Drought is a major limiting factor for the production of rainfed rice in the Mekong Region. Thus, estimation of the length of growing period (LGP) based on estimation of the development of water deficit is essential for sound planning of agricultural development. A recently developed soil-water balance model was used to quantify the availability of water for rice crops and yield reduction due to water deficit. Field water availability is known to be largely affected by soil type and rainfall pattern, and their separate effects were investigated in this study of a spatial analysis of LGP, using inputs for rainfed lowland rice in Savannakhet province in central Laos. The analysis showed that the start of growing period (SGP) and end of growing period (EGP) were affected largely by geographical variations in rainfall and soil clay content, respectively. Also, the areas having relatively short LGP were generally associated with large yield reduction because of low water availability associated with coarse-textured soils. At local scale, field water availability varied from upper to lower positions of a sloping land (toposequence) in the rainfed lowland ecosystem of the Mekong region, causing variation in yield within the toposequence. Using the level of field water determined around flowering time at different toposequence positions in 45 farms, estimated yield reduction was compared in seven main rice-growing districts of the province. Variability of yield loss, associated with variation in water availability, was larger across the toposequence positions than across districts, showing the importance of local variability in determining yield in rainfed lowland rice. The present approach of a combination of simulation model and GIS is adopted for characterisation of the water environment for rainfed lowland rice in other parts of Laos, as well as in neighbouring Thailand and Cambodia.
A daily salt balance model for stream salinity generation processes following partial clearing from forest to pasture