Dry deposition velocity

[Stephen McGuire <SAM2@nrc.gov>]

On March 17, Henriette van Graan asked if the dry deposition velocity factor of
0.001 m/s in the COSYMA code is appropriate for radation environmental
consequent analysis.  What follows is just my opinion.  

The short answer is, yes, it is a good general-purpose value as long as you
understand its limitations.  The long answer is below.

Dry deposition velocity is a somewhat artificial concept that is useful for modelers
but only weakly linked to physical processes.  To use the concept wisely, you really
have to understand the processes.  There are 3 main removal processes for the
removal of particulates from air: gravitational settling, inertial impaction, and
Brownian diffusion.  

Gravitational settling is the dominant process for large particles, roughly 10
microns and larger.  For these particles, whose removal is mostly by gravitational
settling, a settling velocity is an appropriate physical and easily measured quantity. 
If we are considering an accident, we probably will never know the initial particle
size distribution, but it is likely to contain mostly large particles.  These large
particles rapidly deposit within a short distance.  When we have an aerosol with
mostly large particles, a deposition velocity of 0.01 m/s is probably more
appropriate than 0.001 m/s.  (The 0.01 value is often used by NRC in its accident
consequence calculations.)

After the larger particles have fallen out, we are left with small particles, for which
gravational settling is not an important removal mechanism.  Depending on the
initial size distribution, height of the release, heat content of the release, and the
wind velocity, the larger particles should be largely removed by 1/4 to 1 mile from
the release in most circumstances.  At this point, the removal will be dominated by
what the individual particles hit.  Usually, almost all the removal will be by plants
because, even though leaves appear to the eye to be smooth, under a microscope
most leaves have lots of little hairs.  These hairs act like filters for removing
particles.  As a reasonable average removal rate, a deposition velocity of 0.001 m/s
is appropriate, but it can vary.

Consider two cases.  There will be most removal then there is a lot of turbulence,
for example atmospheric stability classes A and B and higher wind speeds.  For this
case, a deposition velocity of 0.01 m/s is probably more realistic.  The other
extreme is very stable F class.  For F stability there is little vertical mixing.  For
the part of the plume close to the ground that passes directly through vegetation,
the removal will be very high initially, and then all the particles will have been
removed.  At that point, the deposition velocity will become zero, essentially no
removal.  For the F stability plume above the vegetation, the deposition velocity
will be zero, essentially no removal.  It is not correct to mix the two pieces of the
plume.  The particles higher in the plume will not be removed.  They will just keep
going until the sun comes out and creates vertical turbulence.

Thus, when all is said and done, for average conditions 0.001 m/s should be a pretty
good value for deposition velocity as long as we do not try to apply it to extremes
(too close to the release point, too stable a meteorology such as F, too turbulent a
meteorology such as A, etc.)