posted 01-03-2009 01:25 PM ET (US)        

Path loss in the real world is more difficult to model theoretically. A general solution proposed is:

pathLoss = 10 x ϒ x log(d)

where

d = distance in meters
ϒ = the path loss slope

The path loss slope, ϒ, is generally varied to suit the intervening terrain. One source comments:

"The propagation over water can be viewed as free-space propagation with ϒ close to 2, where for land mobile communications the path loss slope is typically within the range of 3 to 5. ϒ can increase to higher values such as 7 or 8 in woods and areas with high buildings."

Performance Enhancements in a Frequency Hopping GSM, Nielsen and Wigard, p55.

http://books.google.com/books?id=p-_jcQKb9MMC&pg=PA55&lpg=PA55&dq=path+ loss+calculation+over+water&source=web&ots=jEdq2ojgMg& sig=BxkYqT6i2mEMqX0BK3XDNYuWwu8&hl=en&sa=X&oi=book_result&resnum=6& ct=result

Another source mentions that propagation over water is generally more favorable than over land, referring to water as an enhancement:

"In general, path loss is the decrease, or attenuation, of the power of a signal usually occurring as a result of absorption, reflection, diffusion, scattering, diffraction, or dispersion, from an original level. In a mobile communication network, path loss may be determined from several components. For example, path loss may be a combination of distance dependent path loss, path loss due to terrain obstacles, path gain (or loss) due to sloping terrain, path gain caused by over- water propagation enhancement, path loss due to rain attenuation, and/ or path loss due to street orientation relative to the propagation path."

http://www.wipo.int/pctdb/en/wo.jsp?WO=2006126917&IA=SE2005000771& DISPLAY=DESC

Another source mentions the variation possible in propagation over water:

"Fade magin is extremely important over longer distances, especially those spanning varied terrain, over water, experiencing sun one day and rain the next, or through significant altitude changes where atmospheric conditions can affect a signal drastically...."

I would assume that the variation in propagation over water would be associated with the sea state, the mist and fog present, and any temperature conditions that might lead to unusual atmospheric effects that could influence the radio waves.

Water also can create a propagation path loss due to reflections. In a line-of-sight path over water, there can be signal loss due to the arrival of signals reflected from the water itself. Briefly, when a signal is reflected from a surface the phase of the signal tends to invert. If a signal is arriving by both a direct path and a reflected path, the phase of the two signals may be destructive and cause a decrease in signal strength. This phenomenon is mention as occurring over water:

"An LOS path may have adequate Fresnel zone clearance, and yet still have a path loss which differs significantly from free space under normal refraction conditions. If this is the case, the cause is probably multipath propagation resulting from reflections (multipath also poses particular problems for digital transmission systems - we'll look at this a bit later, but here we are only considering path loss).

"One common source of reflections is the ground. It tends to be more of a factor on paths in rural areas; in urban settings, the ground reflection path will often be blocked by the clutter of buildings, trees, etc. In paths over relatively smooth ground or bodies of water, however, ground reflections can be a major determinant of path loss."

http://www.tapr.org/ve3jf.dcc97.html under subheading "Ground Reflections"