Abstract

This report presents the results of a study by the U.S. Geological Survey, in cooperation with the Missouri Department of Natural Resources, to analyze and compare hydrologic and water-quality characteristics of tallgrass prairie and agricultural basins located within the historical distribution of tallgrass prairie in Missouri and Kansas. Streamflow and water-quality data from two remnant, tallgrass prairie basins (East Drywood Creek at Prairie State Park, Missouri, and Kings Creek near Manhattan, Kansas) were compared to similar data from agricultural basins in Missouri and Kansas.

Prairie streams, especially Kings Creek in eastern Kansas, received a higher percentage of base flow and a lower percentage of direct runoff than similar-sized agricultural streams in the region. A larger contribution of direct runoff from the agricultural streams made them much flashier than prairie streams. During 22 years of record, the Kings Creek base-flow component averaged 66 percent of total flow, but base flow was only 16 to 26 percent of flows at agricultural sites of various record periods. The large base-flow component likely is the result of greater infiltration of precipitation in prairie soils and the resulting greater contribution of groundwater to streamflow. The 1- and 3-day annual maximum flows were significantly greater at three agricultural sites than at Kings Creek. The effects of flashier agricultural streams on native aquatic biota are unknown, but may be an important factor in the sustainability of some native aquatic species.

There were no significant differences in the distribution of dissolved-oxygen concentrations at prairie and agricultural sites, and some samples from most sites fell below the 5 milligrams per liter Missouri and Kansas standard for the protection of aquatic life. More than 10 percent of samples from the East Drywood Creek prairie stream were less than this standard. These data indicate low dissolved-oxygen concentrations during summer low-flow periods may be a natural phenomenon for small prairie streams in the Osage Plains.

Nutrient concentrations including total nitrogen, ammonia, nitrate, and total phosphorus were significantly less in base-flow and runoff samples from prairie streams than from agricultural streams. The total nitrogen concentration at all sites other than one of two prairie sampling sites were, on occasion, above the U.S. Environmental Protection Agency recommended criterion for total nitrogen for the prevention of nutrient enrichment, and typically were above this recommended criterion in runoff samples at all sites. Nitrate and total phosphorus concentrations in samples from the prairie streams generally were below the U.S. Environmental Protection Agency recommended nutrient criteria in base-flow and runoff samples, whereas samples from agricultural sites generally were below the criteria in base-flow samples and generally above in runoff samples. The lower concentrations of nutrient species in prairie streams is likely because prairies are not fertilized like agricultural basins and prairie basins are able to retain nutrients better than agricultural basins. This retention is enhanced by increased infiltration of precipitation into the prairie soils, decreased surface runoff, and likely less erosion than in agricultural basins.

Streamflow in the small native prairie streams had more days of zero flow and lower streamflow yields than similar-sized agricultural streams. The prairie streams were at zero flow about 50 percent of the time, and the agricultural streams were at zero flow 25 to 35 percent of the time. Characteristics of the prairie basins that could account for the greater periods of zero flow and lower yields when compared to agricultural streams include greater infiltration, greater interception and evapotranspiration, shallower soils, and possible greater seepage losses in the prairie basins. Another difference between the prairie and agricultural streams was the duration of extreme low flows (flows within the lowest 10th percentile). The long-term extreme low-flow duration was a median of 112 days for the prairie stream, Kings Creek, but was not greater than a median of 16 days at any of the agricultural sites. This extended duration of low flows indicates a prolonged period of groundwater contributions to streamflow from the prairie stream compared to the agricultural streams.

First posted October 28, 2009