>…Show Climate Responses That Cannot Be Easily Illustrated With Time-Series Graphs Alone
In this post, I’ll illustrate natural heat redistribution from the tropics to the mid-to-high latitudes of the Northern Hemisphere, using the Time-Latitude plot for Lower Troposphere Temperature (TLT) anomalies available from Remote Sensing Systems (RSS). The post provides an introduction to the Time-Latitude plot for those new to it and comparisons of Time-Latitude plots to time-series graphs. It ends with a series of illustrations intended to show the natural causes for the rises in global TLT anomaly since 1979, with an emphasis on the mid-to high latitudes of the Northern Hemisphere.
THE TIME-LATITUDE PLOT
Figure 1 shows a Remote Sensing Systems (RSS) MSU Time-Latitude plot for Lower Troposphere Temperature (TLT) anomalies. It illustrates much more information than a time-series graph, but it is rarely used to show global temperature variations over the past 30 years. Accuweather’s Brett Anderson posts it occasionally on their Global Warming blog.
If you’re not familiar with the RSS TLT Time-Latitude Plot, it’s relatively easy to read. The scale at the bottom shows the range of colors they use to indicate TLT anomalies. The x-axis is time, the same as a time-series graph. The y-axis is latitude, with the South Pole at the bottom and the North Pole at the top. The point in time that stands out in Figure 1 is the big reddish patch in the tropics around 1998. It shows the TLT response to the 1997/98 El Nino.
The graphic is available from RSS here (their Figure 8):
There is a wealth of information contained in the TLT Time-Latitude Plot, but it fails to show some subtleties. In the following, I’ve combined TLT Time-Latitude Plots and time-series graphs in many of the illustrations, using both to clarify what is being presented by the other. Figure 2 shows the Time-Latitude Plot and Time-Series graph of Global TLT anomalies from January 1979 to April 2009. Global TLT anomalies in the Time-Series graph appear to modulate above and below 0 deg C until the mid-to-late 90s, then rise in a step after the 1997/98 El Nino. In the Time-Latitude Plot, it’s easy to see that the majority of the warming after the 1997/98 El Nino took place in the mid-to-high latitudes of the Northern Hemisphere. There was also a comparatively minor warming in the tropics after the 1998/99/00 La Nina, but there was little to no warming in the extreme high latitudes of the Southern Hemisphere.
There are a few items that need to be clarified before discussing the redistribution of heat from the tropics to the mid-to-high latitudes of the Northern Hemisphere. The first are the…
TIMING OF ENSO EVENTS AND VOLCANIC ERUPTIONS
It’s well known that the explosive volcanic eruptions of El Chichon in 1982 and of Mount Pinatubo in 1991 lowered global temperatures temporarily. Figure 3 illustrates the TLT Time-Latitude Plot and a graph of Stratospheric Aerosol Optic Thickness (Sato Index) data as a proxy for volcanic influence. It clarifies the timing of the eruptions, but shows that the volcanic aerosol impact on TLT can be hard to see in the Time-Latitude Plot due to the timing of other natural climate perturbations.
And those other perturbations are El Nino and La Nina events. Refer to Figure 4. In addition to the Time-Latitude Plot, it illustrates a Time-Series graph of NINO3.4 SST anomalies. I’ve also highlighted the ENSO events with ovals in the tropics of the Time-Latitude Plot.
The significant El Nino of 1982/83, though similar in magnitude to the 1997/98 El Nino, did not have the same tropical TLT signature, because the 1982/83 El Nino was suppressed by the 1982 eruption of El Chichon. Refer to Figure 5. Likewise, the eruption of Mount Pinatubo suppressed the El Nino of 1991/92 and the El Nino conditions (not a full-fledged El Nino on the ONI scale) that occurred in 1993.
Let’s also take a look at two latitude bands before illustrating the redistribution of heat of El Nino events.
In Figure 6, I’ve enclosed the tropics (20S-20N) with a green box to isolate it for comparison with the corresponding time-series graph of the tropical TLT anomaly data. The magnitude of the 1997/98 El Nino stands out above all else.
Comparing the tropical TLT anomalies to scaled NINO3.4 SST anomalies, Figure 7, shows the correlation between the two datasets.
The two agree well after the 1997/98 El Nino. This shows how significant NINO3.4 SST anomalies (a small section of the tropical Pacific) are to TLT anomalies for the Tropics. Before the 1997/98 El Nino, there are divergences between the two datasets. The El Chichon and Mount Pinatubo eruptions, as noted earlier, are responsible for a good portion of those differences. Refer to Figure 8. I’ve added the Sato Index of Stratospheric Aerosol Optic Thickness (also scaled) to the comparative graph to illustrate the timing of the volcanic eruptions and their lingering effects.
And the last preliminary discussion is of…
THE MID-TO-HIGH LATITUDES OF THE NORTHERN HEMISPHERE
I’ve highlighted the Northern Hemisphere Mid-To-High Latitudes in Figure 9 for comparison with the time-series graph of the corresponding TLT anomaly data. Looking at the Time-Latitude Plot, the mid-to-high latitudes of the Northern Hemisphere appear to warm in the mid-1990s and remain warm through 2009. But the visual effect is misleading, because the cause of the mid-90s warming is obvious in the time-series graph. The warming in the mid-1990s is simply a rebound from the cooling effect of stratospheric aerosols emitted by Mount Pinatubo.
There is also a minor surge in TLT anomalies for the Northern mid-to-high latitudes in 1995, Figure 10, but this appears to be a response to the 1994/95 El Nino.
THE IMPACT OF THE 1997/98 EL NINO ON THE MID-AND-HIGH LATITUDES OF THE NORTHERN HEMISPHERE
So, in Figure 11, I asked a question (with an obvious answer) about the TLT anomalies of the mid-to-high latitudes of the Northern Hemisphere. If the short period of warming in the mid-90s is simply a rebound from Mount Pinatubo-induced cooling, then what could be the cause of the period of elevated TLT anomalies that began in 1998? The answer: The 1997/98 El Nino, the “Super El Nino”, the “El Nino of the Century”. Before that El Nino, the mid-to-high latitude Northern Hemisphere TLT anomalies averaged -0.03 deg C, and after it, they averaged +0.44 deg C. Nothing else happened at that time that could have caused that sudden rise. The 1997/98 El Nino produced an upward step change in TLT anomalies of the Northern mid-to-high latitudes of approximately 0.47 deg C.
The heat from the 1997/98 El Nino (Figure 12, Cell a) was redistributed poleward to mid-and-high latitudes of both hemispheres, with most migrating north. This El Nino-induced heat was not radiated completely into space over the next four years. Only a portion. The La Nina of 1998/99/00 HAY HAVE reduced the mid-and-high latitude warming, but its effect is not apparent.
Let me reinforce something. El Nino and La Nina events correlate well with TLT of the TROPICS. This indicates that El Nino AND La Nina events have similar impacts on TLT of the tropics. It appears, however, that El Nino and La Nina events do not have equal but opposite effects on mid-to-high latitudes of the Northern Hemisphere. If they had equal but opposite effects, Northern Hemisphere mid-to-high latitude TLT anomalies would have responded fully to the 1998/99/00 La Nina, but, in actuality, they responded very little to that multiyear La Nina event.
The subsequent El Nino events of 2002/03, 2004/05, and 2006/07 (Figure 12, Cell b) redistributed additional heat from the tropics to the mid-and-high latitudes, primarily to the Northern Hemisphere. This maintained the elevated TLT anomalies there. Then, as a result of the 2007/08 La Nina (Figure 12, Cell c), TLT anomalies of the Northern Hemisphere mid-to-high latitudes cooled slightly.
THE 1982/83 AND 1986/87/88 EL NINO EVENTS
Figure 13 is provided solely as a reminder: the 1982/83 El Nino was comparable in magnitude to the 1997/98 El Nino.
So the poleward heat redistribution of the 1982/83 El Nino should have been similar in scale to the 1997/98 El Nino (refer to Figure 14, Cell a), but, again, it was suppressed by the eruption of El Chichon.
In addition to the 1997/98 El Nino event, there was only one other significant El Nino event since 1979 that wasn’t impacted by a major volcanic eruption. It was the multiyear El Nino of 1986/87/88. Refer to Figure 14, Cell b. It also redistributed heat from the tropics to the mid-to-high latitudes to an extent that it’s noticeable in the TLT Time-Latitude plot, but the transport wasn’t as great as the 1997/98 El Nino.
The RSS TLT data is available here:
The ERSST.v3b data used for the NINO3.4 SST anomaly graph is from the KNMI Climate Explorer website: