Despite progresses in representing different processes, hydrological models remain uncertain. Their uncertainty stems from input and calibration data, model structure, and parameters. In characterizing these sources, their causes, interactions and different uncertainty analysis (UA) methods are reviewed. The commonly used UA methods are categorized into six broad classes: (i) Monte Carlo analysis, (ii) Bayesian statistics, (iii) multi-objective analysis, (iv) least-squares-based inverse modeling, (v) response-surface-based techniques, and (vi) multi-modeling analysis. For each source of uncertainty, the status-quo and applications of these methods are critiqued in gauged catchments where UA is common and in ungauged catchments where both UA and its review are lacking. Compared to parameter uncertainty, UA application for structural uncertainty is limited while input and calibration data uncertainties are mostly unaccounted. Further research is needed to improve the computational efficiency of UA, disentangle and propagate the different sources of uncertainty, improve UA applications to environmental changes and coupled human–natural-hydrologic systems, and ease UA’s applications for practitioners.

Reservoir storage‐area‐depth relationships are the most important factors controlling thermal stratification in reservoirs and, more broadly, the water, energy, and biogeochemical dynamics in the reservoirs and subsequently their impacts on downstream rivers. However, most land surface or Earth system models do not account for the gradual changes of reservoir surface area and storage with the changing depth, inhibiting a consistent and accurate representation of mass, energy, and biogeochemical balances in reservoirs. Here we present a physically coherent parameterization of reservoir storage‐area‐depth data set at the global scale. For each reservoir, the storage‐area‐depth relationships were derived from an optimal geometric shape selected iteratively from five possible regular geometric shapes that minimize the error of total storage and surface area estimation. We applied this algorithm to over 6,800 reservoirs included in the Global Reservoir and Dam database. The relative error between the estimated and observed total storage is no more than 5% and 50% for 66% and 99% of all Global Reservoir and Dam reservoirs, respectively. More importantly, the storage‐depth profiles derived from the approximated reservoir geometry compared well with remote sensing based estimation at 40 major reservoirs from previous studies and ground‐truth measurements for 34 reservoirs in the United States and China. The new global reservoir storage‐area‐depth data set is critical for advancing future modeling and understanding of reservoir processes and subsequent effects on the terrestrial hydrological, ecological, and biogeochemical cycles at the regional and global scales.

Key Points: Modeling analysis of future hydrologic droughts over U.S. is provided Relative contributions from climate change and water managements are quantified Future hydrologic drought will be alleviated by reservoirs but intensified by local water extraction

Reservoir operations may alter the characteristics of Flood Frequency Curve (FFC) and challenge the basic assumption of stationarity used in flood frequency analysis. This paper presents a combined data‐modeling analysis of reservoir as a nonlinear filter of runoff routing that alters the FFCs. A dimensionless Reservoir Impact Index (RII), defined as the total upstream reservoir storage capacity normalized by the annual streamflow volume, is used to quantify reservoir regulation effects. Analyses are performed for 388 river stations in the contiguous U.S. using the first two moments of the FFC, mean annual maximum flood (MAF) and coefficient of variations (CV), calculated for the pre and post‐dam periods. It is found that MAF generally decreases with increasing RII but stabilizes when RII exceeds a threshold value, and CV increases with RII until a threshold value beyond which CV decreases with RII. Hence depending on the magnitude of RII, reservoir regulation acts as a filter to increase or reduce the nonlinearity of the natural runoff routing process and alters flood characteristics. The nonlinear relationships of MAF and CV with RII can be captured by three reservoir models with different levels of complexity, suggesting that they emerge from the basic flood control function of reservoirs. However, the threshold RII values in the nonlinear relationships depend on the more detailed reservoir operations and objectives that can only be captured by the more complex reservoir models. Our conceptual model may help improve flood‐risk assessment and mitigation in regulated river systems at the regional scale.

This study investigates the impacts of climate change and water management including agricultural irrigation, water withdrawal, and reservoir regulation on future meteorological, agricultural, and hydrological droughts and their connections. The analysis is based on the simulations from four global hydrological models forced with the projections from five global climate models for historical period 1971–2000 and future period 2070–2099 with and without water management. Three unified drought indices, the standardized precipitation index, standardized soil moisture index, and standardized streamflow index, are adopted to represent the meteorological, agricultural, and hydrological droughts, respectively. The analysis suggests that the climate‐induced drought changes in all three types of droughts are enhanced but in different directions, while water‐management‐induced changes in agricultural and hydrological droughts are more consistent across space and simulations. Overall, water management activities reduce both the duration and intensity of agricultural droughts by roughly 1 order of magnitude, while increase those of hydrological droughts by up to 50%. Basin‐scale analysis reveals that the higher is the intensity of irrigation, the larger will be the water‐management‐induced drought changes. Due to water management activities at some regions, the return periods of extreme agricultural droughts (in terms of severity) can change from 100 to 300 years or even longer, while typical 100‐year hydrological droughts are likely to occur more often for the regions located in 25°N–40°N and 15°S–50°S. This study provides a global view on drought modification in the Anthropocene, which will help improve adaptation strategies for future droughts.

The increasing intensity and frequency of extreme storms pose a growing challenge to stormwater management in highly urbanized areas. Without an adequate and appropriate stormwater system, the storms and associated floods will continue to cause significant damage to infrastructure and loss of life. Low Impact Development (LID) has become an emerging alternative to the traditional stormwater system for stormwater management. This study evaluates and optimizes applications of different combinations of LIDs to minimize flows from a catchment under past and future storm conditions. The Storm Water Management Model (SWMM), forced by observed and downscaled precipitation from Coupled Model Intercomparison Project phase 6 (CMIP6), was used to simulate the runoff and apply the LIDs in the Renton City, WA. The final results show that the performance of LIDs in reducing total runoff volume varies with the types and combinations of LIDs utilized. A 30% to 75% runoff reduction was achieved for the past and future 50 year and 100 year storms. The study demonstrates the effectiveness of LID combinations with conventional stormwater systems to manage the future runoff in the study area, which is expected to increase by 26.3% in 2050.