Search

Studies of spatial and temporal variation of Nitrate (NO3) and Dissolved Organic Carbon (DOC) concentrations in rivers can inform on the control of their exports by hydrological processes, soil biogeochemical processes and catchment characteristics (geomorphology, land use, etc...). However, such studies are often conducted on large river basins, in which in-stream processes obscure soil-to-stream processes and landscape effects. This study has two objectives: 1) characterize NO3 and DOC concentration dynamics on small headwater streams, and particularly their dyssynchrony, 2) identify their controlling factors.We used time series of public dataset on water quality together with spatial information on climate, soil, geology, hydrology, agricultural land, domestic pressures and landscape features from 185 independent headwater catchments located in Brittany, Western France. Rivers and streams have been monitored for NO3 and DOC concentrations between 2007 and 2017, with at least eight samples per year. Catchments sizes vary from 1 to 200 km² (median of 38 km²). We used Generalized Additive Models to describe seasonal variations of NO3 and DOC concentrations and flow rates and extract amplitude and phase metrics. Potential controlling factors were identified using a correlation analysis between seasonal metrics and hydrological and land use variables.Respectively for NO3 and DOC, 85% and 71% catchments present a significant seasonal cycle and amplitudes are 3.3 ± 2.1 mgN/L and 3.1 ± 1.7 mgC/L. Four groups of catchments were identified based on seasonal metrics. In group A, 66 catchments depict an anti-correlation between NO3 and DOC seasonal pattern. NO3 maximum occurs in winter synchronously with DOC minimum and discharge maximum, while NO3 minima and DOC maximum occur in autumn. In group B (12 catchments), NO3 maximum occurs in summer synchronously with DOC minimum and the lowest summer flows. Group C (64 catchments) shows an intermediate pattern with two annual maxima for NO3 (and two minima for DOC) occurring in winter and summer, but sometime with a lag time for the pattern of NO3. Group D (26 catchments) is composed of catchments without significant seasonal variation for NO3 and DOC concentrations. The spatial variability of seasonality was mainly explained by differences in hydrological reactivity and to a minor extend by differences in geological substrate characteristics. In catchments with lowest groundwater contribution and highest hydrological reactivity, NO3 and DOC seasonal fluctuations are high and correspond mainly to group A. In catchments with highest groundwater contributions, NO3 seasonal fluctuations are small and those of DOC strong (mainly associated with group B). These results indicated that water transit time through the catchment as a result of mixing between groundwater and sub-surface runoff is a key element explaining opposite seasonal patterns of NO3 and DOC. [ABSTRACT FROM AUTHOR]

The SCHADEX method (Paquet et al., 2013) is a reference method in France for the estimation of extreme flood for dam safety studies. It is based on a semi-continuous rainfall–runoff simulation process: hundreds of different rainy events, randomly drawn up to extreme values, are simulated independently in the hydrological conditions of each rainy day of a historical record. This allows the generation of an exhaustive set of crossings between precipitation and soil saturation hazards, which leads to compute a complete distribution of flood discharges up to extreme quantiles.Since its introduction twelve years ago, the method has been applied to more than 150 catchments in France and abroad, with two types of rainfall-runoff models:• Lumped models like MORDOR or GR4J. • Semi-distributed models in which the modeled catchment is divided into several altitude bands, like MORDOR-SD, HBV or DDD.To provide a better representation of the hydrological processes in large and heterogeneous catchments, a distributed modeling approach is advisable. In this respect, a distributed version of the MORDOR model has been recently achieved and documented (Rouhier et al., 2017). In the SCHADEX method, the simulated rainy events are drawn and affected to a probability as areal rainfall values, i.e. uniformly on the catchment. The application with a distributed model thus requires these areal values to be disaggregated at an adequate resolution to feed the distributed rainfall-runoff model. This disaggregation is done thanks to a catalog of rain field shapes in which a relevant one is chosen according to the intensity of the simulated event, its season, weather type, etc. This catalog can be built by two methods:• A spatial interpolation of observed rainfall build for each day of the 1948-2012 period (SPAZM, see Gottardi et al., 2012).• A stochastic spatial rainfall generator accounting for spatial and temporal dependencies in daily precipitation fields (Vaittinada et al., 2019).This study compares the extreme quantiles estimations provided by a "classical" application of SCHADEX (both with a lumped and a semi-distributed model), to a distributed application using different strategies for disaggregating the simulated events. Different features of the simulated floods will be compared in order to assess the contributions of a distributed approach (compared to a lumped one) and the sensitivity to the modeling options. It is illustrated with the example of Ardèche catchment at Sauze (south-east of France, 2260 km²). References:Gottardi, F., Obled, C., Gailhard, J., & Paquet, E. (2012). Statistical reanalysis of precipitation fields based on ground network data and weather patterns: Application over French mountains. Journal of Hydrology, 432, 154-167.Paquet, E., Garavaglia, F., Garçon, R., & Gailhard, J. (2013). The SCHADEX method: A semi-continuous rainfall–runoff simulation for extreme flood estimation. Journal of Hydrology, 495, 23-37.Rouhier, L., Le Lay, M., Garavaglia, F., Le Moine, N., Hendrickx, F., Monteil, C., & Ribstein, P. (2017). Impact of mesoscale spatial variability of climatic inputs and parameters on the hydrological response. Journal of Hydrology, 553, 13-25.Vaittinada, P., Blanchet, J., Paquet, E. & Penot, D. (2019). Space-time simulation of precipitation based on weather pattern sub-sampling and meta-Gaussian model. Submitted. [ABSTRACT FROM AUTHOR]