ESTIMATION OF ZERO-TENSION PAN LYSIMETER COLLECTION EFFICIENCY

When using zero-tension pan lysimeters to assess the mass of solute leached, it is necessary to estimate pan collection efficiency (PCE), defined as the proportion of water draining through the root zone captured by the pan lysimeter. To evaluate PCE in a well structured Hagerstown silt loan, a study was designed to estimate PCE using two bromide (Br-) tracer methods (Br- recovery and Br- mass balance methods) and two water budget analyses (Mather water budget and the LEACHM water flow model). The tracer estimates of PCE alw provided the opportunity to compare the accuracy of leachate volumes predicted by the two water flow models. Eighteen zero-tension pan lysimeters (0.465 m2) were installed 1.2 m below the soil surface in an experiment designed to evaluate nitrate leaching from corn (Zea mays L.). A KBr tracer was broadcast over all plots at 100 kg Br-ha-1. Four methods were used to estimate PCE. The Br- recovery method set the maximum Br- recovery in the pan lysimeters to be 100%. Bromide mass collected in each pan was divided by the maximum collected to estimate PCE for each pan. Bromide mass balances were determined by measuring mass of soil Br- to a depth of 1.2 m, plant Br- uptake, and the Br- recovered in each pan. Unaccounted for Br- provided the estimate of PCE. Water budget PCE estimates were determined by dividing the leachate collected by each pan by the amount predicted by the models. Leachate volumes collected were highly variable with coefficient of variation (CV) of 43%; however, leachate collected within individual pan lysimeters remained fairly consistent among years. The amount of Br- recovered in the pan lysimeters ranged from 9 to 76% of that applied, with a CV of 53%, and was highly correlated to the amount of leachate collected (r = 0.72, p < 0.001). Estimates of individual PCE ranged from 13 to 92%, with an overall average of 52%. Averages of all four efficiency methods were similar, ranging from 45 to 58%. Bromide tracer estimates were similar to the water budget estimates, indicating that tracers can be useful to assess PCE. The tracer data also provided a method to compare the two water flow models. The LEACHM model better fit the cumulative leachate collected in 1988 and 1990; in 1989 there was no difference. In years when precipitation exceeded evapotranspiration, the models performed equally well; however, the more complex LEACHM model performed better when evapotranspiration exceeded precipitation. Our estimates of PCE using 0.465 m2 pans were higher than other reports using smaller surface area pans, supporting previous observations that increasing the pan surface area increases the chances of capturing macropore flow or reducing water flow around the pan by capillary attraction.