When will the USGBC come clean about their energy data?

Ever since the U.S. Green Building Council (USGBC) certified its first LEED building, questions have been raised as to whether LEED-certified buildings actually save energy.  For years LEED proponents have attempted to answer these questions by putting forward energy simulations — calculations performed by the design team before a building is ever constructed (or renovated) that demonstrate how much energy the proposed building design should save.

The problem is that intentions do not equal performance, and numerous studies of buildings have demonstrated a wide gap between the actual measured energy performance of a building and its design simulations.

I have undertaken several studies that compare the energy performance of LEED-certified buildings with other, similar buildings.  One of the key barriers to such studies is the difficulty in obtaining measured energy performance data for LEED-certified buildings.  Municipal energy benchmarking disclosure laws are beginning to crack this “green wall of silence” but, even so, you will find energy data for only a few hundred LEED-certified buildings in the literature.

One of my regular end-of-the-year rituals is to download the current version of the LEED Project Database posted by the USGBC.  This database lists all registered LEED projects, including information about the LEED system, certification, number of points received, etc.  Below I will share some interesting statistics calculated for these data.

As of December 26, 2017, there are 23,137 LEED-certified commercial buildings (*) in the U.S., certified in programs that address whole-building energy (NC, EB:OM, CS, School).  This is nearly 100X the aforementioned number of LEED-certified buildings whose annual energy consumption have been studied in the peer-reviewed literature.  Obtaining energy performance data is a critical road block to understanding building energy performance.

To address this, the USGBC, starting in 2009 with its version 3 certification programs, instituted a requirement that all LEED-certified buildings must report to the USGBC for five consecutive years following certification, whole building energy use data.  It was hoped that such data would demonstrate the success of the program in saving energy and would guide future improvements in the LEED standard.

So, what have we learned from these data gathered by the USGBC?  We have learned that the USGBC does not want to publicize these data.  Four buildings were certified in version 3 programs in 2010 — so their first year energy performance data would have been reported in 2011.  That number has grown dramatically in successive years.  The graph below shows the total number of buildings certified in relevant LEED.v3 or LEED.v4 programs as of January 1 of the year shown.  By January 2017 this number had grown to nearly 10,000.  When 2018 arrives these buildings will have another year of energy use data to report to the USGBC.  Moreover, 1,931 of these buildings certified by the first of 2013 should be reporting their fifth year of energy consumption.  Where are the reports that analyze these data?

So why isn’t the USGBC making these data available for analysis?  The answer is simple — the data show that LEED-certification is not saving the 30-35% energy that the USGBC has claimed for years.  This is no different from General Motors suppressing data that show Corvairs are not safe, tobacco companies hiding data that show cigarettes cause cancer, or the Catholic church protecting priests accused of sexual misconduct.  All organizations, first and foremost, care about self-preservation.

But the LEED project data show another interesting trend.  Again, looking at the commercial LEED systems that address whole-building energy, it is interesting to look at the numbers of U.S. buildings that were certified by year.  This graph is shown below.  The graph shows a trend that you can detect when you talk to builders and building managers.  Interest in LEED is waning.  2013 was the peak year for LEED certifications in the US.  Since that peak the annual number of U.S. commercial buildings receiving LEED certification in these programs has steadily declined.  Builders and property owners are catching on to the fact that LEED buildings are not saving energy, and the novelty of certification is wearing off.

The graph above actually over-estimates the number buildings certified each year.  The reason is that some buildings get certified a second, and even a third time.  These certifications are counted above, even though these “re-certifications” do not add new buildings to the list (just new certifications).

The USGBC, of course, does more than just certify U.S. buildings in the whole-building energy systems considered here.  Marketing green is their strength — they have exported their wares to many other countries and they have invented new LEED certification systems that can make small tenants in large buildings feel good (e.g., commercial interiors, CI).  No doubt global USGBC sales continue to rise.

But make no mistake about it — the core product of energy efficiency is falling flat with U.S. commercial building owners because the product is highly flawed.

* The numbers provided from the LEED project database refer to registered projects.  It is not quite right to say each project corresponds to a building.  A

Energy harvesting — the siren’s allure

My wife, Deborah Mills-Scofield monitors dozens of media outlets and forwards articles to me that might be of interest.  One recently came my way about an effort in Portland, ME to harvest hydroelectric energy from its water pipes.  A company, LucidEnergy, has developed turbines that can be installed for this purpose.  The basic idea is to capture free energy in municipal water pipes that would otherwise be wasted.

While I applaud such innovation and creativity, I find the effort is misplaced.  I predict these turbines, like solar panels of the 1970’s and green roofs of this last decade — will soon be removed and abandoned.  This kind of energy harvesting is a fool’s errand.

About a decade ago I learned about another energy harvesting project in Israel — to install piezo-electric tranducers in highways to capture energy from passing trucks.  As heavy vehicles passed over these tranducers the truck weight would cause the transducers to compress and produce electricity.  The promoters of this energy argued that normal road compression represented lost energy — their technology would capture energy that would otherwise be lost.  The installed transducers did, in fact, produce electricity.  But I am confident that careful analysis would show that this energy comes from slight increase in fuel consumption of the vehicles that pass over the transducers.  Highway rolling resistance is mostly due to compression of the tires, not the road surface!

I am not aware of any evidence that water passing through municipal pipes arrives at end destinations with excessive kinetic energy.  Therefore any energy harvested along the way is likely to have to be re-injected by pumps.

And the maintenance issues must be significant.  I envision a few years of testing at the end of which it will be concluded that the cost of maintaining these units far exceeds the value of the energy they generate.  And what about the maintenance of pipes which get plugged due to low flow velocity?

Nature has handed us sunlight, wind, and hydo energy.  Harvesting these abundant resources is proving to be a challenge.  Harvesting efforts should focus on these well-understood and low-maintenance options.

Humans clearly waste a terrific amount of energy.  And there are many different ways that this wasted energy might be harvested.  The problem is cost-effectiveness.