Coastal Climate Extension

An article in the Post and Courier, the daily newspaper for Charleston, SC, is causing quite a stir among their online readers. According to reporter Tony Bartelme, SC’s rate of carbon dioxide emissions has increased by 45% since 1990. This puts SC’s carbon emissions growth behind only Arizona and Colorado in terms of how fast we’ve been emitting carbon dioxide. Without access to the Post and Courier’s data and the methodology they used to do this emissions inventory, I can’t confirm how accurate their numbers are. However, I can go through the online comments on the article and address a few common misconceptions that are cited. I’ve decided to take advantage of this “teachable moment” to pick out a few things people were confused about in their comments, and post on each of them. So this week’s question is: “What’s been happening with global average temperatures over the last 10 years?”

To answer this question, let’s recap the difference between weather and climate. Weather is the state of the atmosphere, right here, right now – currently, outside my office, it’s about 67 degrees Fahrenheit, and according to the National Weather Service, there’s about a 50% chance that it’ll rain tomorrow. Climate is the various statistics about weather over a long period of time. Today, the normal high temperature – or, the average of the high temperatures for October 13 over a 30 year period – is 78 degrees. Climate is one of the reasons we see such diverse plant and animal life across the world. For example, in a desert, the climate is dry (maybe with a rainy season), so cacti and other desert plants need dry conditions, and many have adapted by developing ways to store water. Generally, we use “weather” to refer to atmospheric conditions over a period up to a few weeks. “Climate” happens on a scale of months to years. We also divide climate into “climate variability” and “long-term climate change.” Climate variability is the cycles that contribute to conditions over months to years. Natural factors of climate variability include things like the El Nino-Southern Oscillation (ENSO) cycle. Long term climate change refers to the combination of natural and human factors that contribute to conditions over decades to hundreds of years. This includes questions like what the global average temperature has been over the past 100 years and what that global average temperature will be in 2100. To make matters more confusing, natural climate variability can affect what we see in long-term trends in climate over decades – and some scientists worry that long term climate change may alter the natural climate variability patterns we humans are used to experiencing.

Because of the differences between weather, climate variability, and long-term climate change, to determine whether the earth’s average temperature (i.e., climate change) is warming or cooling in the long term, you need to look at LOTS of data. Generally, climatologists examine at least 30 years of data when we are trying to draw conclusions about the climate. If we’re investigating long-term climate change, we need to look at even more data. So looking at the global temperature for the last 10 years can help us with questions of climate variability – but NOT with questions about long-term climate change. Why? Natural cycles of weather and climate occur IN ADDITION to the overall long term trends, and with only 10 years of data, you can’t use statistics to separate this natural variability “signal” from the long-term trend in temperature. For example, 1998 is tied with 2005 as the warmest years on record in modern times, as determined by the global average temperature (the tie occurs because the difference between the two years isn’t statistically significant – it’s within the difference we might expect due to errors in the data). However, the reason for the record global average temperatures wasn’t the same for both years! In 1998, an exceptionally strong El Nino contributed to a higher global average temperature, indicating that natural variability in our climate system had an effect. In 2005, there was no strong El Nino or other natural factor that contributed to the high global average temperature. When we look at graphs of the global average surface temperature over from 1998-2008 (Figure 1 – please excuse the crude graph), we don’t see a warming trend at all – but this is because 10 years of data doesn’t give us enough information to pick out the long term temperature trend from the natural fluctuations. In fact, the graph looks awfully flat. But if we want to know how the last 10 years compare to the long-term trend, we need to look back farther than 10 years – and when we do (Figure 2), we see that yes, in the long term, the global average temperature is still trending toward higher temperatures.

FIGURE 1: NASA GISS data for global average surface temperature, measured by meteorological stations, for the last 10 years (derived from Global Historical Climate Network data and converted to degrees Fahrenheit - available at http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt). Note that by looking at just this 10 years of data, it appears there's little warming trend - or even that the rate of warming is steady!

FIGURE 2: As in Figure 1, but with for 1901-2008 (the time period used in the Easterling and Wehring paper mentioned below). With more data it's easier for you to see the natural ups and downs in global temperature from natural variability - and a long term trend (trendline not shown).

To be honest, it’s important to note that different groups use different data to calculate global average temperatures – this is why looking at records like this one for trends in measured surface temperatures can be a little misleading. All estimates of global average temperature suffer from problems like changes in measurement locations (moving temperature gauges), changes in measurement methodology (weather stations to satellites), and lack of data from certain regions (e.g., the Arctic). These problems don’t make the data useless – rather, it just means you should be cautious about the strength of the conclusions you draw from the data. In my graphs, I use the GISS data set from NASA, partially because it’s easily accessible to everyone, so you can make the graphs yourselves. It’s also one of the data sets where scientists have gone back and removed as much bias from those problems as possible. Despite corrections, it’s still only based on global records from meteorological stations… so really these are graphs for illustrative purposes, not for a definitive analysis of global surface temperature trends!

One of the ways climatologists can study trends is to look at temperature anomalies, rather than the absolute surface temperatures I discuss above. The anomaly is the difference between a measured temperature and a calculated average temperature. In the data above, the GISS temperature data is reported as anomalies, based on the average global surface temperature from 1951-1980. This reduces some of the problems with variations between areas with lots of meteorological data stations and areas with fewer stations (NASA GISS 2009). It also allows you to see how warm or cool a year is based on the average (0). Figure 3 depicts the global surface temperature anomalies for the last 10 years, and Figure 4 depicts the anomalies for 1901-2008. Figure 3 still has a similar shape to the Figure 1 global average absolute surface temperatures – but it’s clear that these temperatures are still well above the 1951-1980 climatological average. Figure 4 suggests that despite natural fluctuations, since the late 1970s, our global temperatures have been consistently above average.

FIGURE 3: NASA GISS data for global average surface temperature anomalies for the last 10 years (Global Historical Climate Network data converted to degrees Fahrenheit - available at http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt). Note that by looking at just this 10 years of data, it appears there's little trend… but that temperatures have remained well above average nonetheless.

FIGURE 4: NASA GISS data for global average surface temperature anomalies for the last 10 years (Global Historical Climate Network data converted to degrees Farenheit - available at http://data.giss.nasa.gov/gistemp/tabledata/GLB.Ts.txt). Since the late 1970s global surface temperatures have remained above average – even though the size of the anomaly varies from year to year.

New work coming out from David Easterling and Michael Wehner takes this a step further (a draft of the paper here: http://www.cdc.noaa.gov/csi/images/GRL2009_ClimateWarming.pdf). They analyze whether periods like the last 10 years, where warming has not occurred as fast as it has in the past, are consistent with overall increasing global average temperatures. To make a long story short, they look at different data sets and models and conclude that yes, periods like the last 10 years where there is little warming – and maybe even cooling – are consistent with climate models. This makes sense, considering that natural climate variability leads to warmer and cooler periods. However, the overall trend still increases. So the moral to the story: even though the rate of warming over the last 10 years has slowed, we don’t see any evidence that this is the end of climate change. Periods where the warming rate slows or even reverses are not fully inconsistent with climate models, and the global average temperature is still trending upward.