Weather forecasting

Weather forecast is called the probable course of weather phenomena over a given area. The preparation of forecast requires knowledge of the current state of the weather, expressed by the quantity of meteorological elements, the trend of their changes and knowledge of physical processes occurring in the atmosphere. Over the years, forecasts were based on the knowledge and experience of synoptics based on observations from national measurement networks. Numerical and "nowcasting" models are now used for forecasts.

The basis of model work are the so-called initial conditions based on the previous results of forecasts as well as the results of observations and measurements. In numerical models, physical processes occurring in the atmosphere are written in the form of mathematical equations - nonlinear partial differential equations. The models predict the values of meteorological elements for a given grid for different atmospheric levels. All atmospheric processes, that take place on a scale smaller than a given grid, are subjected to parameterization.

"Nowcasting" forecasts are short-term forecasts, usually prepared up to 6 hours ahead. The idea of forecast "for now" is related to the current assimilation of measured data and the assessment of the current state of the atmosphere. Most numerical models allow preparation of a forecast for a few days ahead. Long-term forecasts are usually prepared on the basis of global models (e.g. GFS, NOGAPS, and GEM), however, with the prolongation of the forecasting period, its reliability decreases. Operational forecasts usually use regional models (e.g. WRF, MM5, ALADIN, and COSMO) which spatial resolution usually reaches several kilometres.

On the geoportal page, the forecast results of selected meteorological elements (temperature, relative humidity, atmospheric pressure, wind speed and direction, precipitation and radiation) from the Weather Research and Forecasting Model (WRF ARW) for Poland (https://meteo.plgrid.pl). /index.html) at a 4 km x 4 km spatial resolution are presented. The forecasts are prepared with an hourly time step up to 72 hours ahead. The forecast is based on initial and boundary conditions with the reanalysis of the National Centres for Environmental Prediction (NCEP GFS) available at a 0.5° x 0.5° resolution. The results obtained are presented in the form of maps of the spatial distribution of particular weather elements and biometeological indicators determined on the basis of modelling results.

Elaborated by H. Ojrzyńska, K.Wałaszek