10/03/2016: ERSST v3b production stopped. 09/24/2015: change user name to boyin.huang 05/16/2014: ERSST web page has been changed to https://www.ncdc.noaa.gov/data-access/marineocean-data/extended-reconstructed-sea-surface-temperature-ersst-v3b The old link has been redirected http://www.ncdc.noaa.gov/ersst/ 11/01/13: ERSST data are back to normal from shutdown of government operations. 10/01/13: During the shutdown of government operations, ERSST will continue to update automatically, but will not be monitored by the ERSST team. We will also be unable to answer questions until after government operations resume. 03/04/13: remove oisst download and oiv2mon1t2.f calculation that has been done separately in /home/bhuang/gpfs/oisst/script. The procedure caused problem when oisst did not update correctly and on time. The oisst is used for diagnosis only that has now used the data in /home/bhuang/gpfs/oisst/2x2 01/22/13: change ninos.f: add cj(j) weighting in calculating NINO indices: No impact on NINO3,4,and 3.4 change in NINO1.2 is less than 0.01K change in gtsqc.f, reduce duplication from "since 2007" to "since 2011" change in dati2.upd.f, add a filter of SPV in calculating sqrt(e2), the original code generates error in sqrt(-999.9) regenerate ersst.2000.2009.asc using dati2.upd.2000.f due inconsistency between ascii and binary files. 05/26/11: change med-ice data directory derived in OISST analysis to the directory independently derived by ice processing. ice processing is operational daily. c=========================================================================== 05/10/11: changes in ersstv3b.sh: remove Scp data to eclipse-d2 add a script ersst.ftp.sh that ftp data to ftp0.ncde.noaa.gov add a script ersst.scp.sh that scp data to eclipse-d2 or other place. Not used right now. c=========================================================================== 05/04/11: create a tag of ERSST.v3b at $rep/ersst/tags/v3b.2011.05.04 delete code in $rep/ersst/trunk revise code without v3b and situ in filenames svn archieve at $rep/ersst/trunk c=========================================================================== 03/15/11: gtsqc.f if (rsst.ge.0.0.and.rlat.gt.84.9) then !byh c if (rsst.eq.0.0.and.rlat.gt.84.9) then bug: add i_idid=0; nship=0; nbuoy=0; i_globe=0 to initialize goto 800 !byh: needed. The paragraph is finally commented: meaning quality is actually controlled by following ERSST. these changes make slight difference in ERSST in interior <0.1C, but may be 0.2-0.3C near coasts. otherwise. The final ERSST is same (cmp) as that from Chunying Liu. 03/04/11: gtsqc.f change: gts_in='../sstday/'//ciyear//'/obs/buoyship/NCEP/nq//ciyear2//cimonth into gts_in='../data/buoyship_quarter/coads/nq'//ciyear2//cimonth ICOADS (v2.4) before 2006 (--2006) where? GTS data (2006-present): data/buoyship_quarter/coads/nq???? hfsst.f change filename hfsst.1985.last.wgt.dat into hfsst.1985.last.wgt.dat pdov3_el_nino.f & pdov3_el_nino.fullsst.f change if(iy.le.1997)then into if(iy.le.1985)then for consistency no impacts since there is not output open add nino12 index calculation and output in binary file c=========================================================================== 03/03/11: err.norm.map2.upd.f bug: read(23,end=100,err=100) ac should be changed as read(23,end=100,err=100) ac ac is temp used to determine ALF read(23,end=100,err=100) an !an is obs count bug: dd=(shf-a2(i,j)) should be dd=abs(shf-a2(i,j)) c=========================================================================== 02/25/11: sst2d.f: add output sst2d.1985.last.test.dat for the test run, which will not over written sst2d.1985.last.dat. This is for the purpose of rerun to check code. The operation run is set to rewritten sst2d.1985.last.dat. c=========================================================================== 02/22-24: all code change absolute path into relative path e.g. /raid2/ersst/data/err.norm.map.upd.1985.last.dat ../data/err.norm.map.upd.1985.last.dat c=========================================================================== Extended Reconstructed Sea Surface Temperatures, version 3b(ERSST.v3b), beginning January 1854. ERSST.v3b is an improved extended reconstruction over version 2. Most of the improvements are justified by testing with simulated data. The major differences are caused by the improved low-frequency (LF) tuning of ERSST.v3b which reduces the SST anomaly damping before 1930 using the optimized parameters. One of the additional changes is that the low frequency 15-year filter uses data centered on the current year. If some of data for 15-year period has not yet been observed, it can, of course, not be used. However, with time these data will become available. Thus, there may be slight modifications of recent ERSST fields with time. These changes were first implemented in version3, which also contained satellite data as an input. In the most recent version3b, we have removed satellite data from ERSST product, because it changed the rankings of warmest months, as explained in the webpage.