RDPS

1. Overview of RDPS v9.0.0

  • Implemented on June 11, 2024, at the Canadian Meteorological Centre (CMC).

  • Part of Innovation Cycle 4 (IC-4), integrating upgrades across all CMC prediction systems.

  • Switch from a regional limited-area model (LAM) assimilation strategy to a global 10 km resolution framework, using background fields from GDPS-9.0.0.

  • Provides regional-scale numerical weather predictions (NWP), delivering fine-resolution forecasts for users on a 10 km latitude-longitude grid.

Key Features:

  • Model Framework: Global Environmental Multiscale (GEM) model v5.2.0.

  • Resolution: 10 km horizontal grid with 84 vertical levels.

  • Forecast Frequency: Four forecasts per day at 00, 06, 12, and 18 UTC, extending 48 hours into the future.

  • Primary Purpose: Provides detailed short-term forecasts for meteorologists, emergency planners, and industries reliant on high-resolution weather data.

2. Data Assimilation & Objective Analysis

2.1 Assimilation Strategy

RDPS-9.0.0 adopts a Four-Dimensional Ensemble-Variational (4DEnVar) assimilation strategy. Unlike previous versions, continuous assimilation cycles have been removed, and single analyses are performed every 6 hours.

  • Assimilation Window: 6-hour period centered on each synoptic hour (00, 06, 12, 18 UTC).

  • Data Cutoff: Observations are ingested up to 2 hours after the synoptic time (T+2h cutoff).

  • Background Fields: 15 km GDPS-9.0.0 outputs, generated every 3 to 9 hours with updates every 15 minutes.

2.2 Observational Data Sources

To enhance forecast accuracy, RDPS ingests diverse real-time observations:

  1. Satellite Data

    • Microwave Radiance Sensors: AMSU-A, MHS, SSMIS, ATMS, MWHS-2.

    • Infrared Sensors: IASI, CrIS (AIRS no longer used).

    • Geostationary Satellite Imagers.

    • GPS Radio Occultation (GPS-RO).

  2. Ground-Based Observations

    • Radiosondes (weather balloons) for temperature, wind, and humidity profiles.

    • Surface Weather Stations (SYNOP, METAR, BUOY, SHIP).

    • Aircraft-Based Observations (AMDAR, TAMDAR).

    • Scatterometer Winds from ASCAT.

  3. Quality Control & Bias Corrections

    • Satellite radiance bias correction using GDPS-derived coefficients.

    • Dynamic bias correction applied to aircraft temperature data.

    • Updated hyperspectral infrared quality control to handle albedo changes.

    • Cloud-affected (all-sky) radiance assimilation for AMSU-A, ATMS, and MHS.

3. Forecast Model Configuration

RDPS-9.0.0 leverages the GEM model version 5.2 for regional forecasts.

3.1 Key Model Features

  • Model Grid: Global Yin-Yang grid at 10 km resolution.

  • Numerical Scheme: Semi-Lagrangian iterative-implicit solver (450-second timestep).

  • Output Grid: Rotated latitude-longitude regional grid (covering North America and adjacent oceans).

3.2 Physics & Parameterizations

  • Deep Convection: Updated Kain-Fritsch scheme for more accurate storm forecasting.

  • Shallow Convection: Bechtold mass-flux scheme.

  • Cloud Microphysics: Two-moment scheme for cloud formation & precipitation modeling.

  • Boundary Layer Mixing: Uses Turbulent Kinetic Energy (TKE) approach for atmospheric turbulence simulation.

3.3 Land Surface Modeling

  • Land surface physics simulated with ISBA (Interactions between Soil, Biosphere, and Atmosphere).

  • Includes snow depth, sea ice fraction, and soil temperature predictions.

  • Sea Surface Temperature (SST) and ice thickness derived from GDPS & Global Ice Ocean Prediction System (GIOPS).

4. Computational Performance

  • Runs on 3,600 computing cores, completing each analysis step in ~20 minutes.

  • Nested-grid approach improves regional forecast precision.

  • Parallel computing optimizations reduce processing time.

5. Applications and Use Cases

RDPS-9.0.0 is a critical tool for regional weather prediction, supporting various domains:

  • Severe Weather Prediction: Hurricanes, storms, heatwaves, winter storms.

  • Aviation & Transportation: Provides high-resolution wind, visibility, and turbulence forecasts.

  • Energy & Infrastructure: Supports renewable energy planning (solar, wind), power grid management.

  • Agriculture & Water Resource Management: Forecasts for precipitation, drought risk, frost events.

  • Disaster Management & Emergency Response: Real-time severe weather warnings and impact analysis.

6. Summary & Importance

The Regional Deterministic Prediction System (RDPS) v9.0.0 is a state-of-the-art short-term forecasting model, enhancing Canada’s ability to predict extreme weather events with improved precision. Its high-resolution forecasts, advanced data assimilation techniques, and strong computational efficiency make it an essential tool for meteorologists, emergency planners, and climate adaptation initiatives.

7. References & Further Reading

Conclusion

RDPS v9.0.0 significantly advances regional weather forecasting by leveraging high-resolution modeling, ensemble data assimilation, and real-time observational inputs. The improvements in cloud microphysics, turbulence modeling, and radiance assimilation contribute to more accurate and timely forecasts that are crucial for disaster resilience, energy management, agriculture, and aviation safety.

PreviousGDPSNextRDPS-Vertical

Last updated

Was this helpful?