Strategies for characterizing subsurface hydraulic properties often lack observations on actual water movement in soil. The goals of this study were to investigate the variability of wetting front velocities during a field-scale infiltration experiment at the Maricopa Environmental Monitoring site, Maricopa, Arizona (surface area: 50 by 50 m). Wetting front velocities and soil textural components were analyzed as local and effective parameters. Local representation was done by measuring the wetting front travel time through specific depth increments (normally 0.5 m) to a depth of 3 m. Effective representation was done by measuring travel time from ground surface to the depth of interest in 0.5-m increments from the 1-m to 3-m depths. Textural components were represented similarly. One-way analysis of variance (ANOVA) tests were run on wetting front velocities and soil textural components, with spatial location of neutron probe access tubes and soil sampling points classified as replicates and depth intervals classified as independent treatments. The results showed that mean local velocities varied significantly with depth (range=20--44 cm d-1). Effective front velocities varied much less (range=22--25 cm d-1) and were found to be statistically the same, even though differences in soil texture and strong layering were observed. As the total wetting depth increased, variance in effective wetting front velocity decreased, with a coefficient of variation of wetting front arrival of only 9% at the 3-m depth. The decline in the variance of the front velocity is consistent with previous findings that the variance of effective parameters reduces as the measurement scale increases. ¿ 1999 American Geophysical Union