Validity of Marcott et al. PART II

Validity of Marcott et al. contention that “Global temperatures are warmer than at any time in at least 4,000 years” and “Global temperature….. has risen from near the coldest to the warmest levels of the Holocene within the past century.” A heat spike like this has never happened before, at least not in the last 11,300 years” ”

Guest post by Dr. Don J. Easterbrook

Part I of this series looked at the validity of conclusions for the 11,300 time span covered in Marcott et al. “A Reconstruction of Regional and Global Temperature for the Past 11,300 Years.” This segment (Part II) analyzes conclusions in the Marcott et al. paper in which they contend that “Global temperatures are warmer than at any time in at least 4,000 years” and “Global temperature….. has risen from near the coldest to the warmest levels of the Holocene within the past century.” A heat spike like this has never happened before, at least not in the last 11,300 years”

As in Part I, this segment analyzes the Marcott et al. conclusions using the scientific method of Feynman in which conclusions are checked against well-established data from other sources,. As Feynman points out, if a hypothesis (conclusion) disagrees with observations and data, it is wrong. It doesn’t make any difference how beautiful the hypothesis (conclusion) is, how smart the author is, or what the author’s name is, if it disagrees with data or observations, it is wrong.

So let’s check the Marcott et al. conclusions against several of the best data sets available.

For this exercise, we’ll use (1) the GISP2 oxygen isotope data of Stuiver and Grottes (1997), (2) the GISP2 paleotemperature data of Cuffy and Clow (1997) and Alley (2000), and (3) temperature reconstructions from Chinese tree rings. Among the many data sets that could be used, the GISP2 ice core data have been selected because (1) the ice core data is based on thousands of isotope measurements that quantitatively reflect paleo-temperatures, (2) the chronology is accurate to within about 1-3 years, (3) even small fluctuations of ice core paleo-temperatues can be clearly and unequivocally correlated with advance and retreat of glaciers globally, confirming that the ice core data mimic global temperatures, and (4) Greenland temperatures measured over the past century match global temperature trends almost exactly, confirming that Greenland temperatures march in lock step with global temperatures. Thus, the GISP2 ice core data provides an excellent check against conclusions about global climate—it is quantitative, chronologically accurate, and representative of global climate. Keep in mind, however, that the magnitude of temperature fluctuations generally increases with latitude, i.e., the higher the latitude the greater the temperature fluctuations are likely to be, so Greenland temperature variations are likely to be greater than global averages. It also means that we are more likely to see details of temperature changes in the Greenland data than in global averages.

Figure 1 shows a comparison of the Marcott et al. temperature curve for the past 4,000 years (1A), the Greenland GISP2 temperature curve of Alley (2000), based on the data from Cuffy and Clow (1997) (1B), the Greenland GISP2 and Δ¹⁸O (the ratio of oxygen 18 to oxygen 16 relative to a standard) from isotope data measured by Stuiver and Grootes (1997)(1C), and a temperature reconstruction based on Chinese tree rings (1D) (included as an example of the good correlation of the GISP2 data to places far away from Greenland).

Several things are worth noting about the Greenland data. There are two kinds of temperature data: (1) figure 1B, which shows temperatures from borehole measurements, and (2) figure 1C, which shows variation in oxygen isotope ratios. The significance of this is that temperature variations in both curves are essentially the same, confirming one another. The Little Ice, Medieval Warm Period, Dark Ages Cool Period, Roman Warm Period, and other temperature peaks show up equally well in both types of curve.

Figure 1. Comparison of the Marcott et al. temperature curve (A); the Greenland GISP2 temperature curve of Alley (2000) based on data from Cuffy and Clow (1997) (B); Greenland GISP2 oxygen isotope ratios (delta ¹⁸O) from ice core data measured by Stuiver and Grootes (1997)(C); and temperature reconstruction from Chinese tree rings (D) (Liu et al., 2011).

The Marcott et al. curve shows a nearly vertical line for recent warming, which they claim puts present temperatures above any in the past 4,000years. This nearly vertical part of their curve apparently comes not from their proxy data, but is pasted on from elsewhere and plays a central role in their contention that present temperatures and the rate of warming are ‘unprecedented in the past 4,000 years.’ Let’s test both of these assertions against ice-core and global glacial data.

Both the Greenland GISP2 temperature curve (Figure 1B) and the oxygen isotope curve (Figure 1C) clearly show that except for the Little Ice Age and Dark Ages Cool Period, temperatures for all of the past 4,000 years have been warmer than the end of the ice core (1950 AD). The Medieval Warm Period was 1.1° C warmer than the top of the core (1950) and at least four other warm periods of equal magnitude occurred in the past 4,000 years; four other warm periods were ~1.3°C warmer; two other warm period were 1.8-2.0°C warmer; and one warm period was 2.8°C warmer. At least a dozen periods more than 1°C warmer than 1950 occurred, clearly contradicting the Marcott et al. conclusions.

Figure 2. Peak temperatures warmer than 1950 in the past 4,000 years.

The top of the GISP2 ice core is 1950 AD, so we need to look at more recent temperatures in Greenland in order to get to the ‘present temperature’, i.e., has the temperature in Greenland risen since 1950? Figure 2 shows 1880 to 2004 temperatures in Greenland (Chylek et al., 2004, 2006). Temperatures in 2004 were slightly lower than in 1950, so temperatures at the top of the Greenland ice core are not significantly different than those ‘at present.’

Figure 3. 1880 to 2004 temperatures in Greenland (Chylek et al., 2004, 2006).

The Marcott et al. conclusion that “Global temperatures are warmer than at any time in at least 4,000 years” is clearly contrary to measured real-time data and thus fails the Feynman test, i.e., it is are wrong.

Marcott et al. contend that “Global temperature….. has risen from near the coldest to the warmest levels of the Holocene within the past century.” A heat spike like this has never happened before, at least not in the last 11,300 years. “If any period in time had a sustained temperature change similar to what we have today we would have certainly seen that in our record.” Let us test this conclusion against real-time data. First, their statement that “Global temperature….. has risen from near the coldest to the warmest levels of the Holocene within the past century” is not true. The coldest part of the Little Ice Age occurred about 400 years ago, during the Maunder Minimum, so right off the bat, their conclusion is flawed. They appear to be unaware of the cyclic nature of temperature change and use the low point of the 1880-1915 cool period as their starting point for assessing the rate of warming over the ‘past century,’ rather than 1913-2013. Comparing the depth of cooling in a cool period with a warm period peak is comparing apples and oranges. It distorts the real rate, which should be measured from cool peak to cool peak or warm peak to warm peak. The 1880-1915 cool period was followed by the 1915-1945 warm period, the 1945-1977 cool period, and the 1978-1998 warm period (Figure 4). The rate of warming from 1913 to 2013 is about 0.7°C per century (which is about the same as the warming rate over the past 400 years as we have been thawing out of the Little Ice, long before atmospheric CO₂ began to rise significantly).

Figure 4. Global temperature during the past century.

So let’s compare this rate (0.7°C per century) to rates of temperature increase in the past 11,300 years. Figure 5 shows rates of temperature change in the Greenland GSP2 ice core from the end of the last Ice Age through the Holocene (Figure 4A). Figure 4B shows some of the higher rates of temperature change in Figure 4A. The highest rates occurred at the transition from the Ice Age to Holocene when warming rates in Greenland were 20 to 24°F per century and the huge continental ice sheets that covered large areas of North America and Eur-Asia melted dramatically. As shown in Figure 4B, the rate for the past century (0.7°C) is puny indeed compared to late Ice Age/early Holocene rates.

Figure 4. A. Temperature changes in the Greenland GISP2 ice core from the end of the last Ice Age through the Holocene. (Easterbrook, 2011 modified from Cuffy and Clow, 1997). B. Rates of temperature change. (Easterbrook, 2011)

Holocene rates of warming and cooling were not as profound as those at the end of the last Ice Age, but were nonetheless greater than or equal to recent warming rates. Marcott et al. contend that “If any period in time had a sustained temperature change similar to what we have today we would have certainly seen that in our record” As shown in Figure 4A, we do indeed have a record of warming rates far in excess of those in the past century.

The Marcott et al. conclusion that “Global temperature….. has risen from near the coldest to the warmest levels of the Holocene within the past century.” A heat spike like this has never happened before, at least not in the last 11,300 years” is clearly contrary to measured real-time data and thus fails the Feynman test, i.e., their conclusion is wrong.

Next, in Part III, we’ll analyze the Marcott et al. conclusions that “Over the coming decades we are likely to surpass levels not seen on the planet since before the last ice age.” “Surface temperature reconstructions of the past 1500 years suggest that recent warming is unprecedented in that time.” Our global temperature reconstruction for the past 1500 years is indistinguishable within uncertainty from the Mann et al. reconstruction”