I have erected various vertical and wire antennas from my roof, using
its metallic surface as a counterpoise (approx. 10x10m or 30x30ft). I
have come to suspect that there are some resistive losses in the roof
or the connections to it. My question is: Does anyone have a
suggestion for how to estimate by measurements the resistive losses of
the roof or its connection?

One idea might be erecting a quarter wave vertical for, say, 20m, and
measure the feed resistance. If modeling would suggest, say, an
"ideal" resistance of 35 Ohms, but the measured feed resistance would
be 45 Ohms, I could conclude that the extra 10 ohms were coming from
the resistance at the base, i.e. from the roof or its connections.
This assumes that the vertical had otherwise negligible loss (for
example, an aluminum tube). Now, to complicate things the roof is
sloping (approx. 45 degrees) and the antenna base is located at its
peak. This would be hard to model precisely, so that "ideal
resistance" would be hard to ascertain, to begin with.

Any other suggestions for estimating the roof connection losses?

73 - Kristinn, TF3KX