The influence of micro location on the air freezing index
In this paper are exposed research results of air freezing index values occurring in area of Bosnia and Herzegovina. Theoretic probabilities of air freezing index occurrence are calculated, based on data of 26 meteorological stations, in order to determine its return periods of appearing (the probability of air freezing index appearance average once in period considered). To this purpose five the most frequent in praxis theoretic distribution functions are used: GAUSS, GALTON, LOG-PEARSON, PEARSON and GUMBEL. As the most appropriate function for all meteorogic stations the LOG-PEARSON was selected because it on the best way approximates values of calculated air freezing index. Regarding to influence of micro location on the value of air freezing index some independent parameters of location which are measurable (variables) and have supposed influence, are took in analysis. To this purpose the mathematic model of multi variant regression analysis was used and the regression equation of the associated influence of independent parameters was determined. Using this discovered equation designers of road pavement can for each micro-location cutted by the road line, calculate the air freezing index and check the pavement structure on the harmful freezing effects.