Iceland's timezone

[Adam David]

Sorry about the delay getting back with this information.

> I'm working on a rewrite of `europe' with a large set of historical
> data.  My current draft (see below) claims that Iceland changed its
> timezone in 1968.  I've taken this claim from Thomas G. Shanks, The
> international atlas (3rd edition), San Diego: ACS Publications, Inc.
> (1991).  If you have some better data, please let me know.

In a way, there was a timezone change in 1968, where the summertime change
was applied during the spring, and became the all-year-round time, and there
have been no changes since then, not for DST or any other reason. When looking
back to previous years when there was DST, icelanders tend to refer to the
concept of "winter time", as the summer time had the same offset as the present
all-year-round time. This is why I was led to believe that no timezone change
had taken place in 1968.

> It would be nice to know what time zone abbreviations were used before
> 1968, since Shanks doesn't say.  In the following table I've made up
> the names `IST', `IDT', `RMT', and `LMT' for Iceland Standard Time,
> Iceland Daylight Time, Reykjavik Mean Time, and Local Mean Time
> respectively.

I have no details of any abbreviations, sorry.

> # From Shanks (1991):
> # Rule	NAME	FROM	TO	TYPE	IN	ON	AT	SAVE	LETTER/S
> Rule	Iceland	1908	only	-	Jan	1	0:00	0	S
> Rule	Iceland	1917	1918	-	Feb	20	2:00	1:00	D
> Rule	Iceland	1917	only	-	Oct	25	2:00	0	S
> Rule	Iceland	1918	1919	-	Nov	15	2:00	0	S
> Rule	Iceland	1919	only	-	Feb	19	2:00	1:00	D
> Rule	Iceland	1921	only	-	Mar	19	2:00	1:00	D
> Rule	Iceland	1921	only	-	Jun	22	2:00	0	S
> Rule	Iceland	1941	1942	-	Mar	Sat>=1	2:00	1:00	D
> Rule	Iceland	1941	1942	-	Jul	2	2:00	0	S
> Rule	Iceland	1943	1946	-	Mar	Sun>=1	2:00	1:00	D
> Rule	Iceland	1943	1948	-	Oct	Sun>=22	2:00	0	S
> Rule	Iceland	1947	1962	-	Apr	Sun>=1	2:00	1:00	D
> Rule	Iceland	1949	only	-	Oct	30	2:00	0	S
> Rule	Iceland	1950	only	-	Oct	22	2:00	0	S
> # From Paul Eggert <eggert at twinsun.com> (November 10, 1993):
> # Shanks's ``1953 Oct 26'' is probably a typo for Oct 25.
> # Likewise, correct Shanks's ``1961 Oct 22'' to Oct 29.
> Rule	Iceland	1951	1967	-	Oct	Sun>=23	1:00s	0	S
> Rule	Iceland	1963	1968	-	Apr	Sun>=1	1:00s	1:00	D
> # Zone	NAME		GMTOFF	RULES/SAVE	FORMAT	[UNTIL]
> Zone Atlantic/Iceland/Reykjavik -1:27:24 -	LMT	1837 Jan 1
> 			-1:28	-		RMT	1908 Jan 1
> 			-1:00	Iceland		I%sT	1968 Oct 27 1:00s
> 			0:00	-		GMT
> # This link is for backward compatibility with the old naming convention.
> Link Atlantic/Iceland/Reykjavik Iceland

I don't feel up to messing with this table and submitting a corrected version.
Here I copy a list of periods when summertime was active, and explanatory text.
This material is paraphrased from the 1988 edition of the University of Iceland
Almanak.

>From January 1st, 1908 the whole of Iceland was standardised at 1 hour behind
GMT. Previously, local mean solar time was used in different parts of Iceland,
the almanak had been based on Reykjavik mean solar time which was 1 hour and
28 minutes behind GMT.

Summertime when it was applied was always by putting the clock an hour forward
during the summer months, which made the clock time equal to GMT. The following
list shows when this summertime was in force. The time of the clock change is
specified in GMT.

year	from			to
----	----			--

1908-1916 no summertime

1917	20.Feb. 0:00		21.Oct. 1:00
1918	20.Feb. 0:00		16.Nov. 1:00

1919-1938 no summertime

1939	30.Apr. 0:00		29.Nov. 2:00
1940	25.Feb. 3:00		 3.Nov. 2:00
1941	 2.Mar. 2:00		 2.Nov. 2:00
1942	 8.Mar. 2:00		25.Oct. 2:00

1943-1946 rule (first Sunday in March until first Sunday in winter)

1943	 7.Mar.	2:00		24.Oct. 2:00
1944	 5.Mar.	2:00		22.Oct. 2:00
1945	 4.Mar.	2:00		28.Oct. 2:00
1946	 3.Mar.	2:00		27.Oct. 2:00

1947-1967 rule (first Sunday in April until first Sunday in winter)

1947	 6.Apr.	2:00		26.Oct. 2:00
1948	 4.Apr.	2:00		24.Oct. 2:00
1949	 3.Apr.	2:00	******* 30.Oct. 2:00 (delayed by 1 week)
1950	 2.Apr.	2:00		22.Oct. 2:00
1951	 1.Apr.	2:00		28.Oct. 2:00
1952	 6.Apr.	2:00		26.Oct. 2:00
1953	 5.Apr.	2:00		25.Oct. 2:00
1954	 4.Apr.	2:00		24.Oct. 2:00
1955	 3.Apr.	2:00		23.Oct. 2:00
1956	 1.Apr.	2:00		28.Oct. 2:00
1957	 7.Apr.	2:00		27.Oct. 2:00
1958	 6.Apr.	2:00		26.Oct. 2:00
1959	 5.Apr.	2:00		25.Oct. 2:00
1960	 3.Apr.	2:00		23.Oct. 2:00
1961	 2.Apr.	2:00		22.Oct. 2:00
1962	 1.Apr.	2:00		28.Oct. 2:00
1963	 7.Apr.	2:00		27.Oct. 2:00
1964	 5.Apr.	2:00		25.Oct. 2:00
1965	 4.Apr.	2:00		24.Oct. 2:00
1966	 3.Apr.	2:00		23.Oct. 2:00
1967	 2.Apr.	2:00		29.Oct. 2:00

1968	 7.Apr. 2:00	(actually GMT after this)

"first day of winter" referred to above, means the first day of the 26 weeks
of winter, according to the old icelandic calendar that dates back to the time
the norsemen first settled Iceland. The first day of winter is always Saturday,
but is not dependent on the Julian or Gregorian calendars. This is the reason
for the discrepancy with October 1961.
I can find out more exactly how to determine when it falls, if necessary. It
is a pity I don't have that information handy, otherwise I could include it
here.

There are more serious discrepancies before 1941 which are not obviously
explainable. What to do... (?)

--
Adam David  <adam at veda.is>