Evaluation of the WRF-UCM mesoscale model and ECMWF global operational forecasts over the Paris region in the prospect of tracer atmospheric transport modeling

Jinghui Lian, Lin Wu, François‐Marie Bréon, Grégoire Broquet, Robert Vautard, T. Scott Zaccheo, J. T. Dobler, Philippe CiaisElementa 2018 · DOI

TL;DR

This study evaluates the WRF mesoscale model against ECMWF operational forecasts for simulating atmospheric conditions over Paris, crucial for urban CO2 emission monitoring. The findings suggest that WRF, when enhanced with local data assimilation and urban physics, offers a marginal but significant advantage over ECMWF, particularly for capturing urban-specific atmospheric gradients and wind direction.

What Problem Does This Paper Address?

Accurately quantifying greenhouse gas emissions from cities is vital for climate action. This requires precise atmospheric transport models to interpret ground-based measurements. However, complex urban environments significantly alter local weather patterns (e.g., heat islands, modified wind flow), which coarser global models might not capture adequately. The central question is whether high-resolution mesoscale models, like WRF, incorporating urban effects and local data, can provide a more accurate representation of atmospheric transport than readily available global forecasts, specifically for a city like Paris. This improved accuracy is essential for "top-down" emission estimation methods.

Methods

The researchers employed the WRF (Weather Research and Forecasting) model, specifically version 3.7.1, coupled with the Single-Layer Urban Canopy Model (UCM). They conducted simulations with varying configurations, including:

  • Control Run (REF): A standard WRF setup.
  • WRF with Objective Analysis (WRF_OA): This configuration incorporated multi-nudging techniques (grid, surface, and observation nudging) to assimilate local weather data.
  • WRF without Objective Analysis (WRF_noOA): This served as a baseline to assess the impact of data assimilation.

They also performed sensitivity tests by varying WRF's physics schemes (microphysics, radiation, PBL, etc.) to understand their influence on model outputs. The performance of these WRF configurations was compared against ECMWF's high-resolution operational forecasts (HRES) and ground-based meteorological observations. Key evaluation metrics included correlation coefficient (R), root-mean-square error (RMSE), and mean bias error (MBE).

Datasets

  • Meteorological Observations: Data from Météo France and the SIRTA research platform in the Île-de-France (IdF) region were used. This included hourly air temperature (1.5m), wind speed and direction (10m), and twice-daily radiosonde profiles (temperature and wind) from Trappes.
  • ERA-Interim Global Re-analyses: Provided by ECMWF (0.75° × 0.75° resolution) for boundary and initial conditions for WRF.
  • CORINE Land Cover Dataset (250m): Used to update the land use information in WRF, replacing older USGS data.
  • ASTER Global Digital Elevation Model Version 2 (GDEM v2): Used for higher-resolution topography data.
  • ECMWF High-Resolution Operational Forecasts (HRES): Used as a benchmark for comparison.

Key Insights and Ideas

  • Nudging is Crucial: The study strongly emphasizes that assimilating local weather data through multi-nudging techniques (WRF_OA) is essential for WRF to achieve good agreement with observations, significantly outperforming WRF without such assimilation (WRF_noOA).
  • Urban Effects Matter: WRF's ability to incorporate urban canopy models and higher resolution allows it to better represent urban-specific phenomena like the urban heat island effect and associated temperature gradients, which coarser models like ECMWF struggle with.
  • PBL Schemes Drive Variability: The choice of Planetary Boundary Layer (PBL) schemes within WRF has a substantial impact on simulated wind speed and PBL height, highlighting the importance of selecting appropriate schemes for specific applications.
  • Trade-offs in Nudging: While strong nudging improves agreement with observations, it can also reduce the model's ability to capture natural atmospheric variability and spatial gradients, suggesting a need to balance accuracy with dynamic realism.

Key Results

  • Temperature: Both WRF_OA and ECMWF forecasts reproduce surface temperature well, with correlation coefficients above 0.92 and MBEs within 1°C. WRF_OA shows a slight advantage in capturing temperature gradients between urban and rural areas.
  • Wind Speed: Correlation coefficients for wind speed are similar between WRF_OA and ECMWF (around 0.76-0.85), but ECMWF generally exhibits lower RMSE and MBE. However, WRF_OA shows better performance in representing wind direction.
  • Wind Direction: WRF_OA significantly outperforms ECMWF in reproducing wind direction, especially at suburban and rural sites.
  • PBL Height: WRF's PBL height simulations are highly sensitive to the chosen PBL schemes. WRF_OA generally shows better agreement with observations than ECMWF, though both tend to overestimate PBL height.
  • Spatial Gradients: WRF_OA demonstrates a better ability to reproduce fine-scale horizontal gradients of temperature and wind speed compared to ECMWF, which is crucial for understanding urban atmospheric dynamics.

Limitations and Future Work

  • Limited Spatial/Temporal Scope: The study focused on a single month (February 2015) and the Île-de-France region. While a one-year simulation was conducted for supplementary analysis, a more extensive multi-year evaluation would strengthen the conclusions.
  • Nudging's Impact on Dynamics: The authors acknowledge that strong nudging, while improving accuracy, can suppress the model's inherent dynamics and reduce its sensitivity to physics schemes. They suggest adjusting nudging strengths for optimal balance.
  • Future Work: The authors propose further testing of WRF against ECMWF with prescribed CO2 flux maps to compare modeled CO2 concentrations. They also recommend a long-term (one-year) modeling study of CO2 concentrations in the Paris area using various WRF parameterizations and comparing them with observational data.

Summary updated

Abstract

The quantification of CO2 emissions from cities using atmospheric measurements requires accurate knowledge of the atmospheric transport. Complex urban terrains significantly modify surface roughness, augment surface energy budgets, and create heat islands, all of which lead to lower horizontal winds and enhanced convection over urban areas. The question remains whether these processes should be included in atmospheric transport models that are used for city scale CO2 inversion, and whether they need to be tailored on a city basis. In this study, we use the WRF model over Paris to address the following research question: does WRF runs at a 3 km resolution, including urban effects and the assimilation of local weather data, perform better than ECMWF forecasts that give fields at 16 km resolution? The analysis of model performances focuses on three variables: air temperature, wind and the planetary boundary layer (PBL) height. The results show that the use of objective analysis and nudging tools are required to obtain good agreements between WRF simulated fields with observations. Surface temperature is well reproduced by both WRF and ECMWF forecasts, with correlation coefficients with hourly observations larger than 0.92 and MBEs within 1°C over one month. Wind speed correlations with hourly observations are similar for WRF (range 0.76~0.85 across stations) and ECMWF (0.79~0.84), but the associated RMSEs and MBEs are better for ECMWF. Conversely, WRF outperforms ECMWF forecasts for its description of wind direction, horizontal and vertical gradients. Sensitivity tests with different WRF physics schemes show that the wind speed and the PBL height are strongly influenced by PBL schemes. The marginal advantage of WRF over ECMWF for the desired application is sufficient to motive additional testing with prescribed CO2 flux maps for comparing modeled CO2 concentrations with available observations in an urban environment.
BibTeX
@inproceedings{lian2018evaluation,
  author = {Jinghui Lian and Lin Wu and François‐Marie Bréon and Grégoire Broquet and Robert Vautard and T. Scott Zaccheo and J. T. Dobler and Philippe Ciais},
  title = {{Evaluation of the WRF-UCM mesoscale model and ECMWF global operational forecasts over the Paris region in the prospect of tracer atmospheric transport modeling}},
  booktitle = {Elementa},
  year = {2018},
  doi = {10.1525/elementa.319},
}
Figure