[TowerTalk] Tree attenuation

Jim Lux jimlux at earthlink.net
Mon Nov 7 17:37:19 EST 2005

At 01:24 PM 11/7/2005, Jim Brown wrote:
>On Mon, 07 Nov 2005 12:15:59 -0800, Jim Lux wrote:
> >As far as HF goes, it gets quite a bit more complicated(!), for two reasons:
> >1) The forest scale (height of trees, etc.) is on the same order as the
> >wavelength  (at VHF, trees are MUCH bigger than a wavelength, and leaves
> >are MUCH smaller).
> >2) Most practical applications of a model have the antenna in the forest,
> >where the forest is in the near field of the antenna.
>I skimmed both the Power Point and the longer document.  The analysis that 
>into the Power Point make it considerably more valuable in guestimating what
>might happen at VHF and low UHF,  but neither gives much insight into HF.
>The obvious question is, is anyone aware of anything equally good for the HF
>spectrum?  My future antenna supports are likely to be redwoods, 25-30 meters

The paper you want is by Theodore Tamir, "On Radio-Wave Propagation in 
Forest Environments", IEEE Trans on Ant and Prop, Vol AP-15, #6, Nov 1967, 
pp 806-817

He covers 1-100 MHz, and lots of experimental and theoretical analysis..

His summary:
1) For constant antenna height, received field varies inversely as distance 
squared [RMK note: oddly, irrespective of the terrain when forested]
2) The presence of vegetation produces a constant loss, which seems to be 
independent of the distance between the communication terminals.
3) The transmission loss is reduced by raising either receiving or 
transmitting antenna.  This produces a "height gain" effect which, when 
measured in dB various roughly logarithmically or linearly with antenna 
height. [RMK note: This isn't the same thing as the Longley-Rice height 
effect familiar to VHF/UHF predictions]
4) The received field may be considerably depolarized, relative to the 
orientation of the transmitting antenna.

Some more notes.. below about 12 MHz, the height gain effect was too small 
to be detected.

You can model the forest as a dielectric with epsilon=1.1 and sigma=0.1 

There's a factor you apply that depends on the distance from the antenna to 
the "top" of the forest, and it runs around 0.25 dB/m up to 1.1 dB/m, 
depending on the the dielectric properties.

The middle of the curve starts at around 0.2 dB/m for 1 MHz, rises steadily 
to 0.5 dB/m for about 15 MhHz then sort of flattens out to 0.6 dB/m up to 
400 MHz.  (Figure 8 in the paper)
(that distance is the distance from antenna to TOP of the forest... so if 
your antenna is up 10m and the forest is 15m tall, you'd have about 2.5 dB 
of loss)

Note that this for both receiver and transmitter in forest.  If only one is 
in the forest, cut those in half.  (Tamir's analysis assumes skywave 
propagation, by the way).

>Jim Brown K9YC
>See: http://www.mscomputer.com  for "Self Supporting Towers", "Wireless 
>Weather Stations", and lot's more.  Call Toll Free, 1-800-333-9041 with 
>any questions and ask for Sherman, W2FLA.
>TowerTalk mailing list
>TowerTalk at contesting.com

More information about the TowerTalk mailing list