Observations and Analysis of dynamic non equilibrium effects on soil hydraulic parameters

In the past years, many scientists have observed dynamic non equilibrium effects on water flow in unsaturated soil where the Richard's equation fails to express the relationship between water content and hydraulic potential in the unsaturated zone. Until now, traditional techniques for determining soil hydraulic parameters have been used involving stepwise equilibrium imbibition and drainage experiments. More recently, optimization of the parameters describing the soil hydraulic functions in a transient outflow experiment using conventional methods have proved to be promising to derive soil hydraulic parameters but the estimates from one step outflow experiments using only cumulative outflow data in the objective function are often unreliable and non-unique. This work is an attempt to calculate the soil hydraulic properties using multi step outflow experiment data with different soil samples from Little Washita watershed, Oklahoma. The new dual fraction model (Diamantopolous et. al., 2012) in which water flow is treated as two fractions, one fraction with water in equilibrium with the pressure head whereas water in the second fraction as in non-equilibrium, has been used to ascertain the soil hydraulic parameters. Genetic Algorithm (GA) has been used to optimize the inverse solution of soil hydraulic parameters and to estimate uncertainties of calculated parameters using multi-populations within GA and by considering data and modeling errors. The authors aim to use the present work to be a stepping stone for calculation of dynamic non-equilibrium effects on soil hydraulic parameters estimation on greater (field) scale. The concern about dynamic non equilibrium effect on larger scale is still unanswered satisfactorily, because of conflicting hypothesized reasons such as: 1. Air entrapment is not expected to occur under natural drainage conditions, 2. erratic change in pressure at the system boundary unlike MSO experiments differ from that occurring in natural conditions. Simulated soil moisture values will be validated against observed air borne remotely sensed data soil moisture data.