All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 335 83 20
PDF Downloads 204 48 2

An Explicit Cloud Physics Parameterization for Operational Numerical Weather Prediction

Paul SchultzNOAA Forecast Systems Laboratory, Boulder, Colorado

Search for other papers by Paul Schultz in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

In anticipation of computers that will be able to run weather forecasting models on very fine grids fast enough for real-time purposes, an algorithm for representing water phase change and precipitation processes was developed. The design criteria guiding this development are sufficiency (i.e., providing the required services), computational efficiency, and compatibility within four-dimensional data assimilation systems. The implication of the last criterion is that the model's moisture variables need to be consistent with observable or inferable cloud properties.

The algorithm is compared with a well-documented research microphysics algorithm in terms of computing efficiency and agreement with observations. The weather forecasting model runs much faster using the new package instead of the research algorithm. The agreement between the new algorithm and the research algorithm is much better than the agreement between the observations and the results from either algorithm, which suggests that errors in observations or other errors in the model runs (initialization, boundary conditions) are larger sources of error than the microphysics representation.

Abstract

In anticipation of computers that will be able to run weather forecasting models on very fine grids fast enough for real-time purposes, an algorithm for representing water phase change and precipitation processes was developed. The design criteria guiding this development are sufficiency (i.e., providing the required services), computational efficiency, and compatibility within four-dimensional data assimilation systems. The implication of the last criterion is that the model's moisture variables need to be consistent with observable or inferable cloud properties.

The algorithm is compared with a well-documented research microphysics algorithm in terms of computing efficiency and agreement with observations. The weather forecasting model runs much faster using the new package instead of the research algorithm. The agreement between the new algorithm and the research algorithm is much better than the agreement between the observations and the results from either algorithm, which suggests that errors in observations or other errors in the model runs (initialization, boundary conditions) are larger sources of error than the microphysics representation.

Save