Water and heat fluxes in desert soils. 1. Field studies

Soil physics parameters were monitored in a semiarid site in the Chihuahuan Desert of Texas to evaluate unsaturated flow processes under natural conditions. The bed of an ephemeral stream was instrumented with neutron probe access tubes to monitor water content for 3 years and with thermocouple psychrometers to monitor water potential and temperature for 2 years. Soil samples were collected from stream and interstream settings for laboratory measurement of water content, water potential, water retention, and saturated hydraulic conductivity. Absence of temporal variations in water content in deep (41-m) profiles indicated that within the accuracy of the neutron probe measurements, water pulses were not moving through the system. Penetration of water after rainfall was generally restricted to the uppermost 0.2 m of the sediments. During rain-free periods, the surficial sediments were extremely dry, as indicated by laboratory-measured water potentials as low as -16 MPa. After rainfall, water potentials were highest at the surface and decreased sharply at the base of the wetting front, with gradients as steep as -13 MPa in 0.05 m. Water potentials were out of range of the in situ psychrometers down to 0.8 m throughout most of the monitoring period because the soils were very dry (less than or equal to -7 to -8 MPa). Seasonal fluctuations in water potential were recorded at depths of 1.1 and 1.4 m. Water potentials at greater depths remained fairly uniform with time. Except in the shallow subsurface after rainfall, water potentials generally decreased upward; this trend indicated an upward driving force for liquid and isothermal vapor movement. Soil temperatures displayed large seasonal and diurnal fluctuations. Temperatures generally increased with depth in the winter and decreased with depth in the summer. Downward summer temperature gradients opposed upward water potential gradients and resulted in a downward driving force for thermal vapor movement.