IANA provides a list of timezones on https://www.iana.org/time-zones with their time shift. There are frequent updates.
=> are only the time shifts updates, or can their be new timezones / timezones deleted?
I noticed that the naming of a region should not be tied to a country, but I have to be sure that eventually there is no change :-)
Yes, new time zone identifiers can be added. Indeed, that has to be the case if a region which previously observed the same time and was covered by a single time zone splits into separate time zones.
Additionally, time zone identifiers can sometimes be added for other reasons - for example, in 2022b, the identifier Europe/Kyiv was added; the old Europe/Kiev was converted to be a link to it (in the backward file). Similarly in 2008, Asia/Kolkata was added and Asia/Calcutta was converted to a link.
I don't think I've seen a case where an existing identifier was removed rather than converted into a link - that would be awful for existing data. Of course, if a time zone does split in two as described in the first paragraph, then we can't know which of those two any given piece of data should really refer to, unless it also comes with a location - but that's just inherent complexity.
The theory file refers to this:
Timezone boundaries are not part of the stable interface. For example, even though the Asia/Bangkok timezone currently includes Chang Mai, Hanoi, and Phnom Penh, this is not part of the stable interface and the timezone can split at any time. If a calendar application records a future event in some location other than Bangkok by putting "Asia/Bangkok" in the event's record, the application should be robust in the presence of timezone splits between now and the future time.