No abstract

Abstract. Model intercomparison studies are carried out to test and compare the simulated outputs of various model setups over the same study domain. The Great Lakes region is such a domain of high public interest as it not only resembles a challenging region to model with its transboundary location, strong lake effects, and regions of strong human impact but is also one of the most densely populated areas in the USA and Canada. This study brought together a wide range of researchers setting up their models of choice in a highly standardized experimental setup using the same geophysical datasets, forcings, common routing product, and locations of performance evaluation across the 1×106 km2 study domain. The study comprises 13 models covering a wide range of model types from machine-learning-based, basin-wise, subbasin-based, and gridded models that are either locally or globally calibrated or calibrated for one of each of the six predefined regions of the watershed. Unlike most hydrologically focused model intercomparisons, this study not only compares models regarding their capability to simulate streamflow (Q) but also evaluates the quality of simulated actual evapotranspiration (AET), surface soil moisture (SSM), and snow water equivalent (SWE). The latter three outputs are compared against gridded reference datasets. The comparisons are performed in two ways – either by aggregating model outputs and the reference to basin level or by regridding all model outputs to the reference grid and comparing the model simulations at each grid-cell. The main results of this study are as follows: The comparison of models regarding streamflow reveals the superior quality of the machine-learning-based model in the performance of all experiments; even for the most challenging spatiotemporal validation, the machine learning (ML) model outperforms any other physically based model. While the locally calibrated models lead to good performance in calibration and temporal validation (even outperforming several regionally calibrated models), they lose performance when they are transferred to locations that the model has not been calibrated on. This is likely to be improved with more advanced strategies to transfer these models in space. The regionally calibrated models – while losing less performance in spatial and spatiotemporal validation than locally calibrated models – exhibit low performances in highly regulated and urban areas and agricultural regions in the USA. Comparisons of additional model outputs (AET, SSM, and SWE) against gridded reference datasets show that aggregating model outputs and the reference dataset to the basin scale can lead to different conclusions than a comparison at the native grid scale. The latter is deemed preferable, especially for variables with large spatial variability such as SWE. A multi-objective-based analysis of the model performances across all variables (Q, AET, SSM, and SWE) reveals overall well-performing locally calibrated models (i.e., HYMOD2-lumped) and regionally calibrated models (i.e., MESH-SVS-Raven and GEM-Hydro-Watroute) due to varying reasons. The machine-learning-based model was not included here as it is not set up to simulate AET, SSM, and SWE. All basin-aggregated model outputs and observations for the model variables evaluated in this study are available on an interactive website that enables users to visualize results and download the data and model outputs.

Intensification of the global hydrological cycle and increase in precipitation for some regions around the world, including the northern mid-to high latitudes, is expected in a changing climate. Changes in the amount of seasonal precipitation and the intensity and frequency of extreme precipitation events directly affect the magnitude of seasonal streamflows and the timing and severity of floods and droughts. In this study, the Canadian Regional Climate Model (CRCM) projected changes to streamflow characteristics (i.e., hydrologic regime, mean annual streamflows, and the timing, frequency, and magnitude of extreme flows-low and high) over selected basins in western Canada and assessment of errors associated with these characteristics in the current climate are presented. An ensemble of five current (1961-90) and five future (2041-70) simulations, corresponding to the Special Report on Emissions Scenarios (SRES) A2 scenario, are used in the assessment of projected changes; the ensemble of simulations allows better quantification of uncertainty in projected changes. Results of the study suggest an increase in the magnitude of winter streamflows and an earlier snowmelt peak for the northern basins. In addition, study of selected return levels of extreme flows suggest important changes to the timing, frequency, and magnitude of both low and high flows, with significant increases in 10-yr 15-day winter and fall low flows and 1-day high flows, for all the high-latitude west Canadian basins. The level of confidence in projected changes to mean annual streamflows is relatively higher compared to that for extreme flows for most of the basins studied.

Abstract. This work explores the potential of the distributed GEM-Hydro runoff modeling platform, developed at Environment and Climate Change Canada (ECCC) over the last decade. More precisely, the aim is to develop a robust implementation methodology to perform reliable streamflow simulations with a distributed model over large and partly ungauged basins, in an efficient manner. The latest version of GEM-Hydro combines the SVS (Soil, Vegetation and Snow) land-surface scheme and the WATROUTE routing scheme. SVS has never been evaluated from a hydrological point of view, which is done here for all major rivers flowing into Lake Ontario. Two established hydrological models are confronted to GEM-Hydro, namely MESH and WATFLOOD, which share the same routing scheme (WATROUTE) but rely on different land-surface schemes. All models are calibrated using the same meteorological forcings, objective function, calibration algorithm, and basin delineation. GEM-Hydro is shown to be competitive with MESH and WATFLOOD: the NSE √ (Nash-Sutcliffe criterion computed on the square root of the flows) is for example equal to 0.83 for MESH and GEM-Hydro in validation on the Moira River basin, and to 0.68 for WATFLOOD. A computationally efficient strategy is proposed to calibrate SVS: a simple unit hydrograph is used for routing instead of WATROUTE. Global and local calibration strategies are compared in order to estimate runoff for ungauged portions of the Lake Ontario basin.Overall, streamflow predictions obtained using a global calibration strategy, in which a single parameter set is identified for the whole basin of Lake Ontario, show accuracy comparable to the predictions based on local calibration: the average NSE √ in validation and over seven subbasins is 0.73 and 0.61, respectively for local and global calibrations. Hence, global calibration provides spatially consistent parameter values, robust performance at gauged locations, and reduces the complexity and computation burden of the calibration procedure. This work contributes to the Great Lakes Runoff Inter-comparison Project for Lake Ontario (GRIP-O), which aims at improving Lake Ontario basin runoff simulations by comparing different models using the same input forcings. The main outcome of this study consists in a new generalizable methodology for implementing a distributed hydrologic model with a high computation cost in an efficient and reliable manner, over a large area with ungauged portions, using global calibration and a unit hydrograph to replace the routing component.

Between January 2013 and December 2014, water levels on Lake Superior and Lake MichiganHuron, the two largest lakes on Earth by surface area, rose at the highest rate ever recorded for a 2 year period beginning in January and ending in December of the following year. This historic event coincided with below-average air temperatures and extensive winter ice cover across the Great Lakes. It also brought an end to a 15 year period of persistently below-average water levels on Lakes Superior and MichiganHuron that included several months of record-low water levels. To differentiate hydrological drivers behind the recent water level rise, we developed a Bayesian Markov chain Monte Carlo (MCMC) routine for inferring historical estimates of the major components of each lake's water budget. Our results indicate that, in 2013, the water level rise on Lake Superior was driven by increased spring runoff and over-lake precipitation. In 2014, reduced over-lake evaporation played a more significant role in Lake Superior's water level rise. The water level rise on Lake Michigan-Huron in 2013 was also due to above-average spring runoff and persistent over-lake precipitation, while in 2014, it was due to a rare combination of below-average evaporation, above-average runoff and precipitation, and very high inflow rates from Lake Superior through the St. Marys River. We expect, in future research, to apply our new framework across the other Laurentian Great Lakes, and to Earth's other large freshwater basins as well.

[1] Computational budget is frequently a limiting factor in both uncertainty-based (e.g., through generalized likelihood uncertainty estimation (GLUE)) and optimization-based (e.g., through least squares minimization) calibration of computationally intensive environmental simulation models. This study introduces and formalizes the concept of simulation model preemption during automatic calibration. The proposed "model preemption" method terminates a simulation model early to save computational budget if it is recognized through intermediate simulation model results that a given solution (model parameter set) is so poor that it will not benefit the search strategy. The methodology proposed here is referred to as deterministic model preemption because it leads to exactly the same calibration result as when deterministic preemption is not applied. As such, deterministic preemption-enabled calibration algorithms which make no approximations to the mathematical simulation model are a simple alternative to the increasingly common and more complex approach of metamodeling for computationally constrained model calibration. Despite its simplicity, the deterministic model preemption concept is a promising concept that has yet to be formalized in the environmental simulation model automatic calibration literature. The model preemption concept can be applied to a subset of uncertainty-based and optimization-based automatic calibration strategies using a variety of different objective functions. Results across multiple calibration case studies demonstrate actual preemption computational savings ranging from 14% to 49%, 34% to 59%, and 52% to 96% for the dynamically dimensioned search, particle swarm optimization, and GLUE automatic calibration methods, respectively.

No abstract

No abstract