Dimensional analysis, scaling and fractals

Dimensional analysis refers to the study of the dimensions that characterize physical entities, like mass, force and energy. Classical mechanics is based on three fundamental entities, with dimensions MLT, the mass M, the length L and the time T. The combination of these entities gives rise to derived entities, like volume, speed and force, of dimensions L3, LT-1, MLT-2, respectively. In other areas of physics, four other fundamental entities are defined, among them the temperature θ and the electrical current I. The parameters that characterize physical phenomena are related among themselves by laws, in general of quantitative nature, in which they appear as measures of the considered physical entities. The measure of an entity is the result of its comparison with another one, of the same type, called unit. Maps are also drawn in scale, for example, in a scale of 1:10,000, 1 cm2 of paper can represent 10,000 m2 in the field. Entities that differ in scale cannot be compared in a simple way. Fractal geometry, in contrast to the Euclidean geometry, admits fractional dimensions. The term fractal is defined in Mandelbrot (1982) as coming from the Latin fractus, derived from frangere which signifies to break, to form irregular fragments. The term fractal is opposite to the term algebra (from the Arabic: jabara) which means to join, to put together the parts. For Mandelbrot, fractals are non topologic objects, that is, objects which have as their dimension a real, non integer number, which exceeds the topologic dimension. For the topologic objects, or Euclidean forms, the dimension is an integer (0 for the point, 1 for a line, 2 for a surface, and 3 for a volume). The fractal dimension of Mandelbrot is a measure of the degree of irregularity of the object under consideration. It is related to the speed by which the estimate of the measure of an object increases as the measurement scale decreases. An object normally taken as uni-dimensional, like a piece of a straight line, can be divided into N identical parts, so that each part is a new straight line segment represented in a scale r = 1/N of the original segment. With the indispensable aid of computers, fractal geometry is developing very fast in several areas of knowledge, including arts, as a new tool to better understand nature. In agronomic research it is used to study the dynamic processes that occur in soils (water, solute, heat and gas movements), soil structure, plant architecture and development, drainage of watersheds, etc. Fractal models that simulate soil structure are also used to better understand soil behaviour. The fractal characteristic of several soil attributes has led to the use of these new technologies in substitution to several empirical procedures