This study aims at characterizing spatiotemporal variability of the fate of nutrients in the karst basin of the Loue River in the French Jura Mountains. The long-term temporal analysis (40 years) shows that the NO3 increase from 1970s to the 1990s followed by a no-trend period in 2000s. The changes are linked to the usage of mineral fertilizers. The short-term analysis shows that the degree of aquifer saturation at the beginning of the hydrological cycle is a key factor to assess NO3 mobilization during the recharge events. Contrary to nitrate, the PO4 concentrations are disconnected from agricultural practices and are probably the consequence of point-source contaminations from domestic wastewater. Annual loads were estimated on 5 sub-basins in order to characterize the spatial variability of water contamination. Difference in fluxes for each sub-basin highlighted the most impacted reaches, providing information on hydrological units where anthropogenic pressure is highest. A correlation of NO3–N loads with the surface area of main crops using highest level of fertilization and tillage (field crops, temporary grasslands) is proposed, highlighting the environmental impact of most intensive agricultural practices (inventoried in a small area covering less than 10% of the whole basin). This study illustrates complex interactions between agricultural practice and hydrological function and gives first insight into the fate of nutrients in karst environment.

For a cost-effective survey of water quality in aquifers with a fast chemical response, such as karst aquifers, continuous high-resolution monitoring by an automatic sensor is relevant. We tested the suitability of the s::can UV–visible spectrometer for continuous measuring in a karst environment of the parameters proposed by the probe manufacturer (NO3 and total organic carbon TOC) as well as other parameters, such as total phosphorus (TP). The spectrometer was installed at the Loue Spring (French Jura Mountains), where water was also sampled for chemical analysis of NO3, TOC, and TP at a frequency of 1 to 4 days. A calibration model was developed based on the partial least-squares regression (PLSR) method, applied to the absorption spectra. Our method showed good results. For NO3, both the factory calibration (\(R^{2} {_{\text{VAL}}} = 0.98\)) and our calibration model (\(R^{2} {_{\text{VAL}}} = 0.99\)) are very good. For TOC, except for a slight underestimation of some peaks, the low and high values are better reproduced by our model developed (\(R^{2} {_{\text{VAL}}} = 0.63\), or 0.23 with the factory calibration). For TP, despite a higher background noise, the overall dynamics are well simulated (\(R^{2} {_{\text{VAL}}} = 0.56\)). Finally, our results showed that processing the raw data of the spectrum measured by the spectrometer optimizes the high-frequency monitoring of water quality and provides a better prediction of some parameters, as well as giving promising results for the calibration of non-programmed parameters.

We measured nitrate in water (n = 4762) and nitrogen stable isotopes (δ15N) (n = 353) of macroalgae and macroinvertebrates at different sites in a French karst river (Loue River) which has moderate nitrogen loading (~ 23 kg nitrogen ha−1 year−1). The main objective was to estimate whether nitrate in water and nitrogen stable isotopes of the biota could allow identification of the spatial and temporal variations in agriculture-related nitrogen losses to the river. Highest nitrate concentrations (> 90% quantile) increased significantly over the last three decades but mean nitrate concentrations did not change significantly. Nitrate and biota δ15N values tended to increase from upstream to downstream, although the δ15N values decreased in the most downstream sites. Generalized additive mixed models allowed the identification of clear matching annual patterns of nitrate and biota δ15N values characterized by recurrent autumn to winter maxima, supporting that agricultural nitrogen export and its assimilation within the river biota would mostly occur during the autumn and winter seasons. Overall, our results highlight the high vulnerability of karst river to nitrogen-related eutrophication.

This work focuses on the development of FLAME (Forecasting Landslides induced by Acceleration Meteorological Events) that analyze of the relationship between displacements and precipitations using a statistical approach in order to predict the surface displacement at active landslide. FLAME is an Impulse Response model (IR) that simulates the changes in landslide velocity by computing a transfer function between the input signal (e.g. rainfall or recharge) and the output signal (e.g. displacement). This model has been applied to forecast the displacement rates at Séchilienne (French Alps). The FLAME model is enhanced by achieving the calibration using joint inversion of multiple time series data. We consider that the displacements at two different sensors are explained by the same long-term response of the system to ground water level variations. The parameters describing the long-term response of the system are therefore identical for all sensors. The joint inversion process allows decreasing the ratio between the number of parameters to be inverted and the volume of data and is thus more statically steady. The results indicate that the models are able to reproduce the displacement pattern in general to moderate kinetic regime but not extreme kinetic regime. Our results do not give clear evidence of an improvement of the models performance with joint inversion of multiple time series of data. The reasons which could explain these inconclusive results are discussed in the paper.

An integrated analysis on the relationship between rainfall and displacement in the most active area of the Sechilienne unstable slope was performed. This study combines several techniques and models to adequately reproduce the landslide movement induced by the rainfall. The analysis of available time series shows a long term trend and seasonal variations in the displacement, respectively independent and synchronous to precipitations. In particular wavelet analysis highlights that the movement is rather linked to groundwater recharge than to precipitation (rainfall + snowfall), involving then the importance of groundwater process in the area. A first and simple relationship between the water input and the fluctuations of displacements apart from the general trend is shown using a tank model. Moreover, a seasonal analysis of this relationship was performed, showing that displacement rate follows the behavior of the hydrological cycle. Two different models were applied to the long temporal series of extensometric and precipitation data: the FLAME model, from BRGM and the FORESEES model, from Univ. Lausanne. These tools are based on a combined statistical-mechanical approach to predict changes in landslide displacement rates from observed changes in precipitation amounts. The forecasting tool FLAME associates 1) a statistical impulse response (IR) model to simulate the changes in landslide rates by computing a transfer function between the rainfall and the displacements, and 2) a 1D mechanical (ME) model (e.g. visco-plastic rheology), in order to take into account changes in pore water pressures. The performance of different combinations of models was evaluated against observed displacement rates at the selected pilot study area. Our results indicate that both models are able to reproduce, with a high degree of accuracy, the observed displacement pattern in the general kinematic regime. Finally the variability of the results, depending in particular on the input data, is discussed.

The rainfall threshold determination is widely used for estimating the minimum critical rainfall amount which may trigger slope failure. The aim of this study was to develop an objective approach for the determination of a statistical rainfall threshold of a deep-seated landslide. The determination is based on recharge estimation and a multi-dimensional rainfall threshold. This new method is compared with precipitation and with a conventional ‘two-dimensional’ rainfall threshold. The method is designed to be semiautomatic, enabling an eventual integration into a landslide warning system. The method consists in two independent parts: (i) unstable event identification based on displacement time series and (ii) multi-dimensional rainfall threshold determination based on support vector machines. The method produces very good results and constitutes an appropriate tool to define an objective and optimal rainfall threshold. In addition to shortened computation times, the non-necessity of pre-requisite hypotheses and a fully automatic implementation, the newly introduced multi-dimensional approach shows performances similar to the classical two-dimensional approach. This shows its relevance and its suitability to define a rainfall threshold. Lastly, this study shows that the recharge is a relevant parameter to be taken into account for deep-seated rainfall-induced landslides. Using the recharge rather than the precipitation significantly improves the delineation of a rainfall threshold separating stable and unstable events. The performance and accuracy of the multi-dimensional rainfall threshold developed for the Séchilienne landslide make it an appropriate method for integration into the present-day landslide warning system.

We propose an approach to study the hydro-mechanical behaviour and evolution of rainfall-induced deep-seated landslides subjected to creep deformation by combining signal processing and modelling. The method is applied to the Séchilienne landslide in the French Alps, where precipitation and displacement have been monitored for 20 years. Wavelet analysis is first applied on precipitation and recharge as inputs and then on displacement time-series decomposed into trend and detrended signals as outputs. Results show that the detrended displacement is better linked to the recharge signal than to the total precipitation signal. The infra-annual detrended displacement is generated by high precipitation events, whereas annual and multi-annual variations are rather linked to recharge variations and thus to groundwater processes. This leads to conceptualise the system into a two-layer aquifer constituted of a perched aquifer (reactive aquifer responsible of high-frequency displacements) and a deep aquifer (inertial aquifer responsible of low-frequency displacements). In a second step, a new lumped model (GLIDE) coupling groundwater and a creep deformation model is applied to simulate displacement on three extensometer stations. The application of the GLIDE model gives good performance, validating most of the preliminary functioning hypotheses. Our results show that groundwater fluctuations can explain the displacement periodic variations as well as the long-term creep exponential trend. In the case of deep-seated landslides, this displacement trend is interpreted as the consequence of the weakening of the rock mechanical properties due to repeated actions of the groundwater pressure.

Groundwater-level rise plays an important role in the activation or reactivation of deep-seated landslides and so hydromechanical studies require a good knowledge of groundwater flows. Anisotropic and heterogeneous media combined with landslide deformation make classical hydrogeological investigations difficult. Hydrogeological investigations have recently focused on indirect hydrochemistry methods. This study aims at determining the groundwater conceptual model of the Séchilienne landslide and its hosting massif in the western Alps (France). The hydrogeological investigation is streamlined by combining three approaches: a one-time multi-tracer test survey during high-flow periods, a seasonal monitoring of the water stable-isotope content and electrical conductivity, and a hydrochemical survey during low-flow periods. The complexity of the hydrogeological setting of the Séchilienne massif leads to development of an original method to estimate the elevations of the spring recharge areas, based on topographical analyses and water stable-isotope contents of springs and precipitation. This study shows that the massif supporting the Séchilienne landslide is characterized by a dual-permeability behaviour typical of fractured-rock aquifers where conductive fractures play a major role in the drainage. There is a permeability contrast between the unstable zone and the intact rock mass supporting the landslide. This contrast leads to the definition of a shallow perched aquifer in the unstable zone and a deep aquifer in the intact massif hosting the landslide. The perched aquifer in the landslide is temporary, mainly discontinuous, and its extent and connectivity fluctuate according to the seasonal recharge.

Water chemistry is a very fine signal which allows fine location in time and space of the arrival of infiltration water inducing mechanical instability pulses of the landslide. This tool is designed to understand the complex relationship between chemical weathering, hydromechanical changes and weakening/motion of the unstable rock slope. For this purpose, a hydrogeochemical groundwater monitoring has been established since 2010 on the site of Séchilienne (France). Electrical conductivity is representative of the chemical signal generated by the degradation of the massif. The continuous measurement of this parameter is relevant to the site of Séchilienne and can replace chemical monitoring. The benefit of acquiring this data is threefold: real-time measurements, with a short time step, and inexpensive implementation work, enabling to use it as a tool for risk management.

Rainfall threshold is a widely used method for estimating minimum critical rainfall amount which can yield a slope failure. Literature reviews show that most of the threshold studies are subjective and not optimal. For this study, effective rainfall was considered for threshold definition. Support vector machines (SVM) and automatic event identification were used in order to establish an optimal and objective threshold for the Sechilienne landslide. Effective rainfall does significantly improve threshold performance (misclassification rate of 7.08 % instead of 13.27 % for gross rainfall) and is a relevant parameter for threshold definition in deep-seated landslide studies. In addition, the accuracy of the Sechilienne SVM threshold makes it appropriate to be integrated into a landslide warning system. Finally, the ability to make predictions at a daily time step opens up an opportunity for destabilisation stage predictions, through the use of weather forecasting.

Pore water pressure build-up by recharge of underground hydrosystems is one of the main triggering factors of deep-seated landslides. In most deep-seated landslides, pore water pressure data are not available since piezometers, if any, have a very short lifespan because of slope movements. As a consequence, indirect parameters, such as the calculated recharge, are the only data which enable understanding landslide hydrodynamic behaviour. However, in landslide studies, methods and recharge-area parameters used to determine the groundwater recharge are rarely detailed. In this study, the groundwater recharge is estimated with a soil-water balance based on characterisation of evapotranspiration and parameters characterising the recharge area (soil available water capacity, runoff and vegetation coefficient). A workflow to compute daily groundwater recharge is developed. This workflow requires the records of precipitation, air temperature, relative humidity, solar radiation and wind speed within or close to the landslide area. The determination of the parameters of the recharge area is based on a spatial analysis requiring field observations and spatial data sets (digital elevation models, aerial photographs and geological maps). This study demonstrates that the performance of the correlation with landslide displacement velocity data is significantly improved using the recharge estimated with the proposed workflow. The coefficient of determination obtained with the recharge estimated with the proposed workflow is 78% higher on average than that obtained with precipitation, and is 38% higher on average than that obtained with recharge computed with a commonly used simplification in landslide studies (recharge = precipitation minus non-calibrated evapotranspiration method).

L’eau, par l’intermediaire de la pression de fluides, est un phenomene declencheur majeur de la destabilisation des mouvements de terrain profonds. En consequence, la caracterisation des mecanismes de deformation necessite une bonne comprehension des processus hydrogeologiques controlant la destabilisation. Les milieux fissures et de surcroit les milieux instables presentent de fortes heterogeneites, ce qui rend les etudes hydrogeologiques classiques peu adaptees. De plus, les mouvements de terrain profonds presentent des relations hydromecaniques complexes avec des evolutions significatives dependantes du temps (deformation de type fluage). Cette these s’attache a caracteriser les relations precipitations-deplacement du mouvement de terrain profond de Sechilienne. Un suivi saisonnier de traceurs naturels et artificiels a permis de definir un schema conceptuel d’ecoulement de l’eau souterraine sur l’ensemble du massif malgre un nombre limite de points d’interet hydrogeologiques. Les donnees de pression de fluides etant rarement mesurees, les parametres indirects, tels que la recharge, sont souvent les seules donnees hydrogeologiques qui permettent de caracteriser la relation precipitations-destabilisation. Une methode d’estimation de la recharge basee sur un calcul de bilan du sol a ete developpee afin d’estimer la recharge avec precision. En se basant sur le schema conceptuel d’ecoulement et le calcul de la recharge, une analyse en ondelettes couplee a un modele numerique a permis de caracteriser la relation precipitations-vitesse de deplacement. Cette modelisation tient compte de parametres dependant du temps et permet de simuler une deformation de type fluage (tendance pluriannuelle des vitesses de deplacement), consequence des couplages hydro- mecaniques indirects. La caracterisation des processus hydrogeologiques controlant la destabilisation a permis de definir un seuil statistique d’activation de la destabilisation, base sur une approche multi-dimensionnelle innovante.

Abstract. Pore water pressure, build up by recharge of hydrosystems, is one of the main triggering factors of deep seated landslides. Effective rainfall, which is the part of the rainfall which recharges the aquifer, is a significant parameter. Soil-water balance is an accurate way to estimate effective rainfall. Nevertheless this approach requires evapotranspiration, soil water storage and runoff characterization. Available soil storage and runoff were deduced from field observations whereas evapotranspiration computation is a highly demanding method requiring significant input of meteorological data. Most of the landslide sites used weather stations with limited datasets. A workflow method was developed to compute effective rainfall requiring only temperature and rainfall as inputs. Two solar radiation and five commonly used evapotranspiration equations were tested at Sechilienne. The method was developed to be as general as possible in order to be able to be applied to other landslides. This study demonstrated that, for the Sechilienne unstable slope, the displacement data correlation performance (coefficient of determination) is significantly enhanced with effective rainfall (0.633) compared to results obtained with raw rainfall (0.436) data. The proposed method for estimation of effective rainfall was developed to be sufficiently simple to be used by any non-hydro specialist who intends to characterize the relationship of rainfall to landslide displacements.

Les travaux menes depuis plus d'une dizaine d'annees sur differents massifs instables ont montre les potentialites des observations hydrogeochimiques (suivi a long-terme et experimentations) pour une meilleure comprehension du comportement hydro-mecanique des instabilites de versant et potentiellement l'identification de precurseurs de la rupture ou d'acceleration.

Time series analysis and cross-wavelet analysis are used to characterize the relationship between water input and displacement in the most active zone of the Sechilienne unstable slope. Time series analysis shows a displacement long term trend and seasonal intra-annual variations, respectively independent and synchronous to precipitations. Wavelet analysis has allowed identifying and characterizing the precipitation-detrended displacement relationship which shows that the Sechilienne destabilisation is rather linked to effective rainfall than to raw precipitation (rainfall + snowfall), involving then groundwater process. Seasonal analysis of this relationship was performed, showing that displacement rate follows the behaviour of the hydrological cycle. Finally, trend was analysed and a weakening model approach was developed with an attempt to forecast the next modifications in unstable slope destabilisation behaviour.

La chimie des eaux est un signal tres fin qui permet la localisation dans le temps et dans l'espace de l'arrivee des eaux d'infiltration a l'origine des impulsions mecaniques des instabilites de versant. Il s'agit de comprendre comment s'organise la relation complexe entre alteration chimique, modifications hydromecaniques et fragilisation/mouvements du versant rocheux instable. Pour cela, un suivi hydrogeochimique des eaux souterraines a ete mis en place depuis 2010 sur le site de Sechilienne (France). La conductivite electrique est representative du signal chimique engendre par la degradation du massif. L'acquisition en continu de ce parametre est pertinente pour le site de Sechilienne et peut se substituer au suivi chimique. L'avantage de l'acquisition de cette donnee est triple, ce qui peut en faire un outil pour la gestion du risque : mesure en temps reel, avec un pas de temps tres fin et mise en œuvre peu couteuse.