Publications

Huang, LX; Cai, JB; Zhang, BZ; Chen, H; Bai, LL; Wei, Z; Peng, ZG (2019). Estimation of evapotranspiration using the crop canopy temperature at field to regional scales in large irrigation district. AGRICULTURAL AND FOREST METEOROLOGY, 269, 305-322.

Abstract
Evapotranspiration (ET) is a major component of the hydrologic cycle, and its accurate estimation is essential in water resource management. In this study, field experiments were performed from 2015 to 2016 in the Jiefangzha Irrigation Area, Inner Mongolia, in China. To estimate the daily crop ET (ETd) using the midday canopy and air temperature difference (T-c-T-a) and net radiation (Rn-d), the S-I model (Seguin and Itier, 1983) was calibrated and validated based on observations from maize and sunflower fields. The canopy temperature and meteorological factors were monitored continuously and synchronously from June to August with a Canopy Temperature and Meteorology Monitoring System (CTMS). Parameters a and b in the S-I model were fitted at 10:00 to 16:00 by linear regression analysis between (ETd - Rn-d) and (T-c-T-a), from June to August 2015. Based on the values of a and b, the daily ET can be estimated using the S-I model (named ETM) from 10:00 to 16:00 in 2016, which provided good results for all the statistical parameters. The S-I model has performed reasonably well in predicting the ETd during the maize and sunflower growth period at the field scale. The Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM) was employed to produce and fuse daily Land Surface Temperature (LST) (30 m) data by integrating high-spatial Landsat data and high-temporal MODIS images during the primary growing period of maize and sunflowers from June to August in 2015 and 2016. The fused LSTs exhibited good performance compared with the canopy temperature observed using the CTMS system in situ. Afterwards, the daily scale-field ET (30 m) was estimated by using the validated S-I model during the primary growing periods for maize and sunflowers at 11:00 in 2015 and 2016. The results showed that the estimated values of ET from the S-I model were consistent with the results for the regional water balance, with average relative error (RE) values of 7.0%, 5.0%, 9.4% and 7.3% (absolute value) in two years, in the Huangji, Qinghui, Wula and Yangjiahe sections of the Jiefangzha Irrigation Area, respectively. This finding indicates that as a relatively simple and convenient method, the S-I model could be applied to estimate the ET at a regional scale.

DOI:
10.1016/j.agrformet.2019.02.024

ISSN:
0168-1923