ИСТИНА |
Войти в систему Регистрация |
|
ИСТИНА ЦЭМИ РАН |
||
Solar radiation is a key factor determining the climatic and weather conditions in different regions across the globe. It influences the surface energy budget, carbon dioxide and water vapor exchange at the vegetation - atmosphere interface, thereby affecting not only the plant functioning and growth, but also the concentrations of greenhouse gases in the atmosphere. Over the last few decades, many radiative transfer models of different complexity, from the simplest 1D to more sophisticated 3D models, were developed and applied in various ecological and meteorological studies. Within the framework of the study a 3D Monte Carlo model to derive the radiative transfer within a spatially non-uniform forest canopy was suggested. The spatial vegetation structure was derived using the methods of fractal geometry. Namely a model of a young birch maintaining proportions and shape was built. A fractal tree model was built based on the self-similarity principle by sequentially applying the branching rule using the Wolfram Mathematica mathematical package. At the first step, the algorithm sets tree coordinates, length of trunk and first branches, number of first branches, and also the zenith and azimuth angles of the first-order branches. To rotate the branches, the RotationMatrix function was used, and the length of the branches tree was assumed to be a multiple of the golden ratio φ≈0.61803. Namely when constructing the model tree, a scale factor φ–1 was used for the side branches of the first order, for the second order - φ–2, etc. To simulate the branching of the tree, the function NestList was used. The developed model allows to describe the reflection and transmission of solar radiation within heterogeneous forest canopy taking into account the multiple scattering from plant elements and soil surface. The radiation transfer is modeled as a random Markov chain of sun photons interaction with plant elements (leaves, branches) and soil. To carry out calculations on multiprocessor computing systems, a parallel implementation of all basic operations of the Monte Carlo method, an algorithm using distributed programming techniques of the mathematical package Wolfram Mathematica was developed. Application of the Monte Carlo method to describe the radiative transfer within a spatially non-uniform forest canopy