Just Temperature

Just Temperature:

The U.S. has just experienced an intense heat event with many records falling in the eastern half of the U.S. Here is Chris Burt’s post on the historic event. There is an excellent discussion of this event and its relation to a warming climate by Andrew Freedman at Climate Central. (Global Warming May Have Fueled March Heat Odds) I have a talk to give next week, and I am sure that the heat will contribute to questions. A question that has been put to me frequently in the past weeks is that should we expect such high temperatures in the future?

Usually when I talk about evidence of a warming, I talk about coherent and convergent evidence. That is, one can’t just look at the global surface temperature data and state that the planet has warmed. But if you look at the surface temperature data along with many other sources of data, then one finds that the evidence of warming is overwhelming. If you add the impacts of this warming to ecosystems, for example, the observations that spring is coming earlier over most of the land area in the Northern Hemisphere, then the evidence becomes smothering. For me and many others this evidence of warming is convincing, but it relies on pulling together information from many sources, explaining their relationships, and presentation of the information. So as people have asked me about the heat in Michigan and Maine this past week, I have thought of what I could do with just temperature. Here is the thread that I put together.

The last month when the global mean monthly average was below the 20th century average was February 1985. Here is a picture of the difference from the 100 year average of temperature data from each February. It has been 324 months since there was a month below the global average temperature. (Not 324 Februarys, 324 consecutive months.) Looking at the graph, the Southern Hemisphere, which is dominated by the ocean, goes back into the 1970s. There have been Februarys in the Northern Hemisphere with little blips below average.



Figure 1: February monthly difference from a 20th century average of all Februarys. From the National Climatic Data Center.

The average in this figure is based on the entire 20th century. Therefore, if you look at the record during the 20th century, there is a balance between the warm and the cold months. This fact comes directly from the definition of calculating the differences from an average. There is a famous 1930s warm period. This warm period is present in the February time series, but compared with a later span centered around 1960, this period in not as intense. A prominent characteristic of the graph is that on the left, in the first part of the 20th century, it is cooler than the average and on the right, the here and now, it is warmer.

To go along with the February graph, I have placed the graph from August 2011. The main part of the story, that in 1900 it was cooler than in 2000 remains the same. Here, in the Northern Hemisphere summer, the 1930s warm period is more prominent and more global than in February. In is easy to conclude from this figure that the spatial extent and the temporal persistent of the current warming are both far larger than in the spurt of warmth of the 1930s.



Figure 2: August monthly difference from a 20th century average of all Augusts. From the National Climatic Data Center.


I started this article with the question is the current heat event in the U.S. what we can expect in the future? Taking this simple argument, looking at the average for the past, almost 30 years, it seems reasonable to expect it be warm. And given, the relentless increase of carbon dioxide in the atmosphere, we should expect it to be warmer in the future. To expect otherwise would be betting against the average.

Betting against the average – the next plot, Figure 3, is adapted from a 2009 paper by Jerry Meehl and a host of other authors. (Original Paper, Paper Discussion from NCAR ) What this figure shows, for the U.S., is the number of new record highs divided by the number of record lows – the ratio of highs to lows. In a simplistic, intuitive way, if the average temperature where staying the same, then one would expect the number of new record highs and the number of new record lows to be about the same. What is seen in the figure is as we go from the 1980s to the 1990s to the 2000s, there is trend of record highs out numbering record lows by a factor of 2 to 1. Comparing this with Figures 1 and 2, this evolution of new record highs outpacing new record lows occurs during the time when there has not been a month below the global 20th century average.



Figure 3: Adapted from Meehl et al. (2009) the ratio of U.S. record highs and record lows by decade.

The next figure I show is another version of the global difference figure. This one is calculated as differences from 1950 onwards in order to overlap with the data from the Climate Prediction Center that identify El Nino and La Nina Cycles. El Nino and La Nina are names given to frequently occurring patterns of variation that are concentrated in the tropical Pacific Ocean, but that change the average temperature of Earth for about a year. When there is an El Nino then the globe is warmer and when there is a La Nina the globe is cooler.



Figure 4: Global temperature differences with El Nino (warm) and La Nina (cool) years marked. From National Climatic Data Center.

Looking first at the La Nina years, 1985, the last year when the Earth was cooler that the 20th century average was a La Nina year. One could say that this was the last year when the variation associated with La Nina was strong enough to counter the warming trend enough for the Earth to appear “cool.” What is striking is that the La Nina years in the past three decades are systematically warming. This suggests that in the La Nina cool period, we are seeing a warmer and warmer background, average, temperature evolving.

The warm phase of this variation does not paint as easy a picture. The very strong 1997-1998 El Nino famously raised the Earth’s temperature to a point that many argue was the warmest year observed. The subsequent El Nino events are not as strong as the 1997-1998 El Nino, and each one has temperature maximum that flirts with the 1998 maximum. It is important to note that in 1998 the entire positive anomaly of temperature was not due to the presence of El Nino. The El Nino events take place on a background of increasing temperature, and each event is a burst towards new historic highs in temperature. It is useful to look back earlier in the graph, say 1970 and earlier, to get an idea of the size of variation that can be associated with El Nino and La Nina.

Returning again to the question posed in the beginning, can we expect to regularly see such warm temperatures going forward? Yes, it makes sense that we will see more and more record high temperatures. To not expect that is to bet against the emerging observed trend of warmer and warmer temperatures that is a metric of the warming climate.

I will finish this just temperature story with a map of the Plant Hardiness Zones. Here is the official version from the US Department of Agriculture with an service that lets you pick out your zip code. I show a map of Michigan. In 1990 the green zones, 6, were down around the Ohio River in southern Ohio. This is a measure of not only warming, but also of the definitive changes in the onset of spring. The Washington Post has an excellent graphic that shows the changes between 1990 and 2012.



Figure 5: Plant hardiness zones in Michigan for 2012. From US Department of Agriculture.

We have just experienced in the U.S. a record extreme heat event. This raises the natural questions of climate, weather, and climate change. I have linked a couple of excellent discussions of these issues in the opening paragraph. What I have done in my article is to focus simply on temperature. I have laid out a thread that starts from the globe and the remarkable observation that we have not seen a month below the 20th century global average in more than 25 years. This I followed with the observation that we are in a time when we are setting more than twice as many record highs as record lows. After that I discussed the role of one of the most prominent forms of planetary temperature variations, El Nino and La Nina. The compelling point from this graph was that in the past 30 years during the cool phase, La Nina, the planet shows a warming trend. Finally, I introduce the plant hardiness zones, which show warmer winters, and can be translated to earlier springs. So the question that has been posed to me last week, can we expect such high temperatures in the future? Yes. If we use our experience and observations for the basis of decision making, then the rational answer is yes. We will see more records. We will see an earlier spring. We will see warmer times.


r