I've done some work to find earthquake catalogs for a project from the Institute for Celestial Geodynamics as a consultant. I can share an opinion:
In general, the NOAA database of significant earthquakes is great, but has better coverage for older than newer quakes. For post-1960 earthquakes, SeismiQuery (by IRIS) is immense. GEM or USGS round out the set.
NOAA - List of Significant Earthquakes
SeismiQuery - since 1960
USGS - since 1900
GEM Instrumental - since 1900
GEM Historic Catalog - 1000 - 1903
If the project is post-1960, I'd use SeismiQuery only with confidence. If not, I'd combine SeismiQuery (IRIS), NOAA, GEM and USGS.
I'll include my working notes below, in case you're doing more detailed (or regional) work.
Here are some of the major websites for earthquake catalogs:
Yes, I like what you are doing very much! Nice work!
It looks like IRIS site will only give you about 1/120 of the data at a time. I'd run something like 200 queries over the course of a few days, splitting up the queries by time, and keeping them in order so that you can repopulate a single large datafile. But ... I think better than sending in 200 queries would be to write to someone at IRIS and explain what you are doing and what you want, and they may be able to email you the dataset that you are looking for. I would guess that they are very friendly.
Note that IRIS only has data from 1960 onward. Also make sure to note how the dataset distinguishes different events from different reports of the same event.
Yes. I've had the opposite experience. I've asked questions to the administrator of the GEM and NOAA databases, and also to the Iceland Meteorological Office's administrator, and in every case they were friendly, professional and generous. I've met someone who works at IRIS and I imagine that they will be friendly and professional there as well.
For scientific investigatitions the one of the best catalogues is the global NEIC catalogue of the USGS. It is free available at the page http://earthquake.usgs.gov/neic.world.epic
The representative magnitude in the NEIC catalogue is M=4.5. For the period 1973-2014, for example, the global NEIC catalogue contains information about 205,311 earthquakes with magnitudes M≥4.5.
From USGS I had the answer to "...search for smaller geographic areas or smaller time periods". They have also a nice R tool, but it didn't work this morning. I'll try again.
From IRIS, the answer was "...run your query by year."
So, nothing like what you told me.
Probably you are either more respectable or more convincing (or maybe simply, you are not Greek, who knows?)
Great list of links for lists of earthquakes. What to use depends on your objectives. There are problems with most lists, particularly at lower magnitudes. Completeness in any list needs to be verified, and with older lists it depends on station distribution as well as persons interpreting the records.