Page **1** of **1**

### Molecular Weight of Particles for Dry Deposition Simulation

Posted: **April 27th, 2020, 7:26 pm**

by **rejoyce17**

Hello, I am trying to run a dispersion model for bioaerosols to determine the amount of dry deposition of the particles over ~120 hours. I have input nonzero values for the aerodynamic diameters of the aerosols (3um, 1um, and 0.1um, for separate simulations), the density, and the shape, and want to force the model to estimate the amount of settling (as opposed to putting in a settling velocity). So I have kept all other values at zero for the "deposition" of my particles (including velocity, molecular weight, A-ratio, and D-ratio), as was outlined by the HYSPLIT user manual for situations in which you are modeling particles and do not want to input settling velocities a priori. However, the output of the model when I input a molecular weight of zero is radically different from when I input a nonzero molecular weight. So it would appear that entering a molecular weight for my particles is rather important, contrary to what I have read in supporting HYSPLIT materials.

I am using real data for these models based on aerodynamic particle sizes only. I do not have molecular weight information the bioaerosols that I am trying to model (and it would likely be a very wide range of molecular weights). Is there a way around putting in a molecular weight for particle dispersion to examine their dry deposition, so that the amount of dry deposition is based on particle diameter alone?

And I am using HYSPLIT Version 4

### Re: Molecular Weight of Particles for Dry Deposition Simulation

Posted: **April 28th, 2020, 9:45 am**

by **alicec**

05/07/2020 - This post was edited as the original post provided incomplete/mis-leading information regarding the dry deposition options. Please refer to the later posts for information.

Can you please provide more information.

Could you provide more information on why you think that option is not working as expected?

Attaching your CONTROL files for the different options you are looking at would be helpful.

Attaching output could also be helpful.

Also please specify the version of HYSPLIT that you are using.

### Re: Molecular Weight of Particles for Dry Deposition Simulation

Posted: **April 28th, 2020, 10:02 am**

by **rejoyce17**

Thank you for the quick response! I have attached my setup file, and it sounds like what you have described in option 1. I don't know the dry deposition velocity of my particles, the only information I have on them is diameter, density, and shape. So I input those values into the first dialog box as "3.0 1.0 1.0" and I keep all other values, including dry dep velocity, MW, etc. set to 0.0

My goal is to examine how much dry deposition is occurring during the simulation, which I check the resulting MESSAGE files for. There are no errors in the model that I can see, but the amount of mass that is lost during the 120 hr simulation for 3.0um particles seems very very small (they should be quite heavy!). Because of this, I tried another simulation where I kept all of the same values for diameter, density, and shape, and also set the molecular weight of the particle to a non-zero number (i tried 1 and 10). This time, the amount of mass in the atmosphere that was lost throughout the 120 hr simulation seemed far more believable than when the MW was set to 0. So I am wondering if setting the MW is still important even when using option 1 that you described.

Thanks!

### Re: Molecular Weight of Particles for Dry Deposition Simulation

Posted: **May 4th, 2020, 3:28 pm**

by **alicec**

Please reference the tech memo and references found therein.

https://www.arl.noaa.gov/wp_arl/wp-cont ... rl-224.pdf
If you have the molecular weight set to 0 then see section 5.1 and 5.2 in the tech memo. The dry deposition velocity in 5.2 is simply the settling velocity in 5.1.

Setting the molecular weight to a value greater than 0 will cause the resistance method to be used for particles.

See section 5.5 in the tech memo. Specifically equations 102,103,110.

If particle diameter, density and shape are set, then an above 0 value of molecular weight is really just a flag to use

the resistance method as described in the tech memo.

### Re: Molecular Weight of Particles for Dry Deposition Simulation

Posted: **May 7th, 2020, 10:29 am**

by **alicec**

This post contains information on the dry deposition options in HYSPLIT.

The tech memo contains complete details about the equations used to compute deposition.

https://www.arl.noaa.gov/wp_arl/wp-cont ... rl-224.pdf
Dry deposition is controlled by lines described here

https://ready.arl.noaa.gov/hysplitusersguide/S314.htm
27(1) Diameter, Density, Shape

28(2) Deposition Velocity, molecular weight, surface reactivity, diffusivity ratio, effective Henry’s constant.

**1. For particles:** Setting the diameter, density and shape.

- (a) If diameter, density and shape are set to non-zero, and molecular weight is zero then deposition velocity is set equal to settling velocity. (See section 5.1 and 5.2 in the tech memo).

- (b) If molecular weight is set to non-zero and diameter, density, shape are non-zero then this causes HYSPLIT to calculate dry deposition through the resistance method for particles. This will give a larger deposition velocity than the settling velocity. See section 5.5 in the tech memo, specifically equations 102, 103, 110. Note that the results will not be dependent on the value of the molecular weight set, the molecular weight is simply a flag.

**2. For particle or gas: **The dry deposition velocity can be set directly for each pollutant by entering a non-zero value in the first field (28). In the special case where the dry deposition velocity is set to a value less than zero, the absolute value will be used to compute gravitational settling but with no mass removal.

**3. For gas:** resistance method requires setting the four parameters (molecular weight, surface reactivity, diffusivity, and the effective Henry's constant). Deposition velocity, diameter, density, shape set to 0. See tech memo section 5.5 and references therein.