The effect of soil type and crust cover on non-rainfall water inputs - laboratory and field experiments revealing contradicting results

Non-rainfall water inputs (NRWI) have been proven to be an important water source in arid and semiarid environments. The reciprocal relation between soils and NRWI plays an important role in the total NRWI absorbed by the soil. In general, the ability of a soil to retain water at a given pressure is a function of the soil properties, and particularly its clay content. The presence of a crust layer on a soil further affects the formation/absorption of NRWI. The objective of the research was to quantify the effect of soil type (loess vs. sand) and crust cover (without crust vs. with bio-crust or physical-crust) on NRWI using two different approaches. The first was a field experiment where four undisturbed 50 cm depth soil samples were used (sand and loess with and without crust). The second approach was a lab experiment where very shallow ( 0.5-1 cm deep) samples were used. The loess soil samples were taken from the Jacob Bluestein Institutes for Desert Research (BIDR), Ben-Gurion University of the Negev (30˚51' N, 34˚46' E, 470 m a.s.l); and the sand samples from the Nizzana Sand Dune area (30˚58'N, 34˚24'E, 226 m a.s.l). The field experiment was conducted at BIDR during the dry seasons, May-October 2016. The mass of the samples was continuously recorded. In the laboratory experiment, oven-dry samples were exposed to 85% relative humidity and mass change of the samples was recorded over 6 days. Relative adsorption rates were calculated. The adsorption process in the field started in the late afternoon with the arrival of sea breeze. The daily adsorption amounts of the crusted loess sample adsorbed more water than the crusted sand, and non-crusted loess soil absorbed more water than the non-crusted sand. The crusted samples seemed to adsorb less water compared to the non-crusted, as the crust hindered the adsorption into deeper layers of the soil. A different, somewhat contradicting, result was obtained in the laboratory experiment. While here as well the adsorption by the loess soil samples (with or without crust) was higher than the sand soil samples, the difference between crusted and non-crusted samples was reversed, such that the crust increased the amount of adsorption in both soil types. It is hypothesized that the contradiction is a result of the samples' depth and the length of time they were exposed to conditions favoring adsorption.