DETERMINATION OF CHARACTERISTIC VALUES OF CLIMATIC LOADS BY DATA SMOOTHING OF SURROUNDING WEATHER STATIONS

Abstract

During territorial zoning of climatic loads and impacts, the method of smoothing of the data of nearby weather stations with weight functions is used, which takes into account the amount of data and the distance from each meteorological station to a given design point. The purpose of this research is the mathematical justification of the type and parameters of the weight function for meteorological data smoothing.

For the analysis, the characteristic values of four climatic loads (snow load, ice load, maximum wind pressure and wind pressure during ice loads), determined by results of meteorological observations at 172 flat weather stations of Ukraine. The exponential, triangular and rectangular weighted functions with different values of parameters (smoothing intervals) are analyzed. The best type of weight function and smoothing interval is one that takes into account the largest volume of weather data and provide that this data remain homogeneous. The possibility of combining the results of meteorological observations at surrounding meteorological stations is established on the basis of the dispersion analysis procedure, modified in the previous works of the author.

With an increase of the smoothing interval, the factor variance increases and the F-test index increases accordingly. It is reasonable to consider suchlike smoothing interval, in which F-test index becomes equal to its critical value.

By the example of five project points from different geographical regions of Ukraine it is shown that during the normalization of climatic loads, an exponential or triangular weight function should be used with the recommended smoothing intervals. The results of the research can be used both in determining the characteristic values of climatic loads in a specific design point, as well as in a developing of maps of territorial zoning of Ukraine.