Slope erosion by evaporation

Experimental tests

A simple experimental test of evaporation of wet granular slope was set up. The slope of glass ballotini was prepared with a mean water content equal to W = 1.24%. Then it was placed on an electronic precision balance allowing the measure of mean water content by measuring the weight of the entire glass box device (Figure 1) while a pc-controlled camera recorded the evolution of the slope profile (Figure 2).

Figure 1: Layout of the experimental test of evaporation of a granular slope.

Figure 2: Slope profile during the evaporation process.

The evaporation rate was assumed to be depth-dependent using the empirical relationship (Blight, 2009):

$latex r = r_{max}e^{-c_2d}$

Some experimental tests with cylinder of different heights were performed in order to estimate the evaporation parameters (Figure 3a and 3b).

Figure 3: (a) Experimental tests and (b) calibration of the evaporation parameters.

DEM simulations

The granular slope was reproduced with the same geometry, porosity and particle density as the real one, using 11628 balls and the same amount of water homogeneously distributed among all the contacts with the capillary bridges between sphere and walls also being taken into consideration.

Good accordance was observed with experimental results: typical talus-shape profile was reported (Figure 4). Also the evolution of capillary bridges was monitored (Figure 5) and suction values of about 14000 kPa were measured at the surface of the numerical slope.

Figure 4: Evolution of the simulated slope profile.

Figure 5: Capillary bridges during simulations of evaporation.

For details see

Experimental tests

DEM simulations

References