One-Level Modeling for Diagnosing Surface Winds over Complex Terrain. Part II. Applicability to Short-Range Forecasting

View More View Less
  • 1 Department of Geophysics and Planetary Sciences, Tel Aviv University, Israel
© Get Permissions
Full access

Abstract

The modified one-level sigma-surface model used by Alpert and Getenio (1988) to study the surface-averaged summertime flow in Israel is run once with the best available information, without tweaking any parameters, for four events consisting of two winter cases (a Cyprus low and a Siberian high), and two summer cases. In a statistical verification for a total of 436 observations, the average direction error, &Deltaθ, and speed error, ΔV, were only −2.8 deg and 0.11 m s−1, respectively. However, the averaged absolute direction error, i.e., average |Δθ|, was found to be 49.5 deg. Also, the averaged absolute normalized wind speed error, i.e., average |ΔV|/V0 was found to be 37% for all wind intensifies exceeding 4 m s−1. Hence, the model's overall wind forecasts were accurate, but not very precise. Although the statistics contain one case, the Siberian high, which was not well simulated. the results (particularly the wind directions) were found to be good. Characteristics of the model are studied through an analysis of variances. Model results were compared to three other models that diagnosed surface wind vectors over complex terrain.

Results suggest that the model is not only capable of diagnosing many details of surface mesoscale flow, but might also be useful for various applications which require operative short-range prediction of the diurnal changes of high-resolution surface flow over complex terrain. Examples of such applications are locating wildland fires dispersion of air pollutants and prediction of changes in wind energy or of surface wind for low-level air flights.

Abstract

The modified one-level sigma-surface model used by Alpert and Getenio (1988) to study the surface-averaged summertime flow in Israel is run once with the best available information, without tweaking any parameters, for four events consisting of two winter cases (a Cyprus low and a Siberian high), and two summer cases. In a statistical verification for a total of 436 observations, the average direction error, &Deltaθ, and speed error, ΔV, were only −2.8 deg and 0.11 m s−1, respectively. However, the averaged absolute direction error, i.e., average |Δθ|, was found to be 49.5 deg. Also, the averaged absolute normalized wind speed error, i.e., average |ΔV|/V0 was found to be 37% for all wind intensifies exceeding 4 m s−1. Hence, the model's overall wind forecasts were accurate, but not very precise. Although the statistics contain one case, the Siberian high, which was not well simulated. the results (particularly the wind directions) were found to be good. Characteristics of the model are studied through an analysis of variances. Model results were compared to three other models that diagnosed surface wind vectors over complex terrain.

Results suggest that the model is not only capable of diagnosing many details of surface mesoscale flow, but might also be useful for various applications which require operative short-range prediction of the diurnal changes of high-resolution surface flow over complex terrain. Examples of such applications are locating wildland fires dispersion of air pollutants and prediction of changes in wind energy or of surface wind for low-level air flights.

Save