We have already discussed the effect of humidity on the thermal conductivity of mineral wool. We have shown the following graph, where the measured thermal conductivity is given versus the absorbed humidity. Indeed, the thermal conductivity doubles for a huge humidity absorption.

But the laboratory measures according to the standards with a guarded hot plate or a heat flux meter like already discussed. And these measurements don´t include the heat pipe effect, which can be present on a flat roof with a membrane on top of the thermal insulation. The heat pipe is a very efficient heat conductor and works as follows: a liquid is evaporated at the warm side, the vapour flows to the cold side and condenses. The condens flows back to the warm side. Because water has a very large evaporation / condensation heat value ( 2257 kJ/kg), it can be used (wanted or not wanted) to transport heat in a very efficient way. The principle is shown in the following:

We made a small calculation for a flat roof with 20cm mineral wool at 60kg/m³ density and 0.040 W/mK (declared) thermal conductivity. We assume that the mineral wool contains 1% humidity and that in winter (when we heat) this water evaporates. The vapour migrates to the (cold) water proofing membrane through the permeable insulation and condenses. We assume that this is happening daily: evaporation during the day, condensation during the cold night. In that case, the heat transfer through the thermal insulation has increased with 78%. The laboratory measurement, where the heat pipe effect is not present,  shows an increase of only 5%.

It is an open question whether in a certain mineral wool roof the heat pipe effect is present or not. But it is absolutely sure that it is not present in a GLAPOR cellular glass roof, with a comparable price as the high density mineral wool. The above shows that the academic declared value is giving only little information because nobody can avoid 1% humidity in a roof.