Water transfer and mobile water content measurement in a cultivated crusted soil

In crusted soils, runoff and erosion at the surface are strongly controlled by soil infiltrability. An in situ hydrodynamic characterization of a cultivated crusted soil was conducted to define the factors that reduce infiltrability. Experiments were carried out on three layers of two different profiles in the topsoil: (i) on the surface crust, which was either sedimentary or structural, (ii) within the underlying soil, and (iii) at the plow pan surface. A structural crust is the result of gradual coalescing of aggregates by raindrop compaction, whereas a sedimentary crust is formed by deposition of the particles suspended in overland flow. The purpose here was to characterize water transfer as a function of vertical heterogeneity. A tension disc infiltrometer, along with an 180 solution, was used to create a near-saturated flow. Hydrodynamic properties and mobile water fraction of the soil surface were inferred from the cumulative infiltration and the soil solute concentration at the end of the experiments. Visual observations of X-ray images obtained from thin sections were used to emphasize some of the conclusions about the hydrodynamic characterization. Results of infiltration in soil covered with either more or less developed crusts were compared. Then, comparisons were made between the infiltrability of the underlying soil, which was covered by sedimentary or structural crusts. Finally, estimated values of hydraulic conductivity and the mobile water fraction for each layer of the two profiles provided information on water transfer. Results showed that the fraction of the soil surface covered by sedimentary crusts and structural crusts was an important factor for infiltrability (the cumulative infiltration at t = 5000 s varied between 11.5 mm and 14.8 mm in soil covered by a sedimentary crust, whereas the variation was between 18 nun and 22 nun in soil covered by a structural crust). On the other hand, infiltrability did not depend on the developmental stage of the surface crusts as the differences between the cumulative infiltration in more or less developed crusts were not significantly different at P = 0.05 (according to the Student's t test). The sedimentary crust seemed to protect the underlying soil from aggregate coalescence. Thus, collapsing was less important in the underlying plowed material covered by a sedimentary crust. As a consequence, the mobile water fraction and effective mean pore size estimations showed that in the case of strongly collapsed material, coalescing increased the pore connection: water transferred through small but well connected pores (the effective mean pore size was lambda(m) = 0.105 mm, and the mobile water fraction was f = 0.93). When there was less collapsed underlying soil, the pores participating in transfer were bigger but less connected (lambda(m) = 1.2 mm and f = 0.5). The plow pan did not show strong impermeable behavior because the macropores made by roots were not sealed by plowing.