Abstract: Abstract: Evapotranspiration (ET) and its estimation and partition are important research contents in the soil-plant-atmosphere continuum (SPAC), as it provides abundant information about water and heat transfer and plays an important role in the efficient water use in agriculture. The isotopic technique is considered to be an effective method to study it. Here, the isotopic compositions of soil water, evaporated vapor, transpiration vapor and atmospheric background vapor were measured and 2 isotope-based methods (Iso-CG method, isotope combined with the Keeling plot and Craig-Gordon model; and Iso-WB method, combining isotope conservation and water balance) were applied to estimate and partition ET. The study was conducted in 2015 at Shiyanghe Experimental Station of China Agricultural University, Wuwei City, Gansu Province, China (37°52′ N, 102°51′ E). The results showed the transpiration fraction varied from 0.64 to 0.91 by Iso-WB method and 0.52 to 0.91 by Iso-CG method, with a mean of 0.80 and 0.78 (±0.01), respectively, during the entire growing season of maize. However, the results calculated by Iso-WB were overestimated. The mean ET of maize were 3.95, 5.30 and 4.98 mm/d (±0.15 mm/d) during the earlier stage, middle stage and later stage. The results of ET showed that the transpiration rate of maize changed greatly with the seasons, showing a small transpiration rate at the early growth stage, and a slow decline in the later period. The ET determined by Iso-CG method was robust, but there could be some deviation in Iso-WB due to its measurement accuracy was relatively high and the calculation error was large. It indicated that it was relatively accurate to estimate and distinguish ET by using the Keeling plot and Criag-Gordon combined chamber system method. Compared with conventional methods, such as combining hydrometric measurements (e.g., sap flow) to estimate transpiration with other methods (e.g., weighing lysimeter measurements) to measure evaporation, the stable isotope based method had the advantages of high accuracy and controllability. For example, the calculation accuracy of the 2 isotope methods used in this study was about ± 0.15 mm/d. The study results not only provide valuable information for determining irrigation scheduling of maize farmland and improving water-use efficiency, but also have important significance for exploring deeply the field of oxygen isotope hydrology. Meanwhile, the isotope-based method has its limitations compared to other methods of scaling up from leaf level to canopy scale. The sources of error in this isotopic based methods study may come from the variability of isotopic signatures, weather conditions and sample size and time. Therefore, it is still a great challenge in ET estimation and partition. There are very few direct methods to measure isotopic composition of plant transpiration and soil evaporation. We expect that future refinement of methods will allow for accurate and continuous measurement of transpiration and evaporation isotopic composition. For some limitations of determining evaporation isotopic composition, we recommend that field experiments pay particular attention to sampling precision at the evaporation front in the surface soil and temperature.