[Babase] Notes regarding Amboseli minimum temperatures;
look at when returning to Scott's ms
Lacey Maryott
lacey.maryott at duke.edu
Wed Jan 2 09:14:16 EST 2008
I thought it may be useful to have this archived, so I'm forwarding it
to the list.
-------- Original Message --------
Subject: Notes regarding Amboseli minimum temperatures; look at when
returning to Scott's ms
Date: Mon, 31 Dec 2007 10:01:41 -0500
From: Jeanne Altmann <altj at Princeton.EDU>
To: Susan Alberts <alberts at duke.edu>
CC: Katherine H. Fenn <kfenn at Princeton.EDU>, Lacey Maryott
<lacey.maryott at duke.edu>
Susan,
Am sending this now as it’s been on my list this past month, and I don’t
want to forget for when you get back to Scott’s analyses. This is really
a reminder of something we talked about in Duke and perhaps afterwards
when we were considering what to do about Scott’s finding of the
elevated Max temp for the first 5 yrs in camp and we thought more about
Min temp.
Regarding our recorded minimum temperatures:
Through the years as the baboons moved farther away and descended
earlier, our departure time from camp and, therefore, our reading of the
min-max temp changed. In general, it got earlier, but when we switched
to the ‘half day’ field day system, it became about 0500-0520 on early
field days and a range of times after 0600, but usually by 0830 on late
field days or Sundays (I think that if it’s RSM reading, usually
0600-0700, if the others then more likely 0800 or so). As we found a
couple of years ago with Weatherhawk data, minimum temperature occurs
here about 0700 (because we record on the hour, we don’t know the exact
time, but this is a reasonable time for the min to occur). This means
that what we have in our records for daily min temperature is a mixture
of information, examples as follows I think are right and give a feel
for the issue:
--Temp read 0500-0530 3 days in a row: The min recorded on days 2 and 3
will represent the previous day’s min as intended.
--Temp read 0800 3 days in a row: The min recorded on days 2 and 3 will
represent the min for those days, not for the previous day.
--Temp read at 0800 on day 1, then at 0515 on day 2: The day 2 record
will not represent the min from day1 as the min for day 1 would have
been past when the gauge was rest on day 1. Nor will it represent day 2
min because the temp would continue to do down for another 1-2hrs.
Depending on when the temp is read on day 3,and the relationship between
the true mins on the various days, the day 3 record may represent the
day 2 min, as we assume, or it may not.
We could identify dates before which the min records are as intended. We
could also search for some dates thereafter when the combination is just
right for a day’s record to be the same. This would all be messy,
time-consuming and probably reanalysis. Since Scott uses the average
daily values for the month as the value for analysis, we could also
think about what the effect is likely to be on this value. I think that
in general, the within month variance is affected; I’m less clear on
whether we are likely to have any systematic bias.
jeanne
--
Lacey Maryott
Alberts Lab
Department of Biology
Duke University
ph: 919-660-7306
fax: 919-660-7293
Lacey.Maryott at duke.edu
More information about the Babase
mailing list