Joining list. Calendar reforms.

[Irv Bromberg]

On 2009 Jun 18, at 19:56 , MIKE OSSIPOFF wrote:

A radio weather reporter a few days ago. Respondents to questions at Internet websites. Opinions of people I've spoken to. In Santa Cruz, where I grew up, it was just common knowledge that June was summer. The school systems apparently agree, when they choose early June as the time to start summer vacation.

Where is the date then? In the statements from everyone I've heard expressing an opinion on the subject. Some will then say, "But of course officially it's June 21". But it's always clear that they believe it because authorities couldn't be wrong, not because it's their own perception.

Irv replies:  OK, so you've just proven that you have hardly any data, certainly no objective data, and that there is no reliable way to gather public perception of the thermal seasons because their ideas are so strongly influenced by their knowledge of the dates of the equinoxes and solstices.

You're an astronomical purist, and there's nothing wrong with that. You study minutiae of changes in the earth's orbit, and I don't criticize that. But those changes are not going to invalidate the seasonal calendars that I propose. How long before the obliquity changes enough to change people's seasonal perceptions? Will there still be humans on the Earth?

Irv replies:  Certainly.  At the present rate of perihelion advance perihelion advances about 1° per 59 years and therefore takes about 360° × 59 ≈ 21,240 years to revolve once relative to the northward equinox of the date.  Thus it will take 21,240/12=1770 years to advance through one month of the calendar.  It will actually take less time, in fact progressively less time, because the Earth orbital eccentricity is declining (and will continue to decline for tens of millennia), which will cause perihelion to advance progressively more rapidly.  You certainly will need to adjust month lengths at least that frequently, possibly much more frequently depending on the rules of the calendar.  Also, depending on the rules of the calendar it might work out that although the adjustments are every millennium or so, it could be that the next required adjustment is coming up soon.  Furthermore, as perihelion advances there are 4 possible points in the calendar year where an adjustment may be called for, and those adjustments are unlikely to be synchronized, so it is likely that adjustments will occur at 4 times more often than you might expect from these figures.  Thus every 2-4 centuries it may be necessary to adjust a month length here and there in the calendar, to maintain optimal alignment with the mean seasons.

-- Irv Bromberg, Toronto, Canada

<http://www.sym454.org/seasons/>

[Irv Bromberg]

On 2009 Jun 18, at 19:38 , MIKE OSSIPOFF wrote:

Irv has been defining lag as the time between, say, the summer solstice and the beginning of calendar summer. I say that, for a meaningful measurement of lag, the lag should be measured from the summer solstice to the _middle_ of calendar summer.

Irv replies:  You can't measure from the north solstice to any part of the calendar year without defining the rule that connects the solstice to the calendar.

-- Irv Bromberg, Toronto, Canada

<http://www.sym454.org/seasons/>

[Irv Bromberg]

On 2009 Jun 18, at 19:05 , MIKE OSSIPOFF wrote:

On 2009 Jun 17, at 19:27 , Irv Bromberg wrote:

The thermal lag will also vary with the solar sunspot cycle

Why would that substantially change the timelag? It would change the intensity of sunlight, but would a small change in the intensity sustantially change the lag? By an amount at all comparable to the temperate regions' variation in timelag?

Irv replies:  Data.  Need data.  You don't know that it will affect it or not until you get the data to find out if it does or doesn't.  Or a meteorological literature source that gives the answer.  We're currently at a sunspot / solar radiation minimum.  I would expect that to prolong the time to peak temperature after the north solstice, and to reduce the peak maximum, and it should shorten the time to maximum cold temperature after the south solstice, and make the maximum cold colder.

You could compare year 2008 and 2009 data with data 5-6 years earlier when it was a sunspot / solar radiation maximum.

On the other hand, it isn't possible to predict the sunspot count or to predict when the solar maximum or solar minimum will next occur on the basis of daily local temperature records, at least not with present knowledge.  We're not even sure when it is a solar maximum until after we are a couple of years past it!  And we're not sure that it is a solar minimum until we see very low sunspot counts persisting for a couple of years.

-- Irv Bromberg, Toronto, Canada

<http://www.sym454.org/seasons/>

[Brillig]

Dear Irv and Calendar People,

On Thu, Jun 18, 2009 at 7:54 PM, Irv Bromberg<irv.bromberg@...> wrote:

>  We're not even sure when it is a solar maximum until after we are a couple
> of years past it!  And we're not sure that it is a solar minimum until we
> see very low sunspot counts persisting for a couple of years.

The minimum is identified retroactively, but I believe it's not
identified by sunspot count, but by sunspot latitude. When the ratio
of sunspot count at high latitude to sunspot count at low latitude
flips, that marks the start of a new sunspot cycle. The new sunspot
cycle marks the end of the low.

Victor

[Irv Bromberg]

On 2009 Jun 18, at 21:07 , Victor Engel wrote:

On Thu, Jun 18, 2009 at 7:54 PM, Irv Bromberg<irv.bromberg@...> wrote:

 We're not even sure when it is a solar maximum until after we are a couple

of years past it!  And we're not sure that it is a solar minimum until we

see very low sunspot counts persisting for a couple of years.

The minimum is identified retroactively, but I believe it's not
identified by sunspot count, but by sunspot latitude. When the ratio
of sunspot count at high latitude to sunspot count at low latitude
flips, that marks the start of a new sunspot cycle. The new sunspot
cycle marks the end of the low.

The flip is a magnetic one, alternating between solar hemispheres.  So for example the few remaining low latitude sunspots in the northern solar hemisphere get outnumbered by increasing numbers of sunspots appearing at high latitudes in the southern solar hemisphere, or visa versa.  The total cycle is about 22 years, alternating as 11-year subcycles between the solar northern hemisphere and the solar southern hemisphere.

Perhaps approximately 11-year periodicity has something to do with the 11.86-year orbital cycle of Jupiter?  Or the intervals between heliocentric conjunctions of Jupiter and Saturn?  Additional variations when Uranus is also in heliocentric proximity to Jupiter and/or Saturn?  Jupiter's magnetic field (magnetosphere) is by far the largest "object" in the solar system after the heliosphere.  Sometimes the Jovian magnetosphere encloses Saturn while Saturn is at opposition from Jupiter.  Nevertheless, there can't be any direct magnetic interaction between Jupiter and Sun, because even at perihelion Jupiter is about 740,573,600 km from Sun whereas its magnetosphere extends only about 7,000,000 km in the direction of Sun.

I always thought that we have 12-month years because there are approximately 12 lunar cycles per solar year, and that that lead to the division of the zodiac into 12 signs.

However another possibility is that each time Jupiter is at opposition as seen from Earth it has advanced eastward by about 30° or one zodiac sign.

As each solar sunspot cycle progresses the sunspots appear at lower and lower latitudes, with solar radiation and sunspot count maxima occurring when the majority of sunspots are appearing at latitudes roughly midway between the solar equator and solar pole in the most active hemisphere.

-- Irv Bromberg, Toronto, Canada

<http://www.sym454.org/seasons/>

[Brillig]

Dear Irv and Calendar People,

On Thu, Jun 18, 2009 at 9:34 PM, Irv Bromberg<irv.bromberg@...> wrote:

> The flip is a magnetic one, alternating between solar hemispheres.  So for
> example the few remaining low latitude sunspots in the northern solar
> hemisphere get outnumbered by increasing numbers of sunspots appearing
> at high latitudes in the southern solar hemisphere, or visa
> versa.

This is incorrect. A cycle starts with high latitude sunspots. During
the progression of the sunspot cycle, the location of the sunspots
becomes lower in latitude, approaching the equator from both the north
and the south. When the new sunspots at high latitude (north and
south) outnumber the new sunspots at low latitude, we have begun a new
cycle. See

http://upload.wikimedia.org/wikipedia/commons/1/1e/800px-Sunspot_butterfly_with_graph.gif

Victor

[Irv Bromberg]

On 2009 Jun 18, at 22:44 , Victor Engel wrote:

On Thu, Jun 18, 2009 at 9:34 PM, Irv Bromberg<irv.bromberg@...> wrote:

The flip is a magnetic one, alternating between solar hemispheres.  So for

example the few remaining low latitude sunspots in the northern solar

hemisphere get outnumbered by increasing numbers of sunspots appearing

at high latitudes in the southern solar hemisphere, or visa versa.

This is incorrect. A cycle starts with high latitude sunspots. During
the progression of the sunspot cycle, the location of the sunspots
becomes lower in latitude, approaching the equator from both the north
and the south. When the new sunspots at high latitude (north and
south) outnumber the new sunspots at low latitude, we have begun a new cycle.

Irv replies:  Sorry.  Let me try again.  Quoting from the "Solar Activity" essay by Ken Tapping in "Observer's Handbook 2009" published by the Royal Astronomical Society of Canada, page 194, second paragraph:

"The 'conventional' magnetic configurations for active regions reverse on alternate activity cycles.  For example, during Cycle 22, active regions in the northern solar hemisphere were oriented with their "negative" (i.e. south-seeking) magnetic polarity ends leading and "positive" (north-seeking) ends following, with the reverse situation in the Southern Hemisphere.  In Cycle 23 this arrangement is reversed.  A magnetic activity cycle, which is probably a more realistic description of the rhythms of solar activity, is equal to two of Wolf's activity cycles and takes about 22 years to complete."

-- Irv Bromberg, Toronto, Canada

<http://www.sym454.org/>

[Karl Palmen]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear Irv,  Mike and Calendar People

 

From: East Carolina University Calendar discussion List [mailto:CALNDR-L@...] On Behalf Of Irv Bromberg
Sent: 19 June 2009 01:47
To: CALNDR-L@...
Subject: Re: June-September season

 

On 2009 Jun 18, at 19:38 , MIKE OSSIPOFF wrote:

Irv has been defining lag as the time between, say, the summer solstice and the beginning of calendar summer. I say that, for a meaningful measurement of lag, the lag should be measured from the summer solstice to the _middle_ of calendar summer.

 

Irv replies:  You can't measure from the north solstice to any part of the calendar year without defining the rule that connects the solstice to the calendar.

 

KARL SAYS: I disagree with Irv.

In any properly defined solar calendar you can find which date of any given year the north solstice occurs and then work out how many days that occurs before the middle of the north summer season date. Such a calendar need not have a rule that refers to any solstice at all. What is needed as a calendar rule that defines the calendar north summer season or the middle of north summer date.

This lag may vary from year to year.

 

Karl

 

10(09(26 till noon

--
Scanned by iCritical.

[Karl Palmen]

Dear Mike, Irv and Calendar People

It would have helped if Mike said _middle_ from the outset. Irv seems to
have misunderstood it as _start_.

Karl

10(09(26 till noon

-----Original Message-----
From: East Carolina University Calendar discussion List
[mailto:CALNDR-L@...] On Behalf Of MIKE OSSIPOFF
Sent: 19 June 2009 00:38
To: CALNDR-L@...
Subject: Re: June-September season

 
Dear Karl, Irv, Victor and Calendar People,
 
Karl, you wrote:
 
> I don't understand what Mike means by "the
> year based on solstices and equinoxes".

I reply:
 
Ok, good question. I had some doubts about the usefulness of that
wording when I wrote it.
 
This is all I meant:
 
When reckoning the lag of a seasonal calendar, it's necessary to measure
that lag between the _middle_ of calendar summer or winter and the
solstice that it lags behind. That's because the solstice is the middle
of a high north or south declination period, and is the maximum north or
south declination. So, we would be, and should be, comparing one middle
to another. That's what I mean by saying we should compare corresponding
parts when measuring lag.
 
Irv has been defining lag as the time between, say, the summer solstice
and the beginning of calendar summer. I say that, for a meaningful
measurement of lag, the lag should be measured from the summer solstice
to the _middle_ of calendar summer.
 
 
You continued:
 
> Does he mean a year divided into four seasons (not necessarily
> equal) in which each season has its solstice or equinox in the middle?

I reply:
 
I just meant that, if you're going to reckon lag from the duration
between the summer solstice and some part of calendar summer, then you
should use the _middle_ of calendar summer.
 
But I spoke of another way to reckon lag, and it gives almost the same
result. For the whole-months unequal seasons calendar, with seasons
starting on June 1, October 1, December 1 and April 1, the method
described in the paragraph before this one, and the method described in
the paragraph after this one, both result in the same figure for the
lag: 40 days:
 
The other method that I mentioned was to find the lag such that the
lag-adjusted solar declinations at the beginning and the end of calendar
summer are the same.
 
By "lag-adjusted declination" for a certain date, I mean the declination
at the time preceding that date by the amount of the lag.
 
But, in answer to your question, I meant that, if you're going to
measure lag between the summer solstice and some part of calendar
summer, then the measurement should be between the summer solstice and
the _middle_ of calendar summer.
 
 
I'd said:
 
> To measure time-lag from temperature records, I look at the
> delay between the hottest and coldest months and the corresponding
solstice.

 
You replied:
 
> This also agrees with the assumption, but is not such a good
> idea if the temperatures of the months are grossly asymmetrical as for
El Paso
> as shown in (
http://en.wikipedia.org/wiki/El_Paso#Temperature_statistics
> ) where the hottest month is the first of the three or four hottest
> months by average high.
 
Ok, that's true. The middle of summer is better defined as the day with
equal summed temperatures before and after it, over some pre-chosen
period before and after--as opposed to merely the month with the highest
mean temperature.
 
I was assuming that the hottest month would also be the middle as
defined above. Of course I looked at so many cities that hopefully any
local skewness would tend to cancel out in the overall conclusions.
 
But yes, I too was thinking that it would be better to try to interpret
the monthly temperature averages in a way similar to the method that I
used for Santa Cruz, California (the method described two paragraphs
before this paragraph). That takes longer, and I wanted quick answers,
and so I just looked at the hottest and coldest months. Again, I hope
that, because I looked at so many cities, the conclusion that the lag
varies from .75  to 1.75 still means something.
 

I'd said:

> The June-September summer (or North, for international purposes)
> has advantages. The whole old-calendar months make it easier to
describe and
> simpler to define to people. The time lag of 40 days is so close to 38
days
> that the difference is negligible in comparison to the unavoidable
imprecision
> of the monthly temperature records. The time lag iss 40, even if you
look for
> the lag such that the timelagged solar dec is equal at both ends of
summer
> (unless I made an error). The dec then is .51 of maximum.
>

You wrote:

> The time-lag of 40 leads to a season of 121 days from June 1 to
> September 29 inclusive with June 21 placed 40 days before the middle
day of
> July 31.
 
Ok, but I just meant that (if I didn't make an error) a calendar summer
of June thru September implies a time-lag that rounds off to 40 days--40
days being the closest integer number of days.
 
A terrestrial seasonal calendar with seasons of 17 weeks and 9 weeks,
with the 4-week months and 5-week months, is an appealing way to have a
meaningful seasonal calendar, with the convenience that comes with a
fixed calendar.
 
As I was saying in another posting today, I'd name the months for their
position in a particular season. For instance, this month would be
SummerI or (for international purposes) NorthI.
 
Isaac Asimov didn't use months, but his summer and winter were shorter,
because he had four equal seasons. With the longer summer and winter,
months become more desirable, for convenient and easy determination of
days of the week for particular dates. Finding multiples of 7 greater
than 91 and not more than 119 wouldn't be prohibitively time-consuming,
and would certainly be easier than day-of-week/date calculations now,
but, nevertheless, with 4-week and 5-week months, people can be offered
an easier and more convenient determination of the day of week for any
date.
 
 
Mike Ossipoff
 
_________________________________________________________________
Bing(tm)  brings you maps, menus, and reviews organized in one place.
Try it now.
http://www.bing.com/search?q=restaurants&form=MLOGEN&publ=WLHMTAG&crea=T
EXT_MLOGEN_Core_tagline_local_1x1
--
Scanned by iCritical.

[Karl Palmen]

Dear Mike and Calendar People

-----Original Message-----
From: East Carolina University Calendar discussion List
[mailto:CALNDR-L@...] On Behalf Of MIKE OSSIPOFF
Sent: 18 June 2009 23:12
To: CALNDR-L@...
Subject: June-September, contd.

 
Yes, the terrestrial-seasonal calendar with unequal seasons works very
well as a fixed calendar, with the popular 4-week and 5-week months.
 
I'd name the months according to their place in a calendar season. For
instance, this month would be SummerI or NorthI.
 
Maybe the 4 and 5 week months can be arranged so as to come close to an
alternation of 4 and 5 week months. Myself, I like the idea of starting
each calendar season with a 5-week month. That's simpler. And it gives
the year's first month, the month containing the important Christmas
holiday and the winter solstice, a 5-week length.

KARL SAYS:
Actually a typical year of 52 weeks would require four months of five
weeks, which is one per season. So one could start each season with a
five-week month and then follow by four-week months till next season.
This works with seasons of 13 weeks or seasons of 17 and 9 weeks.
If using a constant week of 7 days, some years need an additional week
(a leap week). This can be added to the last month of the year if it
normally has 4 weeks.
See http://www.hermetic.ch/cal_stud/palmen/lweek1.htm for more about
leap week calendars.

The simplest pattern of 4-week and 5-week months is to have every 3rd
month a five-week month in each 52-week year. This leads to quarters
with exactly the same month pattern. Seasons of 17 weeks and 9 weeks can
be accommodated in such a calendar, if every long season starts on a
four-week month.

Irv has suggested such a calendar (see
http://en.wikipedia.org/wiki/Symmetry454 ) The 12 months have the same
name as in the Gregorian calendar and the middle month of each quarter
(February, May, August and November) have 5 weeks. In a 53-week year
December also has 5 weeks. Starting the seasons on April, June, October
and December would ensure seasons of 17 and 9 weeks as stated before
(except for season beginning with December of a 53-week year). Indeed,
starting the long season on any 4-week month would ensure the 17|9 week
season pattern. An earlier such Calendar was the Bonavian Civil Calendar
http://personal.ecu.edu/mccartyr/bonavian.html .

Note that neither calendar has the 3rd month of the quarter a 5-week
month. This allows the last month of the year to be extended to 5 weeks
in a 53-week year. This is the most convenient month to extend, because
it ensures that each date within a year is the same day of the year
counting from the start, so making calendar calculations easier.

Karl

10(09(27

--
Scanned by iCritical.

[MIKE OSSIPOFF]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Victor,
 
You wrote:
 
This will be my last post on the subject.
 
I reply:
 
...But you said that before. So why should I believe you now? :-)
 
You continued:
 
I'm simply not interested in
> your proposal for reasons I've already mentioned.
 
I reply:
 
Fair enough. As I was saying, my purpose hasn't been to convince the anti-seasonal list members, but only to state the case for a seasonal calendar, preferably a terrestrial seasonal calendar, or at least a declination calendar, for the benefit of anyone other than those who are already firmly opposed to such a thing.
 
You continued:
 
In any case, the data (for Austin)  are available from a number of
> weather-related websites.

I'll check it out. Daily records averaged over a long period are best, but monthly will do.
 
I'd said:
 
> > In Dallas, the timelag is 1.0 month in summer and .75 month in winter.

You replied:
 
> I haven't followed the discussion about the definition of timelag.
 
I reply:
 
For the purposes of determining it from those monthly temperature records in a quick way, it's the time difference between the middle of the hottest month and the summer solstice; and between the middle of the coldest month and the winter solstice. Not precise?  No, but that isn't so bad, when you consider that the timelag varies from .75 month to 1.75 month, throughout the north and south temperate zones. Later I'll find some daily records and determine timelags from them.
 
You continued:
 
I
don't know what the reference zero time is.
 
I reply:
 
As I'm measureing it, timelag is measured from the summer or winter solstice.
 
You continued:
 
That doesn't really
> matter, though, because given the numbers you've just mentioned, there
> is a difference of half a month already for the summer.
 
I reply:
 
The Subjective Seasonal Calendar uses a timelag of 1.25 month. That's the midrange of the variation over nearly everywhere in the temperate zones, nearly always differing from the actual lag by no more than 2 weeks.
 
The summer lags I reported for Dallas and Houston were each only .25 month different from the Subjective Seasonal Calendar's 1.25 month lag.
 
You continued:
 
That proves my
> point.

I reply:
 
Yes, if your point is that the time lag isn't the same everywhere. No, if you're claiming that the .5 month maximum variation from 1.25 month, nearly everywhere in the temperate zones, spoils the value of a terrestrial seasonal calendar.
 
The point isn't to precisely match the seasons everywhere. The point is to name the year-divisions in an un-arbitrary and useful way, a natural way. And what could be more un-arbitrary, useful and natural than year-division-names based on the natural year that all calendars measure.
 
Your city has a lag different from 1.25? That's ok; the solar declination's variation, adjusted by the midrange value of the temperate-region timelag, remains a meaningful basis for naming the year-divisions.
 
You continued:
 
My figures came from weatherunderground.com, which gathers data from a
> number of sources. I believe the maxima I cited are either 50 year or
> 100 year averages. I could be wrong about these numbers, but it is
> abundantly clear that the numbers are averages of many years. You can
> tell immediately by how smooth the graphs are.
 
I'll check it out.
 
You continued:
 
Note that I'm talking
> about what weatherunderground.com refers to as seasonal averages, not
> all time maxima for specific dates. The latter are all over the place
> and not useful for this discussion.
 
I reply:
 
Quite so.
 
I'd said:
 
> > To test any particular date's qualification as the middle of winter:
> >
> > Add up the daily temperatures before and after the date being tested, for some pre-chosen number of days before and after that date. Maybe add up the temperatures for the 60 days before the date being tested, and also for the 60 days after the date being tested.
> >
> > If those sums are equal, or as equal as you can find, then the date being tested is the middle of winter.
>
You replied:
 
> That definition is not well-defined.

I reply:
 
If you mean that it could conceivably fail to specify one and only one day, then yes, that's a valid concern. But not an un-solvable problem. In fact, with records averaged over many years, it is unlikely to be a problem at all.
 
I'd said:
 
>
> > Of course the procedure is the same for finding the middle of summer.
>
> And you've just demonstrated one reason why it's not well defined.
 
I reply:
 
You're implying that, because there will be such a day at midsummer, and also at midwinter, that means we can expect others. I don't think that follows.
 
You continue:
 
You
> are making an assumption, which needs first to be proved (I suggest
> you can't) that there are only two such parts of the year, one for
> summer and one for winter, for all parts of the globe.

I reply:
 
Of course, for any one particular year, with fluctuating temperatures, it would be possible for there to be more than one day such as I spoke of, during the winter.
 
But those fluctuations aren't there, in records averaged over many years.
 
When looking at single years, I'd suggest keeping the answer that recurs. For Santa Cruz, I got only one answer, January 28, for all the years I checked, except for an El Nino year. Therefore, January 28 can be accepted as midwinter, not an anomaly due to fluctuations.
 
Mike Ossipoff
 

---

Lauren found her dream laptop. Find the PC that’s right for you.

[MIKE OSSIPOFF]

Some parts of this message have been removed. Learn more about Nabble's security policy.

 
 
Irv--
You wrote:
 
OK, so you've just proven that you have hardly any data
 
I reply:
 
Just what I've heard from everyone who's expressed an opinion on the subject.
 
You continued:
 
, certainly no objective data
 
I reply:
 
Irv, it's called the Subjective Seasonal Calendar. Subjective. I chose June as the beginning of summer, because that's people's perception. Likewise I've heard that in Australia it's December.
 
You continued:
 
, and that there is no reliable way to gather public perception of the thermal seasons because their ideas are so strongly influenced by their knowledge of the dates of the equinoxes and solstices.
 
I reply:
 
I don't agree there. People usually make it clear whether they're talking about their own perception, or the official date.
 

I'd said:

 

You're an astronomical purist, and there's nothing wrong with that. You study minutiae of changes in the earth's orbit, and I don't criticize that. But those changes are not going to invalidate the seasonal calendars that I propose. How long before the obliquity changes enough to change people's seasonal perceptions? Will there still be humans on the Earth?

You replied:  

 

Certainly.  At the present rate of perihelion advance perihelion advances about 1° per 59 years...

 

I reply:

 

...and that will change by only a day or two the length of the period where the solar declination is above some particular north or south value.

 

Those few days aren't the problem that you seem to think they are.

 

Mike Ossipoff

 

---

Bing™ brings you maps, menus, and reviews organized in one place. Try it now.

[MIKE OSSIPOFF]

Some parts of this message have been removed. Learn more about Nabble's security policy.

 OSSIPOFF wrote:
 

Irv has been defining lag as the time between, say, the summer solstice and the beginning of calendar summer. I say that, for a meaningful measurement of lag, the lag should be measured from the summer solstice to the _middle_ of calendar summer.

Irv replies:  

 

You can't measure from the north solstice to any part of the calendar year without defining the rule that connects the solstice to the calendar.

 

I reply now:

 

I don't know what you mean. I've described two ways of deffining the Summer (or North) calendar season. One of those definitions mentions the summer solstice, but the other doesn't.

 

But, in the current discussion about equal seasons (and I do _not_ advocate equal seasons), the positioning of the quarters has always been specified in terms of a solstice or equinox.

 

Anyway, I've clarified, earlier, two ways to measure the timelag of calendar summer, and one of those is by the duration between the summer solstice and the middle of calendar summer.

 

Mike Ossipoff

 

 

 

---

Microsoft brings you a new way to search the web. Try Bing™ now

[MIKE OSSIPOFF]

Some parts of this message have been removed. Learn more about Nabble's security policy.

 
Dear Karl and Calendar People,
 
True, the leapweek should occur in the last month of the year, so as to leave the rest of the year's dates undisturbed.
 
So then, two possibilities, each with its own advantages, for arranging the 4 week and 5 week months--starting each calendar season with a 5-week month, or having a regular consistent pattern of 4-week and 5-week months.

Though my favorite proposal is the terrestrial seasonal calendar with unequal calendar seasons, the next best, from my point of view, would be a declination calendar, with North and South calendar divisions having the solar declination greater than .5 of its north maximum or greater than .5 of its south maximum. As I was saying, that's virtually only a time-lag away from a seasonally-meaningful terrestrial calendar.
 
But my proposal is a terrestrial seasonal calendar with unequal seasons, in the various versions that we've discussed.
 
Mike Ossipoff
 
 
Mike Ossipoff
 

> Date: Fri, 19 Jun 2009 12:52:53 +0100
> From: karl.palmen@...
> Subject: Re: June-September, contd.
> To: CALNDR-L@...
>
> Dear Mike and Calendar People
>
> -----Original Message-----
> From: East Carolina University Calendar discussion List
> [mailto:CALNDR-L@...] On Behalf Of MIKE OSSIPOFF
> Sent: 18 June 2009 23:12
> To: CALNDR-L@...
> Subject: June-September, contd.
>
>
> Yes, the terrestrial-seasonal calendar with unequal seasons works very
> well as a fixed calendar, with the popular 4-week and 5-week months.
>
> I'd name the months according to their place in a calendar season. For
> instance, this month would be SummerI or NorthI.
>
> Maybe the 4 and 5 week months can be arranged so as to come close to an
> alternation of 4 and 5 week months. Myself, I like the idea of starting
> each calendar season with a 5-week month. That's simpler. And it gives
> the year's first month, the month containing the important Christmas
> holiday and the winter solstice, a 5-week length.
>
> KARL SAYS:
> Actually a typical year of 52 weeks would require four months of five
> weeks, which is one per season. So one could start each season with a
> five-week month and then follow by four-week months till next season.
> This works with seasons of 13 weeks or seasons of 17 and 9 weeks.
> If using a constant week of 7 days, some years need an additional week
> (a leap week). This can be added to the last month of the year if it
> normally has 4 weeks.
> See http://www.hermetic.ch/cal_stud/palmen/lweek1.htm for more about
> leap week calendars.
>
> The simplest pattern of 4-week and 5-week months is to have every 3rd
> month a five-week month in each 52-week year. This leads to quarters
> with exactly the same month pattern. Seasons of 17 weeks and 9 weeks can
> be accommodated in such a calendar, if every long season starts on a
> four-week month.
>
> Irv has suggested such a calendar (see
> http://en.wikipedia.org/wiki/Symmetry454 ) The 12 months have the same
> name as in the Gregorian calendar and the middle month of each quarter
> (February, May, August and November) have 5 weeks. In a 53-week year
> December also has 5 weeks. Starting the seasons on April, June, October
> and December would ensure seasons of 17 and 9 weeks as stated before
> (except for season beginning with December of a 53-week year). Indeed,
> starting the long season on any 4-week month would ensure the 17|9 week
> season pattern. An earlier such Calendar was the Bonavian Civil Calendar
> http://personal.ecu.edu/mccartyr/bonavian.html .
>
> Note that neither calendar has the 3rd month of the quarter a 5-week
> month. This allows the last month of the year to be extended to 5 weeks
> in a 53-week year. This is the most convenient month to extend, because
> it ensures that each date within a year is the same day of the year
> counting from the start, so making calendar calculations easier.
>
> Karl
>
> 10(09(27
>
> --
> Scanned by iCritical.
>

---

Microsoft brings you a new way to search the web. Try Bing™ now

[MIKE OSSIPOFF]

Some parts of this message have been removed. Learn more about Nabble's security policy.

 Say, at the finish of a car-race, car 1 and car 2 cross the finish line with car 1's front bumper 14 feet ahead of car 2's bumper. Does car 1 win? No. Car 2 wins, because car 2's front bumper was 2 feet ahead of car 1's rear bumper when they crossed the finish line.
 
Oh, excuse me, I should have told you that we're comparing car 2's front bumper with car 1's rear bumper.
 
Moral of the story: When comparing the positions of two things, by the locations of particular parts of those things, use the same part of both things. For instance, if you're going to compare to the summer solstice, which is the _middle_ of the high declination season, and you want to know how much calendar summer lags behind, then you want to know how much the _middle_ of calendar summer lags behind the summer solstice.
 
Of course I agree that the year's maximum or minimum temperature, when available only for one year, or just a few years, is no good for measuring a locale's seasonal timelag, because yearly maximum and minimum aren't stable measurements.
 
But, if we have averages over many years, then the year's average maximum or minimum temperature is suitable for the purpose of judging seasonal timelag.
 
In Santa Cruz, I used the method I described (finding the day with equal summed-temperatures for 2 months before and after it) because I didn't have longterm averages and so I needed a more stable measurement. With longterm averages, either would do.
 
Someone pointed out that the average year's maximum could be skewed away from the middle of summer determined in some other way. Sure, but, still, who's to say which is a  better measure?
 
The method I used in Santa Cruz had its own special value, for my particular purposes then, because I was interested in finding out what day would be appropriate to hold a celebration because we were past the middle of the cold days.
 
It was pointed out that the monthly averages, and my use of them, was a crude method for measuring seasonal timelag. True, but, because what I'm measuring, for temperate regions, varies by about two weeks each way from its usual midrange of 1.25 months, that reduces the importance of great precision in measuring seasonal timelag.
 
Still, of course I wouldn't just leave it at that. I'll look for daily (as oppposed to monthly) longterm average temperatures.
 
Also, for better interpretation of monthly longterm averages, I'll approxmate the temperature with a function of time, the function-value representing temperature. A function chosen so that its average value in each of six adjacent months is what the record-book says it was.
 
Multiplying a month's average temperature by the length of the month gives the value of the definite integral of the temperature over that month, and the function would be chosen so that that value is what it should be for each of six successive months.
 
Maybe a polynomical function, but probably better a sum of sine functions. Either a sum of two sine functions with variables of amplitude, phase and period, or a sum of three sine functions all with a 1-year period, with variables of amplitude and phase.
 
I'll try both of those two sums of sine functions, to find which does better in the months adjacent to the six months that determine the function.
 
I'll use Newton's method to numerically solve the system of 6 nonllinear equations. The time-consuming part will be the repeated solution of 6 simultaneous linear equations, till convergence is reached.
 
Especially if such a method is written as a computer program, it would be a way to get better information from the widely-available longterm monthly temperature averages.
 
By the way, I checked the temperature records for Austin and Alpine (the crude way, of course), and I found that Austin isn't anomalous at all. Alpine, for some of its lag values, has a measured lag of only .25 month or less. But the name "Alpine" suggests that it's at a much higher elevation than its surrounding region.  Is it then so unforgivable for such a different place to have a more significantly different seasonal timelag?
 
Anyway, when I find longterm daily temperature averages, &/or when I start using the sine approximation with the longterm monthly averages, I'll take another look at Alpine.
 
Mike Ossipoff
 
 
 
 
 
 
 

---

Insert movie times and more without leaving Hotmail®. See how.

[Karl Palmen]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear Mike, Irv and Calendar People

 

From: East Carolina University Calendar discussion List [mailto:CALNDR-L@...] On Behalf Of MIKE OSSIPOFF
Sent: 23 June 2009 00:37
To: CALNDR-L@...
Subject: Re: Middle vs. Start RE: June-September season

 

 Say, at the finish of a car-race, car 1 and car 2 cross the finish line with car 1's front bumper 14 feet ahead of car 2's bumper. Does car 1 win? No. Car 2 wins, because car 2's front bumper was 2 feet ahead of car 1's rear bumper when they crossed the finish line.
 
Oh, excuse me, I should have told you that we're comparing car 2's front bumper with car 1's rear bumper.
 
Moral of the story: When comparing the positions of two things, by the locations of particular parts of those things, use the same part of both things. For instance, if you're going to compare to the summer solstice, which is the _middle_ of the high declination season, and you want to know how much calendar summer lags behind, then you want to know how much the _middle_ of calendar summer lags behind the summer solstice.

KARL SAYS:

That argument does not work unless you state that the summer solstice is in the middle (of an zero time-lag season). In the west, it is conventional to call the summer solstice day the first day of summer. I think this is what lead Irv to think we were comparing starts rather than middles.

I didn’t fall for that, because I knew the results would not be sensible, so I assumed that each zero-lag seasons has its solstice or equinox in the middle, as suggested by the Chinese solar terms Start of Spring, Start of Summer etc.. I made that assumption (of the middles) explicit in my E-mails and later asked Mike to confirm or deny it. Irv did not make his assumption (of the starts)explicit. I don’t think he was aware of it.

Karl

10(09(30 till noon

 

--
Scanned by iCritical.

[Karl Palmen]

Some parts of this message have been removed. Learn more about Nabble's security policy.

Dear Mike, Irv and Calendar People

 

From: East Carolina University Calendar discussion List [mailto:CALNDR-L@...] On Behalf Of MIKE OSSIPOFF
Sent: 19 June 2009 20:13
To: CALNDR-L@...
Subject: Re: June-September season

 

 OSSIPOFF wrote:
 

Irv has been defining lag as the time between, say, the summer solstice and the beginning of calendar summer. I say that, for a meaningful measurement of lag, the lag should be measured from the summer solstice to the _middle_ of calendar summer.

 

Irv replies:  

 

You can't measure from the north solstice to any part of the calendar year without defining the rule that connects the solstice to the calendar.

 

I reply now:

 

I don't know what you mean. The way I first understood Irv’s statement, I disagreed with it.

For any well defined solar calendar you can convert the day that the solstice occurs into a date in that calendar then measure the number of days between that date and any other date in the calendar. This includes any solar calendar that has no rules that refer to a solstice or equinox.

 

I've described two ways of deffining the Summer (or North) calendar season. One of those definitions mentions the summer solstice, but the other doesn't.

 

But, in the current discussion about equal seasons (and I do _not_ advocate equal seasons), the positioning of the quarters has always been specified in terms of a solstice or equinox.

 

Perhaps what Irv may meant is that if the calendar specifies the positioning of quarters (or unequal seasons) in terms of a solstice or equinox, then it must have a rule that refers to that solstice or equinox; a rule that connects the solstice or equinox to the calendar. It may (for example) be

(1)    A simple rule such as solstice is always June 21, which for a 38-day lag and a 91 week quarter begin the quarter on June 14 (so June 21 is the 8^th day and the day 38 days later is the middle (46^th) day).

(2)    An observation-based rule such that the mid season moment  occurs within 12 hours of the moment 38.25 days after the solstice as observed in Pacific Standard Time.

 

Note that rule (2) is an implicit leap year rule, while rule (1) does not specify any leap years.

         

The rules of the calendar need not refer to the solstice at all, but simply be set up so that the mid season moment occurs about the desired lag after the solstice.

 

Anyway, I've clarified, earlier, two ways to measure the timelag of calendar summer, and one of those is by the duration between the summer solstice and the middle of calendar summer.

 

 

Karl

 

10(10(01

--
Scanned by iCritical.

[Irv Bromberg]

Dear Mike and Calendarists:

Yep, I'm totally confused as to what Mike wants his calendar to do.  It seems to be a design without rules.

Mike, try implementing calendar date conversion functions to interconvert your calendar dates with the Julian Day number, and/or the Gregorian date, and/or a fixed day count from some specified epoch.  Test that you can convert any date back and forth in both directions without returning to the wrong date, and verify that your algorithm can convert any date in either direction without crashing.  Evaluate its performance relative to astronomical yardsticks (such as the equinoxes and solstices, perihelion).

The only way that will be possible is by pinning down all of the necessary specific calendar rules.

-- Irv Bromberg, Toronto, Canada

<http://www.sym454.org/seasons/>

[MIKE OSSIPOFF]

             <677CE4DD24B12C4B9FA138534E29FB1D0670D7BC@...>
             A<BLU134-W218CA9F423BC265627B2A3CF390@...>
 <677CE4DD24B12C4B9FA138534E29FB1D06781964@...>
Content-Type: text/plain; charset="Windows-1252"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-OriginalArrivalTime: 23 Jun 2009 23:46:05.0591 (UTC) FILETIME=[C62F0A70:01C9F45C]


=20
Karl=2C
=20
You wrote:
=20
That argument does not work unless you state that the summer solstice is in=
 the middle (of an zero time-lag season).=20
=20
I reply:
=20
But the summer solstice is definitely in the middle of the time of high dec=
linations=2C and those high declinations are what causes summer=2C and what=
 summer follows.
=20
You continued:
 =20
In the west=2C it is conventional to call the summer solstice day the first=
 day of summer. I think this is what lead Irv to think we were comparing st=
arts rather than middles.
=20
I reply:
=20
Yes.

Mike
=20
_________________________________________________________________
Lauren found her dream laptop. Find the PC that=92s right for you.
http://www.microsoft.com/windows/choosepc/?ocid=3Dftp_val_wl_290=

[MIKE OSSIPOFF]

 
Irv wrote:

> Yep, I'm totally confused as to what Mike wants his calendar to do. It seems to be a design without rules.
 
I reply:
 
I have posted a specific definition for the fixed version of the calendar that I propose. And I have posted several definitions for alternative nonfixed versions of it.
 
I specified how I'd place the calendar with respect to dates in our old calendar. I described the positioning of the beginning of North with respect to old-calendar June 1. My rules for that referred to the solar ecliptic longitude that was the center of the cyclical drift of June 1, in the 2004-2008 leapyear cycle.
 
I've posted all that, and it shouldn't be necessary to post it again. Likewise, I've posted complete definitions of the fixed version, and several nonfixed versions, including the one that Karl suggested, with calendar seasons beginning on June 1, October 1, December 1, and April 1 of our current old calendar. My rule for positioning with respect to June 1 is unchanged.
 
Any specific questions or objections
Mike Ossipoff
 
>
>
_________________________________________________________________
Microsoft brings you a new way to search the web.  Try  Bing™ now
http://www.bing.com?form=MFEHPG&publ=WLHMTAG&crea=TEXT_MFEHPG_Core_tagline_try_bing_1x1