The scaling exponents for statistical topography of random surfaces are found an the fractal dimension of corresponding isolines are determined. An analytical relationship is derived for the exponents describing the light intensity distribution from rough self-affine surfaces including two-and three-dimensional cases. An approach for studying pair dispersion (Richardson law) in fully developed turbulence of compressible flows is proposed. The clustering of sticky particles in turbulent compressible velocity fields is described. It is shown that anisotropy of large-scale fluctuations can propagate along the turbulent cascade. The scaling properties of the time series of asset prices and trading volumes of stock markets are analysed.  The trading volume data obey multi-sacling lenght distribution of low.vcariability periods.