The effect of particle size and mineralogy of soils on the diurnal cycle of atmospheric water absorption

Direct atmospheric water vapor absorption by structureless soils in coastal deserts has been the subject of various field studies, usually carried out with micro-lysimeters (ML’s). In one of these studies the absorption patterns of loess and sand were studied. Given the larger surface area of the loess soil, it was hypothesized that the loess soil would absorb more water vapor than the sand. The results, when the natural crust present on both soils was removed, were surprisingly similar and contrary to expectation.

We hypothesize that one of the reasons for these results was the different pore size distribution of both substrates, the larger pores of the sand allowing deeper penetration of downwelling eddies, thus directly exposing a thicker soil layer to the atmospheric water vapor concentration and hence to enhanced absorption or desorption.

To test this hypothesis, it is necessary to obtain data on the soil water distribution dynamics within the soil profiles of MLs whose soils have different  pore size distribution.

In the present study we report the response of two aggregate sizes of two substrates (aggregated soil and quartz) to the daily fluctuations of atmospheric conditions.

The field study was carried out at the Wadi Mashash Experimental Farm in the Negev Desert, Israel, using four MLs.   The MLs were instrumented with six temperature and relative humidity (RH) sensors (MX2302A, HOBO) inserted at depths of 0.5, 2, 5, 10, 20 and 45 cm. Water retention curves were obtained using a vapor sorption analyzer (Aqualab, Addium) for the driest part of the curve and standard pressure plate for the wetter parts of the curve. Data from soil and meteorological sensors and scales were recorded every 15 minutes and collected for six successive days during late summer.

The ML with large soil aggregates absorbed significantly more atmospheric water than the one with smaller aggregates, while the opposite trend was observed for the quartz particles. The absorption of both quartz MLs was, however, significantly lower than that of the small aggregate ML.

The temporal changes in soil water content distribution with depth were estimated by using temperature and RH to compute the thermodynamic soil water potential and transforming the latter into water contents via the water retention curves obtained for each soil and size fraction. The computed total water absorption and release patterns of the soil profiles within each of the MLs corresponded very well with the total recorded mass changes.

The depth of eddy penetration was indirectly estimated by comparing the fluctuations of water vapor concentration within the soil at various depths to the one measured simultaneously five cm. above the soil surface.  Penetration depth was larger for the large quartz particles when compared to the small ones, but this effect was not so clear for the soil aggregates.

These results highlight the importance of inter- and intra- pore size distribution in determining water vapor absorption and desorption patterns in bare soils.