BEE-L Archives

Informed Discussion of Beekeeping Issues and Bee Biology

BEE-L@COMMUNITY.LSOFT.COM

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Jerry Bromenshenk <[log in to unmask]>
Reply To:
Informed Discussion of Beekeeping Issues and Bee Biology <[log in to unmask]>
Date:
Thu, 4 Feb 2016 13:30:34 -0500
Content-Type:
text/plain
Parts/Attachments:
text/plain (28 lines)
>>>  What, other than barometric pressure, would reliably precede a rainstorm by a day?

I've been dragging my feet on this one.  It's complicated.  We have hard data to show that they can predict incoming thunderstorms, but not a day or two ahead - rather 20-30 minutes.  Our data also shows that it is NOT simply barometric pressure.

From 1995 through 2001, we had 50 Smart Hives on Aberdeen Proving Grounds, with groups of 10 hives at 5 different sites.  The hives were wired, connected to the Internet.  Each of the 50 hives was continuosly monitored for weight changes, internal RH and temperature, bidirectional flight counters, etc. plus we placed a digital weather station at each site, and had access to weather data from the Army base, which included changes in electrical charges in the air (APG has bunkers full of munitions, conduct live fire exercises, etc.).  Base personnel hunkered down when a thunderstorm was approaching - the change in the charge in the air gave them an early warning.  APG is in a zone of frequent thunderstorms.

Robert Seccomb completed and published his graduate thesis using Artificial Neural Networks (ANNs) to compare colony  preformance and bee activity with weather.  His aim was to be able to tell from a remote location whether anyting unusual was happening (the worry was unexpected exposure to a release of toxic gases) from any of APGs numerous military dump sites.  APG is only 16 miles from Baltimore.  If bees went out,  but didn't come back - we'd get an automatic alert from the affected location.  But, we didn't want to set off a panic if it was simply something like it was raining or windy.

Robert showed that he could correlate many colony and bee activity changes with weather (not surprising).  He could determine what the normal parameters should be for any of  these at any moment, given the weather at that time, and he could easily flag anomalies (non-weather induced variations).  He was even able to predict swarm events - although the ANN didn't know what a swarm was.  The program would predict to the exact day when the colony would swarm, but it didn't have sufficient data to know what happened in the aftermath, so the AI in the computer surmised that the bees would come home by nightfall.

We weren't trying to predict weather 1 or more days out, although we didn't see any obvious signs of that happening.  What we did see is that bees headed home 20-30 minutes before a thunderstorm hit.  First time we saw this was late afternoon, we were looking at the live display (on the computer) of the flight activity of 10 colonies at an APG site.  Suddenly, we saw a huge spike in numbers of incoming bees from one hive, then we saw it happen at another hive, and within a few minutes, the bees came racing back to all 10 hives.  It was calm, still sunny, but from the west, a black cloud was approaching, and it about 1/2 hour, it hit.

We also saw that most of the colonies stayed inside after the storm passed.  It was early evening, still daylight.  Except, as the raindrops eased in intensity, one colony went back to work.  Over the years, we saw this over and over - foragers racing home just before a storm hit.  Most colonies shutting down flight for the day if the rainstorm occurred in late afternoon.  But a few of the 50 colonies, would resume flight as soon as the rain slacked off.  Next morning, all of the colonies at a sight would go to out work about the same time.

We had the concurrent weather data.  What was clear was that the return home behavior was NOT simply changes in barometric pressure.  We suspect that the bees were reacting to a variety of indicators - changing light levels, rh, and given anecdotal observations on the base - when a storm was rolling in fast with a lot of  electrical energy, warning sirens would go off - we suspect that they were keying in on this factor.

We know from our own published work that bees carry static charges on their bodies, and others have shown that bees have some control over that charge.  Conclusion - it's not simply barametric pressure, it may be that just like we notice changes in wind, light, smell of the air, they do the same. But if I were a betting man, I'd look at electrical charge in the air.

I don't know how far in advance bees can predict weather.  I do know that they respond to short term events.  I also know that bees in a greenhouse respond to external weather.  We've done years of Procoscis Extension research.  In winter, bee responsiveness tanks when it's snowing outside, even though the bees are free flying in heated greenhouses.  Snowy day, don't bother trying, the assay won't work.




             ***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software.  For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html

ATOM RSS1 RSS2