HRDPS

Overview

The High Resolution Deterministic Prediction System (HRDPS) v7.0.0 is an advanced numerical weather prediction (NWP) system implemented by Environment and Climate Change Canada (ECCC) under the Meteorological Service of Canada (MSC). The HRDPS operates as part of a limited-area model (LAM) configuration within the Global Environmental Multiscale (GEM) model framework.

Key Features:

  • Provides high-resolution forecasts at 2.5 km spatial resolution with 62 vertical levels.

  • Uses a 4D-EnVar data assimilation approach for improved atmospheric analysis.

  • Generates forecasts four times daily with short-term prediction cycles every six hours.

  • Supports early-warning systems for severe weather phenomena such as storms and heatwaves.

  • Data is continuously assimilated, improving forecast accuracy for Days 1 and 2.

1. Data Assimilation and Objective Analysis

Data assimilation is the process of integrating observational data into a numerical model to improve forecasts. The HRDPS v7.0.0 follows a continuous upper-air cycling strategy to refine predictions.

1.1 Assimilation Process

  • Uses 4D-EnVar (Four-Dimensional Ensemble-Variational Data Assimilation), which combines ensemble forecasting with variational data assimilation to improve accuracy.

  • Observations are assimilated every 6 hours (00, 06, 12, 18 UTC) with a 7-hour observation cut-off to ensure real-time updates.

  • Uses a Gaussian grid (~25 km horizontal resolution) to process atmospheric data.

  • Incorporates radiance data from satellite sensors using advanced radiative transfer models.

1.2 Data Sources and Observations

HRDPS ingests a variety of observational data:

  1. Satellite Radiance Data

    • Microwave sensors (e.g., AMSU-A, MHS, ATMS).

    • Infrared sensors (e.g., AIRS, IASI, CrIS).

    • Geostationary imagers.

  2. Ground-Based Observations

    • Radiosondes (weather balloons) provide temperature, pressure, and humidity profiles.

    • Surface weather stations supply real-time temperature, wind, and pressure readings.

    • Aircraft-based observations enhance vertical profiles.

  3. Radar and GPS Data

    • Radar reflectivity helps estimate precipitation rates using Latent Heat Nudging (LHN).

    • GPS-based refractivity data aids in atmospheric moisture profiling.

2. Forecast Model Configuration

The forecasting component of HRDPS 7.0.0 relies on Version 5.2.0 of the GEM model.

2.1 Model Formulation

  • Numerical Grid: Covers Canada, parts of the northern U.S., and adjacent oceans, with a 2.5 km horizontal resolution.

  • Time Integration: Semi-Lagrangian numerical scheme, using a 60-second timestep for efficient computation.

  • Physical Parameterizations:

    • Deep convection: Kain & Fritsch scheme.

    • Shallow convection: Kuo Transient scheme.

    • Cloud and precipitation microphysics: P3 (two-moment scheme).

    • Turbulence & boundary-layer mixing: Based on the Turbulent Kinetic Energy (TKE) approach.

2.2 Land Surface and Hydrology

  • Soil temperature and moisture are provided by the Canadian Land Data Assimilation System (CaLDAS) at 2.5 km resolution.

  • Sea surface temperature and sea-ice data are obtained from the Global Deterministic Prediction System (GDPS-9.0.0).

3. Lateral Boundary Conditions and Initial Surface Conditions

HRDPS requires accurate initial and boundary conditions for reliable forecasting.

  • Lateral Boundary Conditions (LBCs) come from the GDPS-9.0.0, updated hourly at 10 km resolution.

  • Initial Surface Conditions (ISC) are sourced from the CaLDAS assimilation system.

4. Computational Performance

  • The data assimilation step runs on 3,600 computing cores, completing in ~13 minutes per cycle.

  • The forecast model uses a nested-grid approach with finer resolution over key areas.

5. Applications and Impact

HRDPS is critical for operational weather forecasting in Canada. It supports:

  • Short-term severe weather prediction (storms, heatwaves, floods).

  • Air quality and wildfire smoke forecasting.

  • Energy grid demand prediction, especially for extreme heat and cold.

  • Agriculture and hydrology, improving water resource management.

6. Summary and Importance

HRDPS v7.0.0 is a state-of-the-art, high-resolution weather forecasting system that significantly improves Canada’s ability to predict and respond to severe weather events. By using advanced data assimilation, high-performance computing, and real-time observational data, it enhances forecasting for disaster resilience and climate adaptation.

PreviousRDPS-VerticalNextHRDPS-North

Last updated

Was this helpful?