Mesoscale Observing System Simulation Experiments (OSSE)

 

 

GOAL

 

To assess the value of environmental observing systems to operational mesoscale numerical weather forecasts in a controlled software environment.

 

Future observing systems can be tested using projected instrument characteristics.

PROCEEDURES

 

Observations are synthesized from forecasts generated by a sophisticated numerical prediction model that is independent of the operational assimilation system being used. These forecasts are referred to as the "true", or "nature" atmosphere.

 

The forecast model used to produce the "nature" atmosphere must have a known performance history, and must be calibrated against reality.

 

The observations that are synthesized from the "nature" atmosphere must mimic, as close as possible, those observations from the real observing system that is being evaluated.

This OSSE is being conducted over a limited area domain. The assimilating forecast model was isolated from the influence of pre-specified lateral boundaries.

 

 

Pilot Experiment

 

 

HYPOTHESIS:

Information from a geostationary-based interferometer will significantly improve the accuracy of numerical weather forecasts over the current stationary radiometer.

 

Temperature and moisture retrievals are simulated by superimposing estimated observation errors on the "true" profiles generated by the "nature" run.

 

OSSE Design

 

An OSSE can be subdivided into four basic steps:

1) Generate a "nature" atmosphere

2) Compute synthetic observations

3) Assimilate the synthetic observations

4) Assess the impact on the resulting forecast.

 

Each step is performed with the goal of minimizing any external influences, which may compromise the value of the synthesized observations, the assimilation process, or the results of the numerical forecasts.

 

1. Generate a "nature" atmosphere using the University of Wisconsin, Nonhydrostatic Modeling System.

 

Horizontal resolution = 60 km

The horizontal domain is chosen to be as large as practical to isolate the influence of the pre-defined lateral boundary conditions.

 

Boundary conditions = NCEP Eta forecast model, AWIPS 104 grid.

Ideally, the "nature" atmosphere should be two to four times the resolution of the simulated observing system.

 

Vertical resolution = 38 levels

The model vertical resolution is chosen to be a minimum of two-times the resolution of the observing system to be simulated.

 

A 12hr forecast was be generated to allow the model to "spin up".

 

2. Simulate observations

 

Temperature and moisture profiles from the "true" atmosphere are modified using realistic observation errors.

 

Profiles of temperature and moisture are generated at hourly intervals over the 12-hour analysis period.

 

A cloud mask is used to simulate gaps in the coverage.

 

 

3. Assimilate the synthesized observations

 

The operational 40km Rapid Update Cycle (RUC) was used to assimilate the observations at hourly intervals.

 

Boundary conditions: NCEP Eta model, projected onto the AWIPS 211 grid (80km resolution).

 

 

Four assimilation experiments were performed:

 

1) No observations (NO)

 

2) Perfect observation experiment (PO) assimilates profiles extracted directly from the "nature" run

 

3) Geostationary radiometer (GR) experiment assimilates profiles adjusted to emulate a GOES-type system.

 

4) Geostationary interferometer experiment (GI) assimilates profiles from a proposed geostationary interferometer.

 

Note: The NO and PO experiments represent the range of performance that can be expected from the RUC.

 

 

4. Impact Assessment

 

The impact of the observations will be assessed by objectively measuring the ability of each observing system to steer the resulting 12-hour forecasts toward the "true" atmosphere.

 

PRELIMINARY RESULTS

 

The root-mean-square (RMS) error and S1 skill score were computed for each experiment against 'perfect observations' from the 'nature' run.

 

Profile Count Experiment

None No Observation (NO) = NO OBS

 

~ 5000 Geostationary Radiometer (GR) = GEO-R

 

~ 5000 Geostationary Interferometer (GI) = GEO-I

 

~ 6200 Perfect Observations (PO) = OPTIMAL

 

GI results are significantly improved over those from the GR.

 

500 hPa temperature errors are reduced by 0.2 C root mean square (rms) over the extended CONUS (contiguous United States) and 700 hPa relative humidity errors are reduced by 2%.

 

To assess the relative benefit of the geostationary interferometer over the geostationary radiometer a rating score (1 to 10) was computed. The RMS errors for temperature and relative humidity were summed over four layers (700hPa, 500hPa, 400hPa, 300hPa) and normalized between the RMS error sums from the No Observation (NO) run and the Perfect Observation (PO) run. A score of 10 is perfect.

 

FUTURE DIRECTIONS

 

20 km UW-NMS "nature" forecasts

 

14 day test periods (winter and spring)

 

Conventional observing system evaluation

 

Radiance assimilation (3D-Var)

 

Geostationary wind OSSE (IR and interferometer)

 

Low Earth Orbit (LEO) OSSE