Water temperature is an important factor in aquatic environments. Dam construction, especially the construction of multiple dams in rivers, can greatly affect the downstream water temperature. Several dams, including Wudongde, Baihetan, Xiluodu, Xiangjiaba, Three Gorges, and Gezhouba, have been constructed between Panzhihua and Yichang along the Yangtze River. The aim of this paper was to quantify the impact of these dams on the water temperature downstream. One-dimensional and two-dimensional models were used to simulate the water temperatures, and the results showed that the dams had different cumulative effects on it. For example, in January, after the construction of the Xiangjiaba and Xiluodu dams, the discharge water temperature of Xiangjiaba was 3 °C higher than the natural conditions, and after the construction of the Baihetan and Wudongde dams was completed, it increased by a further 2 °C. The natural river ran over 416 km with no dams from the Xiangjiaba dam to the Cuntan Station. With the influence of climate and tributary inflow, the impact of upstream dams on the water temperature was mitigated by more than 48% at Cuntan Station, displaying a recovery. It seemed that the cumulative effects of dams on the discharge water temperature of the Three Gorges decreased with the increase in the upstream storage capacity from March to May, and the construction of dams even had a negative effect. From September to February of the next year, the cumulative effects increased with the increase of the upstream storage capacity, but only the total storage capacity until a certain level, where no further impact was observed.

The hydrological cycle is particularly sensitive to and is greatly affected by global climate change. In addition, runoff change has a strong influence on the hydrological cycle and migration of biogenic substances. The Yarlung Zangbo River basin in China is a typical basin for which climate and hydrological data are lacking. Land surface models can provide data for studying land surface substance and energy circulation, which are meaningful to face climate change. The midstream region of the Yarlung Zangbo River basin, which is strongly affected by climate change, was selected as the study area. First, the observed mean temperature, precipitation and runoff characteristics were analysed. Second, after combining the Global Land Data Assimilation System (GLDAS) and the water balance equation, we simulated climate and hydrological processes for the same time period. Finally, the correlation and error between GLDAS and observed data were analysed to verify applicability of the GLDAS data, and the impacts of climate factors on runoff were discussed. The results revealed that under the background of global warming, precipitation, temperature, and runoff changed significantly and showed strong consistency during the research period. Mean monthly precipitation, temperature and runoff exhibited clear cyclical fluctuations of approximately 12 months, and they all tended to increase. GLDAS is not a good system to describe the land surface conditions of the Yarlung Zangbo River basin all the time. However, within a certain time period, GLDAS data have a good applicability in the basin. Thereinto, the GLDAS mean monthly precipitation was moderately correlated with observed precipitation, with a correlation coefficient of 0.75. GLDAS mean monthly temperature was highly correlated with observed data, with a correlation coefficient of 0.94. Based on the Brunke ranking method, it indicates that GLDAS-Noah-based runoff data were closer to observed runoff data than the three other GLDAS models. Correlation coefficients between precipitation and runoff for the three time scales were higher than those between temperature and runoff. This means that rainfall was the main factor affecting natural runoff change, as opposed to temperature, and it can control the evolution of the river to some degree. This paper indicates the impacts of climate change on runoff and the application of GLDAS for data-limited basins. The results provide a deeper understanding of the Yarlung Zangbo River basin characteristics and can provide a scientific basis for the management of water resources and policy implementation for this basin.

The Qinghai-Tibetan Plateau region has unique meteorological characteristics, with low air temperature, low air pressure, low humidity, little precipitation, and strong diurnal variation. A two-dimensional hydrodynamic CE-QUAL-W2 model was configured for the Pangduo Reservoir to better understand the thermal structure and diurnal variation inside the reservoir under the local climate and hydrological conditions on the Qinghai-Tibetan Plateau. Observation data were used to verify the model, and the results showed that the average error of the 6 profile measured monthly from August to December 2016 was 0.1°C, and the root-mean-square error (RMSE) was 0.173°C. The water temperature from August 2016 to September 2017 was simulated by inputting measured data as model inputs. The results revealed that the reservoir of the Qinghai-Tibetan Plateau was a typical dimictic reservoir and the water mixed vertically at the end of March and the end of October. During the heating period, thermal stratification occurred, with strong diurnal variation in the epilimnion. The mean variance of the diurnal water temperature was 0.10 within a 5 m water depth but 0.04 in the whole water column. The mixing mode of inflow changed from undercurrent, horizontal-invaded flow and surface layer flow in one day. In winter, the diurnal variation was weak due to the thermal protection of the ice cover, while the mean variance of diurnal water temperature was 0.00 within both 5 m and the whole water column. Compared to reservoirs in areas with low altitude but the same latitude, significant differences occurred between the temperature structure of the low-altitude reservoir and the Pangduo Reservoir (P<0.01). The Pangduo Reservoir presented a shorter stratification period and weaker stratification stability, and the annual average SI value was 26.4 kg/m², which was only 7.5% that of the low-altitude reservoir. The seasonal changes in the net heat flux received by the surface layers determined the seasonal cycle of stratification and mixing in reservoirs. This study provided a scientific understanding of the thermal changes in stratified reservoirs under the special geographical and meteorological conditions on the Qinghai-Tibetan Plateau. Moreover, this model can serve as a reference for adaptive management of similar dimictic reservoirs in cold and high-altitude areas.