Time Zone Changes in 2026

Paul Eggert (2026-06-26):

In “Morocco’s GMT+1, a century of shifting time and a debate far from over”
https://en.yabiladi.com/articles/details/191310/morocco-s-gmt1-century-shifting-time
(2026-03-30), Yabiladi’s Latifa Babas reports the following:
* A 1913-10-26 royal dahir established GMT as legal time in Morocco.
* A 1918-05-10 dahir instituted DST on 1918-05-16 at 00:00.
* A September 1939 dahir restarted DST on 1939-09-12.
* A February 1940 dahir restarted DST as early as 1940-02-25.
* Standard time resumed in September 1946.
* A June 1950 decree restarted DST, which ran until as late as October.
* Royal decree 455-67 (1967-06-02) restarted DST on June 3 at noon.
* After 1967 Morocco used DST “during several summers, particularly
throughout the 1970s and 1980s.”
Babas consulted official records that disagree with and are surely
more correct than our pre-2008 timestamp data, which came from the
unreliable Shanks & Pottenger.  Unfortunately, Babas did not provide
enough detail to correct our data.

Attached to: Morocco (africa)

Paul Eggert (2026-06-25):

https://www.moroccoworldnews.com/2026/06/325034/confirmed-morocco-to-restore-gmt-on-september-20-ending-eight-year-gmt1-saga/
Today the Moroccan government adopted Decree No. 2.26.530, which abrogates
the 2018 decree that put it at +01 with daylight saving during Ramadan.
The plan is to go back to +00 without DST on 2026-09-20 at 02:00.

Attached to: Morocco (africa)

Anass Taghjichte (2026-07-03):

https://www.sgg.gov.ma/BO/AR/3111/2026/BO_7521_Ar.pdf

Attached to: Morocco (africa)

Afaf EL MAAYATI (2026-07-06):

https://www.mapexpress.ma/actualite/activite-gouvernementale/conseil-gouvernement-approuve-projet-decret-relatif-au-retour-lheure-legale/

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: Morocco (africa)

N.F. Hase (2026-07-26):

Ettela'at (Tehran daily) printed a daily panel headed "taqvim-e
Ettela'at / sa'at-ha-ye shar'i", attributed to Dr. Riyazi Kermani.
Each panel gives today's maghrib followed by fajr, sunrise, solar noon
and maghrib for the next day, together with the Iranian, Hijri and
Gregorian dates.... 26 May 1979 (5 Khordād 1358)
https://commons.wikimedia.org/wiki/File:Ettelaat13580305.pdf
... That also matches Movahedi's wording, that the government
returned to DST on 26 May, and the note in the 2022 thread that the
change was announced one day in advance on 25 May.

Attached to: Iran (asia)

Paul Eggert (2026-07-27):

Go with Pournader for 1935 through spring 1979 (with the last
transition clarified by Hase), and for timestamps after August 1991;
go with Shanks & Pottenger for other timestamps.
Go with Santoni's citation of the UIT for fall 1977, as 20 October 1977
is 28 Mehr 1356, consistent with the "Mehr" in Pournader's source.
Assume that the UIT's "1930" is UTC, i.e., 24:00 local time.

From Oscar van Vlijmen (2005-03-30), writing about future
discrepancies between cal-persia and the Iranian calendar:
For 2091 solar-longitude-after yields 2091-03-20 08:40:07.7 UT for
the vernal equinox and that gets so close to 12:00 some local
Iranian time that the definition of the correct location needs to be
known exactly, amongst other factors.  2157 is even closer:
2157-03-20 08:37:15.5 UT.  But the Gregorian year 2025 should give
no interpretation problem whatsoever.  By the way, another instant
in the near future where there will be a discrepancy between
arithmetical and astronomical Iranian calendars will be in 2058:
vernal equinox on 2058-03-20 09:03:05.9 UT.  The Java version of
Reingold's/Dershowitz' calculator gives correctly the Gregorian date
2058-03-21 for 1 Farvardin 1437 (astronomical).

Attached to: Iran (asia)

Paul Eggert (2026-01-22):

During World War II, Greenland was effectively independent of Denmark and
observed daylight saving time.  TIME, volume 37, page 23 (1941-04-21)
<https://time.com/archive/6770243/war-peace-greenlands-icy-mountains/> says,
"Penfield and West made their way to the U.S.'s most northerly consulate.
They were astonished to find that Greenlanders, with almost 24 hours of
sunlight a day during the summer, have daylight saving time."
As the details are unknown they are omitted from the data for now.

During World War II, Germany maintained secret manned weather stations in
East Greenland and Franz Josef Land, but we don't know their time zones.
Also, they're likely out of scope for the database
as we lack resources to track every bit of military activity.
My source for this is Wilhelm Dege's book mentioned under Svalbard.

Attached to: Thule (europe)

Heitor David Pinto (2026-02-22):

Soviet Moldovan resolution 132 of 1990 defined the summer time period from
the last Sunday in March at 2:00 to the last Sunday in September at 3:00,
matching the dates used in most of Europe at the time:
https://web.archive.org/web/20211107050832/http://lex.justice.md/viewdoc.php?action=view&view=doc&id=298782&lang=1

It seems that in 1996 Moldova changed the end date to October like most of
Europe, but kept the transitions at 2:00 and 3:00 rather than 1:00 UTC,
which would have been locally 3:00 and 4:00....

The notices in the Moldovan government website and broadcaster showed the
transitions at 2:00 and 3:00 until 2021:
2015 https://old.gov.md/en/node/7304
2016 https://old.gov.md/en/node/12587
2017 https://old.gov.md/en/node/20654
2017 https://old.gov.md/en/content/moldova-upholds-winter-time-night-28-29-october
2018 https://old.gov.md/en/content/moldova-switch-summer-time
2018 https://old.gov.md/en/content/cabinet-ministers-informs-about-switch-winter-time-28-october
2019 https://old.gov.md/en/content/moldova-switch-summer-time-31-march
2019 https://old.gov.md/en/node/31122
2020 https://old.gov.md/en/node/32771
2020 https://old.gov.md/en/node/34497
2021 https://trm.md/ro/social/moldova-trece-in-aceasta-noapte-la-ora-de-vara
2021 https://trm.md/en/social/republica-moldova-trece-la-ora-de-iarna1

However, since 2022, the notices showed the transitions at 3:00 and 4:00,
matching the EU rule at 1:00 UTC:
2022 https://trm.md/en/social/in-acest-weekend-republica-moldova-trece-la-ora-de-vara
2022 https://old.gov.md/en/content/moldova-switch-winter-time
2023 https://moldova1.md/p/6587/ora-de-vara-2023-cum-schimbam-acele-ceasornicelor-si-cand-trecem-la-ora-de-vara
2023 https://old.gov.md/en/node/46662
2024 https://moldova1.md/p/26535/republica-moldova-trece-la-ora-de-vara-in-acest-weekend
2024 https://moldova1.md/p/37768/republica-moldova-trece-in-aceasta-noapte-la-ora-de-iarna
2025 https://moldova1.md/p/46349/republica-moldova-trece-la-ora-de-vara-pe-30-martie-cum-ne-afecteaza-si-ce-recomanda-medicii
2025 https://moldova1.md/p/60469/republica-moldova-trece-la-ora-de-iarna-ceasurile-se-dau-inapoi-cu-o-ora

It seems that the changes to the end date and transition times were just
done in practice without formally changing the resolution. In late 2025, the
government said that the Soviet resolution was still in force, and proposed
a new resolution to replace it and formally establish the EU rule:
... based on the notices, it seems that in practice Moldova already
uses the EU rule since 2022. This was also the year when Moldova applied to
join the EU.

Attached to: Moldova (europe)

Robert Bastian (2026-02-26):

This has been approved and published in the government gazette:
https://monitorul.gov.md/ro/monitorul/view/pdf/3234/part/2#page=27

Attached to: Moldova (europe)

Paul Eggert (2026-02-24):

Also see Svetlana Rudenko, "Moldova abandons the 'Soviet era'", Logos Press,
2026-02-21 <https://logos-pres.md/en/news/moldova-abandons-the-soviet-era/>.

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: Moldova (europe)

Tim Parenti (2026-05-26):

Observance of DST on the Azores and Madeira was explicitly covered by
mainland legislation in:
- Portaria 11767 of 1947-03-28 for 1947,
- Portaria 12286 of 1948-02-19 for 1948, and
- Decreto-Lei 37048 of 1948-09-07 through its revocation in 1966.
(See mainland commentary, above.)  However, most legislation prior to 1947
didn't explicitly call out these "adjacent islands", so we assume that DST
rules on the islands prior to 1947 were also like that of the mainland.

Attached to: Atlantic/Azores (europe)

Tim Parenti (2026-08-26):

Sitka's UT offset changed from +14:58:47 to -9:01:13, so effectuate this
change at 15:30:00 +14:58:47, or 1867-10-19 00:31:13 UT in the Alaskan
locations below.

Attached to: America/Juneau (northamerica)

Paul Eggert (2026-05-08):

For 1916 timestamps America/Winnipeg covers only a small region. See:
Cassidy C. Winnipeg’s 110-year history with daylight time.
Winnipeg Free Press. 2026-05-06.
https://www.winnipegfreepress.com/our-communities/correspondents/2026/05/06/winnipegs-110-year-history-with-daylight-time
Of the 1916 experiment, Cassidy writes: “As rural areas and nearby
urban centres such as Selkirk and Brandon did not adopt DST, the
City of Winnipeg essentially had its own time zone.” Cassidy also
writes that province-wide DST came into effect on 1963-05-12.

Shanks & Pottenger write that Winnipeg did not observe DST in 1964 and 1965.
Although dubious in the light of Cassidy’s article, we lack a better source.
Perhaps S&P’s data are for the train stations, not for the city?
Also, S&P say Manitoba switched at 02:00 (not 02:00s) starting in 1966.
Since 02:00s is clearly correct for 1967 on, assume 02:00s in 1966 too.

Attached to: Winn (northamerica)

Heitor David Pinto (2026-05-14):

In Saskatchewan, a bill was passed to replace the Time Act. It sets UTC-6 all
year in the whole province, including Lloydminster, but allows the government
to issue regulations specifying a different time in localities that request
so:
https://docs.legassembly.sk.ca/legdocs/Bills/30L2S/Bill30-58.pdf
The bill ... received royal assent today.

Attached to: Regina (northamerica)

Tim Parenti (2026-05-14):

In light of Alberta joining Saskatchewan on year-round -06, this simplifies
the prior Act's framework for the many local exceptions to year-round -06 in
border areas, by extending province-wide the notion of "time option areas"
which can be prescribed by regulation "if it is in the provincial interest"
for those areas to observe either "UTC-5, UTC-6 or UTC-7 for all or part of
the year" on at least 30 days' notice.
The new Act comes into force by order of the Lieutenant Governor in Council.

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: Regina (northamerica)

Roozbeh Pournader (2026-04-20):

https://calgaryherald.com/opinion/columnists/bell-alberta-daylight-time-year-round-premier-danielle-smith

Attached to: Edm (northamerica)

Tim Parenti (2026-04-23):

Section 3 of Bill 31, the Red Tape Reduction Statutes Amendment Act, 2026
https://docs.assembly.ab.ca/LADDAR_files/docs/bills/bill/legislature_31/session_2/20251023_bill-031.pdf
would repeal the Daylight Saving Time Act in the Revised Statutes of Alberta
2000 Chapter D-5:
https://web.archive.org/web/20240128134751/https://kings-printer.alberta.ca/documents/Acts/D05.pdf
...and substitutes a new chapter with language that closely parallels the
original.  The new title is the Official Time Act and will be numbered
Chapter O-5.7.  The Act establishes a standard time of UTC−6 without
replacing the language previously used to effectuate DST.

Attached to: Edm (northamerica)

Tim Parenti (2026-06-19):

After receiving Royal Assent on 2026-05-14, Order in Council 204/2026 was
issued on 2026-06-18 proclaiming the relevant section of the bill in force on
the same date.  Order in Council 206/2026, issued the same day, uses the
regulatory authority within the Act to prescribe the official term "Alberta
Time"; we use the traditional abbreviation CST for consistency.
https://kings-printer.alberta.ca/Documents/Orders/Orders_in_Council/2026/2026_204.pdf
https://kings-printer.alberta.ca/Documents/Orders/Orders_in_Council/2026/2026_206.pdf

Since wall clock times do not diverge from past practice until 2026-11-01,
use that transition date for now to work around potential CLDR limitations in
the meantime; see British Columbia, below.

Attached to: Edm (northamerica)

Paul Eggert (2026-07-02):

The temporary hack for Alberta is needed for CLDR 48.2 (2026-03-17)
and earlier, not the CLDR 48.1-and-earlier which drives BC’s temporary hack.
Only a few platforms track minor CLDR releases, though, so the two
temporary hacks have roughly the same effect in practice.

The term “Alberta Time” is legally prescribed from yesterday until
2031-06-30, when the regulation in OiC 206/2026 expires to ensure that the
term is reviewed by then for relevancy and need.  This plan for possible
obsolescence affects neither timekeeping nor TZDB’s data, which do
not contain the string “Alberta Time”.

Attached to: Edm (northamerica)

Tim Parenti (2026-07-21):

The new Official Time Act is now in the Revised Statutes as Chapter O-5.7:
https://kings-printer.alberta.ca/documents/Acts/O05p7.pdf

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: Edm (northamerica)

Arthur David Olson (2026-03-02):

B. C. Gov News: “Adopting permanent daylight saving time: ‘Spring forward’
on March 8 will be the last time change, ending twice-yearly clock changes.”
https://news.gov.bc.ca/releases/2026AG0013-000209

Attached to: Vanc (northamerica)

Paul Eggert (2026-07-02):

The law says that 21 hours after the usual 2026-03-08 02:00 switch from
PST to PDT, the next day inaugurates the new standard time Pacific Time,
i.e., just one clock change but two name changes separated by 21 hours.
PT, the obvious abbreviation for Pacific Time, is one letter too short
to conform to TZDB’s (and POSIX’s) [-+[:alnum:]]{3,6} requirements.
I asked the BC government for advice, with no response. For now, do this:
1.	As a temporary hack, pretend that the BC law takes effect
not on 2026-03-09 at 00:00, but on 2026-11-01 at 02:00.
This pretense works around a limitation in CLDR 48.1 (2026-01-08),
which would otherwise say the interval uses “Pacific Standard Time”.
(Below, this temporary hack is marked “Temporary hack; see above.”)
Strictly speaking this hack is incorrect since the interval uses
standard time, but it does have the right UT offset and it
works around the CLDR limitation.  We should be able to remove
the temporary hack by November when there would be little point
to keeping it anyway.
2.	After the BC law takes effect, model the time as MST sans DST.
We can change this later if another conforming non-numeric abbreviation
for Pacific Time becomes more popular.  Possibilities include:
MST - the most compatible with existing software and practice,
and already used in parts of BC and in Yukon
PDT - almost as software-friendly, but confusing because it implies
it is DST and is paired with PST, whereas PT is standard time
PST - straightforward but even more confusing,
and will likely break much software that assumes PST is -08
-07 - accurate and clear in itself, but makes BC look odd vs neighbors
PacT - straightforward but novel abbreviation for Pacific Time

Attached to: Vanc (northamerica)

Tim Parenti (2026-08-21):

In the southeast of the province, the Regional District of East Kootenay
(RDEK), Columbia Shuswap Regional District’s Electoral Area A (Rural Golden),
and the Town of Golden have all historically been aligned with neighboring
Alberta and use America/Edmonton.

Shortly after the BC provincial government announced its adoption of
year-round -07, the RDEK Board of Directors on 2026-03-13 initially voted
narrowly, 8–7, that it would fall back one last time to rejoin the rest of BC
on -07 from November 2026.  However, citing the rushed decision and lack of
feedback from residents, Mayor Steve Fairbairn of Elkford used his power to
ask the question be called a second time, remarking, “Pardon the pun, but
this is not a time-sensitive issue,” and the initial decision was rescinded
by the RDEK Board at its 2026-04-10 meeting.
https://www.cbc.ca/news/canada/british-columbia/what-time-is-it-in-the-east-kootenay-debate-9.7132624
In the ensuing months, after Alberta announced and implemented its similar
shift to “permanent DST” (see above), RDEK coordinated a public survey across
all of the relevant BC communities.  The survey found 89% of respondents in
favor of ending seasonal time changes, with 66% of that group preferring to
remain aligned with Alberta time on year-round -06 instead of the rest of BC.
In line with those results, the RDEK Board adopted year-round -06 at its
2026-08-14 meeting.
https://www.cbc.ca/news/canada/british-columbia/east-kootenay-stops-changing-clocks-9.7308201

Attached to: Vanc (northamerica)

Tim Parenti (2026-09-10):

Although on 2026-09-01 the Town of Golden voted instead to split its
timekeeping practice from RDEK and rejoin BC on year-round -07, this was met
with significant pushback from residents in Golden and CSRD Area A.  In light
of these public comments, CSRD voted on 2026-09-10 to defer its decision on
timekeeping in Area A “to gather and assess further information before a
decision is implemented”.  As such, at time of writing, it seems likely that
Golden may soon revisit its decision as well.

For now, these areas remain aligned with America/Edmonton, but if they
diverge, a new zone with a name like 'America/Brit_Columbia/Golden' would be
needed.

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: Vanc (northamerica)

Tim Parenti (2026-04-21):

"[Northwest Territories] Premier R.J. Simpson announced Monday that the
territory will move to end seasonal time changes and will adopt a year-round
time standard instead. ... Simpson has previously said the territory wouldn't
end seasonal time changes until Alberta does."
https://www.cbc.ca/news/canada/north/nwt-ends-daylight-saving-9.7170964

Attached to: NT_YK (northamerica)

Tim Parenti (2026-07-24):

Now that Alberta has adopted year-round -06, NWT plans to follow through with
the same.  Bill 13 of the 19th Assembly's 2nd Session, assented to
2021-03-31, provides the NWT Commissioner in Executive Council with authority
to fix a year-round standard time:
https://www.ntlegislativeassembly.ca/sites/default/files/legacy/bill_13_-_reprint_-_interpretation_a.pdf

Attached to: NT_YK (northamerica)

Paul Eggert (2026-08-21):

The regulations were registered 2026-08-18 and came into force today:
https://www.justice.gov.nt.ca/en/files/legislation/interpretation/interpretation.r2.pdf
Use a temporary hack like Alberta’s.

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: NT_YK (northamerica)

Paul Eggert (2026-08-11):

Chile’s timekeeping regime is renewed by Decreto 98 of 2026 of the
Ministry of the Interior, Undersecretariat of the Interior,
promulgated 2026-06-30 and published 2026-07-02:
https://www.bcn.cl/leychile/navegar?idNorma=1225724
Although this decree stops at fall 2029, assume that
similar DST rules will continue thereafter.

Rule	NAME	FROM	TO	-	IN	ON	AT	SAVE	LETTER/S

Attached to: Chile (southamerica)

Paul Eggert (2026-07-17):

The 1992 spring forward was May 2 at 00:00.  See:
Jiménez FA. La noche en que Colombia perdió una hora para encarar el apagón,
¿se repetirá la historia en 2026? El Colombiano. 2026-06-21.
https://www.elcolombiano.com/negocios/colombia-perdio-una-hora-para-evitar-apagon-horario-gaviria-LF37983109

Attached to: CO (southamerica)

Paul Eggert (2026-07-24):

These zones are for backward compatibility with Unix System V (1983).
They are ahistorical, in that they pretend standard time extends
backward indefinitely into the past.
They use USback (not US) rules so that they conform to POSIX instead
of using abbreviations like EWT for Eastern War Time during World War II.
The zones are here, not in the northamerica file, to lessen
confusion downstream by making it clear that the zones are present
only for backward compatibility.
New software should use settings like TZ='America/New_York' instead
of these backwards-compatibility zones.

Zone	NAME		STDOFF	RULES	FORMAT	[UNTIL]

Attached to: PST8PDT (backward)