This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
kb:sources [2017/11/10 09:25] external edit
kb:sources [2018/05/07 16:26] (current)
Line 1: Line 1:
-====== ​Sources ​in ENVI-met ====== +====== ​Pollutant sources ​in ENVI-met ======
-FIXME  This section is very outdated and needs fixing+
-===== Using Sources in ENVI-met ​===== +The simulation the dispersion of passive and active pollutants is one of the most complex tasks when using ENVI-met. The calculation of the transport ​and deposition of pollutants requires everything a full microclimate simulation needs to be run successfully plus additional information concerning ​the pollutant emissions. 
-Since version 2.5 ENVI-met provides the option to include sources of particulate matter (PM) or inert gases in the model domain. \\ +This Knowledgebase should help you to find your way in this process.
-The calculation of the dispersion of PM and gas includes ​the following processes:+
-  * Calculation of emission volume based on 24 hourly single values provided ​ 
-  * Advection and mixing using an upstream-advection scheme ​ 
-  * Sedimentation of particles depending on size and weight ​ 
-  * Deposition of PM and gases at solid surfaces (roof, ground surfaces, water) ​ 
-  * Deposition of PM and gases at leaf surfaces 
-In general there are three groups of possible (chemical) components treated in ENVI-met: +{{ :kb:​knowledgebase_sources.pdf | This KB is supplied as PDF for the moment.}}
-  - Particulate Matter +
-  - CO<​sub>​2</​sub>​ +
-  - all other Gases+
-For **particles**,​ typical aerodynamic parameters such as the settling velocity are calculated. A particle is supposed to be deposed as soon as it reaches a surface (leaf or ground), no re-release rates are calculated. Also, no maximum load of surfaces is taken into account. \\  
-**CO<​sub>​2</​sub>​** is a special component as it is already treated by the vegetation model. Additional sources of CO<​sub>​2</​sub>​ will be added to the vegetation model CO<​sub>​2</​sub>​ prognostic system. \\ 
-All **other gases** are treated as inert substances which means that they do not react with other species or dissolve etc. \\ 
-To calculate the absorption of gases at the soil surface, on water and at plant leafs, additional information such as the //molecular diffusivity//​ or the //​Henry'​s Law Constant// are needed. These data are coded inside ENVI-met and automatically selected depending on the type of gas chosen. In the recent version ENVI-met supports (see also [[SOURCES|here]]): ​ 
-  Carbon oxide 
-  Carbon dioxide 
-  Nitrogen oxide 
-  Nitrogen dioxide 
-  Sulfur dioxide 
-  Ammonia 
-  Hydrogen peroxide 
-If your component is not included, please chose a gas with a similar behaviour. 
-===== Numerical Treatment and Accuracy ===== 
-A lot of effort is taken recently to make the numerical solution of the pollutant advection and dispersion equations as accurate as possible. In ENVI-met a simple upstream advection scheme is used to calculate the pollutant dispersion. We know, that this scheme has several drawbacks such as numerical diffusion. On the other hand we have to face the fact, that the calculated wind field in microscale (which, after all is the basis of the advection process) is not free of divergence. Whatever spatial resolution one chooses, there are still sections in the flow field (e.g. at sharp corners, in the center of a vortex) where the flow cannot be resolved correctly resulting in a divergence of the wind field (more mass is entering a grid cell than leaving it, or the other way round). In combination with advection, a divergence in the flow field will always result in a local source or sink of the transported substance. With this "​feature"​ in mind, the sense of introducing more accurate numerical methods to avoid e.g. numerical diffusion is limited because there is a much bigger problem in the flow field than in the numerical methods. ​ 
-In order to realise a model that is able to simulate complete diurnal cycles over several days on a normal computer, numerical accuracy can only be followed up to a certain level. If you need to have an extremely accurate simulation of pollutant dispersions,​ we recommend to use a finite-element software package. After all, when talking about accuracy of specific modules, remember that the environment as it is represented in ENVI-met consists of blocks and flat roofs only, which is probably not realistic, too.  
 ---- ----