Last Updated on July 24, 2026 by John Berry
The refractivity of the atmosphere changes. As I note in another page, the refractivity, given by dN/dh, varies between about -300 and +30 N units per kilometre of height (for from 1% to 99% of time. The median value is -40. This is great science, but how does it help radio amateurs make a tropospheric prediction?
High values of dN/dh are bad news for radio hams. When the refractivity goes that high, the Earth’s bulge protrudes more into the path between two stations. As a result, the diffraction loss rises and makes the path unworkable.
At the other extreme, at -300 N units/km, the Earth’s bulge diminishes, and the diffraction loss reduced to a minimum. DX operation is possible when values of dN/dh reduce below about -100 and tropospheric lifts occur.
So, the question is, how do we predict when dN/dh is on the way down, and likely to reduce below -100? And hence how will we know there are likely to be good tropo conditions?
Predicting good tropo conditions
From Rec. ITU-R P.453-14, the radio refractivity, N (in N units) is given by:
Where:
Pd is dry atmospheric pressure (hPa);
e is the water vapour pressure (hPa); and
T is the absolute temperature.
This equation reduces to:
Where:
P is the total atmospheric pressure (hPa).
So, the radio refractivity is a function of temperature and pressure of the gases in the troposphere.
Rec. ITU-R P.453-14 goes on to relate N to the height profile of the troposphere.
The question then is how, from available data about the atmosphere, can we predict P, T, and e, and hence calculate the height profile, and calculate the refractivity at any time for any path?
Making a prediction
The simple answer is that we, in the ham radio world, can’t. We don’t have access to the necessary huge worldwide data sets of temperature and pressure. We must turn to those in the meteorological industry.
Online, there are two sites with interpreted data: f5len.org, and dxinfocentre.com. The former is managed by Pascal Grandjean, F5LEN, in France, and the latter by William Hepburn in Canada. The two have similarities and are likely based on the same core stream of meteorological data. The following image is from F5LEN.org.
Both these sites generalise the maths above and likely give an interpretation of dN/dh values as traffic-light colours from ‘minor’, through ‘good’ to ‘vy. strong’. William Hepburn’s gives added scale descriptors from ‘intense’ to ‘extreme’. Pascal Grandjean suggests an upper scale colour of ‘duct’.
There is no engineering explanation given from the site managers to show what’s behind either refractivity map.
Guessing, I’d suggest that anything better than very strong on these maps is indicating k>3 – my arbitrary tropospheric DX state described when dN/dh is -100 N units per kilometre or less. I deal with ducting on another page.
Making a prediction
So, simply, can I expect that on the 14th July 2026 I will be able to work a path from southern Scotland to the south of England? A distance of about 600km. Answer, possibly. The map above from Pascal Grandjean, F5LEN, suggests a ‘strong’ tropospheric lift for much of the path,
To have the necessary tropospheric conditions to work that path, we would need a dN/dh of better than -100, say, and these sites would need to colour the region through which the path passes as ‘vy. strong’ or better to suggest a dN/dh of approaching -157N units/km.
Of course, if we were in the Mediterranean, attempting to communicate across the water from eastern Spain to Corsica, we might have more success.
Click here for more on making a tropo prediction over what would ordinarily be a diffracted path.