K.E. Schilling and C.A. Thompson
Iowa Department of Natural Resources, Geological Survey Bureau
Technical Information Series 39, 1999, 169 p.
The Walnut Creek Monitoring Project was established in April 1995 as a nonpoint-source (NPS) monitoring program in relation to the watershed habitat restoration and agricultural management changes implemented at the Neil E. Smith National Wildlife Refuge in Jasper County, Iowa. The project includes a treatment watershed, Walnut Creek (12,895 acres), and a control watershed, Squaw Creek (11,622 acres), as well as an upstream/downstream assessment on Walnut Creek. Five basic components comprise the monitoring project: 1) tracking land cover and land management changes within the basins, 2) stream gaging for discharge and suspended sediment at two locations on Walnut Creek and one on Squaw Creek, 3) biomonitoring of aquatic macroinvertebrates and fish in Walnut and Squaw creeks, 4) surface water quality monitoring of Walnut and Squaw creeks, and 5) groundwater quality and hydrologic monitoring. This report summarizes data collected during the first three years of the project and includes water years 1995, 1996 and 1997.
The U.S. Fish and Wildlife Service owns 33.7% of the Walnut Creek watershed (4,343 acres). From 1992 to 1997, land use changes were implemented on 19.4% of the watershed. These changes included restoration of 1,729 acres from row crop to native prairie (13.4%) and conversion of 773 acres to a cash-rent basis (6.0%). During this time, applied nitrogen loads were reduced by 18.1% in the watershed. Pesticide loads, including atrazine and cyanazine, were reduced by 28%.
Measurement of discharge and suspended sediment of Walnut Creek and Squaw Creek began in 1995 and includes two full years and one partial year of data. In both watersheds, most suspended sediment discharge occurred during winter snowmelt periods (February and March) and during occasional storm events in late spring/early summer (May and June). Discharge and sediment movement through the Walnut and Squaw Creek watersheds was very flashy most of the suspended sediment was discharged during intermittent high flow events. During a single day in 1996, 28% of the total annual sediment load was measured at the downstream Walnut Creek gage and 46% of the total annual sediment load was measured at Squaw Creek. In 1997, the percentages were slightly lower (18% and 33%, respectively). Sediment loads in Walnut and Squaw creeks were similar in 1996 (approximately 14,500 tons each), but nearly double in Walnut Creek during 1997 (9,400 tons compared to 5,000 tons).
Biomonitoring data indicate that both stream communities are impaired with at least some level of organic pollution. Walnut and Squaw creeks are characterized by macroinvertebrate communities dominated by relatively few taxa with occasional new taxa appearing at low frequencies and abundances. Seasonal trends in macroinvertebrate populations were similar and both streams were comparable in community structure and population. Walnut and Squaw creeks rate in the lower quartile of similar streams in the same ecoregion with respect to metric indicators of macroinvertebrate community health (EPT taxa, total number of taxa, percent dominant taxa). Fish communities in both watersheds were dominated by tolerant species, including bigmouth shiners, creek chubs, and bluntnose minnows; however, less tolerant species were sporadically found due to the proximity of the streams to larger bodies of water. The Index of Biotic Integrity (IBI) scores for both Walnut and Squaw creeks were less than the mean IBI score for the ecoregion, implying impaired fish communities.
In surface water samples, atrazine was the most frequently detected pesticide in Walnut and Squaw creeks (as is true across Iowa), with frequencies of atrazine detections ranging from 71% to 88% in the main stems. No significant differences were noted in atrazine loads between Walnut and Squaw Creek during the years 1995 to 1997, although atrazine loads from the lower portion of the Walnut Creek watershed (including the prairie restoration area) may be decreasing in relation to the upstream untreated component of the watershed. The frequencies of atrazine and cyanazine detections appear to be related to the percentage of row crop in the Walnut or Squaw Creek basins and subbasins. Nitrate-N concentrations measured in both Walnut and Squaw Creek watersheds were similar, with mean nitrate-N concentrations ranging from 7.8 to 8.3 mg/L at the downstream Walnut Creek station (WNT2) and from 8.1 to 8.5 mg/L at the downstream Squaw Creek stations (SQW2). Both basins showed a similar pattern of detection and an overall reduction in nitrate-N concentrations from upstream to downstream monitoring sites. Highest nitrate-N concentrations were measured in the headwaters of both watersheds, areas containing a high percentage of row crop (76-83%). Median annual fecal coliform counts varied widely among sampling sites and water years, ranging from 80 counts/100ml at WNT6 to 8,600 counts/100ml at SQW2. Highest levels of fecal coliform bacteria typically occurred in spring and early summer months during high stream flow periods associated with rainfall runoff. Turbidity values fluctuated widely with higher median values and greater variability noted at downstream sites compared to upstream sites. The inorganic water chemistry was similar for both creeks, although concentrations of total inorganics and specific conductance were slightly lower in Walnut Creek.
Seven monitoring wells were installed in 1995 at four sites along a transect in the central portion of the refuge, from an upland recharge position to the valley bottom. Stratigraphy consists of Wisconsinan loess, colluvium and alluvial deposits overlying 50 to 100 feet of pre-Illinoian till. Continuous water level measurements showed water levels ranging from 9.1 feet below ground surface in the uplands to 0.7 feet above the land surface in the alluvium. Groundwater flows west at the transect from higher landscape positions towards Walnut Creek. Baseflow percentages calculated for 1996 indicated that groundwater and tile discharge comprised a fraction (about 40%) of the total stream discharge in both Walnut and Squaw creeks. Groundwater sampling conducted on four occasions showed one pesticide metabolite, deethylatrazine, consistently detected at WC5B at concentrations ranging from 0.15 to 0.28 µg/L. Concentrations of nitrate-N in the monitoring wells ranged from less than the detection limit (<0.5 mg/L) in several wells to 32 mg/L at WC5A in September 1997. Nitrate-N concentrations increased from May 1997 to September 1997. Low nitrate-N and sulfate concentrations at well WC6B may result from nitrate and sulfate reduction occurring in an organic-rich peat screened by this well.