digital edition

Stormy Weather

We use climate data to inform design every day, but what if that data doesn’t tell the whole story?

September 1, 2010

By Russell Fortmeyer

Relying on history to help predict future events is an almost reflexively human tendency. For decades, banks have mined historical data to structure their investments—a blind reliance on such mortgage and housing data is widely considered one of the leading causes of the housing market collapse and current recession. Applying an old, low-risk model to high-risk homebuyers turned out not to be so wise a decision.

Stormy Weather
Illustration by Oliver Munday
Rate this project:
Based on what you have seen and read about this project, how would you grade it? Use the stars below to indicate your assessment, five stars being the highest rating.
----- Advertising -----

Design is not so different. There are a heap of assumptions, based on varying levels of risk, underpinning any new building project. The riskier the design, the more testing and analysis we undertake to prove its basis and predict its performance. But might the historical data we use to establish one of the fundamental contexts of architectural design—climate—prove as unreliable as mortgage data proved for banks? Ultimately, could it be as risky to our industry?

Project design often begins with a climate analysis. Someone, usually the mechanical engineer or environmental consultant, summarizes the site’s annual temperature ranges, wind velocities and directions, humidity levels, solar exposure and insolation, rainfall and snowfall, all with an eye toward informing the design so that it responds to weather conditions inherent to the site.

There are many simple tools we can use to do this, such as Climate Consultant, developed and freely offered by the University of California in Los Angeles. This tool relies on weather data collected by the National Renewable Energy Laboratory (NREL), which collates hourly statistics provided by the National Climatic Data Center (NCDC) from weather stations located throughout the world. These statistics are then generalized over 30 or so years to form a “typical meteorological year” set of data (the most representative conditions for a given hour from those 30 years are then used as the typical condition). The Department of Energy formats this data and makes it freely available for use with its EnergyPlus energy simulation software or for other programs. The most recent data set includes years up to 2005.

Engineers use this annual climate data to predict how a building’s systems or envelope may perform, while the actual sizing of equipment like a chiller is based on maximum and minimum conditions, sometimes called “design day” temperatures, among other things. The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) provides design-day information in its fundamentals handbook, which forms the basis for practically all mechanical design in this country.

This is not to say that all of this represents an exact science that is blindly followed. There are countless other sources for climate information and there will always be good, old-fashioned hearsay, which although generally of little value, cannot always be dismissed outright. Uncertainty resides at the core of any set of climate data, but accepting that uncertainty and placing it in a reasonable context that minimizes risk is what designers do every day. This may be the best approach we have for addressing those building owners and operators who are skeptical about the capability of building systems to address those hot days that seem to be increasingly frequent.

To better address this uncertainty, we can start by not only analyzing our buildings with historical climate data, but applying a morphing technique to the data based on anticipated climate change. This approach has been applied in the UK, where the Chartered Institution of Building Services Engineers (CIBSE) developed a technique for morphing weather data files based on the conditions likely to occur if the world progressed according to a specific climate change scenario proposed by the Intergovernmental Panel on Climate Change (IPCC) and models developed by the country’s Met Office, which is similar to our own NCDC. Using CIBSE’s technique, we can use the typical meteorological year data for a given location to understand how a building system or envelope might perform in 2080.

Naturally, there are challenges with such an approach given that the IPCC has several climate change scenarios available. Currently left with little industry guidance, consultants will have to make some assumptions. Although building equipment installed in 2010 is not likely to be in operation in 2080, such analysis could influence decisions about system robustness or the use of passive strategies.

Confronted with this avalanche of data and a skeptical industry, it’s clear that we need to spend more time in the early stages of design talking about climate—on all projects, not only those highlighting sustainability or enjoying healthy budgets. It’s not enough to insert a clause into a contract or specification absolving the engineer of unforeseen building failures due to climate change.

Perhaps we could learn from the way we classify flood risks for a given location and assign a “climate risk.” However that plays out, we have the opportunity now to develop a set of professional services to address these concerns—reviewing alternative climate data sets, morphing existing data based on predicted climate change or the potential for extreme weather events, discussing risk and professional obligations between consultant and client. If the building industry is not forthright about climate risks today, the lawyers will certainly be glad to step up tomorrow.

Russell Fortmeyer is a Sydney-based journalist and engineer who frequently contributes to GreenSource.

This article appeared in the September 2010 print issue of GreenSource Magazine.

 Reader Comments:

Sign in to Comment

To write a comment about this story, please sign in. If this is your first time commenting on this site, you will be required to fill out a brief registration form. Your public username will be the beginning of the email address that you enter into the form (everything before the @ symbol). Other than that, none of the information that you enter will be publically displayed.

We welcome comments from all points of view. Off-topic or abusive comments, however, will be removed at the editors’ discretion.

----- Advertising -----
Click here to go to product info Page
Sweets, Search Building Products
Reader Feedback
Most Commented Most Recommended
Rankings reflect comments made in the past 14 days
Rankings reflect comments made in the past 14 days
Recently Posted Reader Photos

View all photo galleries >>
Recent Forum Discussions

View all forum discusions >>