Numerical modelling is now used routinely to make predictions about the behaviour of environmental systems

Vidal J.-P. and Hendrickx, F. Impact of climate change on hydropower: Ariège River, France. In: Modelling the Impact of Climate Change on Water Resources (F. Fung, A. Lopez et M. New, éds.), Blackwell Publishing, submitted

Abstract

The assessment of regional climate change requires the development of reference long-term retrospective meteorological datasets. This article presents an 8 km resolution atmospheric reanalysis over France performed with Safran gauge-based analysis system for the period 1958-2008. Climatological features of Safran 50-year analysis—long-term mean values, interannual and seasonal variability—are first presented for all computed variables: rainfall, snowfall, mean air temperature, specific humidity, wind speed, solar and infrared radiation. The spatial patterns of precipitation, minimum and maximum temperature are compared to another spatialisation method, and the temporal consistency of the reanalysis is assessed through various validation experiments with both dependent and independent data. These experiments demonstrate the overall robustness of Safran reanalysis and the improvement of its quality with time, in connection with the sharp increase in the observation network density that occurred in the 1990s. They also show the differentiated sensitivity of variables to the number of available ground observations, with precipitation and air temperature being the more robust ones. The comparison of trends from the reanalysis with those from homogenized series finally shows that if spatial patterns are globally consistent with both approaches, care must be taken when using literal values from the reanalysis and corresponding statistical significance in climate change detection studies. Safran 50-year atmospheric reanalysis constitutes a long-term forcing datasets for land surface schemes and will thus enable simulating the past 50 years of water resources over France.

Vidal J.-P., Martin É., Baillon M., Franchistéguy L. and Soubeyroux J.-M., A 50-year high-resolution atmospheric reanalysis over France with Safran system. Submitted to the International Journal of Climatology

Abstract

Vidal J.-P. and Wade S. D., Probabilistic assessment of the impact of climate change on UK river flows – Part 1: Baseline hydrological modelling with structure and parameter uncertainty. Submitted to the Journal of Hydrology

Abstract

Assessing the impact of climate change on river flows requires a prior detailed evaluation of the uncertainty in the rainfall-runoff modelling process. This article first proposes a framework for hydrological modelling which includes uncertainty on both model structure and model parameters. This framework is then applied to 70 UK catchments covering a large range of climatic and geological characteristics. The Generalized Likelihood Uncertainty Estimation (\textsc{Glue}) methodology is combined with two conceptual model structures---PDM, a lumped model, and Catchmod, a semi-distributed model---in order to provide ensembles of catchment rainfall-runoff models as well as an assessment of uncertainty in river flow regime over the thirty-year 1961-1990 baseline period. Ensemble results are evaluated in terms of model performance and model convergence, and examined with respect to catchment characteristics: baseflow index (BFI), standard average annual rainfall (SAAR), and area. Results first show that catchment characteristics cannot explain the ability of a given structure to provide efficient catchment models. Model ensembles tend to overestimate autumn flows in very permeable catchments and summer flows in dry catchments, and underestimate autumn flows in impervious catchments and summer flows in wet catchments. Dispersion of model ensembles shows a marked seasonal pattern with a peak in summer for baseflow-dominated catchments and in winter in more flashy catchments. Winter dispersion is strongly and negatively correlated with SAAR. An important result of this study is that the uncertainty in model structure has the same order of magnitude than the uncertainty in model parameters. Model ensembles succeed in simulating the very low flows recorded during the benchmark 1976 drought and will be used to assess the impact of climate change in UK river flows.

Vidal J.-P. and Wade S. D. (2009), A multimodel assessment of future climatological droughts in the United Kingdom. International Journal of Climatology, in press. DOI: 10.1002/joc.1843

Abstract

This paper presents a detailed assessment of future rainfall drought patterns over the United Kingdom. Previously developed bias-corrected high-resolution gridded precipitation time series are aggregated to the scale relevant for water resources management, in order to provide 21st-century time series for 183 hydrologic areas, as computed by six General Circulation Models (GCMs) under two emissions scenarios. The control run data are used as a learning time series to compute the Standardized Precipitation Index (SPI) at four different time scales. SPI values for three 30-year future time slices are computed with respect to these learning time series in order to assess the changes in drought frequency. Multimodel results under the A2 scenario show a dramatic increase in the frequency of short-term extreme drought class for most of the country. A decrease of long-term droughts is expected in Scotland, due to the projected increase in winter precipitation. The analysis for two catchment case studies also showed that changes under the B2 scenario are generally consistent with those of the A2 scenario, with a reduced magnitude in changes. The overall increase with time in the spread of individual GCM results demonstrates the utility of multimodel statistics when assessing the uncertainty in future drought indices to be used in long-term water resources planning.

Soubeyroux J.-M., Martin É., Franchistéguy L., Habets F., Noilhan J., Baillon M., Regimbeau F., Vidal J.-P., Le Moigne P. and Voirin-Morel S. (2008), Safran-Isba-Modcou (SIM) – Un outil pour le suivi hydrométéorologique opérationnel et les études. La Météorologie, 63: 40-45. hdl.handle.net/2042/21890

Abstract

The Safran-Isba-Modcou model combines an analysis system of the atmospheric forcing, a land surface scheme, and a hydrogeological model. Advanced research has been done around this suite during the last fifteen years. Since 2003, it has been transferred to Météo-France real-time operational environment and has been progressively supplied with an extended climatology. Its operational use now covers water resources monitoring, flood forecasting, and drought assessment. It is also particularly useful for studying the impact of climate change on the water cycle.

Vidal J.-P. and Wade S. D. (2008), Multimodel projections of catchment-scale precipitation regime. Journal of Hydrology, 353 (1-2): 143-158. DOI: 10.1016/j.jhydrol.2008.02.003.

Abstract

Climate change impact studies are often based on coarse scale projections of General Circulation Models (GCMs) under greenhouse gases emissions scenarios. Outputs from GCMs have then to be downscaled to obtain the information relevant to hydrologic studies. This paper investigates the uncertainty in catchment-scale precipitation scenarios due to (1) the emissions scenario, (2) the configuration of the global climate model, and (3) the downscaling method. Two emissions scenarios, six global climate models and four downscaling methods are used to build climate change scenarios for three catchment case studies in UK. Missing combinations are reconstructed thanks to a variance decomposition algorithm. The highly increasing variance towards the end of the century is shown to be largely due to the spread of results from different GCMs. Multi-model averaged results at the catchment scale generally show an increase in the seasonal pattern in precipitation, with wetter winters and drier summers. Hydrologists and water resources planners make use of downscaled climate scenarios, often with little regard for the performance of scenario construction methods, for informing decisions on water resources and flood risk management policies and projects. This paper shows that both the downscaling method and the multimodel building scheme applied have a significant impact on the seasonal precipitation regime that may in turn lead to quite different conclusions in impacts assessments.

Vidal J.-P. and Wade S. D. (2008), A framework for developping high-resolution multi-model climate projections: 21st century scenarios for the UK. International Journal of Climatology, 28 (7): 843-858, DOI: 10.1002/joc.1593.

Abstract

This paper proposes a simple and efficient framework for building consistent climate projections from an ensemble of Global Circulation Models (GCMs) at the local scale required for impact studies. The proposed method relies on a fine-scale gridded baseline climatology and consists of the following steps: (1) building appropriate precipitation and temperature time series from land areas covered by GCM sea cells; (2) correction of GCM outputs inherent biases through “quantile-based mapping”; and (3) disaggregation of bias-corrected outputs with monthly spatial anomalies between GCM-specific and observed spatial scales. The overall framework is applied to derive 21st century seasonal climate projections and interannual variability for the UK based on an ensemble of 6 GCMs run under two different emissions scenarios. Results show a large dispersion of changes within the multi-GCM ensemble, along with a good comparison between scenarios from individual ensemble members and from previous UK and European studies using dynamically downscaled outputs from corresponding GCMs. The framework presented in this paper provides appropriate outputs to take account of the uncertainty in global model configuration within impacts studies that are influencing current decisions on major investments in flood risk management and water resources.

Vidal J.-P., Moisan S., Faure J.-B., and Dartus D., River model calibration, from guidelines to operational support tools. Environmental Modelling and Software, 22 (11): 1628-1640. DOI: 10.1016/j.envsoft.2006.12.003

Abstract

Numerical modelling is now used routinely to make predictions about the behaviour of environmental systems. Model calibration remains a critical step in the modelling process and different approaches have been taken to develop guidelines to support engineers and scientists in this task. This article reviews currently available guidelines for a river hydraulics modeller by dividing them into three types: on the calibration process, on hydraulic parameters, and on the use of hydraulic simulation codes. The article then presents an integration of selected guidelines within a knowledge-based calibration support system. A prototype called CaRMA-1 (Calibration of River Model Assistant) has been developed for supporting the calibration of models based on a specific 1D code. Two case studies illustrate the ability of the prototype to face operational situations in river hydraulics engineering, for which both data quality and quantity are not sufficient for an optimal calibration. Using CaRMA-1 allows the modeller to achieve the calibration task in accordance with good calibration practice implemented in the knowledge base. Relevant reasoning rules can easily be added to the knowledge base to extend the prototype range of applications. This study thus provides a framework for building operational support tools from various types of existing engineering guidelines.

Vidal J.-P., Moisan S., Faure J.-B., and Dartus D. (2005), Towards a reasoned 1-D river model calibration. Journal of Hydroinformatics, 7 (2): 91-104. →http

Abstract

Model calibration remains a critical step in numerical modelling. After many attempts to automate this task in water-related domains, questions about the actual need for calibrating physics-based models are still open. This paper proposes a framework for good model calibration practice for end-users of 1D hydraulic simulation codes. This framework includes a formalisation of objects used in 1D river hydraulics along with a generic conceptual description of the model calibration process. It was implemented within a knowledge-based system integrating a simulation code and expert knowledge about model calibration. A prototype calibration support system was then built up with a specific simulation code solving subcritical unsteady flow equations for fixed-bed rivers. The framework for model calibration is composed of three independent levels related, respectively, to the generic task, to the application domain and to the simulation code itself. The first two knowledge levels can thus easily be reused to build calibration support systems for other application domains, like 2D hydrodynamics or physics-based rainfall-runoff modelling.

Vidal J.-P., Moisan S., and Faure J.-B. (2003), Knowledge-based hydraulic model calibration. Lecture Notes in Artificial Intelligence, vol. 2773 (Knowledge-Based Intelligent Information and Engineering Sytems: 7^th International Conference KES'2003 ; V. Palade, R. J. Howlett et L. C. Lain, eds.), p. 676-683. DOI: 10.1007/b12002

Abstract

Model calibration is an essential step in physical process modelling. This paper describes an approach for model calibration support that combines heuristics and optimisation methods. In our approach, knowledge-based techniques have been used to complement standard numerical modelling ones in order to help end-users of simulation codes. We have both identified the knowledge involved in the calibration task and developed a prototype for calibration support dedicated to river hydraulics. We intend to rely on a generic platform to implement artificial intelligence tools dedicated to this task.