Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
4479566 | Agricultural Water Management | 2009 | 10 Pages |
Paddy is commonly grown under flooded or submerged condition in which substantial amount of water is lost by different processes. Puddling is traditionally done to reduce water loss from lowland rice fields. Since the very step of puddling, rice root zone undergoes structural changes leading to the formation of a layered profile having a hydraulically less-conductive plow sole below the root zone. However, studies have shown that soil cracking and the presence of preferential flow paths in puddle fields defeat this purpose. Description of soil water regime in such a dynamic soil requires an in situ measurement method for soil hydraulic properties. A field experiment was conducted in twelve 30 m2 plots during the rainy seasons (June to October) of 2004 and 2005 to evaluate a suitable method for estimating soil hydraulic properties of lowland paddy soil. Results showed that piezometric (pressure) heads installed in different soil layers responded to the drying and wetting cycles typically followed in transplanted rice and are observed as a part of monsoon climate in eastern India. The Marquardt–Levenberg algorithm built in the HYDRUS-1D simulation environment was used to inversely estimate soil hydraulic parameters. Estimated parameters revealed larger hydraulic conductivity for the compacted plow sole than those published in literature, which may have resulted from alternate wetting and drying typically observed under monsoon climate and earthworm burrows observed in our experimental field. Results from simulation studies suggest that both the single- and dual-porosity models could simulate water flow considerably well in lowland paddy field although the latter described pressure head time series data slightly better in about 50% of simulations. Similar performance of the single- vs. dual-porosity model may have resulted from estimating a seasonally mean soil hydraulic properties which include the effect of both preferential flow and matrix flow as the specific soil and boundary conditions prevailed. While water may have preferentially transported through the macropores during the wetting cycles in a near-saturated soil, it would have dominantly moved through soil matrix during the drying cycles. This study shows that simple piezometers may be combined with a simulation model to estimate hydraulic properties of different soil layers in a lowland paddy field.