Garbage In, Garbage Out

Posted on May 18, 2011 by Shari Shapiro

There is a corner of the Federal government that, unless you are as data obsessed as I am, you never knew existed.  For the part 25 years, the Energy Information Administration (EIA) has collected  baseline data on commercial building energy usage, known as CBECS.  CBECS is the only government source of statistical data for energy consumption and related characteristics of commercial buildings.   The data EIA has collected forms the underlying data for programs like Energy Star and LEED, and laws like Energy Independence and Security Act of 2007 (EISA 2007).

Two pieces of bad news were released today.  First, the EIA tried to cut costs by contracting out its building energy information gathering.  Unfortunately, the data gathered by its contractor was so shoddy that the EIA is refusing to release the data. 

According to a press release from the EIA:

EIA regrets to report that the 2007 Commercial Buildings Energy Consumption Survey (CBECS) has not yielded valid statistical estimates of building counts, energy characteristics, consumption, and expenditures. Because the data do not meet EIA standards for quality, credible energy information, neither data tables nor a public use file will be released.

Worse still, according to the EIA, the budget cuts made in this year's budget negotiations have reduced the EIA budget so severely that it will suspend collection of 2011 data.  In other words, the best building data we have to work with at this point is almost 10 years old, and no further data is being collected. 

Because approximately 30% of all energy is used by buildings, without a good baseline assessment of current building energy use, it will be difficult to:

  1. Accurately benchmark current building performance; and
  2. Accurately model building energy efficiency efforts. 

If energy models are based on bad or old or unreliable data, the results in practice may not live up to the predictions, and it will be easier to dismiss efforts to comprehensively transform building energy usage.  In other words, garbage in, garbage out.   

To address this situation, there are two choices. 

 A relatively unbiased private organization, like ASHRAE or the International Code Council, could collect the data.  Although the standard setting organizations will have their own internal and external customers to serve, and questions will inevitably arise as to the bias and validity of the data, they possess the relevant expertise and are currently relied upon to provide input into laws (like building codes).  The other option is for the Department of Energy or other government agency to recognize data gathering as a priority, and reallocate funds for the purpose of building energy efficiency data collection. 

In any event, a reliable party must step up to fill the data void, or future efforts to actualize the most cost effective method of reducing energy use and greenhouse gases will be squandered.  We will pay now through unnecessarily high energy costs and the price tag for participating in conflicts in the Middle East, and future generations will pay for damage to the environment.  Garbage in, garbage out. 

 

Tags: , , , , , , , , , , ,

The 50% Rule or Why Emails and Statistics Don't Matter

Posted on December 8, 2009 by Shari Shapiro

We have heard a chorus of voices over the past few days raising the moribund concept that climate change is not happening, and is some global liberal conspiracy to devalue oceanfront property in Palm Beach. 

At the center of raising the hydrahead of the Palm Beach Conspiracy was the discovery of  some emails from the University of East Anglia where climate change scientists were engaging in the age-old academic practice of arguing with one another.  For a "pro" climate change perspective, Gawker explains the situation here, for an "anti" climate change perspective, the Weekly Standard provides this analysis.

I was guest lecturing at Princeton a few weeks ago, and I used the opportunity to propogate one of my favorite ideas--I call it the 50% Rule. It can be used to explain the Palm Beach Conspiracy, statistics about climate change, and as a means of deflating your brother-in-law's wild stories about catching a 45 foot trout during holiday meals. Here it goes--when you hear a statistic or a scandal or a wild trout fishing tale, assume the information is off by 50%.  One-half.  Then determine whether the information still matters.  If your brother's trout was only 22.5 feet, not 45, that's still a mighty large fish.  Similarly, with climate change, if scientists' statistics about sea level rise or drought are off by 50%, we are still looking at a serious problem.  The result? We still need to do something about it.  

With respect to the Palm Beach Scandal, Micheal Oppenheimer from Princeton on NPR explained it beautifully. The consensus of hundreds of scientists, using many different methodologies, all in competition with one another have reached a consensus that climate change is real and caused largely by man's actions.  Even if 50% of the data is wrong or subject to bias or manipulation,  that is still hundreds of the world's best scientists coming to a consensus (which if you have ever had two scientists in a room is a feat in and of itself) coming to the same conclusion.

Finally--here are the choices. Assume climate change is not real, and roll the dice on droughts, wars, starvation, dependence on foreign oil, continued economic stagnation and incalculable human suffering.  Assume climate change is real, take action, create new jobs, industry, reduce pollution and human health risks from carbon emissions in general, reduce dependence on foreign fuel regimes and potentially keep polar bears from extinction.  Strikes me as not much of a choice.

Tags: , , , , , ,