The hydrological projections provided by the outputs of Global Climate Models (GCMs) combining hydrological models include multi‐source uncertainties, which may challenge the formulation of relevant adaption and mitigation policies. In this paper, the overall uncertainty and the relative contribution of each uncertainty component were investigated for hydrological projections over 408 watersheds in China by using 3 shared socioeconomic pathway emission scenarios (SSP1‐2.6, SSP2‐4.5, and SSP5‐8.5), 21 GCMs, 8 bias correction methods, 4 hydrological models, and 2 sets of optimized hydrological model parameters. The results show that the total uncertainty (T) is mainly contributed by uncertainty related to global climate models (G), with the mean percentage ranging from 60.4 to 64.1%, followed by the interaction uncertainties among all components, with the mean percentage ranging from 22.0 to 26.4%. The uncertainty contribution of hydrological models (H) (6.1–9.4%) ranks third, followed by emission scenarios (S) (2.9–5.9%) and bias correction methods (B) (0.2–1.1%). The uncertainty contribution of the optimized hydrological model parameters (P) (0.2–0.3%) is almost negligible. In terms of spatial variability, the relative contribution of uncertainty related to global climate models (G) is the highest in the near future for northern China (67.5–70.6%) and in the far future for southern China (66.1–66.7%). However, it was found to be lower for the Tibetan Plateau and northwestern China (45.3–57.9%) in the near and far future. The relative contribution of hydrological model uncertainty is higher for southwestern and northwestern China and the Tibetan Plateau (7.2–19.5%) and lower for northern, eastern, and southern China (2.5–6.6%). This study highlights the importance of including multiple GCMs and hydrological models in hydrological impact studies to consider their overall uncertainty. The development of global climate models and hydrological models is still the best way to reduce the uncertainty of climate change impact studies.

Desert riparian vegetation forms an ecological corridor in extremely arid environments, and ecological water conveyance is an important measure of vegetation restoration and biodiversity conservation in desert riparian zones. Studying the responses of vegetation to ecological water conveyance and changes in this process in arid desert riparian zones and assessing the comprehensive benefits of ecological water conveyance are highly significant for ecological conservation and restoration in addition to the formulation of water transfer policies. Previous studies mainly used a single indicator to evaluate the ecological restoration of the mainstream Tarim River in Northwest China; thus, systematic and comprehensive assessments based on multiple indicators have not been conducted. In the present study, remote sensing data and field surveys were used to analyze the ecological restoration status of the Tarim River during 2015–2021 in terms of hydrological responses, vegetation responses, and ecological water conveyance benefits. The results showed that groundwater levels and soil moisture in the mainstream area of the Tarim River increased significantly from 2015 to 2021. The amount of groundwater storage also increased. Ecological water conveyance has created good hydrological conditions for groundwater recharge and ecological restoration on both sides of the mainstream area of the Tarim River. Desert forest ecosystems, mainly comprising Populus euphratica and Tamarix ramosissima, have been saved and rejuvenated in water conveyance areas. After ecological water conveyance, the Simpson and Shannon–Wiener indices increased significantly, but the diversity level began to decline and then stabilize with the increase in water conveyance frequency. The overall habitat status improved and the quality of the ecological environment below the Wusiman section of the middle reaches of the Tarim River improved significantly.

Incorporating 50 years of flood data for the Manas River Kenswat Hydrological Station from 1957 to 2006, the Pettitt test and Mann–Kendall trend test are used to analyse non‐stationarity of the flood characteristic sequences. Moreover, the Pearson type‐III (P‐III) distribution, the mixed distribution (MD) and conditional probability distribution (CPD) models are employed to analyse frequency and to calculate the design flood process line. The results showed that the annual maximum peak discharge and the annual maximum flood volume are most likely to change in 1993. The MD model considering the non‐stationarity of the flood sequence is more accurate than the CPD model and the traditional P‐III distribution model. There are significant differences in the design flood process lines of the 1996 typical flood process obtained by the three methods using the same frequency scaling method. In addition, under different design standards, the design value of the MD model is 20–53% smaller than the design value approved in 2008 (approved by China Renewable Energy Engineering Institute) and 4–48% higher than the traditional P‐III distribution design value. The results can provide a new reference for the management of non‐stationary floods in Manas River.

With the acceleration of human economic activities and dramatic changes in climate, the validity of the stationarity assumption of flood time series frequency analysis has been questioned. In this study, a framework for flood frequency analysis is developed on the basis of a tool, namely, the Generalized Additive Models for Location, Scale, and Shape (GAMLSS). We introduced this model to construct a non-stationary model with time and climate factor as covariates for the 50-year snowmelt flood time series in the Kenswat Reservoir control basin of the Manas River. The study shows that there are clear non-stationarities in the flood regime, and the characteristic series of snowmelt flood shows an increasing trend with the passing of time. The parameters of the flood distributions are modelled as functions of climate indices (temperature and rainfall). The physical mechanism was incorporated into the study, and the simulation results are similar to the actual flood conditions, which can better describe the dynamic process of snowmelt flood characteristic series. Compared with the design flood results of Kenswat Reservoir approved by the China Renewable Energy Engineering Institute in December 2008, the design value of the GAMLSS non-stationary model considers that the impact of climate factors create a design risk in dry years by underestimating the risk.

Bias correction and spatial disaggregation (BCSD) is widely used in coupling general circulation models (GCMs) and hydrological models. However, there are some disadvantages in BCSD, such as only one GCM being selected, correcting biases through quantile-mapping (QM), and downscaling through interpolation. Then a combined approach of canonical correlation analysis filtering, multi-model ensemble, and extreme learning machine (ELM) regressions (CEE) was advanced. The performance of CEE and BCSD was evaluated with Manas River Basin as a study area. Results show it is unreasonable to correct biases through QM as it implies that the climate remains unchanged. Multi-model ensemble provides additional information, which is beneficial for regressions. CEE performs better than BCSD in temperature and precipitation rate downscaling. In CEE, the residual in temperature forecasting can be lower than 0.05 times temperature range and that in precipitation rate can be 0.33 times precipitation rate range. The performance of CEE in temperature downscaling in plains is better than mountainous areas, but for precipitation rate downscaling, it is better in mountainous areas. Increasing rate of temperature in the basin is 0.0254 K/decade, 0.1837 K/decade, and 0.5039 K/decade, and that of precipitation rate is 0.0028 mm/(day × decade), 0.0036 mm/(day × decade), and 0.0022 mm/(day × decade) in RCP2.6, RCP4.5, and RCP8.5, respectively.