Jost Page 1
Appendix (as supplementary online material)
Verification of internal runoff as dominant runoff process
For the Leptosols as IP II it can safely be assumed that internal runoff is the dominating runoff component (based on soil surveys and geological surveys. Figurea-1 shows that this assumption also holds for the Cambisols/StagnosolsIP I. Three water-table wells at IP I show very distinct responses to rainfall. While two wells show no response to water input (note the different soil depth), the third well shows a temporary water-table with drainage times varying between several days to about two weeks (interval of chemical sampling and analysis). From the high variability in water-table well responses it can be safely assumed that a uniform watertable is unlikely to develop over the entire area of the plot (50x50 m). Instead, water finds sink holes in the range of a few meters, so that it can be safely assumed that internal runoff is the dominating runoff process not only at IP II but also at IP I.
Hydrological modeling
Table a-1 lists measures of agreement between Brook90-simulated soil water storage and net precipitation and observations. Modeled net precipitation (and hence canopy interception) is in good agreement with measured values. As expected from the heterogenic soils, modeled soil water storage for all 3 soil layers atIP I show some divergence from measured values. However, aggregated to biweekly intervals, the chemical sampling interval, the model performance is in the acceptable range (bold numbers in Table a-1).
Table a-1. Goodness of fit statistics between modeled and observed net precipitation and soil water storage at IP I (3 soil layers) and IP II (1 soil layer). Bold numbers refer to biweekly statistics (corresponding to sample interval).
Annual bulkprecipitation between 1993 and 2006 at the Zöbelboden watershed amounts to about 1500 mm in most years (Table a-2). 2002 was the wettest year in the observation period (2128 mm), 2003 was an exceptionally dry year with only 1185 mm of precipitation. Evapotranspiration seems to be less affected by mean annual precipitation then water flow: low precipitation years result in low discharge rates, whereas evapotranspiration remains fairly constant at about 750 mm/year at IP I and 550 mm/ year at IP II (Table a-2). The lower evapotranspiration at IP II is caused by lower interception losses and lower transpiration rates at the shallow soils.
Table a-2. Modeled annual water balances from 1993 to 2006 for IP I and IP II.a
aP0 = Precipitation, TF = Throughfall, St = Stemflow, I = Intercepted rain evaporation, T = Transpiration, SE = Soil evaporation, Seep = Seepage (internal runoff), S = Soil water balance.
"Nitrogen seepage in a karst watershed"