SAMPLING PERIOD INTERVALS FOR MONITORING WATER TABLE DYNAMICS IN PERCHED WATER TABLE SYSTEMS

Abstract

Accurate water table measurement in the near-surface zone is important for siting various land uses, soil taxonomic classification, and transport of contaminants, such as sediment and nutrients, to streams. Because of cost and logistical constraints, many water table studies measure water tables weekly or less frequently and use interpolation methods to simulate daily monitoring. The objective of this study was to evaluate the reliability of weekly measurements in a perched water table system in soils containing fragipans. Daily measured water table levels were compared with daily simulated water table levels derived from weekly measurements using two simulation methods: the linear and the constant interpolation methods. Water table data for 1 year at four landscape positions on a hillslope underlain by fragipan soils were used for the comparison. Both simulated methods underestimated significantly the presence of water tables in the upper part of the soil profile. Differences decreased with depth in the soil profile. The constant simulated method was more accurate than the linear method for all hillslope positions. Up to 89% of the actual days of saturation were not predicted by the simulated water table data in the upper part of the soil profile. The largest differences were found where the average saturation event duration was less than the sampling interval. Most of the saturation events were of less than 3 days duration at those depths (e.g., 20 cm). The scale of sampling interval should be matched to the scale of the water table fluctuations. In many hydrological settings where short-term saturation events are prevalent, such as perched water table systems, accuracy in quantifying water tables with weekly or bimonthly measurements depends on previous knowledge of the frequency and duration of short-duration saturation events.