Part 2 of “On Sallenger et al (2012) – Hotspot of Accelerated Sea Level Rise on the Atlantic Coast of North America”

Guest post by Bob Tisdale

This is a follow-up to my recent post On Sallenger et al (2012) – Hotspot of Accelerated Sea Level Rise on the Atlantic Coast of North America. It confirms a comment I made there and also takes a quick look at the satellite-based sea level trend maps. Sallenger et al (2012) also referred to climate models as if they have value, so I wanted to add a comment or three about the models used by the IPCC. Last, the post exposes the blunder of unusual size in the Scientific American post about Sallenger et al (2012).

INTRODUCTION

There’s been lots chatter around the blogosphere about sea level rise in response to legislation in North Carolina. Refer to the WattsUpWithThat posts here and here. It escalated last week with the release of the Sallenger et al (2012) paper Hotspot of Accelerated Sea Level Rise on the Atlantic Coast of North America. RealClimate had a post titled Far out in North Carolina. SkepticalScience had back-to-back sea level posts; the first North Carolina Lawmakers Turning a Blind Eye to Sea Level Reality? was followed a day later by Madness over sea level rise in North Carolina. Scientific American also jumped into the mix with North Carolina Sea Level Rises Despite State Senators. Not one of those posts investigated tidal gage sea level data for the region that was the focus of Sallenger et al (2012). In a preliminary investigation, I presented the sea level data from one North Carolina tidal-gage station and from seven stations with long-term, reasonably complete data that were included in the Sallenger et al (2012) east coast hot spot, the hotspot that wound up not looking too hot after all. Refer also to the comments in the cross post at WattsUpWithThat.

LONG-TERM VERSUS SHORT-TERM TRENDS

In my post, I presented a time-series graph of the average of the “Hotspot+1” sea level data. And stated that it appeared the average sea level anomalies for the “Hotspot+1” stations have been relatively flat since about 1996. Figure 1 confirms that comment. The long-term trend (1935-2008) of the “Hotspot+1” sea level data is about 3 cm/decade, while the short-term trend from 1996 to 2008 is less than 10% of the long-term trend at 0.24 cm/decade. I’ve also included the trend for the “Hotspot+1” data starting at the 1990 break point discovered by the Sallenger et al(2012) analysis. It shows an increase from the long-term trend of about 20%.

Figure 1

The more I look at Figure 1, the more it appears that I’m presenting a classic example of trend comparisons with cherry-picked start years. It might appear to some that Sallenger et al (2012) cherry-picked 1990 to front load the short-term data with the impact of the eruption of Mount Pinatubo, which would have driven the trend up of the data after 1990, while I picked 1996 to front load the data with the impact of the 1997/98 El Niño, which would drive the trend of the data after 1996 down . I’ll admit to my cherry-picked start year. Will Sallenger et al (2012) admit to theirs and its impact on their claim of accelerated sea levels? Sallenger et al (2012) briefly mentioned volcanic aerosols toward the end of their paper:

Aerosols may also play a role in explaining variations in NEH SLRDs. The mid-century low (Fig. 4) may have been forced by volcanic aerosols reflecting radiation and lowering air temperatures²⁵ and slowing¹⁴ SLR.

But they never attempted to account for them, in the paper, to see how much the catastrophic eruption of Mount Pinatubo in 1991 could have impacted their claim of accelerating sea levels.

SATELLITE-BASED SEA LEVEL DATA

I had wanted to use the University of Colorado’s Interactive Sea Level Time Series Wizard to determine the satellite-based near-coast sea level rise along the east coast hotspot to see whether it confirmed the tidal gage data, similar to a post by Steve Goddard. Unfortunately, the Sea Level Time Series Wizard wasn’t working this weekend. So I downloaded the University of Colorado’s sea level 1993-2012 trend map in pdf form, zoomed in on the east coast of United States, and overlaid the color-coded trend scale—once again doing something similar to another recent post by Steve Goddard. See Figure 2. The trend of the satellite-based sea level data appears to confirm that the short-term tidal gage-based trend may be about right, but note how the trend is higher toward New York and Boston and relatively low around North Carolina. Too bad the Sea Level Time Series Wizard wasn’t working this weekend.

Figure 2

CLIMATE MODELS

Sallenger et al (2012) refer to climate models more than 20 times in a 2900 word document. It looks as though, if one were to delete all sentences with the word model or some form thereof, that there’d be little left of the paper, maybe 20% to 25% of the original word count. Apparently, the authors of Sallenger et al (2012) believe that climate models have skill at being able to hindcast and project ocean-related variables.

In a couple of posts over the past year (see here and here) and in my recent book, I’ve shown that the climate models used by the IPCC in their 4^thAssessment Report (AR4) show no skill at being able to simulate satellite-era sea surface temperature anomalies. Yes, I understand that sea surface temperature and sea level are different datasets. I’m using sea surface temperatures as an example of an ocean climate variable, since model outputs and observed satellite-based sea level data are not available through the KNMI Climate Explorer for the term we’re interested in discussing.

Figure 3 illustrates the linear trends from January 1982 to April 2012 for the Pacific Ocean on a zonal (latitudinal) mean basis. (Not presented in an earlier post.) The trends for the Southern Ocean (near Antarctica) portion of the Pacific (125E-90W) are shown to the left, and to the right are the trends up to the latitude of the Bering Strait. The equator is at zero latitude. We can see that the trends for the average of the models used by the IPCC in AR4 to simulate sea surface temperatures bear no similarities to the observed trends. The observations portray a pattern associated with how the El Niño-Southern Oscillation (ENSO) redistributes warm water from the tropics toward the mid-latitudes, where it can release heat to the atmosphere more efficiently. But the models appear to portray a zonal mean pattern associated with the annual average sea surface temperatures (not anomalies) of the Pacific—warmer in the tropics than at the poles—as if the models are warming at a faster rate in the tropics in response to the warmer sea surface temperatures there and at slower rates toward the poles because it’s cooler there. That really looks odd.

Figure 3

The comparison of satellite-era trends for the Atlantic on a zonal mean basis, Figure 4, shows the model mean of the climate models used in the IPCC’s AR4 aren’t any better there for the past 30 years, the satellite era. The pattern of warming in the models again appears to represent the annual average sea surface temperatures (not anomalies), while the observations portray a pattern associated with the Atlantic Multidecadal Oscillation (AMO).

Figure 4

Why do I say the observations portray a pattern associated with the AMO? Let’s switch to a longer-term sea surface temperature dataset (HADISST), and look at the Atlantic Ocean sea surface temperature anomaly trends from 1944 to 1976 and from 1976 to 2010 on a zonal mean basis. Refer to Figure 5. We can see that the cooling pattern in the trends of North Atlantic sea surface temperature anomalies, from 1944 to 1976, opposes that of warm trends from 1976 to 2010. Notice also how the two curves diverge, not at the equator, but at the southern end of the tropics.

Figure 5

Referring back to Figure 4, you’ll note that the models do a reasonable job of matching the observed satellite-era sea surface temperature trends for the latitudes of about 33N-43N, which are the latitudes of the “hotspot” discussed in Sallenger et al (2012). So let’s compare the observed sea surface temperature trends for those latitudes to those of the IPCC AR4 climate model outputs on a meridional (longitudinal) basis from the east coast of the United States to the west coasts of Europe and Africa. Refer to the map in Figure 6. We’ll be looking first at the trends of the modeled and observed sea surface temperature anomalies from 1982 to 2011 for each of those grids. The grid farthest to the west captures the sea surface temperature anomalies from the coasts of Georgia/South Carolina north to Maryland and Delaware. The grid bordered by the coordinates of 33N-43N, 75W-70W captures the sea surface temperatures along the coasts from Delaware/New Jersey north to Massachusetts/New Hampshire.

Figure 6

For the period of 1982 to 2011, the trends of the IPCC’s AR4 models of sea surface temperatures between the latitudes of 33N-43N show a pretty uniform warming from west to east, but the observations do not. See Figure 7. In fact, toward the coasts of the Carolinas and Virginia, the observed sea surface temperature anomaly trends are negative. That is, sea surface temperature anomalies have cooled there. And the longitudes that contain the New Jersey, New York and New England shorelines show a positive (a warming) trend, but it’s about half the modeled trend.

Figure 7

The models don’t look so good at these latitudes during the last 30 years. And they performed poorly on a zonal mean basis, too. I wonder what they do simulate correctly. It certainly isn’t land surface temperatures on a regional basis around the globe. In general, the models performed poorly at simulating the regional land surface temperature trends in the Americas, Australia, Southern Africa and Southeast Asia.

One last graph: Now let’s start the meridional mean trend comparison for the latitudes of 33N-43N in 1990, Figure 8, which was the break year in the tidal gage-based sea level data found by Sallenger et al (2012), which, by some strange coincidence, just happens to be right before the largest explosive volcanic eruption of the 20^thCentury. With this start year, the modeled trend agrees with the observed trend for the longitudes of the northern portion of the “hotspot”.

Figure 8

I’ll let readers speculate about that.

THE SCIENTIFIC AMERICAN BLUNDER

As noted in the opening, Scientific American published a post North Carolina Sea Level Rises Despite State Senators about the Sallenger et al (2012) paper. The subtitle is so blatantly wrong it’s laughable. It reads “Less than two weeks after the state’s senate passed a climate science-squelching bill, research shows that sea level along the coast between N.C. and Massachusetts is rising faster than anywhere on Earth.” Anywhere else on Earth? Tell that to the people along the coasts of the Indo-Pacific Warm Pool. See Figure 9.

Figure 9

CLOSING

I included the discussion of climate models with hope that someone from the climate science community would state something to the effect of, climate models have known problems and cannot simulate climate on regional or on short-term (multidecadal) bases.  Because, then I’ll point them to Sallenger et al (2012) and ask why that paper discusses climate models as if they do have value on regional bases and over short time spans. The people of North Carolina are patiently waiting for that discussion.

ANOTHER SHAMELESS BOOK PLUG

My book If the IPCC was Selling Manmade Global Warming as a Product, Would the FTC Stop their deceptive Ads? is available in pdf and Kindle editions. An overview of my book is provided in the above-linked post. Amazon also provides a Kindle preview that runs from the introduction through a good portion of Section 2. That’s about the first 15% of it. Refer also to the introduction, table of contents, and closing in pdf form here.

I still plugging along on my upcoming book about El Niño-Southern Oscillation and hope to publish pdf and Kindle editions by late July, early August 2012.

SOURCES

The Reynolds OI.v2 sea surface temperature data was retrieved from the NOAA NOMADS website. The multi-model mean for the CMIP3 (20C3M/SRES A1B) outputs of sea surface temperature (TOS) were retrieved from the KNMI Climate Explorer.