Parameter transferability between multiple gridded input datasets challenges hydrological model performance under changing climate

Mapping Intimacies ◽

10.5194/egusphere-egu21-8782 ◽

2021 ◽

Author(s):

Moctar Dembélé ◽

Bettina Schaefli ◽

Grégoire Mariéthoz

Keyword(s):

West Africa ◽

Large Scale ◽

Hydrological Model ◽

Meteorological Data ◽

Model Performance ◽

Hydrologic Model ◽

Hydrological Modelling ◽

Climate Change Scenarios ◽

Depth Analysis ◽

Input Dataset

The diversity of remotely sensed or reanalysis-based rainfall data steadily increases, which on one hand opens new perspectives for large scale hydrological modelling in data scarce regions, but on the other hand poses challenging question regarding parameter identification and transferability under multiple input datasets. This study analyzes the variability of hydrological model performance when (1) a set of parameters is transferred from the calibration input dataset to a different meteorological datasets and reversely, when (2) an input dataset is used with a parameter set, originally calibrated for a different input dataset.The research objective is to highlight the uncertainties related to input data and the limitations of hydrological model parameter transferability across input datasets. An ensemble of 17 rainfall datasets and 6 temperature datasets from satellite and reanalysis sources (Demb&#233;l&#233; et al., 2020), corresponding to 102 combinations of meteorological data, is used to force the fully distributed mesoscale Hydrologic Model (mHM). The mHM model is calibrated for each combination of meteorological datasets, thereby resulting in 102 calibrated parameter sets, which almost all give similar model performance. Each of the 102 parameter sets is used to run the mHM model with each of the 102 input datasets, yielding 10404 scenarios to that serve for the transferability tests. The experiment is carried out for a decade from 2003 to 2012 in the large and data-scarce Volta River basin (415600 km2) in West Africa.The results show that there is a high variability in model performance for streamflow (mean CV=105%) when the parameters are transferred from the original input dataset to other input datasets (test 1 above). Moreover, the model performance is in general lower and can drop considerably when parameters obtained under all other input datasets are transferred to a selected input dataset (test 2 above). This underlines the need for model performance evaluation when different input datasets and parameter sets than those used during calibration are used to run a model. Our results represent a first step to tackle the question of parameter transferability to climate change scenarios. An in-depth analysis of the results at a later stage will shed light on which model parameterizations might be the main source of performance variability.Demb&#233;l&#233;, M., Schaefli, B., van de Giesen, N., & Mari&#233;thoz, G. (2020). Suitability of 17 rainfall and temperature gridded datasets for large-scale hydrological modelling in West Africa. Hydrology and Earth System Sciences (HESS). https://doi.org/10.5194/hess-24-5379-2020

On the selection of precipitation products for the regionalisation of hydrological model parameters

Hydrology and Earth System Sciences ◽

10.5194/hess-25-5805-2021 ◽

2021 ◽

Vol 25 (11) ◽

pp. 5805-5837

Author(s):

Oscar M. Baez-Villanueva ◽

Mauricio Zambrano-Bigiarini ◽

Pablo A. Mendoza ◽

Ian McNamara ◽

Hylke E. Beck ◽

...

Keyword(s):

Hydrological Model ◽

Model Performance ◽

Temporal Distribution ◽

Hydrological Regime ◽

Hydrologic Model ◽

Spatial Proximity ◽

Model Parameters ◽

Ungauged Catchments ◽

Water Balance Components ◽

Feature Similarity

Abstract. Over the past decades, novel parameter regionalisation techniques have been developed to predict streamflow in data-scarce regions. In this paper, we examined how the choice of gridded daily precipitation (P) products affects the relative performance of three well-known parameter regionalisation techniques (spatial proximity, feature similarity, and parameter regression) over 100 near-natural catchments with diverse hydrological regimes across Chile. We set up and calibrated a conceptual semi-distributed HBV-like hydrological model (TUWmodel) for each catchment, using four P products (CR2MET, RF-MEP, ERA5, and MSWEPv2.8). We assessed the ability of these regionalisation techniques to transfer the parameters of a rainfall-runoff model, implementing a leave-one-out cross-validation procedure for each P product. Despite differences in the spatio-temporal distribution of P, all products provided good performance during calibration (median Kling–Gupta efficiencies (KGE′s) > 0.77), two independent verification periods (median KGE′s >0.70 and 0.61, for near-normal and dry conditions, respectively), and regionalisation (median KGE′s for the best method ranging from 0.56 to 0.63). We show how model calibration is able to compensate, to some extent, differences between P forcings by adjusting model parameters and thus the water balance components. Overall, feature similarity provided the best results, followed by spatial proximity, while parameter regression resulted in the worst performance, reinforcing the importance of transferring complete model parameter sets to ungauged catchments. Our results suggest that (i) merging P products and ground-based measurements does not necessarily translate into an improved hydrologic model performance; (ii) the spatial resolution of P products does not substantially affect the regionalisation performance; (iii) a P product that provides the best individual model performance during calibration and verification does not necessarily yield the best performance in terms of parameter regionalisation; and (iv) the model parameters and the performance of regionalisation methods are affected by the hydrological regime, with the best results for spatial proximity and feature similarity obtained for rain-dominated catchments with a minor snowmelt component.

Small-scale processes with large-scale impacts: Investigating canopy structure controls on energy fluxes to the forest snowpack

10.5194/egusphere-egu2020-19027 ◽

2020 ◽

Author(s):

Giulia Mazzotti ◽

Richard Essery ◽

Johanna Malle ◽

Clare Webster ◽

Tobias Jonas

Keyword(s):

High Resolution ◽

Land Surface ◽

Large Scale ◽

Meteorological Data ◽

Canopy Structure ◽

Model Performance ◽

Turbulent Fluxes ◽

Temporal Variations ◽

Small Scale ◽

Forest Stands

Forest canopies strongly affect snowpack energetics during wintertime. In discontinuous forest stands, spatio-temporal variations in radiative and turbulent fluxes create complex snow distribution and melt patterns, with further impacts on the hydrological regimes and on the land surface properties of seasonally snow-covered forested environments.As increasingly detailed canopy structure datasets are becoming available, canopy-induced energy exchange processes can be explicitly represented in high-resolution snow models. We applied the modelling framework FSM2 to obtain spatially distributed simulations of the forest snowpack in subalpine and boreal forest stands at high spatial (2m) and temporal (10min) resolution. Modelled sub-canopy radiative and turbulent fluxes were compared to detailed meteorological data of incoming irradiances, air and snow surface temperatures. These were acquired with novel observational systems, including 1) a motorized cable car setup recording spatially and temporally resolved data along a transect and 2) a handheld setup designed to capture temporal snapshots of 2D spatial distributions across forest discontinuities.The combination of high-resolution modelling and multi-dimensional datasets allowed us to assess model performance at the level of individual energy balance components, under various meteorological conditions and across canopy density gradients. We showed which canopy representation strategies within FSM2 best succeeded in reproducing snowpack energy transfer dynamics in discontinuous forests, and derived implications for implementing forest snow processes in coarser-resolution models.

EMO-5: Copernicus pan-European high-resolution meteorological data set for large-scale hydrological modelling

10.5194/egusphere-egu2020-21551 ◽

2020 ◽

Author(s):

Vera Thiemig ◽

Peter Salamon ◽

Goncalo N. Gomes ◽

Jon O. Skøien ◽

Markus Ziese ◽

...

Keyword(s):

High Resolution ◽

Real Time ◽

Large Scale ◽

Interpolation Method ◽

Meteorological Data ◽

Hydrological Modelling ◽

Data Set ◽

Average Temperature ◽

Time Period ◽

Meteorological Observations

We present EMO-5, a Pan-European high-resolution (5 km), (sub-)daily, multi-variable meteorological data set especially developed to the needs of an operational, pan-European hydrological service (EFAS; European Flood Awareness System). The data set is built on historic and real-time observations coming from 18,964 meteorological in-situ stations, collected from 24 data providers, and 10,632 virtual stations from four high-resolution regional observational grids (CombiPrecip, ZAMG - INCA, EURO4M-APGD and CarpatClim) as well as one global reanalysis product (ERA-Interim-land). This multi-variable data set covers precipitation, temperature (average, min and max), wind speed, solar radiation and vapor pressure; all at daily resolution and in addition 6-hourly resolution for precipitation and average temperature. The original observations were thoroughly quality controlled before we used the Spheremap interpolation method to estimate the variable values for each of the 5 x 5 km grid cells and their affiliated uncertainty. EMO-5 v1 grids covering the time period from 1990 till 2019 will be released as a free and open Copernicus product mid-2020 (with a near real-time release of the latest gridded observations in future). We would like to present the great potential EMO-5 holds for the hydrological modelling community.&#160;footnote: EMO = European Meteorological Observations

Observational Uncertainty in Hydrological Modelling using Data Depth

Global NEST Journal ◽

10.30955/gnj.002354 ◽

2017 ◽

Vol 19 (3) ◽

pp. 489-497

Keyword(s):

Hydrological Model ◽

Meteorological Data ◽

River Basin Management ◽

Hydrological Modelling ◽

Data Depth ◽

Basin Management ◽

Flow Generation ◽

Observational Uncertainty ◽

Using Data ◽

Quantifying Uncertainty

For any river basin management, one needs tools to predict runoff at different time and spatial resolutions. Hydrological models are tools which account for the storage, flow of water and water balance in a watershed, which include exchanges of water and energy within the earth, atmosphere and oceans and utilise metrological data to generate flow. There are several sources of error in meteorological data, namely, through measurement at point level, interpolation, etc. When an erroneous input is passed to a model, one cannot expect an error free output from the prediction. Every prediction is associated with uncertainty. Quantification of these uncertainties is of prime importance in real world forecasting. In this study, an attempt has been made to study uncertainty associated with hydrological modelling, using the idea of data depth. To see the effect of uncertainty in rainfall on flow generation through a model, the input to a model was altered by adding an error and a different realisation was made. A Monte Carlo simulation generated a large number of hydrological model parameter sets drawn from the uniform distribution. The model was run using these parameters for each realisation of the rainfall. The parameters which are good for different realisations are more likely to be good parameters sets. For each parameter set, data depth was calculated and a likelihood was assigned to each parameter set based on the depth values. Based on this, the frequency distribution of the likelihood was analysed as well. The results show that uncertainty in hydrological modelling are multiplicative. The proposed methodology to assign prediction uncertainty is demonstrated using the ‘TopNet’ model for the Waipara river catchment located in the central east of the South Island, New Zealand. The results of this study will be helpful in calibration of hydrological model and in quantifying uncertainty in the prediction.

Underestimation of temperature variability in weather generators and implications for the representation of extreme temperatures in downscaled climate change scenarios

10.5194/egusphere-egu2020-4572 ◽

2020 ◽

Author(s):

Pierluigi Calanca

Keyword(s):

Climate Change ◽

Large Scale ◽

Extreme Values ◽

Climatic Variability ◽

Model Performance ◽

Maximum Temperature ◽

Climate Change Scenarios ◽

Modes Of Variability ◽

Weather Generators ◽

Total Variability

Stochastic weather generators are still widely used for downscaling climate change scenarios, in particular in the context of agricultural and hydrological impact assessments. Their performance is in many respects satisfactory, except perhaps for the fact that they fail to represent climatic variability in an adequate way. This has implications for the representation of extreme values and their statistics. Concerning precipitation, different approaches for amending this situation have proposed in the past, including using more sophisticated models to better simulate the persistence of wet and dry spells, conditioning rainfall-generating parameters on indices of the large-scale atmospheric circulation, or employing autoregressive models to represent year-to-year variations in annual precipitation amounts. With regard to (minimum and maximum) temperature, efforts to address the question of why weather generators underestimate total variability have been less systematic. Based on results obtained with a well-known weather generator (LARS-WG), this contribution aims to discuss which modes of variability are missing and why, elaborate on the implications of underrepresenting temperature variance for the simulation of temperature extremes in downscaled climate change scenarios, and suggest options to tackle the problem and improve the model performance.

Using altimetry observations combined with GRACE to select parameter sets of a hydrological model in a data-scarce region

Hydrology and Earth System Sciences ◽

10.5194/hess-24-3331-2020 ◽

2020 ◽

Vol 24 (6) ◽

pp. 3331-3359 ◽

Cited By ~ 1

Author(s):

Petra Hulsman ◽

Hessel C. Winsemius ◽

Claire I. Michailovsky ◽

Hubert H. G. Savenije ◽

Markus Hrachowitz

Keyword(s):

Cross Section ◽

Hydrological Model ◽

Model Performance ◽

Hydrological Modelling ◽

Google Earth ◽

Water Levels ◽

Discharge Data ◽

Model Parameter ◽

Virtual Station ◽

River Cross Section

Abstract. Limited availability of ground measurements in the vast majority of river basins world-wide increases the value of alternative data sources such as satellite observations in hydrological modelling. This study investigates the potential of using remotely sensed river water levels, i.e. altimetry observations, from multiple satellite missions to identify parameter sets for a hydrological model in the semi-arid Luangwa River basin in Zambia. A distributed process-based rainfall–runoff model with sub-grid process heterogeneity was developed and run on a daily timescale for the time period 2002 to 2016. As a benchmark, feasible model parameter sets were identified using traditional model calibration with observed river discharge data. For the parameter identification using remote sensing, data from the Gravity Recovery and Climate Experiment (GRACE) were used in a first step to restrict the feasible parameter sets based on the seasonal fluctuations in total water storage. Next, three alternative ways of further restricting feasible model parameter sets using satellite altimetry time series from 18 different locations along the river were compared. In the calibrated benchmark case, daily river flows were reproduced relatively well with an optimum Nash–Sutcliffe efficiency of ENS,Q=0.78 (5/95th percentiles of all feasible solutions ENS,Q,5/95=0.61–0.75). When using only GRACE observations to restrict the parameter space, assuming no discharge observations are available, an optimum of ENS,Q=-1.4 (ENS,Q,5/95=-2.3–0.38) with respect to discharge was obtained. The direct use of altimetry-based river levels frequently led to overestimated flows and poorly identified feasible parameter sets (ENS,Q,5/95=-2.9–0.10). Similarly, converting modelled discharge into water levels using rating curves in the form of power relationships with two additional free calibration parameters per virtual station resulted in an overestimation of the discharge and poorly identified feasible parameter sets (ENS,Q,5/95=-2.6–0.25). However, accounting for river geometry proved to be highly effective. This included using river cross-section and gradient information extracted from global high-resolution terrain data available on Google Earth and applying the Strickler–Manning equation to convert modelled discharge into water levels. Many parameter sets identified with this method reproduced the hydrograph and multiple other signatures of discharge reasonably well, with an optimum of ENS,Q=0.60 (ENS,Q,5/95=-0.31–0.50). It was further shown that more accurate river cross-section data improved the water-level simulations, modelled rating curve, and discharge simulations during intermediate and low flows at the basin outlet where detailed on-site cross-section information was available. Also, increasing the number of virtual stations used for parameter selection in the calibration period considerably improved the model performance in a spatial split-sample validation. The results provide robust evidence that in the absence of directly observed discharge data for larger rivers in data-scarce regions, altimetry data from multiple virtual stations combined with GRACE observations have the potential to fill this gap when combined with readily available estimates of river geometry, thereby allowing a step towards more reliable hydrological modelling in poorly gauged or ungauged basins.

Large-scale distributed watershed modelling for reservoir operations in cold boreal regions

Canadian Journal of Civil Engineering ◽

10.1139/l06-142 ◽

2007 ◽

Vol 34 (4) ◽

pp. 525-538 ◽

Cited By ~ 4

Author(s):

M E St. Laurent ◽

C Valeo

Keyword(s):

Large Scale ◽

Model Performance ◽

Hydrologic Model ◽

Frozen Ground ◽

Deficit Model ◽

River Watershed ◽

Spring Freshet ◽

Improved Model ◽

Taylor River ◽

Single Constant

The macroscale deterministic hydrologic model, SLURP, was modified and tested on two large watersheds in northern Manitoba, the Taylor River watershed (899 km2) and the upper Burntwood River watershed (6959 km2). Calibration and validation of the model on both watersheds between 1985 and 2000 identified a number of model deficiencies and recommendations for improvement. Date-dependent snowmelt rates were replaced with a single constant snowmelt rate, helping to decrease the parameterization of the model. A snowpack temperature deficit model was also incorporated to simulate the effects of snow ripening. These two modifications provided the modelling flexibility needed to control the timing of initial snowmelt and the rate of snowmelt. Annual spring freshet lasts roughly 2 weeks in this region; however, improved model performance was observed well beyond the spring freshet period. These modifications also provided a better representation of the physical processes that delay snowmelt once the air temperature exceeds 0 °C.Key words: frozen ground, boreal forest, hydrological modelling, snow ripening, snowmelt.

Hourly hydrologic modeling in the upper basin of the Fuerte River, Sinaloa, Mexico

10.5154/r.inagbi.2020.11.085 ◽

2021 ◽

Vol 13 (1) ◽

Author(s):

Edgar Merino-Jiménez ◽

◽

Laura Alicia Ibáñez-Castillo ◽

Mario Alberto Vázquez-Peña ◽

◽

...

Keyword(s):

Hydrological Model ◽

Extreme Rainfall ◽

Curve Number ◽

Hydrologic Model ◽

Climate Change Scenarios ◽

Hydrological Models ◽

Unit Hydrograph ◽

Extreme Rainfall Events ◽

Rainfall Events ◽

Hydrometric Station

Introduction: Mexico uses hydrological models to determine floods, evaluate land use change scenarios, evaluate climate change scenarios, and define federal zones, among other applications. However, the models are rarely calibrated beforehand, which increases uncertainty in the design of structures and hydraulic standards. Objective: To build a hydrological model for the watershed of the Fuerte River, Mexico, of extreme rainfall events occurred in 2009, 2011, 2015, 2016 and 2017. Methodology: Five extreme rainfall events were considered for this study. The hydrologic model was design using the HEC-HMS program, and calibrated at the Tubares hydrometric station. The runoff curve number methodology and the Clark unit hydrograph were used. Results: The results collected in four of the five events were positive; the Nash-Sutcliffe efficiency (NSE) ranged between 0.22 and 0.52. The temporal behavior of the river, at times when it moved away from the flow value, preserved the variation trend. Limitations of the study: The study reaches the Tubares hydrometric station, Chihuahua, without including the downstream dams in Sinaloa. Originality: There are few hydrological studies that generate a calibrated and, therefore, reliable hourly model. Conclusions: The hourly hydrologic model had an acceptable performance in four of the five predicted events in terms of NSE and root mean square error (RMSE).

Distributed hydrological modeling in a large-scale watershed of Northern China: multi-site model calibration, validation, and sensitivity analysis

Hydrology and Earth System Sciences Discussions ◽

10.5194/hessd-9-5697-2012 ◽

2012 ◽

Vol 9 (5) ◽

pp. 5697-5727 ◽

Cited By ~ 3

Author(s):

S. Wang ◽

Z. Zhang ◽

G. Sun ◽

P. Strauss ◽

J. Guo ◽

...

Keyword(s):

Model Calibration ◽

Large Scale ◽

Hydrological Modeling ◽

Model Performance ◽

Northern China ◽

Calibration Method ◽

Hydrologic Model ◽

Hydrological Process ◽

Prediction Errors ◽

Calibration Protocol

Abstract. Model calibration is a complex task for large watersheds, especially for those in a heterogeneous mountain environment where multi-objective calibration strategy is essential. That may improve a model's capability to capture the spatial variations of the internal hydrologic variables. This study used the physically-based distributed hydrologic model, MIKESHE, to contrast a lumped calibration protocol that uses data measured at one single outlet to a multi-site calibration method which employed streamflow measurements at three separate stations within the large Chaohe River basin in Northern China. The results showed that, the single-site calibrated model was able to sufficiently simulate the hydrographs for two of the three stations (Nash-sutchliffe coefficient of 0.65–0.75, and correlation coefficient 0.81–0.87 during the testing period), but model performance was poor at the third station (EF only 0.44). By using the multi-site measurements model calibration reached a compromise between the different stations, the model reasonably representing the hydrographs of all three stations with EF ranging from 0.59–0.72. The modeling calibration results suggested that the dominant hydrological processes varied across the large watershed with upstream area dominated by slow groundwater and middle- and down-stream areas dominated by relatively quick interflow. We conclude that to account for the different hydrological process of watershed with large heterogeneity, it is necessary to employ a multi-site calibration protocol to reduce prediction errors.

Multivariate and spatially calibrated hydrological model for assessing climate change impacts on hydrological processes in West Africa

10.5194/egusphere-egu2020-9143 ◽

2020 ◽

Author(s):

Moctar Dembélé ◽

Sander Zwart ◽

Natalie Ceperley ◽

Grégoire Mariéthoz ◽

Bettina Schaefli

Keyword(s):

Climate Change ◽

Soil Moisture ◽

West Africa ◽

Spatial Patterns ◽

Hydrological Model ◽

Climate Change Impacts ◽

Hydrologic Model ◽

Data Availability ◽

Hydrological Processes ◽

Climate Vulnerability

Robust hydrological models are critical for the assessment of climate change impacts on hydrological processes. This study analysis the future evolution of the spatiotemporal dynamics of multiple hydrological processes (i.e. streamflow, soil moisture, evaporation and terrestrial water storage) with the fully distributed mesoscale hydrologic Model (mHM), which is constrained with a novel multivariate calibration approach based on the spatial patterns of satellite remote sensing data (Demb&#233;l&#233; et al., 2020). The experiment is done in the large and transboundary Volta River Basin (VRB) in West Africa, which is a hotspot of climate vulnerability. Climate change and land use changes lead to recurrent floods and drought that impact agriculture and affect the lives of the inhabitants.Based on data availability on the Earth System Grid Federation (ESGF) platform, nine Global Circulation Models (i.e. CanESM2, CNRM-CM5, CSIRO-Mk3-6-0, GFDL-ESM2M, HadGEM2-ES, IPSL-CM5A-MR, MIROC5, MPI-ESM-LR and NorESM1-M) available from the CORDEX-Africa initiative and dynamically downscaled with the latest version of the Rossby Centre's regional atmospheric model (RCA4) are selected for this study. Daily datasets of meteorological variables (i.e. precipitation and air temperature) for the medium and high emission scenarios (RCP4.5 and RCP8.5) are bias-corrected and used to force the mHM model for the reference period 1991-2020, and the near- and long-term future periods 2021-2050 and 2051-2080.The results show contrasting trends among the hydrological processes as well as among the GCMs. The findings reveal uncertainties in the spatial patterns of hydrological processes (e.g. soil moisture and evaporation), which ultimately have implications for flood and drought predictions. This study highlights the importance of plausible spatial patterns for the assessment of climate change impacts on hydrological processes, and thereby provide valuable information with the potential to reduce the climate vulnerability of the local population.&#160;ReferenceDemb&#233;l&#233;, M., Hrachowitz, M., Savenije, H., Mari&#233;thoz, G., & Schaefli, B. (2020). Improving the predictive skill of a distributed hydrological model by calibration on spatial patterns with multiple satellite datasets. Water Resources Research.