Answers are grounded in this paper and cite their sources.
A winter colony manufactures water as surely as it manufactures heat: oxidising the sugars in honey yields 0.48 kilograms of metabolic water for every kilogram of honey consumed, so a colony eating 100 grams of honey a day produces 48.0 grams of water a day that must leave the hive as vapour or arrive somewhere as liquid. This study computes where that water goes. Cavity air is taken at 10 degrees Celsius; at 80 percent relative humidity its dew point is 6.71 degrees Celsius. The inner surface of an uninsulated 19 millimetre pine wall falls to 1.78 degrees Celsius when the outside temperature is -10 degrees Celsius, which is 4.93 degrees below the dew point, and the estimated condensation rate on such a wall is 618.1 grams per square metre per day. The same wall carrying 50 millimetres of expanded polystyrene holds its inner surface at 8.52 degrees Celsius, which is 1.81 degrees above the dew point, and the computed condensation rate is 0.0 grams per square metre per day. At an outside temperature of 0 degrees Celsius the uninsulated surface still condenses at 115.8 grams per square metre per day while the insulated surface remains dry. Ventilation can substitute for insulation only at a price: holding the cavity at 60 percent relative humidity against the stated water production requires 8.91 litres of air per minute at -10 degrees Celsius outside and 19.02 litres per minute at 0 degrees Celsius, flows that also carry away heat. The combination rule that follows is to insulate first, so that wall surfaces stay above the dew point, and then to provide a modest top vent as insurance for humid spells rather than as the primary moisture defence. The familiar belief that low temperature by itself is what winter colonies fail to survive is not supported by this arithmetic: at the stated cavity temperature the dry insulated hive presents no wetting at all, while the warmer-feeling but bare-walled hive rains condensate onto the space around the cluster. This study is simulated, not measured.
No takes yet. Share an insight, caveat, or question.
William Alexander Aldred (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: