ESTIMATING WATER CONDUCTIVITY OF SAND-BASED ROOT ZONE MATERIALS FROM PARTICLE SIZE DISTRIBUTION: EVALUATION OF MODELS
Soil hydraulic properties are very important for managing irrigation, drainage, solute movement, soil temperature, and soil aeration. Turf quality is greatly affected by these properties. Methods involved in the measurement of basic soil properties, such as particle size distribution, bulk density, and organic matter content, are relatively simple. However, measurements of water conductivity and water retention are more time consuming and tedious. The objective of this study was to estimate saturated water conductivity (Ks) using particle size distribution, bulk density, and organic matter content of sand materials. A total of 292 samples tested in a commercial laboratory (Dakota Analytical, Inc.) were used for this study, which represent 200 locations representing 40 states in America and two provinces of Canada. The range of peat content in those sand mixtures was 0-20 % (v/v). Stepwise multi linear regression (MLR) method was used for model building in this study. Twelve physicoempirical models selected from published literature were also used to predict Ks. The Geometric mean error ratio (GMER) and geometric standard deviation of the error ratio (GSDER) were used for model evaluation along with other statistics. The MLR model in this study is represented by Log10(Ks)=5.34070.5286ρb1.2846CP0.0442c+0.061250.609510+0.08595 (RW2 =0.60, Radj2=0.59). In the equation, ρb is bulk density (g cm-3), CP is capillary porosity, c is clay content (%), and d is grain diameter in unit derived from particle size distribution curve with weight less than d %. The GMER and GSDER parameters of the MLR model are 1.03 and 1.69 respectively. Organic matter content failed to be selected as a regressor due to its narrow range of content on weight basis (0 1.2%). Of the 12 physicoempirical models the Brakensieks performs the best judged by the GMER (0.85) and GSDER (1.91). However, none of the existing models was as effective in predicting Ks of sand materials as the MLR model developed in this study.
Li, D., Kuehl, D. and Wenjuan Fang, 2008. ESTIMATING WATER CONDUCTIVITY OF SAND-BASED ROOT ZONE MATERIALS FROM PARTICLE SIZE DISTRIBUTION: EVALUATION OF MODELS. Acta Hort. (ISHS) 783:133-146
water movement, prediction, physical property