Seasonal super(18)O variations and groundwater recharge for three landscape types in central Pennsylvania, USA

Seasonal super(18)O variations in precipitation, soil water, snowmelt, spring flow and stream baseflow were analyzed to characterize seasonal dynamics of groundwater recharge in three central Pennsylvania catchments. The catchments represented three common landscape types: Valley and Ridge-shale (Mahantango Creek), Valley and Ridge-carbonate (Buffalo Run), and Appalachian Plateau- sandstone (Benner Run). Samples were collected on a biweekly basis from May 18, 1999 to May 9, 2000. Precipitation, soil water, and baseflow isotopic composition data indicated that a seasonal recharge bias existed for these catchments, most recharge occurred in the fall, winter, and spring months. An altitude effect of-0.16 to-0.32/100 m change in elevation was discernible in precipitation, soil water, and stream baseflow isotopic compositions. Soils effectively damped seasonal variations of recharge super(18)O composition after depths of 1.62-2.85 m. The greatest damping of the annual isotopic composition signal occurred in the shallow soil layers (0-15 cm). In these and similar landscapes with thick soils the annual isotopic composition signal may be completely damped prior to reaching the stream as baseflow. Isotopic variations measured in stream baseflow are more likely to be caused by the shallow flowpath water relatively close to the streams. Baseflow stable isotope variations found on the basins studied suggested that residence times for subsurface waters to reach channels were much longer than the annual seasonal cycle of super(18)O in precipitation. Damping depths were similar for the three different catchments but it is not certain how spatially variable damping depths were within each catchment. This information would be useful in determining areas within catchments that contribute to short term isotopic composition fluctuations within streams ('new water'). Predictive models that determine isotopic damping depth from meteorological, soil and vegetation/land-use data can help develop a better understanding of the variability of recharge and attenuation of contaminants across the landscape.