Guest Post by Bob Tisdale
Why is that a big deal? 70% of NOAA’s land+ocean surface temperature dataset is sea surface temperature data.
NOAA recently published the first of the papers describing the latest update of their ERSST dataset. The new version is called, logically, ERSST.v4. The paper is Huang et al. (2014) Extended Reconstructed Sea Surface Temperature version 4 (ERSST.v4), Part I. Upgrades and Intercomparisons. The paper is paywalled, which is quite remarkable for a scientific paper documenting a U.S. government-funded dataset.
The abstract reads (my boldface):
The monthly Extended Reconstructed Sea Surface Temperature (ERSST) dataset, available on global 2°×2° grids, has been revised herein to version 4 (v4) from v3b. Major revisions include: updated and substantially more complete input data from the International Comprehensive Ocean-Atmosphere Data Set (ICOADS) Release 2.5; revised Empirical Orthogonal Teleconnections (EOTs) and EOT acceptance criterion; updated sea surface temperature (SST) quality control procedures; revised SST anomaly (SSTA) evaluation methods; updated bias adjustments of ship SSTs using Hadley Nighttime Marine Air Temperature version 2 (HadNMAT2); and buoy SST bias adjustment not previously made in v3b.
Tests show that the impacts of the revisions to ship SST bias adjustment in ERSST.v4 are dominant among all revisions and updates. The effect is to make SST 0.1°C-0.2°C cooler north of 30°S but 0.1°C-0.2°C warmer south of 30°S in ERSST.v4 than in ERSST.v3b before 1940. In comparison with the UK Met Office SST product (HadSST3), the ship SST bias adjustment in ERSST.v4 is 0.1°C-0.2°C cooler in the tropics but 0.1°C-0.2°C warmer in the mid-latitude oceans both before 1940 and from 1945 to 1970. Comparisons highlight differences in long-term SST trends and SSTA variations at decadal timescales among ERSST.v4, ERSST.v3b, HadSST3, and Centennial Observation-Based Estimates of SST version 2 (COBE-SST2), which is largely associated with the difference of bias adjustments in these SST products. The tests also show that, when compared with v3b, SSTAs in ERSST.v4 can substantially better represent the El Niño/La Niña behavior when observations are sparse before 1940. Comparisons indicate that SSTs in ERSST.v4 are as close to satellite-based observations as other similar SST analyses.
Curiously, NOAA is using satellite-based sea surface temperature data as a reference for their ship inlet-, bucket- and buoy-based dataset.
As of now, there is very little preliminary data available in an easy-to-use format. More on that later.
WHAT CHANGES MIGHT WE EXPECT?
Regarding comparisons to satellite-based data, the only dataset similar to ERSST.v4 that’s available from the KNMI Climate Explorer is ERSST.v3b—inasmuch as they do not use satellite data and they infill grids with missing data. (HADISST infills missing data but uses satellite-based data starting in 1982, and, while the HADSST3 dataset excludes satellite data, it is not infilled.)
As a reference, Figure 1 compares the current ERSST.v3b global data to the Reynolds OI.v2 satellite-enhanced data. They have remarkably similar warming rates. It will be interesting to see how closely the new ERSST.v4 matches.
Regarding the buoy-bias adjustments, we have as a reference the upgrade of the Hadley Centre’s datasets from HADSST2 to HADSST3. Keep in mind that HADSST data are not infilled. That is, if a 5-deg latitude by 5-deg longitude grid doesn’t have data in a given month, the Hadley Centre inserts no data in that grid…it’s a blank. And that’s a problem in the mid-to-high latitudes of the Southern Hemisphere, where observations even today are relatively sparse. With that in mind, for HADSST data, the new buoy-bias adjustments added a very small increase to the warming rate from the early 1980s to the end of 2012 (when KNMI stopped updating the HADSST2 data at the Climate Explorer). See Figure 2. Will the same hold true with the new ERSST.v4 data?
AS A REFERENCE
We mentioned Hadley Centre’s satellite-enhanced sea surface temperature dataset HADISST above. For those interested, the global HADISST data are compared to the Hadley Centre’s primary dataset, HADSST3, in Figure 3. That’s quite a remarkable difference in warming rates, considering that HADISST is used quite often in long-term and short-term climate studies. No wonder the Hadley Centre is upgrading HADISST. It makes the HADSST3 data look like it has a bias toward warming. (Note: HADSST3 is running a number of months late at the KNMI Climate Explorer, which is why it ends in July 2014 in Figure 3.)
For those who want an early peek at small part of the ERSST.v4 data, data for sea surface temperature-based ENSO and PDO indices are available here. That’s the first result provided by Google in a search for ERSST.v4.
As soon as the new ERSST.v4 data are added to the KNMI Climate Explorer, I’ll provide a few comparisons.
PS: Did Richard Reynolds retire? I was surprised not to see his name as one of the co-authors of Huang et al. (2014). He’s been involved with NOAA’s sea surface temperature datasets for decades. There are a few climate scientists I would enjoy chatting with, as soon as they retire. Richard Reynolds of the NOAA is one.