Subject: | |
From: | |
Reply To: | |
Date: | Wed, 31 Mar 2021 23:44:08 +0000 |
Content-Type: | text/plain |
Parts/Attachments: |
|
|
If I have a hive that has 1% phoretic mites and manage to randomly sample 300 bees I will get the following results:
0 mites 5% of the time
1 mite 15% of the time
2 mites 22% of the time
3 mites 23% of the time
4 mites 17% of the time
5 mites 10% of the time
6 mites 5% of the time
7 mites 2% of the time
8 mites 1% of the time
All %s rounded to the nearest % obviously. Obviously the results curve shows the expected skewness towards high mite counts as it is impossible to have a count less than zero. The results also show that 62% of the time the result will be 2, 3 or 4 mites from a 300 mite sample. For a fast and easy screen this strikes me as close enough often enough to be reasonably useful if you are sampling several hives and have a treatment threshold of 1%. All you need to see is one or more mites off of three or four hives sampled to know you need to treat.
Now, what happens if my threshold is say 3%? On a random sample of 300 bees you would expect the following results from a 3% hive:
3 mites 1% of the time
5 mites 6% of the time
7 mites 12% of the time
9 mites 13% of the time
11 mites 10% of the time
13 mites 5% of the time
15 mites 2% of the time
17 mites 1% of the time
If you treat every hive that washes 5 or more mites you will be over 95% confident you have treated all hives at or above a true 3% mite load. Now, some of those treated hives could have been only 1% mite load. So, you have to consider what your objective is before you use these numbers. Is your objective to treat essentially all hives that have a 3% mite load? Or is your objective to treat as many hives with a 3% or higher mite load but not treat more than 5% or those hives that have only a 1% mite load?
Is any of this going to give you a perfect estimate of what needs treatment? Nope. And neither will anything else in my opinion, short of washing samples of several thousand bees and that is not acceptable either. Another idea might be to wash a 300 bee sample from four hives and add the numbers together. Above some threshold you simply treat everything. That of course risks missing a mite bomb which would be important if you were using oxalic acid vapor but not so important for apivar or thymol or formic acid.
These numbers are dirt easy for anyone to calculate. Simply google "binomial calculator" and go to the first hit and use the calculator. Everyone should know how to use a binomial calculator anyhow and use one a few times a week in their regular life to estimate risk.
Dick
***********************************************
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
|
|
|