One-dimensional Markov chain simulation of vertical change of soil texture in middle reaches of Heihe river, northwest China

In the Linze County, located in the middle reaches of the Heihe River, northwest China, the geomorphic types include alluvial-proluvial plain, plain of denudation, and desert grassland in the fringe of an oasis. The main landscapes of this region are composed of desert, oasis farmland and wetland, all of which distribute irregularly and patchily. The differentiation of soil textures was observed both horizontally and vertically. In a vertical direction, sand and clay layers in soil usually sandwich each other, even when there are some impermeable layers in some profiles. The heterogeneity of soil profile textures with depth significantly influences water movement and solute transport. In this study, all 2,000 soil samples were collected with a 5-cm diameter auger to a depth of 300 cm at 100 points in an area of 100 km2. After being taken into a laboratory and having been wind-dried, the soil mechanical composition was measured using a Malvern Laser particle size analyzer. Then the vertical change of soil textural layers was analyzed, and a MC-LN (Markov chain- lognormal distribution) model was constructed to simulate the soil textural profiles. Results showed that there were seven soil textural types occurring in the study area, namely, sand, loamy sand, sandy loam, loam, clay loam, silty clay loam and silt clay, respectively. Compared with the non-occurrence of silty clay layers in the surface soil, another six types of textural layers all occurred in the surface soil of the study area, while sand layers occurred with a much higher probability than the others. The layer thickness of each textural type in the study area was characterized as a lognormal distribution, with relatively thick sand and silty clay loam layers, and relatively thin sandy loam, loam and clay layers. For a certain textural type, layers occurred beneath it were mainly the two types which adjoined it, especially the one which contained more fine particles. The uncertainty analysis of soil textural layer transitions indicated that the formation of loamy sand layers was much strongly dependent on the lower layers, whereas the clay loam layers had a key effect on the formation of the upper layers. Loamy sand and loam layers had a relative high probability to have sandy loam layers as upper layers, while silty clay loam layers had relative high probability to occur as upper layers of both clay loam and silty clay layers. None of the seven textural layers had the same combinations of upper and lower layers simultaneously. Markov characteristics and the stability of the vertical change of textural layers were verified by χ2 test using the TPMs of the entire samples, the subintervals and the sub-regions. A one-dimensional MC-LN model could quantitatively describe the vertical change of textural layers. The simulated TPM approached to the measured TPM, and the simulated layer thickness of each textural type was close to the measured value. The main combinations of textural layers along each profile were sand-loam, loam-sand, loam-clay, and clay-loam. Quantitative description of soil textural profiles will benefit further study on water cycles, transformation and solute transfer in soil profiles. This, in turn, will serve to improve the water use efficiency、the relieve and the salinization of the study area.