Comparisons of energy balance and evapotranspiration between flooded and aerobic rice fields in the Philippines

The seasonal and annual variability of sensible heat flux (H), latent heat flux (LE), evapotranspiration (ET), crop coefficient (Kc) and crop water productivity (WPET) were investigated under two different rice environments, flooded and aerobic soil conditions, using the eddy covariance (EC) technique during 2008-2009 cropping periods. Since we had only one EC system for monitoring two rice environments, we had to move the system from one location to the other every week. In total, we had to gap-fill an average of 50-60% of the missing weekly data as well as those values rejected by the quality control tests in each rice field in all four cropping seasons. Although the EC method provides a direct measurement of LE, which is the energy used for ET, we needed to correct the values of H and LE to close the energy balance using the Bowen ratio closure method before we used LE to estimate ET. On average, the energy balance closure before correction was 0.72 ± 0.06 and it increased to 0.99 ± 0.01 after correction. The G in both flooded and aerobic fields was very low. Likewise, the energy involved in miscellaneous processes such as photosynthesis, respiration and heat storage in the rice canopy was not taken into consideration. Average for four cropping seasons, flooded rice fields had 19% more LE than aerobic fields whereas aerobic rice fields had 45% more H than flooded fields. This resulted in a lower Bowen ratio in flooded fields (0.14 ± 0.03) than in aerobic fields (0.24 ± 0.01). For our study sites, evapotranspiration was primarily controlled by net radiation. The aerobic rice fields had lower growing season ET rates (3.81 ± 0.21 mm d-1) than the flooded rice fields (4.29 ± 0.23 mm d-1), most probably due to the absence of ponded water and lower leaf area index of aerobic rice. Likewise, the crop coefficient, Kc, of aerobic rice was significantly lower than that of flooded rice. For aerobic rice, Kc values were 0.95 ± 0.01 for the vegetative stage, 1.00 ± 0.01 for the reproductive stage, 0.97 ± 0.04 for the ripening stage and 0.88 ± 0.03 for the fallow period, whereas, for flooded rice, Kc values were 1.04 ± 0.04 for the vegetative stage, 1.11 ± 0.05 for the reproductive stage, 1.04 ± 0.05 for the ripening stage and 0.93 ± 0.06 for the fallow period. The average annual ET was 1301 mm for aerobic rice and 1440 mm for flooded rice. This corresponds to about 11% lower total evapotranspiration in aerobic fields than in flooded fields. However, the crop water productivity (WPET) of aerobic rice (0.42 ± 0.03 g grain kg-1 water) was significantly lower than that of flooded rice (1.26 ± 0.26 g grain kg-1 water) because the grain yields of aerobic rice were very low since they were subjected to water stress. The results of this investigation showed significant differences in energy balance and evapotranspiration between flooded and aerobic rice ecosystems. Aerobic rice is one of the promising water-saving technologies being developed to lower the water requirements of the rice crop to address the issues of water scarcity. This information should be taken into consideration in evaluating alternative water-saving technologies for environmentally sustainable rice production systems.