Hive scale graph: flow switched off literally overnight — sharing the data
For the data nerds: my scale hive in Skåne gained 1.3–2.1 kg a day for eighteen straight days through the lime and late clover. Then on the night of the 17th the graph just... flattens. Next three days: −0.28, −0.31, −0.35 kg. No weather change, no rain. The flow didn't taper, it stopped like someone closed a valve.
Screenshot of the month attached. The daily sawtooth is still there — morning dip as foragers leave, but now they come home barely heavier than they left. Colony is now eating roughly 300 g a day.
Eight years keeping and the scale still teaches me something every season. Anyone else's region flip to dearth this week? Curious how sharp the cutoff looks elsewhere.
#hive-scale#dearth#nectar-flow#data
▲ 14 votes💬 4 replies
4 replies
LavenderSophie · Jul 21, 2026
Provence flipped about ten days before you — lavender finished and my scale went from +1.5 kg days to −0.4 almost identically. That sharp cliff is so typical of a single dominant flow ending. The sawtooth flattening is my cue to put entrance reducers back in before robbing starts.
PasiekaOlek · Jul 22, 2026
Same picture in Małopolskie after the lime finished last week. One thing your graph is telling you that inspections won't: with −0.3 kg/day, a colony sitting on 15 kg has about 7 weeks before it's in trouble if nothing else blooms. I now plan my harvest amounts off exactly that arithmetic — take the surplus, leave the runway.
HokkaidoHachi · Jul 22, 2026
Hokkaido is opposite right now — we're mid flow, scale climbing daily. Your post is a good reminder of how local this all is. Question: do you correct the graph for rain soaking into the woodenware? My scale shows fake 0.5 kg 'gains' on wet days that confused me my whole first season.
SkaneBees · Jul 22, 2026
HokkaidoHachi — I don't correct it, I just annotate rainy days and ignore them; a wet roof and hive stand can easily hold half a kilo. LavenderSophie's right about reducers, they went in this morning. The bees were already testing the neighbouring nuc's entrance, so the graph called it before I did.