites (561 sites).
A Water year is the period from October I of a given year through September 30 of the succeeding year and is designated by the calender year in which it ends.Data were collected relatively uniformly throughout the year and range of streamflows; however, most samples were collected in areas integrating the effects of various environmental and anthropogenic factors. Few surface water samples were collected in headwater areas with relatively uniform land use, surficial deposits, and bedrock types, and in the southwest and northwest parts of the study unit. Ground-water samples were collected from 9l9 wells, dissolved nitrite plus nitrate was the constituent sampled at the most wells (789 wells). Ground-water samples were collected primarily in the southern two-thirds of the study unit; however, samples analyzed for total ammonia, total phosphorus, and total orthophosphate were collected mostly in Upper Michigan. Ground-water sites were generally sampled only once during the entire period.
Water-Quality Conditions
Land use was the primary factor affecting the distribution of nutrient and suspended-sediment con centrations and total export in surface water. Total nitrogen and total phosphorus concentrations were directly related to the input of nutrients associated with the specific land use in the drainage basins, especially from fertilizers, manure, and additional fixation (not accounted for in manure inputs); concentrations decreased from agricultural to urban to forested areas. Nutrient concentrations in ground water were also related to land use; however, well depth and texture of surficial deposits also were important. Nutrient concen trations were inversely related to well depth; generally higher in clayey surficial deposits than in sand and gravel deposits, with the exception of dissolved nitrite plus nitrate, for which the opposite was true.
The total export of phosphorus, nitrogen, and suspended sediment was directly related to land use: the agricultural areas in the southern half of the study unit contributed approximately 80 percent of the total load of phosphorus,70 percent of the total load of nitrogen, and 75 to 90 percent of the total load of sus pended sediment. Although point sources of nutrients have been significantly controlled, point sources of phosphorus may still contribute a significant propor tion of the total export, especially during dry years in the Fox and Milwaukee River Basins.
Concentrations of nitrite plus nitrate in agricul tural and urban areas (median concentrations of 0.92 and 0.44 mg/L, respectively) in the surface water of the WMIC study unit were similar to the "national aver age" surface-water concentrations (NAC) (median concentrations of 0.72 and 0.47 mg/L, respectively); however, the concentrations in forested areas (median concentration of 0.09 mg/L) were lower than the NAC (0.21 mg/L). Concentrations exceeding the U.S. Envi ronmental Protection Agency (USEPA) 10-mg/L Maximum Contamination Level (MCL) for dissolved nitrite plus nitrate were found only in agricultural areas. Only two samples from mixed (downstream, integrator) land-use areas exceeded the l-mg/L MCL for dissolved nitrite.
Total phosphorus concentrations were less than the NAC for their respective land uses. Median concentrations were 0.13 mg/L for agricultural areas (0.23 mg/L, NAC), 0.11 mg/L for urban areas (0.20 mg/L, NAC), and 0.02 mg/L for forested areas (0.05 mg/L, NAC). Concentrations exceeding the 0.1-mg/L sug gested limit by the USEPA for total phosphorus were commonly found in all land-use categories, except forested areas, where this limit was sporadically exceeded.
Suspended-sediment concentrations in surface water were highest in urban areas, moderate in agricul tural areas, and lowest in forested areas. These concentrations were quite different from the NAC. Median concentrations were 148 mg/L (25 mg/L, NAC) in urban areas,25 mg/L (131 mg/L, NAC) in agricultural areas, and 4 mg/L (19 mg/L, NAC) in forested areas. The differences in the concentrations from the forested and agricultural areas of the WMIC study unit and those from the NAC may have been a result of upstream dams on many of the major rivers through these areas within the WMIC.
Dissolved nitrate and organic nitrogen were the primary forms of nitrogen in surface water for all land use categories, except urban, where ammonia was also an important fraction. Concentrations of dissolved nitrate and organic nitrogen were highest in agricul tural areas, moderate in urban areas, and lowest in forested areas. Concentrations of total and dissolved ammonia and dissolved nitrite were highest in urban areas, moderate in agricultural areas, and lowest in for ested areas. Dissolved nitrate was the primary form of nitrogen found in ground water in forested and agricul tural/forested areas; however, dissolved ammonia was also important in urban and agricultural areas. Forested areas also had significant fractions as organic nitrogen and particulate ammonia.
Dissolved inorganic (dissolved reactive) phosphorus and particulate phosphorus were the primary forms of phosphorus in surface water for all land-use categories, except agriculture, where dissolved organic phosphorus also was important. Concentrations of dis solved phosphorus, and total orthophosphate and dis solved orthophosphate were highest in agricultural areas, moderate in urban areas, and lowest in forested areas. The few phosphorus samples precluded the partitioning of phosphorus for each land-use category for ground water.
To determine how various environmental and anthropogenic factors affect the distribution of nutrients and suspended sediment, samples were subdivided into areas of similar land use, which were further subdivided into areas of similar texture of surficial deposit and bedrock type. Therefore, samples either represented specific indicator areas referred to as "relatively homogeneous units" (RHU's) or integrator areas representing several RHU's. Samples from specific RHU's indicated no significant difference in concentration caused by texture of surficial deposit and bedrock type for dissolved nitrite plus nitrate, Kjeldahl nitrogen, and total ammonia. Significantly higher dissolved nitrite concentrations were found in surface water surrounded by agriculture on sandy deposits than those surrounded by agriculture on clayey deposits. Significantly lower concentrations of dissolved ammonia were found in areas with clayey surficial deposits than in areas with sandy deposits, possibly caused by clayey deposits preferentially adsorbing ammonia, which is positively charged at most ambient pHs, and therefore, may be less readily transferred to streams than other nitrogen species.
No statistical differences were detected in surface water from RHU's of similar land use for concentrations of dissolved, dissolved inorganic, and total orthophosphate. Total phosphorus and dissolved orthophosphate concentrations were significantly higher in areas with sandy deposits than in areas with clayey deposits, possibly because of differences in the biological communities, which were not examined. Total phosphorus and dissolved orthophosphate concentrations were also higher in areas with carbonate bedrock than areas with sandstone bedrock, possibly because of the differences in the phosphorus content of the bedrock.
Trends in Water Quality
During 1971-90 water years, no significant (p < 0.10) trends were found in dissolved nitrate plus nitrate concentrations. In general, Kjeldahl nitrogen concentrations increased at almost all of the sites examined; however, this upward trend was significant (p < 0.15) only for the Ford, Fox, Manitowoc, and Milwaukee Rivers. The increase in Kjeldahl nitrogen concentrations in association with no change in dissolved nitrite plus nitrate concentrations indicates an increase in total-nitrogen concentrations in rivers draining these areas.
During 1980-90, few changes in total phosphorus concentrations were found across the study unit; however, when the 1971-79 data were included, significant (p < 0.10) downward trends in concentrations were found for the Popple and Milwaukee Rivers. Significant (p < 0.01) downward trends were also found in dissolved phosphorus concentrations for the Fox and Milwaukee Rivers. Similar downward trends were found in dissolved orthophosphate concentrations, but these were not statistically significant. The downward trend in phosphorus concentrations, which were most significant in the southern part of the study unit, may have been caused by the reduction in phosphorus in detergents and by improvements in sewage-treatment facilities.
During 1980-90, suspended-sediment concentrations increased or remained unchanged at all of the sites, except the Ford River; however, the only significant (p < 0.10) increase occurred in the Popple River. When the 1971 -79 data were included, the increases in the southern half of the study unit (agricultural and urban areas) became more significant, but the increase in the Popple River was no longer apparent.
Implications for Future NAWQA Sampling
The results and preliminary conclusions from this study were used to assist in the initial selection of locations for future sampling to be done within the WMIC NAWQA study unit. Eight Basic Fixed Sites (BFS's) representing specific RHU's were chosen for further examination of the observed differences in sur face-water nutrient concentrations caused by differences in land use and surficial-deposit texture. Three BFS's representing an integration of various factors were chosen to represent the diverse conditions throughout the study unit, as well as the major contributors of nutrients and suspended sediment. The location of two land-use surveys were chosen to examine how agricultural land use affects shallow ground-water quality. The location of a flow-path study was chosen to examine changes in nutrients and pesticides along a ground-water flow path in an agricultural setting.