# $Id: harmonics_boilerplate.txt 4391 2012-03-03 00:15:50Z flaterco $ # # Definitions and constants for XTide # # NOTE: This file should not be used as-is to provide a constituent # set to harmgen because it contains multiple constituents with the # same frequency. A better constituent set is included in the harmgen # distribution (see the README). # # # ********* NOT FOR NAVIGATION ******** # # *** DO NOT RELY ON THIS DATA FILE FOR DECISIONS THAT CAN RESULT IN *** # *** HARM TO ANYONE OR ANYTHING. *** # # Highly trustworthy tide predictions cannot be achieved on a zero # budget. If you need guaranteed results, don't use these data! # Contact the tide authority for your region (NOAA in the U.S.). # # This data file is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without any warranty that its use will NOT # INFRINGE ANY RIGHTS, and without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. # # WHOEVER CHOOSES TO REDISTRIBUTE THESE DATA ASSUMES ALL LIABILITY # ASSOCIATED THEREWITH. NEITHER DAVID FLATER NOR ANYONE ELSE INVOLVED # SHALL BE LIABLE PERSONALLY OR OTHERWISE FOR ANY LOSSES THAT THE # REDISTRIBUTOR MIGHT SUFFER AS A CONSEQUENCE OF ANY CLAIMS, BOGUS OR # OTHERWISE, THAT ARE MADE TO THESE DATA. # # # --- Notes to contributors --- # # After getting legal threats from the U.K. Hydrographic Office, I am # no longer willing to accept any data without an explicit statement # from the copyright owner or eminent authority either granting # permission for non-commercial use with XTide or explaining that such # permission is not required. The absence of a copyright notice is # not sufficient; however, a blanket permission for non-commercial use # is sufficient. # # For more technical reasons, I am not prepared to accept data under # conditions that require me to track uses of the data, to limit the # uses of the data to a particular time period, or to verify that # every user has read the legalese. XTide has many different # interfaces by which users can obtain results, and such requirements # are impossible to implement in a reasonable and consistent fashion # across all interfaces. For a reasonable compromise, see what was # done for the POL data: a reference to legalese is included in the # location name field. # # # --- Copyright statements --- # # United States: Data published by the U.S. government is by law # non-proprietary. Stations in the U.S., territories of the U.S., # commonwealths associated with the U.S., and countries with a Compact # of Free Association with the U.S. have been included. A few other # stations available from the NOS web site have been dropped. # # Notice available on the NOS web site at # http://tidesandcurrents.noaa.gov/disclaimers.html as of # 2007-05-30: # # The information on government servers are in the public domain, # unless specifically annotated otherwise, and may be used freely by # the public. Before using information obtained from this server, # special attention should be given to the date and time of the data # and products being displayed. This information shall not be # modified in content and then presented as official government # material. # # The user assumes the entire risk related to its use of these # data. NOS is providing these data "as is," and NOS disclaims any # and all warranties, whether express or implied, including (without # limitation) any implied warranties of merchantability or fitness # for a particular purpose. In no event will NOS be liable to you or # to any third party for any direct, indirect, incidental, # consequential, special or exemplary damages or lost profit # resulting from any use or misuse of this data. # # NOS requests that attribution be given whenever NOS material is # reproduced and re-disseminated. Pursuant to 17 U.S.C. 403, third # parties producing copyrighted (compilation) works consisting # predominantly of material created by Federal Government employees # are encouraged to provide notice with such work(s) identifying the # U.S. Government material incorporated and stating that such # material is not subject to copyright protection. # # Canada: The harmonic constants used to perform tide predictions for # locations in Canada are derived from sea level data made available # by Marine Environmental Data Services, Fisheries and Oceans Canada, # for public, non-commercial use. The predictions are not the same as # those computed by the Canadian Hydrographic Service, which uses its # own sets of harmonic constants. # # Copyright notice for the sea level data at # http://www.dfo-mpo.gc.ca/notices-avis-eng.htm (rev. 2009-07-17): # # Non-Commercial Reproduction # # Information on this site, other than government symbols, has been posted # with the intent that it be readily available for personal and public # non-commercial use and may be reproduced, in part or in whole and by any # means, without charge or further permission from Fisheries and Oceans # Canada, provided that: # # * due diligence is exercised in ensuring the accuracy of the information # reproduced; # * Fisheries and Oceans Canada is identified as the source department; and # * the reproduction is not represented as an official version of the # information reproduced, nor as having been made in affiliation with, or # with the endorsement of, Fisheries and Oceans Canada. # # Netherlands: The Netherlands data was contributed directly by Koos # Doekes of the Rijkswaterstaat RIKZ/ZDI, who grants permission by his # own authority: # # Herewith I grant permission for non-commercial use # of all harmonic constants for the Netherlands by # users of Xtide or Xtide-based products. # # This covers the use of the constants for Dutch # ports in the Xtide-database by designers # of web applications, and the use in commercial # tidal prediction software in which the constants # themselves are not stored - i.e. the user has # to copy them from the Xtide-database himself. # # It does NOT cover the use in commercial software # in which the constants are incorporated, which is # considered commercial use of the constants here. # Any request for permission of use of Dutch constants # in software of this kind should be redirected to # RIKZ. # # Germany: The Germany data was contributed directly by Wolfgang # Lange of the Bundesamt fuer Seeschifffahrt und Hydrographie (BSH), # who grants permission by his own authority: # # As long as you make your software using these tidal data # freely available, I grant permission for non-commercial use # of the harmonic constants for the 12 reference stations at # the german North Sea coast on the same conditions and rules # layed down in the letter from Koos Doekes, RIKZ/ZDI, # concerning the Netherlands harmonic constituents. # # UK: Received 2003-11-15 DWF. See also http://www.flaterco.com/pol.html. # # Proudman Oceanographic Laboratory # Natural Environment Research Council # # Bidston Observatory # Bidston Hill # Prenton # Merseyside # CH43 7RA # United Kingdom # David Flater # 13320 Country Ridge Drive Tel +44 (0) 151 653 8633 # Germantown, MD 20874 Fax +44 (0) 151 653 6269 # United States of America www.pol.ac.uk # # # 07 November 2003 # # Dear David, # # This letter is to confirm that we are happy for you to download the # sea level data from the British Oceanographic Data Centre website and # use this in the generation of harmonic constants using the HarmGen # software. However there are a couple of things that we would politely # ask that you do. # # 1. Limit the number of harmonic constants you use in the software to # a maximum of 40 for any port. This should easily be enough to create # very accurate tidal predictions as for nearly all locations, any # harmonics left out (i.e. the 41st most significant constants onwards) # will have amplitudes of the order of a centimetre. # # 2. State that the data is derived from sea level data obtained from # the British Oceanographic Data Centre based at the Proudman # Oceanographic Laboratory, Liverpool. # # 3. Make it clear that the predictions however are not the same as # those computed by the Proudman Oceanographic Laboratory which uses its # own sets of harmonic constants. # # 4. Continue to make your software using the data freely available. # Please inform us if you intend to produce a software product that you # will be making a charge for. # # I hope these conditions are acceptable to you and it will be good to # see the UK back on the map. # # Yours sincerely, # Colin Bell # Head of Applications Group # # # Changelog # --------- # # 2012-03-02 End of maintenance for non-free data # # As a result of the combination of lack of time, commercialism, and abuses, # I have decided to officially end maintenance of the non-U.S. data, which # were subject to a non-commercial-use-only license. This will be the final # entry in the changelog in the monolithic, legacy formatted # harmonics_boilerplate.txt file. Future updates will be logged more legibly # in a new ChangeLog file bundled with the public domain data. # # # 2011-12-30 # # Imported NOS data from snapshot 2011-12-11. Subordinate tide stations came # from the new web site; subordinate currents have not yet been migrated, so # the old process was used for them. # # A small percentage of subordinate station links were broken at the time of # the snapshot, so a few stations will be down for the year. Most subordinate # tide stations that had complications attached to them disappeared in the # migration to the new web site and are probably gone for good. The new web # site does not [yet] serve up station notes, as far as I can tell. # # The following four current reference stations stopped matching NOS # predictions and were dropped along with their subs: # # Throgs Neck, 0.3 n.mi. NE of, Long Island Sound, New York Current (15d) # The Race, 0.6 n.mi. NW of Valiant Rock, New York Current (38d) # North Inian Pass, Cross Sound, Alaska Current # Unimak Pass (off Scotch Cap), Alaska Current # # The meridian adjustment for College Pt was retained to make predictions agree # with 2012 sub station 8517276. Riverdale was fixed upstream (agrees with # 2012 sub station 8518905). NOS predictions for Fortescue Creek and Texaco # Dock were unavailable for comparison, so they were again adjusted to agree # with their 2008 results. # # The following two stations shifted *west* by 7 hours (i.e., a meridian of # +7 is needed to make the new data match the 2010 predictions): # 9465234 Hagemeister Island (north end), Alaska # 9465249 Estus Point, Hagemeister Strait, Alaska # NOAA predictions were unavailable for comparison; assumed OK. # # # 2011-04-10 # # Imported up-to-date NOS harmonic constants for Wrangell Narrows Current # contributed by Mitch Cline via Dean Pentcheff. Reran import of subordinate # stations from December's web scrape to include all of the inside passage # currents that depend on Wrangell Narrows. # # # 2010-12-30 # # Imported NOS data from snapshot 2010-12-19. # # Compared data with 20081201 snapshot to find recently broken meridians. # Albany was fixed. Santa Cruz, Coyote Pt. Marina, and Moss Landing went # away. Adjusted meridians of the following to agree with 20081201 results: # College Pt, Ft. of 110th Street, Long Island, New York # Riverdale, Hudson River, New York # Fortescue Creek, New Jersey # Texaco Dock, Hackberry Bay, Louisiana # # Fixed miscellaneous spelling errors and nits. # # # 2010-05-29 # # Mark Randolph checked the rest of California and found one other station # that broke after December 2008 in the same manner as Santa Cruz: General # Fish Company Pier, Moss Landing, California. Fixed that and thanks again # to Mr. Randolph. # # # 2010-05-22 # # Santa Cruz, Monterey Bay, California now has the same problem as Albany and # Coyote Point Marina (supposedly UTC data were LST). It broke between # December 2008 and December 2009. Thanks to Mark Randolph for reporting the # problem. # # # 2009-12-27 # # Due to significant, uncorrectable differences with authoritative # predictions, the data for Germany were retired. # # Imported NOS data from snapshot 2009-12-22. # # Removed the current reference station Akutan Pass which no longer matches # published tables (maximum velocities have decreased). # # Reapplied the meridian fixes for Albany and Coyote Point Marina. The # discrepancy for Albany versus the tide tables is now being reproduced by # the new NOAA tide predictor. # # The LST harmonic constants for Johnston Atoll and Kwajalein changed # meridians this year. XTide does not use the LST constants, but there might # be some impact on agreement with NOAA predictions for the subordinate # stations that use those reference stations. # # # 2008-12-28 # # Imported NOS data from snapshot 2008-12-01. # # Reapplied the meridian fixes for Albany and Coyote Point Marina. # Corrected latitude of Monte Carlo Island, AK. # Fixed spelling of Monterey and Chapoquoit. # # In light of the continued absence of harmonic constants for currents on # the NOS web site, I have reinstated 32 reference stations from CURCON00.US # that still match NOS' published 2009 predictions, together with applicable # subordinate stations from the 2008-12-01 web scrape. As always, there is # plenty of opportunity for improvement in the station names. States were # determined heuristically and some will be wrong. # # The database now includes a state field that provides the 2-letter USPS # abbreviation for the state in which each tide station is located, when # applicable. # # All TCD files and dumps are now being distributed in tarballs together # with the boilerplate. # # # 2007-12-28 # # Imported NOS data from snapshot 2007-12-17. # # Reapplied the meridian fixes for Albany and Coyote Point Marina. # Coordinates of I-526 bridge were fixed upstream. # # A comparison of all reference stations versus the 2008 tide tables # published on the web revealed discrepancies for Anchorage, AK and Port # O'Connor, TX. The Anchorage discrepancies are small enough to be # explained by a data update; however, the Port O'Connor discrepancies are # major and are not helped by the meridian fix that worked for Albany and # Coyote Point Marina. Local observations are required to determine whether # the Port O'Connor predictions are any good at all. This is complicated by # the fact that it is a diurnal station with small tidal range, making # eyewitness confirmation that much more unreliable. Subordinate stations # depending on these reference stations will also fail to match the # published tables. # # # 2007-11-22 # # To facilitate compliance with Linux distribution packaging policies # and to draw more attention to the acceptable use restriction on # non-U.S. data, the "free" and "nonfree" data are now being exported # to separate TCD files. At this time, all "free" data are public # domain and all "nonfree" data are merely restricted to # non-commercial use only. # # Replaced the data set for Harwich, England with one calculated from # a longer time series. Proudman still does not yet publish # predictions for this location. # # # 2007-03-18 # # Updated approximated MLLW datums with improved estimates. # # Put IGNORE HEIGHTS in the legalese fields of the 14 subordinate # stations where NOS zeroed out the heights. # # # 2007-02-22 # # Applied renamings as updated and improved by Walt Bilofsky (thanks # Walt). # # # 2007-02-11 # # Interim release, without renamings. # # Imported NOS data from snapshot 2007-02-07. # # (9414449) Coyote Point Marina, San Francisco Bay, CA: corrected meridian. # (Data claimed UTC, was UTC-8) # (8518995) Albany, Hudson River, NY: corrected meridian. # (Data claimed UTC, was UTC-5) # I-526 bridge, Ashley River, South Carolina: fixed longitude. # Pascagoula, Mississippi Sound, Mississippi latitude was fixed upstream. # # In progress: renamings. # # Since there were two reference stations with incorrect meridians # (one that was reported last time and one that was just discovered), # it is reasonable to expect that there will be more. # # # 2006-07-08 # # Coyote Point Marina, San Francisco Bay, California: corrected meridian. # Delaware City, Delaware River, Delweare: fixed typo in name. # Pascagoula, Mississippi Sound, Mississippi: fixed latitude. # I-526 bridge, Ashley River, South Carolina: fixed longitude. # # # 2006-03-05 # # Installed renamings from AH for the 102 stations that lacked states. # # # 2006-02-05 # # Refreshed U.S. data from NOS web site (snapshot 2006-01-04). AH has # dealt with new and modified station names to maintain the naming # convention installed 2005-11-27. Still to be fixed is a list of # stations that do not include their state in their name. Most or all # of those have been that way since November. They didn't stick out # much until an unrelated Harmbase query happened to group them # together. # # An e-mail exchange with Todd Ehret of NOS CO-OPS this time led me to # believe that the UTC constants for the Alaskan reference stations # with weird time zones should be correct, so I have retained them. A # survey of affected subordinate stations shows that most should be # correct if their reference station is correct (they all use the same # one). Most doubtful are the six subordinate stations on # St. Lawrence Island. I do not even have reliable intelligence on # what the de facto time zone of St. Lawrence Island is. It was not # mentioned in the statute that defines U.S. time zones. # # # 2005-11-27 # # Installed renamings from August Hahn to achieve something like the # naming convention previously followed in legacy data. These # renamings will be maintained through subsequent updates if the churn # is not too bad. # # # 2005-09-05 # # Generated and installed harmonic constants for 49 Canadian stations # using up to 10 years of hourly data (the maximum permitted download) # from Marine Environmental Data Services (MEDS), Fisheries and Oceans # Canada. # # MEDS publishes water level data for 962 stations. Of those, only 92 # are currently active. Of those 92, 43 are non-tidal, being located # on or near the great lakes. # # # 2005-06-05 # # Generated and installed fresh harmonic constants for U.K. stations # using all available data for the 19 year epoch from 1986 to 2004. # Quality assurance information is available at # ftp://ftp.flaterco.com/xtide/harmonics-dwf-2005-06-05-QA.txt # # Station Harwich, England is new with only 8 months of data. # Authoritative predictions for this station are not yet available for # comparison. # # # 2005-01-20 # # Deleted the following Alaskan reference stations, which were # rendered incorrect by technical difficulties confirmed by Todd Ehret # of NOS CO-OPS, and all dependent sub stations. They may return # corrected in future revisions. # # - 9460150 Attu, Massacre Bay # - 9465261 Clarks Point, Nushagak Bay # - 9468132 St. Michael, Norton Sound # - 9497649 Prudhoe Bay #2 # - 9497778 Cross Island # # Also deleted the following station which is listed as superseded. # # - 9497652 Prudhoe Bay, Dock #1 # # # 2004-12-10 # # Refreshed all NOS data from NOS web site. # # # 2004-10-18 # # No data changes -- just resyncing and renaming all the different # editions of the database. # harmonics-dwf-2004-10-18-v2.tcd Built with libtcd 2.0 # harmonics-dwf-2004-10-18-v1.tcd Built with libtcd 1.100 # harmonics-dwf-2004-10-18.sql Postgres 7.4.2 # harmonics-dwf-2004-10-18-dump txt/xml restored from -v2.tcd # harmonics-dwf-2004-10-18-compat txt/xml years 1970-2037 only # # # 2004-10-08 DWF # # Updated RIKZ/MII references to new name, RIKZ/ZDI. # # # 2004-10-05 DWF # # Rebuilt using libtcd 1.99, which corrects a problem where # constituents with zero phases were being dropped. The affected # reference stations are Delfzijl, Netherlands; The Battery, New York # Harbor, NY; and South Pass, LA. # # # 2004-09-14 DWF # # Imported subordinate stations available from the NOS web site as of # 2004-09-09. Compare against # http://co-ops.nos.noaa.gov/tide_pred.html. Some stations will # disagree moderately because the reference stations were updated and # don't match the published tables. # # The smattering of sub stations in non-U.S. territory was suppressed. # # There are time zone problems in Alaska. The LST used by the NOS for # stations in Alaska does not always match the local statutory time # zone (which I have followed). For reference stations, the problem # was detectable and correctable because the constants were available # calibrated with UTC as well as LST. But for the sub stations, it is # impossible to tell whether a disagreement with NOS predictions # results from a disagreement about time zone, incorrect calibration, # or both. # # A few names suffixed by (A) and (B) reflect sub stations on the NOS # web site with the same names and close coordinates, where it is far # from obvious why there are two of them. Other name clashes were # resolved by disambiguation. # # (Nit) Corrected months_on_station for German ref stations from 13 to # 12. (It's a time zone / rounding issue, on list to be fixed.) # # # 2004-09-07c DWF # # Put the following three stations on Eastern time (they were calibrated # to Central time on the NOS web site): # Johns Island, Chassahowitzka Bay, FL # Chassahowitzka River, FL # Crystal River, Kings Bay, FL # # # 2004-09-07b DWF # # Corrected heuristic for Alaskan time zones. A few stations in the # Aleutians were affected. # # # 2004-09-07 DWF # # Added notes for stations in Alaska where the LST used by NOS does not # match the apparent statutory time zone. Presently these notes just # get attached to the comments field in the TCD file, but the next major # revision to the TCD format should enable XTide to display them more # prominently. # # # 2004-09-01 DWF # # Most reference stations available from the NOS web site as of # 2004-08-26 have been imported. Compare against # http://co-ops.nos.noaa.gov/tide_pred.html. # # MLLW benchmarks were not available for all stations. For stations # where a MLLW datum was not available, the datum has been set to LAT. # # I know there are typos in the location names. # # Currents are not available at this time. According to Mr. Ehret, they # will become available when the new CO-OPS currents database is # complete. # # The following stations, which are located in jurisdictions that are # neither territories of the U.S., nor commonwealths associated with the # U.S., nor countries with a Compact of Free Association with the U.S., # have been dropped. Their data are distributed without legalese from # the NOS web site; however, that does not necessarily mean that anyone # other than the NOS has permission from the respective governments to # use or redistribute the data. This may seem paranoid, but the # situation with the U.K. data that were pulled in 2001 was very # similar. # # Apia, Samoa # Bermuda Esso Pier, St. Georges Island, Bermuda # Fare Ute Point, Papeete Harbor, French Polynesia # La Unión, El Salvador # Madero, Tampico Harbor, Tamaulipas, Mexico # Puerto Corinto, Nicaragua # Puerto Cortés, Honduras # Puerto Quetzal (San Jose), Guatemala # Settlement Point, Grand Bahama I., Bahamas # Suva Harbor, Suva, Fiji # # # 2004-08-19 DWF # # This is an interim release as I am starting over from scratch using # Harmbase2. All legacy data have been dumped. # # All U.S. data are temporarily down; the 2004-08-09 version of this # file remains usable for those. Mr. Ehret has indicated that there # would be fees involved in obtaining harmonic constants and offsets # in the format previously received, so I am abandoning all that and # gearing up to scrape data off of the web. According to Mr. Ehret, # the set of locations on the web site is a superset of those used in # published prediction tables. The discrepancies that occur are # because the harmonic constants on the web are updated continuously # while the harmonic constants used for the published tables are fixed # in advance. Thus there seems little to be lost by replacing all of # the legacy data with a snapshot of the web data. # # Data presently included: # # Netherlands reference stations computed from hourly levels 1999 # through 2002, as received from Koos Doekes on 2004-02-03. Compare # against http://www.getij.nl/engels/. # # German reference stations computed using Harmgen from year 2001 # observations sent by Wolfgang Lange on 2004-08-19. The datums are # adjusted to LAT. Compare against # http://www.bsh.de/en/Marine%20data/Forecasts/Tides/index.jsp. (The # web site is still using the MLWS datum, so the heights will disagree # strongly until the web site is updated. However, even adjusting for # that, the agreement is not as good as could be hoped. The BSH # predictions are made using a non-harmonic method, so some # discrepancies are inevitable.) # # German subordinate stations as sent by Wolfgang Lange on 2004-08-10. # # U.K. reference stations computed using Harmgen from at least four # years of observations ending 2002-12-31, as downloaded from the BODC # web site 2003-11. Compare against http://www.pol.ac.uk/ntslf/tides/. # # # 2004-08-09 DWF # # harmonics-dwf is a new database that emphasizes responsiveness to # updates and traceability to authoritative sources instead of maximum # coverage of locations and historical predictions. All data known to # be old and expired and all data for which no source of updates # exists have been purged. This will orphan many users who have been # getting good results from old data. For these, please continue to # use RMK's database. # # harmonics-dwf now contains: # - U.S. reference stations provided by Todd Ehret of NOAA circa # 2004-02-04. # - Netherlands reference stations provided by Koos Doekes of the # Rijkswaterstaat RIKZ/ZDI (then RIKZ/MII) on 2004-02-03. # - U.K. reference stations derived by Harmgen from BODC data, # 2003-11. # - The most recent data for other U.S. reference stations that # were in RMK data rev. 2004-04-16 23:30 but not included in # the distribution from Todd Ehret. # # All sub stations are gone temporarily while I update my software to # import the sub station data sent by Todd Ehret in February. I will # also be contacting Mr. Ehret to see if I can obtain a complete dump # of all stations in current use so that I can build a legacy-free # database.