Comparing the Results of Precipitation Forecast Based on Mesoscale Models on the Territory of Iran during the Cold Season

M. Azadi, E. Taghizadeh, M. H. Memarian, and L. R. Dmitrieva-Arrago

Presented are the results of the comparison of categorical precipitation forecasts for the territory of Iran using three mesoscale models: HRM, MM5, and WRF-ARW. Given is a short description of the models. The results of the precipitation forecasts for 24, 48, and 72 hours are compared for the period from November 2008 to June 2009 for the whole territory of Iran and for nine separate regions. To estimate the forecast errors, the standard system based on the contingency table is applied. Computed are the criteria of success, false alarm parameters, and hit rate characteristics. The forecast success characteristics of these models differ insignificantly (by 0.01–0.03). All models have a trend towards overestimating the number of cases with precipitation except some cases of moderate and heavy precipitation underestimated by the HRM model. The maximum values of hit rate characteristics and the largest number of false alarms are obtained for the WRF model. The computation of the Pierce criterion demonstrates that the quality of precipitation forecast by the WRF model is higher than that provided by MM5 and HRM models.

Joomla templates by a4joomla