Comparison of empirical models and laboratory saturated hydraulic conductivity measurements

Abstract

Numerous methods for estimating soil saturated hydraulic conductivity exist, which range from direct measurement in the laboratory to models that use only basic soil properties. A study was conducted to compare laboratory saturated hydraulic conductivity (Ksat) measurement and that estimated from empirical models. Soil samples for the study were collected from four sites at varying depths (15cm, 30cm, 45cm and 60cm) at the Faculty of Agriculture Teaching and Research Farm, University of Maiduguri. The Ksat value for each sample was determined in the laboratory using the falling head permeameter method. Soil physical properties (bulk density, porosity, gravimetric water content, % sand and % silt) required by the models were also determined. A refined Kozeny-Carman model and model developed from multiple regression analysis were used to predict Ksat which were compared with the results obtained from laboratory  measurement. The developed model predicted values of 0.0065, 0.0010, 0.0965 and 0.0048cm/s at 15cm, 30cm, 45cm and 60cm, respectively, that is closer to the value of Ksat measured in the laboratory (0.0061, 0.0054, 0.0050 and 0.0048cm/s at 15cm, 30cm, 45cm and 60cm, respectively) while Kozeny-Carman model predicted a value of 0.2208, 0.2161, 0.2020 and 0.1974cm/s at 15cm, 30cm, 45cm and 60cm, respectively, that is far above the one measured in the laboratory. Therefore, Ksat estimating models could not fit for all locations very well.

Key Words: Saturated hydraulic Conductivity, Empirical models, Laboratory measurements