>I regularly mention in ENSO posts that there was little SST sampling in the eastern equatorial Pacific prior to the opening of the Panama Canal in 1914, that the reader needs to keep that in mind when viewing early NINO3.4 data.
BUT THE SAME HOLDS TRUE FOR ALL SST DATA
The farther back in time we explore, the less accurate the SST data becomes because of the minimal sampling in early years.
I’m working on a post about a curious dataset created by JISAO, the Global-SST ENSO Index.
http://www.jisao.washington.edu/data_sets/globalsstenso/
Attached to it are listings of the number of monthly tropical and extratropical ICOADS SST readings (the last two columns here)…
http://www.jisao.washington.edu/data_sets/globalsstenso/globalsstenso18002008.ascii
…which I’ve broken down into five graphs. It’s very easy to see why the Hadley Centre and the NCDC don’t present SST anomaly data before 1850.
http://i42.tinypic.com/2w53hip.jpg
1800 to 2008
############
http://i41.tinypic.com/3502ka1.jpg
1800 to 1849
############
http://i44.tinypic.com/1zevwv8.jpg
1850 to 1899
############
http://i43.tinypic.com/2m6jrwg.jpg
1900 to 1949
############
http://i42.tinypic.com/14tlvmq.jpg
1950 to 2008
Bob Tisdale - Climate Observations 
