scholarly journals Technical Note – RAT: a Robustness Assessment Test for calibrated and uncalibrated hydrological models

2021 ◽  
Author(s):  
Pierre Nicolle ◽  
Vazken Andréassian ◽  
Paul Royer-Gaspard ◽  
Charles Perrin ◽  
Guillaume Thirel ◽  
...  
Keyword(s):  
Climate Change ◽  
Climate Change Impact ◽  
Hydrological Model ◽  
Technical Note ◽  
New Method ◽  
Hydrological Models ◽  
Split Sample ◽  
Sample Test ◽  
Change Impact ◽  
Robustness Assessment

Abstract. In this note, we present RAT, a new method to assess the robustness of hydrological models. RAT can be seen as an alternative to the classical split-sample test widely used in hydrology. And because the RAT method does not require multiple calibrations, we suggest that it can be applied even to uncalibrated models. The RAT method can be used to determine whether a hydrological model is "safe" for being used for climate change impact studies.

RAT (Robustness Assessment Test): a straightforward evaluation of hydrological model robustness to a changing climate

2021 ◽  
Author(s):  
Vazken Andréassian ◽  
Léonard Santos ◽  
Torben Sonnenborg ◽  
Alban de Lavenne ◽  
Göran Lindström ◽  
...  
Keyword(s):  
Climate Change ◽  
Hydrological Model ◽  
Hydrological Models ◽  
Climate Variables ◽  
Data Set ◽  
Hydrological Characteristics ◽  
Changing Climate ◽  
Split Sample ◽  
Sample Test ◽  
Model Robustness

<p>Hydrological models are increasingly used under evolving climatic conditions. They should thus be evaluated regarding their temporal transferability (application in different time periods) and extrapolation capacity (application beyond the range of known past conditions). In theory, parameters of hydrological models are independent of climate. In practice, however, many published studies based on the Split-Sample Test (Klemeš, 1986), have shown that model performances decrease systematically when it is used out of its calibration period. The RAT test proposed here aims at evaluating model robustness to a changing climate by assessing potential undesirable dependencies of hydrological model performances to climate variables. The test compares, over a long data period, the annual value of several climate variables (temperature, precipitation and aridity index) and the bias of the model over each year. If a significant relation exists between the climatic variable and the bias, the model is not considered to be robust to climate change on the catchment. The test has been compared to the Generalized Split-Sample Test (Coron et al., 2012) and showed similar results.</p><p>Here, we report on a large scale application of the test for three hydrological models with different level of complexity (GR6J, HYPE, MIKE-SHE) on a data set of 352 catchments in Denmark, France and Sweden. The results show that the test behaves differently given the evaluated variable (be temperature, precipitation or aridity) and the hydrological characteristics of each catchment. They also show that, although of different level of complexity, the robustness of the three models is similar on the overall data set. However, they are not robust on the same catchments and, then, are not sensitive to the same hydrological characteristics. This example highlights the applicability of the RAT test regardless of the model set-up and calibration procedure and its ability to provide a first evaluation of the model robustness to climate change.</p><p> </p><p><strong>References</strong></p><p>Coron, L., V. Andréassian, C. Perrin, J. Lerat, J. Vaze, M. Bourqui, and F. Hendrickx, 2012. Crash testing hydrological models in contrasted climate conditions: An experiment on 216 Australian catchments, Water Resour. Res., 48, W05552, doi:10.1029/2011WR011721</p><p>Klemeš, V., 1986. Operational testing of hydrological simulation models, Hydrol. Sci. J., 31, 13–24, doi:10.1080/02626668609491024</p><p> </p>

scholarly journals Implication of calibration period on modelling climate change impact on future runoff

2015 ◽  
Vol 371 ◽  
pp. 3-6 ◽  
Author(s):  
F. H. S. Chiew ◽  
H. Zheng ◽  
J. Vaze
Keyword(s):  
Climate Change ◽  
Climate Change Impact ◽  
Hydrological Model ◽  
Historical Data ◽  
Hydrological Models ◽  
Calibration Period ◽  
Calibration Dataset ◽  
Change Impact ◽  
Hydrologic Responses ◽  
Data Period

Abstract. This paper explores the consideration and implication of calibration period on the modelled climate change impact on future runoff. The results show that modelled runoff and hydrologic responses can be influenced by the choice of historical data period used to calibrate and develop the hydrological model. Modelling approaches that do not take this into account may therefore underestimate the range and uncertainty in future runoff projections. Nevertheless, the uncertainty associated with the choice of hydrological models and consideration of calibration dataset for modelling climate change impact on runoff is likely to be small compared to the uncertainty in the future rainfall projections.

scholarly journals Is bias correction of Regional Climate Model (RCM) simulations possible for non-stationary conditions?

2012 ◽  
Vol 9 (11) ◽  
pp. 12765-12795 ◽  
Author(s):  
C. Teutschbein ◽  
J. Seibert
Keyword(s):  
Climate Change ◽  
Bias Correction ◽  
Climate Change Impact ◽  
Climate Model ◽  
Regional Climate ◽  
Climate Conditions ◽  
Impact Studies ◽  
Split Sample ◽  
Sample Test ◽  
Change Impact

Abstract. In hydrological climate-change impact studies, Regional Climate Models (RCMs) are commonly used to transfer large-scale Global Climate Model (GCM) data to smaller scales and to provide more detailed regional information. However, there are often considerable biases in RCM simulations, which have led to the development of a number of bias correction approaches to provide more realistic climate simulations for impact studies. Bias correction procedures rely on the assumption that RCM biases do not change over time, because correction algorithms and their parameterizations are derived for current climate conditions and assumed to apply also for future climate conditions. This underlying assumption of bias stationarity is the main concern when using bias correction procedures. It is in principle not possible to test whether this assumption is actually fulfilled for future climate conditions. In this study, however, we demonstrate that it is possible to evaluate how well bias correction methods perform for conditions different from those used for calibration. For five Swedish catchments, several time series of RCM simulated precipitation and temperature were obtained from the ENSEMBLES data base and different commonly-used bias correction methods were applied. We then performed a differential split-sample test by dividing the data series into cold and warm respective dry and wet years. This enabled us to evaluate the performance of different bias correction procedures under systematically varying climate conditions. The differential split-sample test resulted in a large spread and a clear bias for some of the correction methods during validation years. More advanced correction methods such as distribution mapping performed relatively well even in the validation period, whereas simpler approaches resulted in the largest deviations and least reliable corrections for changed conditions. Therefore, we question the use of simple bias correction methods such as the widely used delta-change approach and linear scaling for RCM-based climate-change impact studies and recommend using higher-skill bias correction methods.

scholarly journals Uncertainty assessment of future projections on water resources according to climate downscaling and hydrological models

2018 ◽  
Vol 20 (3) ◽  
pp. 597-607 ◽  
Author(s):  
Moon-Hwan Lee ◽  
Deg-Hyo Bae
Keyword(s):  
Climate Change ◽  
Climate Change Impact ◽  
Hydrological Model ◽  
Low Flow ◽  
Hydrological Models ◽  
Future Projections ◽  
Dominant Source ◽  
Change Impact ◽  
Dam Inflow ◽  
Selection Of

Abstract Quantifying the uncertainty of future projection is important to assess the reliable climate change impact. In this sense, this study is aimed at investigating the uncertainty sources of various water variables (seasonal dam inflow, 1-day maximum dam inflow, and 30-day minimum dam inflow) according to downscaling methods and hydrological modeling. Five regional climate models (RCMs), five statistical post-processing methods and two hydrological models were applied for the uncertainty analysis. The changes for seasonal dam inflow are 0.1, 58.8, 5.1, and 1.1 mm for the SWAT model and 2.1, 76.1, −8.5, and −2.9 mm for the VIC model in spring, summer, autumn, and winter, respectively. The effects of the hydrological model is smaller than that of RCM for future projections of the seasonal dam inflow. The changes of annual 1-day maximum dam inflow vary according to the selection of RCM whereas the changes of annual 30-day minimum dam inflow are sensitive to the selection of hydrological model. The RCM is the dominant source of uncertainty of all seasonal dam inflow (except for winter) and high flow, whereas the hydrological model is the dominant source of uncertainty in winter dam inflow and low flow. Considering these results, the appropriate multi-model ensemble chain according to target variable will be necessary for reliable climate change impact assessment.

scholarly journals Technical note: RAT – a robustness assessment test for calibrated and uncalibrated hydrological models

2021 ◽  
Vol 25 (9) ◽  
pp. 5013-5027
Author(s):  
Pierre Nicolle ◽  
Vazken Andréassian ◽  
Paul Royer-Gaspard ◽  
Charles Perrin ◽  
Guillaume Thirel ◽  
...  
Keyword(s):  
Technical Note ◽  
Evaluation Framework ◽  
Hydrological Models ◽  
Climate Conditions ◽  
Evaluation Approach ◽  
Physically Based ◽  
Sample Test ◽  
Change Impact ◽  
Robustness Assessment ◽  
Past Conditions

Abstract. Prior to their use under future changing climate conditions, all hydrological models should be thoroughly evaluated regarding their temporal transferability (application in different time periods) and extrapolation capacity (application beyond the range of known past conditions). This note presents a straightforward evaluation framework aimed at detecting potential undesirable climate dependencies in hydrological models: the robustness assessment test (RAT). Although it is conceptually inspired by the classic differential split-sample test of Klemeš (1986), the RAT presents the advantage of being applicable to all types of models, be they calibrated or not (i.e. regionalized or physically based). In this note, we present the RAT, illustrate its application on a set of 21 catchments, verify its applicability hypotheses and compare it to previously published tests. Results show that the RAT is an efficient evaluation approach, passing it successfully can be considered a prerequisite for any hydrological model to be used for climate change impact studies.

scholarly journals Comparing global and local calibration schemes from a differential split-sample test perspective

2015 ◽  
Vol 52 (11) ◽  
pp. 990-999 ◽  
Author(s):  
Étienne Gaborit ◽  
Simon Ricard ◽  
Simon Lachance-Cloutier ◽  
François Anctil ◽  
Richard Turcotte
Keyword(s):  
Climate Change ◽  
Climate Change Impact ◽  
Climatic Conditions ◽  
Local Calibration ◽  
Global Calibration ◽  
Spatial Consistency ◽  
Split Sample ◽  
Sample Test ◽  
Change Impact ◽  
Parameter Values

This work explores the performances of the hydrologic model Hydrotel, applied to 36 catchments located in the Province of Quebec, Canada. A local calibration (each catchment taken individually) scheme and a global calibration (a single parameter set sought for all catchments) scheme are compared in a differential split-sample test perspective. Such a methodology is useful to gain insights on a model’s skills under different climatic conditions, in view of its use for climate change impact studies. The model was calibrated using both schemes on five non-continuous dry and cold years and then evaluated on five dissimilar humid and warm years. Results indicate that, as expected, local calibration leads to better performances than the global one. However, global calibration achieves satisfactory simulations while producing a better temporal robustness (i.e., model transposability to periods with different climatic conditions). Global calibration, in opposition to local calibration, thus imposes spatial consistency to the calibrated parameter values, while locally adjusted parameter sets can significantly vary from one catchment to another due to equifinality. It is hence stated that a global calibration scheme represents a good trade-off between local performance, temporal robustness, and the spatial consistency of parameter values, which is, for example, of interest in the context of ungauged catchments’ simulation, climate change impact studies, or even simply large-scale modeling.

scholarly journals Is bias correction of regional climate model (RCM) simulations possible for non-stationary conditions?

2013 ◽  
Vol 17 (12) ◽  
pp. 5061-5077 ◽  
Author(s):  
C. Teutschbein ◽  
J. Seibert
Keyword(s):  
Climate Change ◽  
Bias Correction ◽  
Climate Change Impact ◽  
Climate Model ◽  
Regional Climate ◽  
Model Errors ◽  
Climate Conditions ◽  
Split Sample ◽  
Sample Test ◽  
Change Impact

Abstract. In hydrological climate-change impact studies, regional climate models (RCMs) are commonly used to transfer large-scale global climate model (GCM) data to smaller scales and to provide more detailed regional information. Due to systematic and random model errors, however, RCM simulations often show considerable deviations from observations. This has led to the development of a number of correction approaches that rely on the assumption that RCM errors do not change over time. It is in principle not possible to test whether this underlying assumption of error stationarity is actually fulfilled for future climate conditions. In this study, however, we demonstrate that it is possible to evaluate how well correction methods perform for conditions different from those used for calibration with the relatively simple differential split-sample test. For five Swedish catchments, precipitation and temperature simulations from 15 different RCMs driven by ERA40 (the 40 yr reanalysis product of the European Centre for Medium-Range Weather Forecasts (ECMWF)) were corrected with different commonly used bias correction methods. We then performed differential split-sample tests by dividing the data series into cold and warm respective dry and wet years. This enabled us to cross-evaluate the performance of different correction procedures under systematically varying climate conditions. The differential split-sample test identified major differences in the ability of the applied correction methods to reduce model errors and to cope with non-stationary biases. More advanced correction methods performed better, whereas large deviations remained for climate model simulations corrected with simpler approaches. Therefore, we question the use of simple correction methods such as the widely used delta-change approach and linear transformation for RCM-based climate-change impact studies. Instead, we recommend using higher-skill correction methods such as distribution mapping.

Communicating water-related climate change hazards to local stakeholders

2021 ◽  
Author(s):  
Laura Müller ◽  
Petra Döll
Keyword(s):  
Climate Change ◽  
Water Management ◽  
Climate Change Impact ◽  
Biosphere Reserve ◽  
Hydrological Models ◽  
Model Ensemble ◽  
Hydrological Hazards ◽  
Local Stakeholders ◽  
Change Impact ◽  
The Impact

<p>Due to climate change, the water cycle is changing which requires to adapt water management in many regions. The transdisciplinary project KlimaRhön aims at assessing water-related risks and developing adaptation measures in water management in the UNESCO Biosphere Reserve Rhön in Central Germany. One of the challenges is to inform local stakeholders about hydrological hazards in in the biosphere reserve, which has an area of only 2433 km² and for which no regional hydrological simulations are available. To overcome the lack of local simulations of the impact of climate change on water resources, existing simulations by a number of global hydrological models (GHMs) were evaluated for the study area. While the coarse model resolution of 0.5°x0.5° (55 km x 55 km at the equator) is certainly problematic for the small study area, the advantage is that both the uncertainty of climate simulations and hydrological models can be taken into account to provide a best estimate of future hazards and their (large) uncertainties. This is different from most local hydrological climate change impact assessments, where only one hydrological model is used, which leads to an underestimation of future uncertainty as different hydrological models translate climatic changes differently into hydrological changes and, for example, mostly do not take into account the effect of changing atmospheric CO<sub>2</sub> on evapotranspiration and thus runoff.   </p><p>The global climate change impact simulations were performed in a consistent manner by various international modeling groups following a protocol developed by ISIMIP (ISIMIP 2b, www.isimip.org); the simulation results are freely available for download. We processed, analyzed and visualized the results of the multi-model ensemble, which consists of eight GHMs driven by the bias-adjusted output of four general circulation models. The ensemble of potential changes of total runoff and groundwater recharge were calculated for two 30-year future periods relative to a reference period, analyzing annual and seasonal means as well as interannual variability. Moreover, the two representative concentration pathways RCP 2.6 and 8.5 were chosen to inform stakeholders about two possible courses of anthropogenic emissions.</p><p>To communicate the results to local stakeholders effectively, the way to present modeling results and their uncertainty is crucial. The visualization and textual/oral presentation should not be overwhelming but comprehensive, comprehensible and engaging. It should help the stakeholder to understand the likelihood of particular hazards that can be derived from multi-model ensemble projections. In this contribution, we present the communication approach we applied during a stakeholder workshop as well as its evaluation by the stakeholders.</p>

scholarly journals The importance of hydrological uncertainty assessment methods in climate change impact studies

2014 ◽  
Vol 18 (8) ◽  
pp. 3301-3317 ◽  
Author(s):  
M. Honti ◽  
A. Scheidegger ◽  
C. Stamm
Keyword(s):  
Climate Change ◽  
Climate Change Impact ◽  
Hydrological Model ◽  
Uncertainty Assessment ◽  
Assessment Method ◽  
Calibration Data ◽  
Relative Importance ◽  
Impact Studies ◽  
Uncertainty Sources ◽  
Change Impact

Abstract. Climate change impact assessments have become more and more popular in hydrology since the middle 1980s with a recent boost after the publication of the IPCC AR4 report. From hundreds of impact studies a quasi-standard methodology has emerged, to a large extent shaped by the growing public demand for predicting how water resources management or flood protection should change in the coming decades. The "standard" workflow relies on a model cascade from global circulation model (GCM) predictions for selected IPCC scenarios to future catchment hydrology. Uncertainty is present at each level and propagates through the model cascade. There is an emerging consensus between many studies on the relative importance of the different uncertainty sources. The prevailing perception is that GCM uncertainty dominates hydrological impact studies. Our hypothesis was that the relative importance of climatic and hydrologic uncertainty is (among other factors) heavily influenced by the uncertainty assessment method. To test this we carried out a climate change impact assessment and estimated the relative importance of the uncertainty sources. The study was performed on two small catchments in the Swiss Plateau with a lumped conceptual rainfall runoff model. In the climatic part we applied the standard ensemble approach to quantify uncertainty but in hydrology we used formal Bayesian uncertainty assessment with two different likelihood functions. One was a time series error model that was able to deal with the complicated statistical properties of hydrological model residuals. The second was an approximate likelihood function for the flow quantiles. The results showed that the expected climatic impact on flow quantiles was small compared to prediction uncertainty. The choice of uncertainty assessment method actually determined what sources of uncertainty could be identified at all. This demonstrated that one could arrive at rather different conclusions about the causes behind predictive uncertainty for the same hydrological model and calibration data when considering different objective functions for calibration.

scholarly journals Climate change impact on water resource extremes in a headwater region of the Tarim basin in China

2011 ◽  
Vol 15 (11) ◽  
pp. 3511-3527 ◽  
Author(s):  
T. Liu ◽  
P. Willems ◽  
X. L. Pan ◽  
An. M. Bao ◽  
X. Chen ◽  
...  
Keyword(s):  
Climate Change ◽  
Water Resources ◽  
Ground Water ◽  
Climate Change Impact ◽  
Snow Accumulation ◽  
Tarim River Basin ◽  
Hydrological Models ◽  
Snow Melting ◽  
Change Impact ◽  
Ipcc Sres

Abstract. The Tarim river basin in China is a huge inland arid basin, which is expected to be highly vulnerable to climatic changes, given that most water resources originate from the upper mountainous headwater regions. This paper focuses on one of these headwaters: the Kaidu river subbasin. The climate change impact on the surface and ground water resources of that basin and more specifically on the hydrological extremes were studied by using both lumped and spatially distributed hydrological models, after simulation of the IPCC SRES greenhouse gas scenarios till the 2050s. The models include processes of snow and glacier melting. The climate change signals were extracted from the grid-based results of general circulation models (GCMs) and applied on the station-based, observed historical data using a perturbation approach. For precipitation, the time series perturbation involves both a wet-day frequency perturbation and a quantile perturbation to the wet-day rainfall intensities. For temperature and potential evapotranspiration, the climate change signals only involve quantile based changes. The perturbed series were input into the hydrological models and the impacts on the surface and ground water resources studied. The range of impact results (after considering 36 GCM runs) were summarized in high, mean, and low results. It was found that due to increasing precipitation in winter, snow accumulation increases in the upper mountainous areas. Due to temperature rise, snow melting rates increase and the snow melting periods are pushed forward in time. Although the qualitive impact results are highly consistent among the different GCM runs considered, the precise quantitative impact results varied significantly depending on the GCM run and the hydrological model.

Sign in / Sign up

Export Citation Format

Share Document