Estimating consumptive water use of rice in lowland paddy fields of Tanjung Karang, Malaysia

A study was conducted to determine the Consumptive Water Use of rice using micro-lysimeter (in-situ) in Tanjung Karang paddy fields. Two estimation methods for evapotranspiration (ET) using FAO Penman-Monteith and weather ground radar data were evaluated and compared with rice crop ET measurements taken during 2011 and 2012 paddy irrigation seasons. Twenty non-weighing microlysimeters (60 cm x 20.3 cm) were installed to measure ETc and deep percolation (DP). The study covered eight compartments in the irrigation service areas (ISA) of the Tanjung Karang Rice Irrigation Scheme (TAKRIS). A total of 1900 ETc data were collected in the study site. Preliminary analysis was done on the field data and no violation of normality and linearity was observed. The results of measured mean ETc for mid-season (April-August 2011) were between 5.9 mm/day, 7.1 mm/day and 5.1 mm/day for the vegetative, reproductive and maturity stages of paddy growing season, respectively. For the main season (August to February 2011) the mean evapotranspiration obtained were 5.1 mm/day, 6.0 mm/day and 5.1 mm/day for the initial, mid and last stages of the growing season, respectively. In the off season (January to May 2012) the mean evapotranspiration for Sawah Sempadan in block C were 5.4 mm/day, 6.6 mm/day and 5.3 mm/day for the first, mid and last stages of paddy growing season, respectively. The mean values of ETcw for mid-season 2011 obtained from CROPWAT software were 4.6 mm/day, 4.8 mm/day and 3.6 mm/day for the first, mid and last stages of paddy growing season, respectively. In the wet-season the mean ETcw found were 4.4 mm/day, 5.0 mm/day and 3.9 mm/day for the vegetative, reproductive and maturity stages of the paddy growing season, while in the off season (2012) ETcw also ranged between 4.4 mm/day, 5.4 mm/day and 4.1 mm/day for the first, mid and the last growth stages of the growing season, respectively. The predicted ET obtained using weather radar data for 21 days (October/November) ranged from 3- 6 mm/day, 3.3-6.3 mm/day and 4.2-6.9 mm/d on the three ISA’s of TAKRIS. The mean deep percolation was between 1.7-6.3 mm/d during the initial growth stage, 1.6-4.1 mm/d at development stage and 2.5-6.5 mm/d at end growth stage period. The mean values of DP during off season irrigation activity for the three growth stages ranged between 2.0-3.6 mm/d, 1.4-3.5 mm/d and 2.2-4.6 mm/d respectively. Eight statistics were used for assessing the goodness of fit and spatial crossvalidation. The statistical model performance for in-situ rice crop ET and CROPWAT ET obtained are RMSE and MAE with values that ranged from 1.34-2.5 mm/d and (-0.62)-0.00 mm/d. They depict the accuracy between measured and computed ET values. The results of model degree of agreement, uniformity coeffi- cient and simulation efficiency lies between (-0.07)-0.45, (-5.8)-(-0.9) and 0.13-0.21 respectively. The results of model performance for weather radar predicted ET was: MBE (-0.004-1.64 mm/d), RMSE (0.54-1.94 mm/d) and MAE (0.44-1.64). The dimensionless coefficient values are dr=0.03-0.68, E=-29.3-0.23 and U=0.77- 0.11 respectively. The rainfall amount observed by the weather radar for the micro-lysimeter sites showed that higher Z-reflectivity values reflect an increase in rainfall and decrease in evapotranspiration. The FAO CROPWAT under-estimate while the values of ET predicted obtained using weather radar data are closer to ET values measured from the field using micro-lysimeter. More research work is needed in obtaining adequate and accurate radar weather data; and better models to accurately predict ET rates for rice crop.

Preview Text FK 2014 80IR.pdf Download (1MB) | Preview

Actions (login required)

View Item