Climate Change — Just a Data Problem?

2712986388_30e3c433f2_b
By Jason Trager

Data and statistics, when used properly, can provide key insights to help companies make sense of a complicated competitive landscape.

Unfortunately, statistics are often misused. This is, in part, the lamentable result of leaders either improperly selecting or simply fabricating statistics to justify a foregone conclusion.  At best, this tends to result in gross mischaracterizations of critical trends and, at worst, flat-out falsehoods.

Take for example, the bold statement made by Ted Cruz, a Republican junior Senator from Texas and U.S. presidential hopeful: “The satellite data demonstrate that there has been no significant warming whatsoever for 17 years. Now that’s a real problem for the global warming alarmists. Because all those computer models on which this whole issue is based predicted significant warming, and yet the satellite data show it ain’t happening.”

Cruz’s refusal to accept climate change as a highly reproducible scientific result is based on his particular statistical interpretation of the data.  And, unfortunately, his nonprofessional interpretation leads to a false conclusion, that “it ain’t happening.”  Just as a taster to illustrate how much expertise is required to correctly interpret this type of data, I highly recommend reading this article from the Washington Post.

In order to avoid politically fractious and polarizing opinions on key topics like climate change and energy efficiency, we need to dig deeper into the definition and interpretation of data and statistics.

“Data” is not the same as “statistics”

Data are representations of real-world phenomena.  We can use these representations to learn about the world around us and identify a particular trend or event hidden in the data. These representations are deciphered with statistics.

Statistics is the study of collecting, analyzing, interpreting, presenting and organizing data. Before using large quantities of data for practical applications, such as enabling and implementing positive environmental remedies, useful statistics must first be gathered that inform our understanding of the data.

The importance of choosing the appropriate statistical metric can be illustrated by a simple example with temperature data.  First, consider a single-year average of global temperatures, which does in fact show several years over the last decade during which time global surface air temperatures declined.

Temperatures Worldwide, 1901 to 2013
Temperatures Worldwide, 1901 to 2013

This could potentially be useful, as it provides a somewhat focused summary of trends over a large geography and variability in global temperatures from year to year. However, the story changes quite a bit if instead we choose 10-year averages for our statistic. Looking at temperatures in this way shows a clear upward trend.

The Global Climate 2001 to 2010
World Meteorological Organization: The Global Climate 2001 to 2010

This upward trend is manifestly clear despite the short-term cooling or warming drifts that have occurred between recent individual years.  This applies both temporally and geographically: Average temperatures globally are rising, but there’s snow in Washington, D.C.  These are not conflicting pieces of information.  We cannot learn about long-term or global trends with short-term or local statistics.

In other words, data represents events that occur in the world, and statistics provide meaningful insight about these events. However, fully understanding and implementing lessons learned from large datasets requires well-considered statistics. By using the wrong statistics we are answering the wrong questions. This means we might be missing or misrepresenting the story that the data tells.

Unfortunately, the statistics we use to understand global climate phenomena are telling a tragic tale, and we must find a way to change these trends. I believe that a large part of the solution to this process involves using mass production methodologies to roll out efficiency at a rapid and global scale. This methodology will be covered in my next post.

Image credit: Craig Mayhew and Robert Simmon, NASA GSFC

Jason Trager is CEO and co-founder of Persistent Efficiency, maker of the stick-on energy sensor. He is an energy scientist and sustainability engineer with a PhD in Mechanical Engineering from UC Berkeley. An experienced team leader, project coordinator, and fundraiser, Jason founded the Art Rosenfeld fellowship for energy efficiency at UC Berkeley for which his team raised over $600,000 in endowment funding. In addition to the fellowship program, Jason has created several full-time, recurring job positions at UC Berkeley for undergraduates through his Sustainability Champions Internship program and works with energy companies as a consultant.

3p Contributor

TriplePundit has published articles from over 1000 contributors. If you'd like to be a guest author, please get in touch!