Abstract. Canada's water cycle is driven mainly by snowmelt. Snow equivalent (SWE) the snow-related variable that most commonly used in hydrology, as it expresses total quantity of (solid and liquid) stored snowpack. Measurements SWE are, however, expensive not continuously accessible real time. This motivates a search for alternative ways estimating from measurements are more widely available continuous over can be calculated multiplying snow depth bulk density Regression models proposed...

The objective of flood frequency analysis (FFA) is to associate intensity with a probability exceedance. Many methods are currently employed for this, ranging from statistical distribution fitting simulation approaches. In many cases the site interest actually ungauged, and regionalisation scheme has be associated FFA method, leading multiplication number possible available. This paper presents results wide-range comparison families different schemes based on regression, or spatial physical...

Snow water equivalent (SWE) is among the most important variables in hydrological modelling of high latitude and mountainous areas. While manual snow surveys can directly provide SWE measurements, they are time consuming costly, especially compared to automated depth measurements. Moreover, strongly correlated depth. For this reason, several empirical equations relating have been proposed. The present study investigates potential artificial neural networks for estimating from commonly...

Abstract. Data assimilation is an essential component of any hydrological forecasting system. Its purpose to incorporate some observations from the field when they become available in order correct state variables model prior phase. The goal ensure that forecasts are initialized as representative reality possible, and also estimate uncertainty variables. There several data methods, particle filters increasingly popular because their minimal assumptions. baseline idea produce ensemble...

SHYREG method is a regionalized for rainfall and flood frequency analysis (FFA). It based on processes simulation. couples an hourly generator with rainfall-runoff model, simplified enough to be regionalized. The has been calibrated using all hydro meteorological data available at the national level. In France, that represents about 2800 raingauges of French Weather Service network 1800 stations hydrometric National Bank network. Then, provide flow quantiles database. An evaluation was...

The development and expanded application of large-scale hydrological models has produced forecasts that often overlap with more targeted, regional forecasts. Here the possibility is explored for using simple methods to combine from a model, Great Lakes portion National Surface River Prediction System (NSRPS), system, Système de Prévision Hydrologique (SPH) which covers southern Quebec, improve Outputs two forecasting systems are combined multiple methods, including mean, weighted average in...

Abstract. Canada's water cycle is driven mainly by snowmelt. Snow equivalent (SWE) the snow-related variable that most commonly used in hydrology, as it expresses total quantity of (solid and liquid) stored snowpack. Measurements SWE are, however, expensive not continuously accessible real time. This motivates a search for alternative ways estimating from measurements are more widely available continuous over can be calculated multiplying snow depth with bulk density Regression models...

<p>In snow-prone regions, snowmelt is one of the main drivers runoff. For operational flood forecasting and mitigation, spatial distribution snow water equivalent (SWE) in near real time necessary. In this context, situ observations SWE provide a valuable information. Nonetheless, high variability snowpack characteristics makes it necessary to implement some kind modelling get spatially continuous estimation. Data assimilation thus useful approach combine information from both...

<p>Global or large-scale hydrological forecasting systems covering entire countries, continents and even the planet are growing in popularity. As more emerge, it is likely that they will co-exist with pre-existing local systems. It case for instance Canada, where most provinces have their own streamflow system, while new NSRPS eventually cover whole country using a 1km by grid. Those province, Quebec, built on models configured river catchments rather than regular Using this...

<p>Particle filtering is interesting for snow data assimilation because of its minimal assumptions. However, implementing a particle filter over large spatial domain challenging many reasons. For instance, the number required particles rises exponentially as size increases. Another important issue when spatializing creation discontinuities resampling at locations where observations are available. In this presentation, we will describe how implemented spatialized portion...

<p>In recent years, the number of large-scale hydrological forecasting systems has been steadily growing. This may lead to regions having numerous models spatially overlapping each other. Some these have what we will refer as a regional, more specialized, model for area that performs generally better than their counterpart, considering coarser spatial resolution and sometimes lack calibration latter. Our work explored possibility using simple methods retrieve information from...

Abstract. The use of particle filters for data assimilation is increasingly popular because its minimal assumptions. Nevertheless, implementing a filter over domains large spatial dimensions remains challenging, as the number required particles rises exponentially domain size increases. A common solution to overcome this issue localize and consider collection local applications rather than single regional one. Although can solve dimensionality limit, it also create some discontinuity inside...