WATER QUALITY MONITORING AT GREAT PLAINS NATIONAL PARKS

Northern Great Plains Network (NGPN), as part of the National Park Service's (NPS) Inventory & Monitoring Program, is measuring status and trends of core water quality parameters in NPS units in the Dakotas, eastern Wyoming and Nebraska. To accomplish this objective we must optimize efficiency, effectiveness, and quality of inference from this monitoring effort while considering budget constraints. The impact of multiple sources of variation regarding statistical power analyses must be considered. We need to take into account the impact that within-park spatial variation as well as systematic and random diurnal, seasonal, and yearly temporal variation have on statistical precision and power. For this effort we are examining pilot and historic data to address 3 key questions. First, for which parks is spatial variability in core parameters minimal so that one monitoring site (or group of subsites) is adequate for representing the river/stream reaches of interest? Examination of pilot data indicate that for NGPN parks with relatively short stretches of rivers, one monitoring site per park will represent core parameters adequately. In addition, pilot data suggest that widely separated sites in a park show similar values in water temperature, pH, and dissolved oxygen. However, data indicate higher spatial variability for conductivity. Additional pilot data will be collected to refine estimates of spatial variation at the parks. Second, regarding expected variability at diurnal, weekly, and monthly scales within a season, is monitoring effort best focused on repeated field visits to periodically measure core parameters or on deployment of continuous monitors to collect measurement throughout the season? Pilot data from continuous probes deployed at 10 locations during 2008-2009 are being used to compare expected accuracy between periodic field visits and continuous monitoring. Third, what effect do alternative sampling strategies and revisit intervals have on our expected power to assess water quality trends at each park? To address this question, we are examining long-term historic data (10+ years) to quantify variation in core parameters for select rivers/streams.