Search

In arid and semi-arid areas, water is a major limitation factor for agricultural production. Indeed, these areas are characterized by a short rainy season and strong irregularity in time and space of precipitation events. This induces more frequent annual and intra-seasonal droughts. Evapotranspiration that characterizes plant water use and water stress are needed to better manage water resources and agrosystem health. They both can be simulated by a dual source energy balance model that relies on meteorological variables (air temperature, relative humidity, wind speed and global radiation) and satellite data (surface temperature, NDVI, albedo and LAI). These variables might be simulated for a long period in order to be adequate for drought studies purposes. However, available meteorological observations may often be insufficient to account for the temporal variability present in the study area (sparsity of gauged networks, the lack of long observation periods and the presence of numerous gaps). Our first objective is then to adapt a stochastic weather generator "MetGen" driven by large-scale reanalysis data (about 31 km of spatial resolution) to semi-arid climates and to the sub-daily resolution. MetGen serves to fill in missing data and to provide a temporal extension of multiple meteorological variables. It is compared with two state-of-the-art bias correction methods, univariate and multivariate methods, applied to large-scale reanalysis data. The surrogate series that are either produced by MetGen and the bias correction methods or taken as the un-processed reanalysis data, are evaluated in terms of their ability (1) to reproduce the statistical properties of the meteorological observations and (2) to reproduce energy balance outputs when constrained by observations series. The evaluation of these different statistical methods is performed on a validation period which included the observation period (2011-2016). Then, we used MetGen and the unprocessed reanalyses data to generate meteorological data during the whole study period (2000-2019). These surrogate series are used therefore to constrain the dual-source model Soil Plant Atmosphere and Remote Evapotranspiration (SPARSE) in order to simulate water stress indices SI(SWG) and SI(ERA5) from MetGen and ERA5 reanalyses successively. Stress index anomalies retrieved from SPARSE are then compared to anomalies in other wave lengths in order to assess their consistency, reliability and capacity to detect incipient water stress and early droughts at the kilometer resolution. Those are the root zone soil moisture at low resolution derived from the microwave domain, active vegetation fraction cover deduced from NDVI time series and a uniformized precipitation index UPI as a reference for these analyses. Both thermal stress indices show a good performance to detect water status, especially using SI(SWG) which show more precision and ability to identify incipient water stress. Our analyses are carried on in the Kairouan area in central Tunisia which is subject to semi-arid climate.; La surveillance des sécheresses dans les régions arides et semi-arides est cruciale car ses conséquences pour l'agriculture peuvent être dramatiques. Afin d'aider les décideurs à établir de bonnes pratiques de gestion de la ressource en eau et d'atténuation du risque des sécheresses, nous nous intéressons à l'analyse des indices de stress hydriques. À cette fin, un modèle de bilan d'énergie à double source permet, en combinant de l'information satellitaire (température de surface, NDVI, albédo et LAI) et de l'information météorologique (température de l'air, humidité relative de l'air, vitesse du vent et rayonnement global), de simuler l'évapotranspiration ainsi que le stress hydrique. Ces deux variables doivent être fournies d'une façon continue et sur une longue période temporelle pour une analyse adéquate des périodes de sécheresses. Or, les réseaux d'observations météorologiques sont parfois insuffisants (faible densité des sites instrumentés et périodes d'observation courtes et souvent non-concomitantes). Notre premier objectif est alors de simuler des scénarios de différentes variables climatiques afin de les prolonger. Nous avons adapté un générateur de conditions météorologiques "MetGen" qui permet de combler les lacunes présentes sur une période d'observation et de projeter des scénarios sur une période distincte de la période d'observation. MetGen exploite parmi ses co-variables, les données de réanalyses qui fournissent des variables à faible résolution spatiale (environ 31 km), comme source d'information importante. Nous comparons cette méthode avec des méthodes de correction de biais (univariée et multivariée) qui exploitent également les données de réanalyses. Cette approche statistique est validée selon deux volets : l'évaluation de la capacité (1) à bien reproduire les variables météorologiques et (2) à bien restituer les variables de bilan d'énergie. Les analyses, menées avec les données des stations météorologiques du système d'observations, ont permis de valider MetGen sur une période de validation (2011-2016). Nous avons utilisé alors cette méthode afin de simuler des données climatiques sur toute la période d'étude (2000-2019). Cette série ainsi que celle provenant des réanalyses brutes sont utilisées comme forçages climatiques du modèle d'énergie à double source SPARSE, afin de simuler deux indices de stress thermiques SI(SWG) et SI(ERA5) issus du générateur et des réanalyses ERA5 respectivement, à une échelle kilométrique. Ces deux indices sensibles aux anomalies de température de surface, sont comparés avec d'autres indices standardisés issus de différentes longueurs d'onde : le NDVI issu du visible/proche infrarouge, SWI du micro-onde et un indice standardisé de précipitations UPI qui est utilisé comme une référence pour notre analyse. Cette analyse est effectuée en termes de pertinence, de cohérence et de précocité pour la détection d'une sécheresse agronomique. Les deux indices thermiques ont montré des bonnes performances pour la détection du stress, notamment SI(SWG) qui a montré plus de précision et de capacité à détecter le stress hydrique d'une façon précoce. Ces analyses et tous ces approches statistiques sont effectuées au niveau du bassin versant de Merguellil situé au centre de la Tunisie et qui présente un modèle typique des régions semi-arides.

Water resources management and hydrological risks are based, to a large extent, on our competence to generate relevant hydrological scenarios under present and future climatic situations. In alpine areas, characterized by highly complex topography, the availability of discharge data is generally very limited and the hydrological regime is frequently affected by hydraulic infrastructures (e.g. dams for hydro-power production). Flood scenario estimation is thus a complicated task; classical methods are badly adapted to such a context. The present work proposes a global concept based on the simulation of meteorological weather variables transformed subsequently in flood scenarios using a rainfall-runoff model. The meteorological scenarios are obtained from simulation using a stochastic weather generator conditioned by both atmospheric circulation indices and meteorological variables. Atmospheric circulation indices are derived from reanalysis data set provided by the National Centre for Environmental Prediction (NCEP) for the period 1982-2001. Appropriate circulation indices and meteorological series were selected by means of sensitivity analysis. The stochastic generator developed within the framework of this research combines two approaches widely used in recent prediction purposes. The first uses a set of generalized linear models – that extend the classical linear models – to explain regional and daily meteorological variables using the appropriate set of predictors. The second uses nearest neighbor method, based on the hypothesis that similar synoptic situations will produce similar meteorological situations, to disaggregate daily regional meteorological variables into hourly and locally distributed variables. The generation process was applied to the Upper Rhone River basin, an alpine catchment located in the Valais canton – Switzerland (upstream to Lake Leman) to generate a series of hourly meteorological scenarios (precipitation, temperature and lapse rate temperature) at multiple sites (48 raingauges and 11 thermal stations). The ability of the stochastic generator to reproduce statistics of key hydrological variables and extreme weather events was evaluated at multiple spatial (ranging from 100 to 5500 km2) and temporal (ranging from 3 h to 3 days) scales. These statistics are classical mean, standard deviation and lag-1 correlation coefficient for total and liquid precipitation, as well as temperature. Other associated variables, such as the snow/rain line altitude have been studied. In general, statistics related to meteorological scenarios show moderate to very good agreement between simulations and observations. For extreme events, distributions of annual maxima precipitations and of seasonal maximum and minimum temperatures are also well reproduced, even if results obtained at both hourly and locally scales show less good – but still satisfactory – performance than those obtained at both daily and regional scales. This is definitely not surprising since the calibration of the different model parameters has been performed at those latter scales. The meteorological scenarios are next used to simulate 20-year discharge time series at a number of discharge stations throughout the Rhône watershed using an appropriate rainfall-runoff model. The catchment was divided into elevation bands in order to be able to take into account the variation of temperature with altitude; this is of high importance, especially in areas showing highly complex topographical features, which is usually the case of the alpine watersheds. Here again, statistical properties of the hydrological scenarios show good accordance with observations, especially for flood events. Finally, a number of flood scenarios, corresponding either to floods localized in some specific sub-basins or to floods generalized over the whole basin, were extracted from the simulated discharge time series; they can be used to study the hydrological behavior of the basins submitted to various meteorological solicitations.