Evaluation of Some Estimating Equations for Scaling regarding Soil-water Characteristic Curves

Abstract

Sixty soil samples were taken from a plow layer within Shengyang Agriculture Univ. experimental field using the mesh sampling method. These samples were classfied as “BROWN EARTH” in the Chinese soil classification system. Soil-water retention tests were carried out to obtain the soil-water characteristic curves.
These soil-water characteristic curves were scaled by the following four functions with linear or nonlinear regression.
(1) h=Σ[a_k (1-s^k)]/s, (2) s= [1+ (ah) ⁿ] ^-m, (3) s=ah^-b, (4) θ=ch^-d where h is suction, s the degree of saturation, θ is volumetric water content, a, b, c, d, m, n are constants.
The major results are as follows.
(a) Functions (2), (3) and (4) were suitable to estimate these soil-water characteristic curves for Brown Earth.
(b) Function (4) was the most useful function for applying the scaling technique to practical problems.
It is because the volumetric water content (θ) could be directly calculated instead of its function.
(c) It was found that the relationship between cumulative probability and scaling factors were approximately log-normal and normal distribution.
(d) Scaling factors that were obtained from functions (1)-(4) were not equal to each other, but had a strong correlation to the lineal or exponential functions.
(e) These scaling factors can be mainly determined by bulk density or porosity, soil texture and the degree of saturation before a test.
The greater the bulk density, the content of soil particle size of <0.01 mm and the degree of saturation in soil sample before test, or the smaller the content of soil particle size of> 0.25 mm, the smaller the scaling factors became.
(f) In the same suction, the water content estimated from soil-water characteristic curves using the scaling technique was smaller than it from soil-water characteristic curves obtained by the normal method of least squares.