Search Results

You are looking at 1 - 10 of 10 items for

  • Author or Editor: Bernard Bilodeau x
  • All content x
Clear All Modify Search
Jean-François Mahfouf and Bernard Bilodeau

Abstract

The adjoint version of the Global Environmental Multiscale model including a comprehensive package of simplified and linearized physical processes (large-scale condensation, deep moist convection, vertical diffusion, and subgrid-scale orographic effects) is used to evaluate the sensitivity of surface precipitation to initial conditions for up to 24 h for two meteorological systems: a midlatitude front and a tropical cyclone. Such diagnostics are useful to improve the understanding on variational assimilation of precipitation data. In agreement with a similar study, the largest sensitivity is found with respect to the temperature field for both stratiform and convective precipitation. Close to the observation time and for stratiform precipitation, the sensitivity with respect to specific humidity is rather large, which corroborates conclusions from previous one-dimensional variational data assimilation experimentations. The sensitivity is then reduced significantly after the observation time. The sensitivities of surface precipitation to the wind components and to specific humidity are comparable and are at a maximum at the observation time. The sensitivity to the surface pressure is always much smaller than the sensitivity to the other variables. In general, sensitivities are largest at the observation time and then decrease. However, for the midlatitude perturbation, the sensitivity is enhanced after 12 h for stratiform precipitation and also for convective precipitation using a scheme based on the moisture convergence closure. This results from a dynamical coupling upstream of the area of interest through baroclinic instability as evidenced by vertically backward-tilted sensitivities. Such enhancement is not observed for the tropical case. The tangent-linear approximation remains acceptable for accumulated precipitation up to 24 h but is rather poor for instantaneous rain rates. The variational assimilation of accumulated precipitation should thus be favored.

Full access
Marco L. Carrera, Stéphane Bélair, and Bernard Bilodeau

Abstract

The Canadian Land Data Assimilation System (CaLDAS) has been developed at the Meteorological Research Division of Environment Canada (EC) to better represent the land surface initial states in environmental prediction and assimilation systems. CaLDAS is built around an external land surface modeling system and uses the ensemble Kalman filter (EnKF) methodology. A unique feature of CaLDAS is the use of improved precipitation forcing through the assimilation of precipitation observations. An ensemble of precipitation analyses is generated by combining numerical weather prediction (NWP) model precipitation forecasts with precipitation observations. Spatial phasing errors to the NWP first-guess precipitation forecasts are more effective than perturbations to the precipitation observations in decreasing (increasing) the exceedance ratio (uncertainty ratio) scores and generating flatter, more reliable ranked histograms. CaLDAS has been configured to assimilate L-band microwave brightness temperature TB by coupling the land surface model with a microwave radiative transfer model. A continental-scale synthetic experiment assimilating passive L-band TBs for an entire warm season is performed over North America. Ensemble metric scores are used to quantify the impact of different atmospheric forcing uncertainties on soil moisture and TB ensemble spread. The use of an ensemble of precipitation analyses, generated by assimilating precipitation observations, as forcing combined with the assimilation of L-band TBs gave rise to the largest improvements in superficial soil moisture scores and to a more rapid reduction of the root-zone soil moisture errors. Innovation diagnostics show that the EnKF is able to maintain a sufficient forecast error spread through time, while soil moisture estimation error improvements with increasing ensemble size were limited.

Full access
Natacha B. Bernier, Stéphane Bélair, Bernard Bilodeau, and Linying Tong

Abstract

A high-resolution 2D near-surface and land surface model was developed to produce snow and temperature forecasts over the complex alpine region of the Vancouver 2010 Winter Olympic and Paralympic Games. The model is driven by downscaled operational outputs from the Meteorological Service of Canada’s regional and global forecast models. Downscaling is applied to correct forcings for elevation differences between the operational forecast models and the high-resolution surface model. The high-resolution near-surface and land surface model is then used to further refine the forecasts. The model was validated against temperature and snow depth observations. The largest improvements were found in regions where low-resolution (i.e., on the order of 10 km or more) operational models typically lack the spatial resolution to capture rapid elevation changes. The model was found to better reproduce the intermittent snow cover at low-lying stations and to reduce snow depth error by as much as 3 m at alpine stations.

Full access
Marco L. Carrera, Stéphane Bélair, Vincent Fortin, Bernard Bilodeau, Dorothée Charpentier, and Isabelle Doré

Abstract

To improve the representation of the land surface in their operational numerical weather prediction (NWP) models, the Meteorological Research Division of Environment Canada (EC) is developing an external hydrometeorological modeling and data assimilation system. The objective of this study is to verify the improvement in simulating snow cover extent (SCE) and snow water equivalent (SWE) over the Canadian Rockies with this new modeling system. This study will be an important first step in determining the optimal configuration of the land surface model and atmospheric forcing for a future operational implementation. Simulated SCE is compared with the Interactive Multisensor Snow and Ice Mapping System (IMS) analysis, while simulated SWE values are verified against a series of manual snow survey sites located within the Canadian Rockies. Results show that land surface model simulations of SCE and SWE were sensitive to precipitation forcing. Simulations at both low and high resolution forced with EC’s experimental precipitation analysis were found to underestimate SCE and SWE values. Mountain snowpack retreated too early during the spring melt period. Precipitation forcing derived from EC’s short-range NWP model resulted in improved values for both SCE and SWE, which also contributed to higher contributions to streamflow. Terrain adjusting the atmospheric forcing data was found to be important for properly modeling local extreme SWE values. A comparison with available precipitation observations over the Canadian Rockies region found EC’s experimental precipitation analysis to possess a negative precipitation bias that increases with increasing elevation.

Full access
Stéphane Bélair, Ross Brown, Jocelyn Mailhot, Bernard Bilodeau, and Louis-Philippe Crevier

Abstract

The performance of a modified version of the snow scheme included in the Interactions between Surface–Biosphere–Atmosphere (ISBA) land surface scheme, which was operationally implemented into the regional weather forecast system at the Canadian Meteorological Centre, is examined in this study. Stand-alone verification tests conducted prior to the operational implementation showed that ISBA's new snow package was able to realistically reproduce the main characteristics of a snow cover, such as snow water equivalent and density, for five winter datasets taken at Col de Porte, France, and at Goose Bay, Newfoundland, Canada. A number of modifications to ISBA's snow model (i.e., new liquid water reservoir in the snowpack, new formulation of snow density, and melting effect of incident rainfall on the snowpack) were found to improve the numerical representation of snow characteristics.

Objective scores for the fully interactive preimplementation tests carried out with the Canadian regional weather forecast model indicated that ISBA's improved snow scheme only had a minor impact on the model's ability to predict atmospheric circulation. The objective scores revealed that only a thin atmospheric layer above snow-covered surfaces was influenced by the change of land surface scheme, and that over these regions the essential behavior of the atmospheric model was not significantly altered by improvements to the treatment of snow cover. It was shown that this lack of response was most likely related to the treatment of the snow cover fraction in each atmospheric model grid tile. The estimation of snow cover fraction relied on simple formulations that were dependent on poorly known parameters, such as the fractional coverage of vegetation. Results showed that uncertainties of only 15% in vegetation fractional coverage could be responsible for uncertainties of as much as 1–1.5 K in screen-level air temperature. This indicates that some care must be exercised in the specification of vegetation and snow cover fractional coverage.

Full access
Maziar Bani Shahabadi, Stéphane Bélair, Bernard Bilodeau, Marco L. Carrera, and Louis Garand

Abstract

A new ensemble-based land surface data assimilation (DA) system is coupled with the atmospheric four-dimensional ensemble-variational data assimilation (4D-EnVar) system with the goal of improving the analyses within Environment and Climate Change Canada’s Global Deterministic Prediction System. Since 2001, the sequential assimilation of surface variables is used to generate the initial conditions to launch the Global Environmental Multiscale (GEM) coupled forecast model. The work presented here is to replace the sequential DA with an independent surface DA system, the Canadian Land Data Assimilation System (CaLDAS) assimilating screen-level observations, and to compare assimilation experiments with CaLDAS run in uncoupled and weakly coupled modes. In the uncoupled mode, CaLDAS is used to initialize the forecast without interacting with the 4D-EnVar system. In the coupled mode, the analyses generated from CaLDAS and 4D-EnVar are used to initialize the forecast model. The analyses and forecasts from uncoupled and coupled runs are evaluated against surface and radiosonde observations over different subdomains to conclude the impact of coupling CaLDAS with 4D-EnVar. Results indicate a statistically significant reduction in bias and standard deviation at the surface for screen-level temperature and dewpoint temperature on the order of 0.1 K, and in the lower troposphere between 1000 and 500 hPa on the order of 0.1 dam for geopotential height and 0.1 K for air temperature and dewpoint depression in the coupled DA runs. The positive impact persists up to 5 days over some subdomains. It is concluded that the coupled DA approach generally performs better than the uncoupled version.

Open access
Stéphane Bélair, Louis-Philippe Crevier, Jocelyn Mailhot, Bernard Bilodeau, and Yves Delage

Abstract

The summertime improvement resulting from the operational implementation of a new surface modeling and assimilation strategy into the Canadian regional weather forecasting system is described in this study. The surface processes over land are represented in this system using the Interactions between Soil–Biosphere–Atmosphere (ISBA) land surface scheme. Surface variables, including soil moisture, are initialized using a sequential assimilation technique in which model errors of low-level air temperature and relative humidity are used to determine analysis increments of surface variables.

It was found that the magnitude and nature of the analysis increments applied to the surface variables depended on the surface and meteorological conditions observed in each region. In regions characterized by weak meteorological activity (i.e., no clouds or precipitation), model errors of low-level air characteristics are more likely to be related to an incorrect representation of surface processes due to either erroneous initial conditions or inaccurate parameterizations in the land surface scheme. In other regions characterized by more frequent and more intense precipitation events, surface corrections are mainly associated with inaccurate atmospheric forcing.

Objective evaluation against observations from radiosondes and surface stations showed that the amplitude of the diurnal cycle of near-surface air temperature and humidity is larger with the new surface system, in better agreement with observations. This type of improvement was found to extend higher up in the boundary layer (up to 700 hPa) where cold and humid biases were significantly reduced by introducing the new surface system. The model precipitation was also found to be significantly influenced by the new representation of surface fluxes. The problematic increase of a positive bias in precipitation with integration time was found to be significantly reduced with the new system, due to the warmer and drier boundary layer.

Full access
Stéphane Bélair, André Méthot, Jocelyn Mailhot, Bernard Bilodeau, Alain Patoine, Gérard Pellerin, and Jean Côté

Abstract

The objective and subjective evaluations that led to the implementation of the Fritsch and Chappell (FC) convective scheme in the new 24-km Canadian operational regional model are described in this study. Objective precipitation scores computed for a series of 12 benchmark cases equally distributed throughout all seasons and for a parallel preimplementation run of the new version of the model during summer 1998 show the positive impact of increasing the horizontal resolution and of including the FC scheme (instead of the Kuo scheme used in the previous version of the operational model). The comparison is particularly in favor of the FC configuration for the summertime parallel preimplementation run, with improved biases and threat scores, while it is nearly neutral for the 12 benchmark cases comprised mostly of large-scale weather systems.

Examination of a summertime case study confirms the superiority of FC over Kuo for the numerical representation of the structure and evolution of mesoscale convective systems. A wintertime case study, on the other hand, reveals that precipitation patterns with the two model configurations are quite similar, even though the FC scheme is essentially inactive for weather systems organized on such large scales. In contrast with the Kuo simulation, most of the precipitation occurs on the grid scale when using FC. This different partitioning of precipitation into implicit and explicit components is more consistent with the mesoscale-resolving capabilities of the model. It is also observed that the new model physics gives rise to more realistic deepening of coastal large-scale depressions.

The different implicit/explicit partitioning for Kuo and FC is clearly exposed with precipitation statistics from the 12 benchmark cases. With Kuo, it is found that implicit precipitation is produced over areas as large as (and even larger than) that associated with grid-scale precipitation; it is also shown that with this configuration most of the precipitation occurs at weak rates and is mainly produced by the implicit scheme. The results with FC are more realistic, in the sense that convective precipitation only covers a small fraction of the model domain (i.e., 1%–2%) and that both precipitation schemes are dominant in their respective areas, that is, weak precipitation for the explicit scheme and more intense precipitation for the implicit scheme.

Full access
Marco L. Carrera, Bernard Bilodeau, Stéphane Bélair, Maria Abrahamowicz, Albert Russell, and Xihong Wang

Abstract

This study examines the impacts of assimilating Soil Moisture Active Passive (SMAP) L-band brightness temperatures (TBs) on warm season short-range numerical weather prediction (NWP) forecasts. Focusing upon the summer 2015 period over North America, offline assimilation cycles are run with the Canadian Land Data Assimilation System (CaLDAS) to compare the impacts of assimilating SMAP TB versus screen-level observations to analyze soil moisture. The analyzed soil moistures are quantitatively compared against a set of in situ sparse soil moisture networks and a set of SMAP core validation sites. These surface analyses are used to initialize a series of 48-h forecasts where near-surface temperature and precipitation are evaluated against in situ observations. Assimilation of SMAP TBs leads to soil moisture that is markedly improved in terms of correlation and standard deviation of the errors (STDE) compared to the use of screen-level observations. NWP forecasts initialized with SMAP-derived soil moistures exhibit a general dry bias in 2-m dewpoint temperatures (TD2m), while displaying a relative warm bias in 2-m temperatures (TT2m), when compared to those forecasts initialized with soil moistures analyzed with screen-level temperature errors. Largest impacts with SMAP are seen for TD2m, where the use of screen-level observations leads to a daytime wet bias that is reduced with SMAP. The overall drier soil moisture leads to improved precipitation bias scores with SMAP. A notable deterioration in TD2m STDE scores was found in the SMAP experiments during the daytime over the Northern Great Plains. A reduction in the daytime TD2m wet bias was found when the observation errors for the screen-level observations were increased.

Open access
Camille Garnaud, Stéphane Bélair, Marco L. Carrera, Chris Derksen, Bernard Bilodeau, Maria Abrahamowicz, Nathalie Gauthier, and Vincent Vionnet

Abstract

Because of its location, Canada is particularly affected by snow processes and their impact on the atmosphere and hydrosphere. Yet, snow mass observations that are ongoing, global, frequent (1–5 days), and at high enough spatial resolution (kilometer scale) for assimilation within operational prediction systems are presently not available. Recently, Environment and Climate Change Canada (ECCC) partnered with the Canadian Space Agency (CSA) to initiate a radar-focused snow mission concept study to define spaceborne technological solutions to this observational gap. In this context, an Observing System Simulation Experiment (OSSE) was performed to determine the impact of sensor configuration, snow water equivalent (SWE) retrieval performance, and snow wet/dry state on snow analyses from the Canadian Land Data Assimilation System (CaLDAS). The synthetic experiment shows that snow analyses are strongly sensitive to revisit frequency since more frequent assimilation leads to a more constrained land surface model. The greatest reduction in spatial (temporal) bias is from a 1-day revisit frequency with a 91% (93%) improvement. Temporal standard deviation of the error (STDE) is mostly reduced by a greater retrieval accuracy with a 65% improvement, while a 1-day revisit reduces the temporal STDE by 66%. The inability to detect SWE under wet snow conditions is particularly impactful during the spring meltdown, with an increase in spatial RMSE of up to 50 mm. Wet snow does not affect the domain-wide annual maximum SWE nor the timing of end-of-season snowmelt timing in this case, indicating that radar measurements, although uncertain during melting events, are very useful in adding skill to snow analyses.

Open access