The practicability of using Richards' equation for general purpose soil-water dynamics models