The ESRL website will be unavailable for 24 hours starting Friday, March 6th at 5:00pm MT due to building maintenance.

Validation and Improvement of Forecasting Models

Surface momentum and thermal fluxes

PACJET will provide opportunities to evaluate and improve physics in operational forecasting models related to property transport between the surface and the air.

For example, quantitative precipitation forecasts in the vicinity the US West Coast are sensitive to the water vapor transfer that occurs horizontally (from offshore to coast) and vertically (between the sea surface and the air). It has been recognized that the capability of numerical models in describing water vapor transfer at the air-sea interface is limited by insufficient knowledge of physical property exchange between the surface and the air associated with strong weather events.

When a weather system approaches the coast, its interaction with the sea and coastal topography is very dynamic. Offshore in light wind conditions, the sea surface and air-sea interaction is relatively simple because there is little wave breaking, but under high wind conditions or near the coast, wave breaking, sea spray, surface currents, and differences between wave and wind directions all contribute to the air-sea interaction and create a complex challenge to modeling. On land, the topography and land surface characteristics are highly irregular on many spatial scales from meters to kilometers.

Properly simulating the complex interaction of the weather system with the sea and the coast is a major challenge in meteorology. PACJET will combine airborne and ship-based flux measurements and will include intensive wave observations. The intercomparison of airborne and ship measurements require different assumptions about the spatial and temporal sampling of fluxes. A similar intercomparison will be performed on airborne scanning radar altimeter measurements and ship borne measurements of waves.

Cloud physics

The skill of numerical models in fog and quantitative precipitation forecasts near the US West Coast is limited by uncertainties in the understanding and parameterization of cloud microphysics when moist maritime air flow interacts with complex coastal mountains. Of particular interest is the quantification from CALJET that warm rain processes can produce rain rates in the coastal mountains that are capable of producing flooding. PACJET will further explore this phenomena using an enhanced observational approach. These results will improve understanding of microphysical behavior, and will feed back into parameterization testing.

Using Ensembles in Quantitative Precipitation Forecasting

One approach to improve coastal storm and rainfall prediction is the so-called ensemble prediction. This approach is based on the assumption that the skill of a single deterministic model forecast is limited by many uncertainties, particularly in initial conditions and in model formulations, while the average skill of an ensemble of different model predictions will be better than any of the individual members. PACJET will provide comprehensive data sets that can be used to evaluate different ensemble techniques.

Assimilation of Targeted Observations

It has been recognized that targeted observations in areas where routine observations are not available can improve the accuracy of weather prediction for a specific region and lead time.

However, many questions remain.

  • How best identify and sample a target?
  • How large the impact can be?
  • How sensitive the results are to the temporal and spatial scales of the observations?
  • What are the best data assimilation approaches for ingesting the targeted observations into numerical models?

PACJET will provide observations that can be used for answering these questions. This will be done in quasi-real time using parallel runs of key models, including the MRF run at NCEP, an experimental version of the RUC that covers the eastern Pacific and is run at FSL, and a version of MM5 run at ETL. For each model, the difference between a run with and without the experimental data will be calculated, and thus the impact of the data on NWP will be assessed.