Assessment of Water-Quality Data from Long Lake National Wildlife Refuge, North Dakota—2008 Through 2012

Abstract

Long Lake National Wildlife Refuge, located in south-central North Dakota, is an important habitat for numerous migratory birds and waterfowl, including several threatened or endangered species. The refuge is distinguished by Long Lake, which is approximately 65 square kilometers and consists of four primary water management units. Water levels in the Long Lake units are maintained by low-level dikes and water-control structures, which after construction during the 1930s increased the water-storage capacity of Long Lake and reduced the frequency and volume of flushing flows downstream. The altered water regime, along with the negative precipitation:evaporation ratio of the region, may be contributing to the accumulation of water-borne chemical constituents such as salts, trace metals, and other constituents, which at certain threshold concentrations may impair aquatic plant, invertebrate, and bird communities of the refuge. The refuge’s comprehensive conservation planning process identified the need for water-quality monitoring to assess current (2013) conditions, establish comparative baselines, evaluate changes over time (trends), and support adaptive management of the wetland units. In 2008, the U.S. Geological Survey, U.S. Fish and Wildlife Service, and North Dakota Department of Health began a water-quality monitoring program at Long Lake National Wildlife Refuge to address these needs. Biweekly water-quality samples were collected for ions, trace metals, and nutrients; and in situ sensors and data loggers were installed for the continuous measurement of specific conductance and water depth.

Long Lake was characterized primarily by sodium, bicarbonate, and sulfate ions. Overall results for total alkalinity and hardness were 580 and 329 milligrams per liter, respectively; thus, Long Lake is considered alkaline and classified as very hard. The mean pH and sodium adsorption ratio for Long Lake were 8.8 and 10, respectively. Total dissolved solids concentrations averaged approximately 1,750 milligrams per liter, and ranged from 117 to 39,700 milligrams per liter. Twelve of the 14 trace metals detected in the water samples had established North Dakota water-quality standards for aquatic life, and only aluminum and copper consistently exceeded these criteria. Aluminum is considered harmful to aquatic biota in acidic (pH less than 5.5) systems and most of the copper standard exceedances were collected from highly concentrated waters because of evaporation and seasonally low water levels. Concentrations for various forms of nitrogen and phosphorus generally were similar to reported regional values.

Specific conductance of Long Lake varied seasonally and annually both within and among management units, with values ranging from less than 500 to nearly 40,000 microsiemens per centimeter at 25 degrees Celsius. Long Lake was characterized by consistent seasonal patterns of increasing specific conductance from spring (March and April) to fall (September and October), with levels stabilizing through the end of the sampling season (November). These seasonal patterns in specific conductance were associated with decreasing water levels throughout the summer due primarily to evaporation and continuous water releases through the Unit 1 outlet structure, which resulted in the concentration of salts. Specific conductance of each unit, along with water levels, also varied among years. Overall, specific conductance levels were greatest during the drier year of 2008 when water levels were low. Specific conductance levels were lowest during the spring of 2009 following above-average volumes of fresh water from snowmelt runoff. Comparisons of specific conductance among sample sites that were spatially distributed within each management unit suggested that spatial variability within units was low except for areas associated with local inflows.

Data collected during this study revealed consistent seasonal patterns and low within-unit spatial variability of specific conductance. Based on these data results, future sample collection efforts may be reduced, as well as the number of sample locations, to limit sampling costs. Water-quality samples collected monthly or seasonally during the growing season (spring, summer, and fall) from a single representative location within each water-management unit should provide sufficient data to assess seasonal changes in water-quality over time and provide information for Long Lake management decisions.

First posted December 16, 2013