The typical plant species Cinnamomum camphora in the Needle-broad leaf mixed forest in Changsha was used as research object. Based on the stable isotopic composition of precipitation, bulk soil water, and twig xylem water from March 2017 to September 2019, the seasonal variations of water uptake layers of C. camphora was analyzed by the graphical inference, water-line inference, Iso-Source model and MixSIR model. In addition, the advantages and disadvantages of four methods in determining plant water use are compared, in order to provide new perspectives for the following analysis of plant water use. The results show that: four methods of determining the water uptake layers of C. camphora are basically identical. Pronounced seasonal variation of water use of C. camphora is observed during the observation period. During the wet period consisting of March to June 2017, October 2017 to June 2018 and October 2018 to June 2019, C. camphora mainly uses the soil water at 0—20, 0—40 and 0—20 cm depth, respectively. During the dry periods (July to September) in 2017, 2018 and 2019, C. camphora mostly uptakes the soil water at 20—60, 0—60 and 60—100 cm depth, respectively. If there is no obvious difference of the stable isotopic composition of soil water at each depth, the water-line inference could determine the water uptake layers of C. camphora other than others. The main water uptake layers and utilization ratios of C. camphora calculated by Iso-Source model based on different isotopic composition are both different, however, calculated by the MixSIR model based on different isotopic composition are both largely identical. In conclusion, it is appropriate to combine graphical inference and water-line inference in the qualitative analysis of plant water use, and MixSIR model is more reliable than Iso-Source model in the quantitative research of plant water use.