Water use efficiency of flooded rice fields II. Percolation and seepage losses

Abstract

The concept of a constant seepage and percolation (SP) rate in monitoring the water balance of flooded rice fields, as often used in e.g. irrigation system design and management, was investigated. First, magnitude and variability of percolation rate were studied for different combinations of soil-hydraulic properties and hydrologic conditions using the validated water balance model SAWAH. Percolation losses from fields with relatively low subsoil permeability ( k s,sub < 10 −1 cm d −1) are either limited by a poorly permeable plow sole ( k s, top < 10 −2 cm d −1) or by the low hydraulic conductivity of the subsoil itself. Typical percolation losses of 0–0.5 and 1–1.5 cm·d −1 respectively are hardly affected by ponded water depth, subsurface water content and depth of ground water table. Percolation losses from fields with relatively high subsoil permeability ( k s,sub10 1 cm·d −1) may vary from 0–0.5 cm d −1 with a poorly permeable plow sole, to 1–5 cm·d −1 or more for a relatively permeable plow sole ( k s, top10 −2 cm d −1). Only in the latter case, percolation rates are largely affected by the depth of ponded water. Next, the constancy of combined SP rates was studied in a field experiment on a permeable subsoil. Simple book-keeping of the water balance using a fixed SP rate proved accurate to predict the depth of ponded water in time in case of a poorly permeable plow sole and a small seepage component. A decision tree was suggested based on soil-hydraulic properties and characteristics of bunds to estimate the magnitude and variation of SP rates, and to decide whether book-keeping with a fixed SP rate is an appropriate tool in monitoring the water balance of paddy fields.