Non-local Controls on Spatial and Temporal Variability of Soil Water Content in Heavy Clay Soils

Soil water content (SWC) is a key variable for numerous physical, chemical and biological processes that take place at or near the soil surface. Understanding the spatial and temporal variability of SWC at the field scale is of prime importance for implementing efficient measurement strategies in applications. An experimental field in SW Spain, where conventional (CT) and no-till (NT) management on a heavy clay soil are being compared since 1983, was sampled for gravimetric SWC on 38 occasions during 2008 and 2009. Topsoil clay content across the six plots is on average 55%, with a standard deviation of 2.7%. The soil profile was sampled at 54 locations, evenly distributed over the three CT and NT plots, at depths of 0-10, 25-35, and 55-65 cm. In addition, at 18 of the 54 points PVC access tubes were installed and SWC was measured on 64 occasions during the same period, using a Diviner 2000 capacitance probe (Sentek Sensor Technologies, Stepney, Australia). Measurements were made with 10-cm increments down to a depth of 1 m. Topsoil water retention curves (SWRC) were determined in the laboratory on undisturbed soil samples from each of the 54 locations. Topsoil SWRCs showed generally a higher SWC in NT, although differences were only significant within the -10--50 kPa range. An average cumulative difference of 0.08 g/g was found between CT and NT over the entire pressure range. Also gravimetric and probe-measured SWC was significantly higher in the NT plots, especially in the upper soil layers and during the drying periods. Relationships between standard deviation, coefficient of variation, and spatially averaged SWC were determined for both soil management systems, with maximum variability at intermediate SWCs of 0.10-0.12 g/g. The standard deviation of probe-measured SWC was approximately twice the standard deviation of the gravimetric measurements. In addition, the standard deviation of SWC in NT was approximately 50% higher than in CT. For both types of SWC measurements, time-stable SWC patterns were inferred, as well as the points that best represented the field-average SWC throughout the monitoring period. The measurement method affected the time stability patterns. Map comparison methods appeared to be efficient in characterizing soil water content patterns in this work. Seasonality in time stability as well as the recurrence in SWS patterns was analyzed; both appeared to be strongly affected by soil management. Spatial and temporal SWC patterns could be related to soil management system, topography and slope orientation. Overall, analysis of spatio-temporal patterns provides insights into hydrologic functioning of the different domains within the field.