hysplit和flexpart模式的区别在哪儿

jeson90

本人一直在做拉格朗日模式hysplit的工作，最近想看看其他拉格朗日模式的工作，看了不少中文英文的关于flexpart的文章。因为在 flexpart方面没有具体运行操作过，所以还是一枚小白，想请问诸位大神hysplit和flexpart的区别？

超台十八

The HYSPLIT model is a complete system for computing simple air parcel trajectories to complex dispersion and deposition simulations. The intial development was a result of a joint effort between NOAA and Australia's Bureau of Meteorology. Recent upgrades include enhancements provided by a number of different contributors. Some new features include improved advection algorithms, updated stability and dispersion equations, continued improvements to the graphical user interface, and the option to include modules for chemical transformations. Without the additional dispersion modules, Hysplit computes the advection of a single pollutant particle, or simply its trajectory.


FLEXPART (“FLEXible PARTicle dispersion model”) is a Lagrangian transport and dispersion model suitable for the simulation of a large range of atmospheric transport processes. Applications range from the dispersion of radionuclides or air pollutants, over the establishment of flow climatologies, to the analysis of Earth’s water cycle. FLEXPART also produces output suitable for inverse determination of emission sources, e.g., of greenhouse gases or volcanic ash.

超台十八

还有这个 STILT:

STILT, the Stochastic Time-Inverted Lagrangian Transport model, is a Lagrangian particle dispersion model (LPDM) for atmospheric transport. Its primary purpose is to derive the upstream influence region on atmospheric measurement locations. In other words, it is a fast tool to retrieve the adjoint of tracer transport, i.e. the sensitivity of atmospheric tracer mixing rat一io measured at receptor point with respect to upstream surface fluxes. STILT is driven by meteorological fields (most important wind fields) from a variety of weather prediction models (ECMWF, WRF, RAMS, ...), for both analysis of past observations or for measurement planning purposes using forecasts. STILT includes turbulence as a stochastic process.

超台十八

功能应该算一样了(看你想做哪方面?)。哪一个好就难说，看个案吧，例如这个 2010 山火研究说 flexpart 好:

http://repository.lib.ncsu.edu/ir/handle/1840.16/6293

但这 2013 文章又说三个结果差不多:

http://journals.ametsoc.org/doi/abs/10.1175/JAMC-D-13-0125.1



反正这些都不难用吧，你全都试试跟大家报告一下吧。