An Analysis of Historic Flows in the Satilla River Using Two Statistical Methods

As development proceeds in the Coastal Plain, increased interest has been directed toward the flow of fresh water into coastal ecosystems. As part of a larger effort investigating changes that may have taken place in three Georgia Estuaries, a historic analysis of freshwater flows into the estuary of the Satilla River was conducted. The annual mean flow on the Satilla River at Atkinson for the period of record (19311998) shows a slight trend upward (p= 0.1.) A further analysis of the historic flows on the Satilla River measured at the USGS station at Atkinson was performed using two different methods that used higher resolution datasets to identify more subtle changes in the hydrograph. The first analysis was performed using the Indicators of Hydrologic Alteration (IHA) trend-analysis method, developed by Richter, et al, 1996, for The Nature Conservancy. This method generates 34 metrics of alteration using daily-flow data, and significant changes were observed in maximum and minimum flows during the winter months. A second analysis was performed using a hydrologic yield calculation modeled after the method of Chagnon, et al, 1996, and Moglen and Beighley, submitted, have used to asses the impacts of urbanization of runoff characteristics for a basin. The hydrographic yield (a ratio of runoff to precipitation) after typical storm events was calculated for storms between 1948 and 1998. The ordered set of these values was then analyzed on a seasonal basis and, again, the most striking results were observed for winter storms. While the range of yield values for storms in spring, summer, and fall was reasonably consistent, there is a marked increase in the variability of yield values for winter storms. As hydrographic yield is strongly influenced by land use, this pattern suggests that seasonally changing land uses (or land uses in which the land cover changes on a seasonal basis) may significantly be affecting runoff patterns in the Satilla basin. INTRODUCTION Development and land use change affect natural systems and may alter the cycles and flows of energy and materials in these systems. The Georgia Coastal Plain has several pockets of significant urban development and much larger areas have undergone land use change due to agricultural or silvicultural practices. Population trends indicate that this region will likely face even more significant changes in the near future. Though the effects of development on the groundwater resources in these areas are well documented, the broader issue of how land use change affects surface runoff in these areas is less well understood. This paper presents two analyses of a long-term surface water dataset for the Satilla River that seek to determine whether a change has occurred in the amount or timing of runoff in this system. Preliminary analysis of the annual mean flow in the Satilla River at Atkinson, GA, indicates a slight increase (p=.1), but the significance this finding and level of detail is low. Though this is a dynamic system, organisms may be most affected by acute events (floods, droughts) rather than gradual annual variation. The first flow analysis below, the IHA method, uses more `biologically relevant,' measures. The second, a hydrographic yield calculation, measures the extent to which rainfall is detained as is runs off the landscape and provides some clues as to the drivers of change in the hydrologic regime. BACKGROUND AND RELATED WORK Richter, et al, 1996, have developed an analytical approach to river flow analysis that assess the humanderived impacts on 64 flow statistics or Indicators of Hydrologic Alteration (IHA). This method, though designed to highlight the effects of a discrete system stressor, such as a dam or intense groundwater pumping, can also be used for broader trend analysis. The