Comparison of verification results of the Numerical Weather Prediction models operated at ICM

At ICM in quasi-operational form two non-hydrostatic numerical weather prediction models are operated, Unified Model developed by UK Meteorological Office and COAMPS (Coupled Ocean/Atmosphere Mesoscale Prediction System) developed by US Naval Research Laboratory. Unified Model has horizontal reso-lution of 4 km and 38 levels in vertical. COAMPS model is configured with three nested grids and horizontal resolutions of 39, 13 and 4.3 km. UM and the second grid of COAMPS cover the area of Central Europe. In this paper results of both NWP models, i.e. COAMPS computed on 13 km mesh resolution and UM were verified against observations from 60 Polish synoptic stations. In the verification process surface observations were compared with the nearest gridpoint forecasts. Only results from 00 UTC model runs were used.For the following meteorological elements: 2m air temperature, wind speed and wind direction at 10 m, mean sea level pressure ME, MAE and RMSE are presented. In case of 12 hour accumulated precipitation contingency tables for selected thresholds were constructed and their results were used to obtain FBI, ETS, POD and FAR. The verification sample covers period since June 2009 to November 2010.     ...

Verification of selected empirical methods for the calculation of minimum and mean flows in catchments of the Dunajec basin

The study was aimed to verify the selected empirical formulas to calculate SNQ, SSQ and Qminp% flows in four catchments of the rivers: Cicha Woda-Biały Dunajec, Czarny Dunajec, Ochotnica and Wielki Rogoźnik which are located in the Dunajec basin. The verification was aimed to assess the usefulness of the analyzed empirical formulas to calculate minimum and mean flows in relation to current hydrological data. The data were obtained from the Institute of Meteorology and Water Management - National Research Institute and include data string from the period of 1980 - 2009. Hydrometric data were evaluated considering their homogeneity and independence using Mann- Kendall-Sneyers test and the trend analysis was performed using Mann-Kendall Statistic S. The Gumbel's method was used to calculate the probability of not exceeding minimum flow thresholds. Based on the obtained results the comparative analysis was conducted between the calculated values of SNQ, SSQ and real flows from the period of 1980 - 2009. The calculations have shown significant differences between the SNQ, SSQ and Qminp% values calculated by empirical formulas and the values determined based on hydrological data. This justifies the need for verification and updating the empirical formulas used previously for the calculation of minimum and ...

Verification of 10- and 20-day forecasts of the standardized precipitation index SPI

In the paper the verification of precipitation condition forecasts are presented. The forecasts have been developed in the monitoring system of water deficiency and excess conducted by the Institute of Technology and Life Sciences. Analyses were performed for seven meteorological stations within Kujawy Region, south-eastern part of Pomerania and western Wielkopolska in the months of the growing season (April-September) in 2013-2014. Actual precipitation deficit and surplus and their 10-day and 20-day forecasts were determined every 10/11 days for the period of 30/31 days using the standardized precipitation index SPI. Two verification criteria were adopted: the difference between actual and predicted SPI cannot be greater than 0.5 and the agreement of the precipitation category classes. The verifiability is determined as the relative frequency of the forecasts meeting these criteria. It was found that the average verifiability of 10-day forecasts in the study area was high and equal to 80% for the assumed criteria. In the case of 20-day forecasts average verifiability was respectively 52%. ...