Swedish noise model

The model is very similar to the ISO-9613, and thus it should be quite easy to implement. I have "translated" the notations as far as I could, from the Swedish ones to the ISO ones.

Up to 1000m distance, the sound leven from a source is calculated with:
LfT(DW)=LWA_corr-8-20*log(d/d0)-0.005*d

Over 1000m distance, this formula is used (basically the same as the ISO one):
LfT(DW)=LWA_corr-10-20*log(d/d0)-dLa

dLa=10*log(sum(10^((LfT(j)+Af(j))/10))-10*log(sum(10^((LfT(j)+Af(j)-a*d/d0)/10))

j=octave band.

In both cases:
LWA_corr=LW+k*dvh

dvh=vh*(ln(H/z0)/ln(h/z0)*ln(h/0.05)/ln(H/0.05)-1)

z0 is the site roughness length, h=10 m, H= hub height. vh = wind speed at 10 magl as stated for the sound level (ie, usually 8 or 10 m/s).

k=wind speed dependance of sound emission level.

Addition of sources are done as in ISO-9613.

Using the same notations for A as in the ISO standard (and your code), I believe that things could be expressed this way:

Below 1000m:
Adiv = 20.0*log10(d)+11.0; //As before
Agr = -3 //Result if G=0, and q=0 (ok unless you use a very low turbine)
Aatm=0.005*d //Some good mean atmospheric absorption I guess.
Abar = 0
Amisc = 0
Afol = 0

Over 1000 m:
Adiv = 20.0*log10(d)+11.0; //As before
Agr = -3 //Result if G=0, and q=0
Abar = 0
Amisc = 2 //My guess is that this is to make things continous
Afol = 0
Aatm = as stated in previous post.