Winterising a hive in central Kentucky

Every spring, beekeepers open hives that did not make it and find the same thing: a small, dead cluster, often with honey a few inches away. It is easy to read that as cold. It almost never is. A healthy colony with food above it will sit through a Kentucky winter without any help from us — bees have been doing it since long before boxes. What kills a hive in January is decided in August and September, and by the time the snow is on the lid the outcome is mostly already written.

So this is the fall, for a hive in central Kentucky: what should already have happened, what to do this week, and what to leave alone once it is done.

The colony that dies in January was killed in August. Cold gets the blame; mites did the work.

August: the winter bees, and the mites that eat them

The bees flying in July live about six weeks. The bees raised from late August into October are different animals — fat-bodied, long-lived, built to sit in a ball for four or five months and then raise the first brood of spring. Everything depends on them, and they are being made right now, out of larvae that a varroa mite either did or did not feed on. A mite-fed larva becomes a bee with a shortened life and, usually, a virus load. Enough of those and the cluster shrinks through December until it is too small to keep itself warm, and then it dies of cold, technically.

So the first job of fall is a number. Take about half a cup of bees — three hundred or so — from a brood frame, wash them in rubbing alcohol, and count the mites that drop out. It is the one sample that tells you the truth; a sugar roll is kinder to the bees and undercounts. In late summer the line most extension services draw is around three mites per hundred bees. Over it, treat; near it, treat; well under it, count again in three weeks, because the mite population is still climbing while the colony's is falling and the ratio moves fast.

Which treatment is a label question, and the label is the law. Amitraz strips go in for six to eight weeks with no honey supers on. Formic acid works under the cappings, which nothing else does, but wants daytime highs between about fifty and eighty-five and can cost you a queen if it is hotter. Thymol needs warmth and weeks. Oxalic acid kills only the mites riding on adult bees, so it is nearly useless in August and nearly perfect in December — which is why it comes back later in this post. Whatever you use, use it now if you did not in August, because the whole point is to raise the winter bees with the mite load already down, and a treatment in October protects bees that have already been damaged.

September: the flow, the queen, and weight

Central Kentucky gets a real fall flow. Goldenrod comes on in late August and the asters follow, and for a few weeks the hive smells like old gym socks, which is goldenrod nectar curing and is fine. Bees put on weight fast in September if the colony is strong enough to work it, and a strong colony in a good year needs little feeding at all.

September is also the last honest look at the queen. Open the hive on a warm afternoon and look for eggs and young larvae — not the queen herself, just proof she was there in the last three days. A colony that is queenless in mid-September can sometimes still be saved with a bought queen; by October nobody is selling them and the colony cannot raise one, because there are no drones left to mate her. The fix for a weak or queenless colony in fall is usually to combine it with a strong one: a sheet of newspaper between the boxes, a few slits in the paper, and the bees chew through and merge over a couple of days. It feels like giving up. It is taking the loss in September instead of in January, when it costs a second colony's worth of stores to find out.

Then heft. Stand behind the hive and lift the back edge an inch. A colony that will make it through to March here needs somewhere around fifty to sixty pounds of honey — a full deep of capped stores above the cluster, or a medium plus the outer frames of the brood box. If it is light, feed two parts sugar to one part water by weight, as much as they will take, from a top feeder so nothing spills and starts a robbing frenzy at the entrance. Not one-to-one: that is spring syrup, it stimulates brood rearing you do not want now, and the bees have to fan off the water. Feed until the hive is heavy, not until the calendar says stop.

October: the configuration

Once the first frost is close — around the twentieth of October in Lexington, some years earlier — the hive gets set up for winter and then, mostly, left alone. In order:

Reducer to its smallest opening, and a mouse guard across it. Mice move into hives when nights turn cold, and a mouse in a hive for the winter eats comb, urinates on everything, and leaves a colony that smells so bad the bees abandon it in March. A strip of half-inch hardware cloth over the entrance is the whole fix, and it has to go on before the mice do — which is before the first frost, not after.

Queen excluder off. The cluster eats its way upward through the winter, and if there is an excluder between the brood box and the honey the workers move up and the queen cannot. She dies alone below the grid, and the colony dies later of having no queen in February. This is the single most common preventable loss I know of, and it takes ten seconds.

Empty boxes off. The colony should occupy roughly the space it can cover. An empty super above the cluster is a cold attic the bees have to heat the air in and that small hive beetles love. Two deeps, or a deep and a medium, is the winter shape here; a big colony can carry more, a small one should be squeezed down to one.

Tilt it forward. An inch of shim under the back of the stand. The killer in a winter hive is not cold but wet: the cluster breathes out warm, moist air, it rises, hits the cold underside of the lid, and rains back down as condensation. A wet bee at thirty degrees dies; a dry bee at thirty degrees is fine. The tilt sends water that does condense down the front wall and out the entrance instead of onto the ball.

A warm ceiling and a small upper opening. A sheet of one- or two-inch foam board over the inner cover, under the outer lid, keeps the ceiling the warmest surface in the box, so moisture condenses on the walls — where it does no harm and the bees can even drink it — rather than directly above the cluster. The inner cover's notch, turned up, lets some moist air out and gives the bees a second door if snow blocks the first. In central Kentucky that is the whole of the insulation question. We do not wrap hives here; a windbreak on the north and west sides — a fence, a row of bales, a hedge — is worth more than a wrap, and a brick on the lid stops a February gust taking it off.

Syrup feeding ends on its own around now: below about fifty degrees the bees stop taking it and cannot cure it anyway. If the hive is still light in late October, switch to dry sugar or a candy board on the top bars. It is not as good as honey they made themselves. It is far better than the alternative.

November and December: the broodless window, then hands off

The cluster forms once the air is under about fifty-seven degrees, and by late November the queen has stopped laying and there is no capped brood in the hive. That is the window oxalic acid was made for. With no brood, every mite in the colony is out on an adult bee where the acid can reach it, and a single dribble or vaporisation on a still day above freezing knocks the population down to almost nothing for the start of spring. Follow the label for the method and the dose; the timing is the part beekeepers get wrong, and it is simply: when there is no brood, which here means roughly Thanksgiving to New Year.

Put a candy board or a couple of pounds of dry sugar on a sheet of newspaper on the top bars if you have any doubt about weight, close the hive, and then do the hardest thing in beekeeping, which is nothing. Every time the lid comes off in winter the cluster loses heat it spent honey to make. The colony heats itself, not the box — the walls of a hive in January are barely warmer than the air outside, and the core of the cluster is in the nineties only while there is brood to keep warm. Do not open it below fifty degrees. Put your ear to the side and knock; a hum means they are there.

January and February: the hungry months

Here is the thing that surprises new beekeepers most. The colony uses very little honey in December. It uses a great deal in February, because the queen starts laying again in the middle of winter, weeks before anything blooms, and brood has to be held at ninety-three degrees whatever the weather is doing. Consumption climbs just as stores are lowest, and a cluster in a long cold snap cannot move sideways to reach honey two frames away. That is how a hive starves inches from food, and February is when it happens.

So on any day above fifty degrees from January on, heft the back of the hive. If it feels light, lift the lid for the ten seconds it takes to lay sugar or fondant directly on the top bars over the cluster — directly above, because above is the one direction a cold cluster can always reach. Then close it and wait for the maples, which bloom here in late February and are the first real food of the year.

The list

None of this is difficult, and almost all of it is timing. The mite count is the one that matters, and it matters in August, which is why a post that publishes in late September opens by telling you to do it this week if you have not. The rest is a mouse guard, a sheet of foam, an inch of shim and the discipline to leave the lid on. Do those, and the cold can do what it likes.