Submitted to: Soil Science Society of America Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/4/1995
Publication Date: N/A
Citation: Interpretive Summary: Large pores open to the soil surface are important for infiltration of water and air exchange. Persistence of these large pores may be indicative of soil quality. A study was conducted to examine the variation of these pores across a landscape at different times. Measurements were made on four different dates along a line that went up and down the landscape. The measurements were made on a wide range of soils that would be typical of a central Iowa landscape. There was as much variation among the measurement dates as there was across the landscape. There was no consistent trend across the landscape for the measurement times. This variation was due to creation and settling of the pores due to tillage operations, formation and disruption of surface seals, and the activity of roots, earthworms, and ants on the formation of surface pores. The variation in the measurements of infiltration decreased with time after tillage. Understanding the temporal changes in the soil surface across a landscape and after tillage operations will improve the ability to manage the soil and to show how changes occur as a result of different soil management practices. Influence of the different factors on water movement will assist researchers who must understand how soil management practices influence water movement and potential water quality problems.
Technical Abstract: Pore arrangement in the surface soil varies over space and time. Ponded and tension infiltration measurements indirectly indicate functional soil pores. This study used ponded and tension infiltration measurements to quantify spatial variability across a landscape at different times. Paired small-base infiltrometers (76 mm diameter) were used at four pressure heads (+5, -30, -60, and -150 mm) across a transect four times. July, 1991 date followed a cultivation; soil was dry for the August, 1991 date; no tillage had occurred for 8 months prior to the April, 1992 date; and the May, 1992 date was 3 weeks after disking and planting. Mean water contents were 0.20, 0.30, and 0.27 m/m for the August, 1991, April, 1992, and May, 1992 dates, respectively. Ponded conductivity (K) data was best transformed by a natural log transformation, but tension data was best transformed by 1n(K+1). No consistent trend was apparent across all dates for landscape positions. A moving average for K at a head of -150 mm for the April, 1992 date revealed periodic behavior at an interval of approximately 150 m. Shorter interval periodicity was indicated for the July, 1991 date. A moving split window technique divided the July, 1991 data into nine units across the transect, but only into five units for the April, 1992 date. For ponded K, only within-pair data were significantly correlated (0.6 to 0.8 m). At a head of -150 mm, K was spatially correlated over distances of 6.6, 16.8, and 0.6 m for the dates July, 1991, April, 1992, and May, 1992, respectively. The half cycle autocorrelation pattern was similar to that shown by moving averages and moving split window technique.