Seasonal oxygen-18 variations in precipitation, throughfall, soil water, spring¯ow and stream base¯ow were analysed to compare the hydrology of two forested basins in West Virginia (WV) (34 and 39 ha) and one in Pennsylvania (PA) (1134 ha). Precipitation and throughfall were measured with funnel/bottle samplers, soil water with ceramic-cup suction lysimeters and spring¯ow/base¯ows by grab and automatic sampling during the period March 1989 to March 1990. Isotopic damping depths, or depths required to reduce the amplitude of subsurface oxygen-18¯uctuations to 37% of the surface amplitude, were generally similar for soil water on the larger PA basin, and base¯ows and headwater spring¯ows on the smaller WV basins. Computed annual isotopic damping depths for these water sources averaged 49 cm using soil depth as the¯ow path length. The equivalent annual mean hydraulic diusivity for the soil¯ow paths was 21 cm 2 d À1 . Mean transit times, based upon an assumed exponential distribution of transit times, ranged from 0 . 2 y for soil water at a depth of 30 cm on the larger catchment, to 1 . 1±1 . 3 y for most spring¯ows and 1 . 4±1 . 6 y for base¯ows on the smaller catchments. Base¯ow on the larger PA basin and¯ow of one spring on a small WV basin showed no detectable seasonal uctuations in oxygen-18, indicating¯ow emanated from sources with mean transit times greater than about 5 y. Based upon this soil¯ow path approach, it was concluded that seasonal oxygen-18 variations can be used to infer mean annual isotopic damping depths and diusivities for soil depths up to approximately 170 cm.

No abstract

The differences between 6 l 8 0 in throughfall and open rainfall were studied for 16 selected spring and summer storms in deciduous, pine and spruce forests in central Pennsylvania, USA. Throughfall 6 l 8 0 averaged 0.17, 0.32 and 0.24% breater than 6I80 of open rainfall for all storms at the deciduous, pine and spruce sites, respectively. Throughfall 0 enrichment was greater in low intensity spring rainfall events than higher intensity growing season storms and greater in the coniferous stands than the deciduous hardwood stand. Maximum throughfall l 8 0 enrichment of 1.61% occurred in the spruce forest during one spring event. The differences between rainfall and throughfall I8O observed in this study for individual storm events may have important implications for isotope flow separation studies.

A study was conducted to quantify the source of increased dissolved aluminum concentrations during acidic storm flows on a small Pennsylvania stream. Data for six episodes during fall 1986 and spring 1987 showed depressions in stream pH and increases in sulfates, conductivity, dissolved organic carbon, and dissolved aluminum. Flow separation analyses were conducted using 180 as a tracer in a three-component mass balance tracer model. Results showed that soil water and groundwater are the dominant flow sources, accounting for approximately 20 and 75% of total flow during storms, respectively. Channel precipitation generally provided less than 5% of total flows. Hydrograph separation using aluminum agreed with 180 results, while other i:hemical parameters produced unsatisfactory results. The data support Hewlett's (1982) variable source area concept of storm flow generation with inputs of older, deep circulating groundwater from low-elevation source areas early in an event and later inputs of younger soil water and possibly shallow groundwater from expanding source areas at higher elevations. The results suggest that the most toxic' ,rt•, 0ff events for aquatic life occur during large storms when the greatest inputs of soil water cauSerelevated stream dissolved aluminum concentrations. Reductions in storm flow acidity anal dissolved aluminum concentrations on this catchment will be most dependent upon changes in soil water and/or groundwater chemistry.

showed predicted average changes in mean annual streamflow ranging between + 24 and -49% for the specific climate change scenarios tested (precipitation changes -20 to +20%, temperature changes 0øC to +4øC). Urbanization increased mean annual streamflow in rough proportion to average cumulative changes in population density on the basins, equivalent to an average flow increase of 103% with complete watershed urbanization. Urbanization also appeared to reduce the sensitivity of mean annual streamflow to temperature changes compared to mean flow response on rural basins. No significant regional differences in mean flow response to climate change and urbanization were found. Despite the uncertainty in predicting future streamflow with models based upon past records, urbanization appears potentially capable of significantly offsetting flow declines or augmenting flow increases caused by climate change.