Bridging the Gap

Scott Lewis

The July 26 Sustainable Industries article entitled “Closing the Gap” raises important questions created by the frequent disconnection, or “gap,” between how buildings are designed and built, and how they perform once completed and occupied.  Citing a range of data from the New Buildings Institute and the U.S. Green Building Council, the article observes that buildings often use more energy than predicted by the computer models used during the design cycle, concludes that the modeling is fundamentally flawed and that designing to Leadership in Energy and Environmental Design (LEED) specifications often does not deliver expected results.  In an era of urgent ecological, economic and social sustainability challenges, we must carefully examine the real sources of the design-performance gap and then focus our efforts where they will have the greatest impact.

What is really at the root of the problem? Is it that LEED is fundamentally flawed?  Or perhaps, the energy models are “fake,” as one observer suggested?  After working on well over 100 high performance building projects in markets around the world—buildings of all types and sizes ranging from million square-foot skyscrapers to K-12 schools—we at Brightworks have developed a point of view on this question that might help inform this important discussion.

Unfortunately, no one-size-fits-all explanation sufficiently answers the question. But to blame LEED or the energy modeling programs would vastly oversimplify the issue. 

To begin with, until LEED v3 was released late last year, a building could achieve LEED Platinum without necessarily achieving a high degree of energy performance.  LEED awards a range of “green” attributes, from water efficiency to recycled materials content to energy performance.  A high LEED score does not imply, by any definition, energy efficiency far beyond code. So to say that “LEED buildings do not pass Energy Star benchmarks” is a little like saying “these apples make terrible lemonade.” 

Should LEED put even greater emphasis on energy performance?  In an era of Deepwater Horizon-type disasters and climate change, the faster we can move to a renewable energy economy the better, and LEED could be a more forceful driver to that end.  But the fact that LEED buildings don’t always get great Energy Star scores does not itself imply that the buildings are not performing as designed. This is an important fact to understand.

Next to consider is the issue of computer modeling for energy performance.  Computer models are a tool.  A good tool, but imperfect.  They focus primarily on the orientation, shell (windows, walls and roofs) and mechanical systems (HVAC and lighting) of a building.  Plug loads, as noted in the “gap” article, are included in the model, but they are entirely speculative.  By the end of the design phase, a team will know what kind of glass will be in the windows and what kind of insulation will be in the walls, but how many CPUs will be plugged in on a given floor can change up to and after the day of occupancy. 

In fact, in the Oregon Health & Science University (OHSU) Center for Health and Healing—the building discussed in the article—this was a major source of the gap between model and performance.  The tenant made major program changes after the design and computer models were done.  The building, as operating, has higher occupant density and more load-hungry lab equipment than the design team was given as input to the modeling process.  This happens.  But even so, it is still an exceptionally high-performing building and is perhaps one of the highest performing existing buildings in the country.  If the actual occupancy and use of the completed building aligned with the initial plan, the building may well have performed as well as, or better than, the models predicted.  The problem isn’t the models, it’s our expectations. 

As a counter-example to the theme of the gap story, consider the Henry condominiums in Portland, Oregon. This is the first LEED-Gold certified condominium in the world.  The Henry’s energy model predicted savings of about 45 percent compared to the Oregon energy code.  A post-occupancy study of the Henry, done a year after occupancy, showed the building was actually beating code by more than 50 percent. It was outperforming the model by nearly 10 percent.  Why the divergence in this outcome and that of the OHSU building?  Simple. The OHSU building had significant program changes after the model was done, the Henry did not.  Steve Domries, an architect with the firm that designed the Henry—GBD —told me that two years after moving into the building, he had never needed to turn on the heat or air conditioning.  In this case, the correlation between LEED and outstanding building performance had no gap whatsoever, and the energy model actually had a “reverse gap,” underestimating the savings.

No computer model will be able to anticipate changes in occupancy or use of a building, so let’s focus on where the real opportunities lie. Perhaps the single most important suggestion would be for building owners or managers to adopt specific, systematic measures–this means policies and procedures–to ensure their buildings perform to their design potential.  Even the fastest Ferrari won’t stay on the road without a regular tune-up.  LEED for Existing Buildings, Operations and Maintenance (EBO&M), for example, includes nine specific policies and procedures and sequence of operations requirements to ensure buildings operate to the potential of their design and systems, taking into account program and occupancy. The second suggestion is to do careful, complete financial analysis to help create the buy-in and support needed to ensure the needed commitment for your EBO&M program to succeed.  In our experience, an ROI analysis typically shows rapid payback from the easily measured strategies (energy and water savings, typically), leaving the harder-to-quantify benefits (product differentiation in a competitive market, which leads to faster sales and leasing and premium pricing) as gravy (pdf).

None of this is to say there are not real cases where buildings simply do not perform as designed.  But in exploring for leverage points to improve our design, construction and operating practices to push the greatest possible performance from our buildings, it is essential we don’t confuse correlation with causation, and thereby focus on fixing problems that don’t exist or are less vexing than they might appear.  Remember: LEED and Energy Star are only distantly related measures, because energy only comprises a fraction of the total LEED score, and secondly, program changes can throw off energy models in buildings that still perform to or exceed their design expectations.  By focusing effort after construction on the key leverage points – we can ensure our buildings deliver the energy and dollar savings they were designed and built to provide.

Scott Lewis is the founder and CEO of Brightworks Sustainability Advisors.