A Ray Tracing Algorithm Based on the Discrete Anisotropic Radiative Transfer Model
Photosynthetically active radiation(PAR)is an important covariate in crop productivity models,terrestrial ecosystem models,and biogeochemical models.At present,different types of models are available to simulate LiDAR waveforms in woods,i.e.,LiDAR waveforms can be generated by computer model,however,semi-empirical and geometric models are not accurate enough.Monte Carlo ray tracing(MCRT)models have high accuracy but are computationally demanding.In this paper,we proposed a new LiDAR waveform simulation model based on the introduction of a quasi-Monte Carlo ray tracing method in the Discrete Anisotropic Radiative Transfer(DART)model.The combination of two new strategies,the"box method"and the"ray-tracing method"was realized for simulating the propagation of photon in complex environments.In the discrete radiative transfer,simulated photons were added to simulate small samples of light energy;the Russian roulette concept was added to the Monte Carlo ray tracing in order to reduce the large amount of computational requirements while meeting the accuracy requirements;and the model was dissected into grids,which allows for more accurate ray tracing.The experimental results showed that the average relative error between estimated and actual measured values of surface solar radiation from mid-June to July was 3.8%to 4.2%,while it was 3.5%to 3.9%for PAR,indicating the good estimation effect.The accuracy was improved by 1.6%compared with the traditional Monte Carlo algorithm.Compar-ison of the actual measurements with the interception efficiencies simulated by this algorithm,the error rate ranged from 1.66%to 2.72%,which is a high degree of accuracy.The experimental results are impor-tant for the study of light-dependent plant physiological processes.
photosynthetically active radiationBox methodray Carlo methodray-tracingMonte Carlo
万方数据
维普
