Climate Data and PHIUS+ 2015

 

Adam2smAdam Cohen is a principal at Passiv Science in Roanoke, Va, a PHIUS CPHC®, a PHIUS Builder Training instructor, the builder/developer of multiple successful passive building projects, and a member of the PHIUS Technical Committee. With the release of the PHIUS+ 2015 climate-specific standard, Adam weighs in on the importance of climate data sets.

Project teams have always needed to be discerning about climate data sets they use in energy modeling.  Whether it’s WUFI Passive, Energy Plus, PHPP or any other software, the old adage garbage in = garbage out applies. Project teams always must analyze and make a call as to how accurate the climate file is.

For example, I worked on a Houston, Texas project a number of years ago and there were several climate datasets that were close and one that was very different. As a team, we had to decide how to approach this in the most logical and reasoned way.

Recently as I analyzed a Michigan project, I determined that my two dataset choices were “just not feeling exactly right” so I asked PHIUS’ Lisa White and Graham Wright to generate a custom set. I can’t know that this one is exactly right, but I know that it’s as accurate and “right” as we can make it.

Note that when multiple data sets are candidates, it is not just altitude that matters, but location of weather station (roof, ground, behind a shed, etc.). Ryan Abendroth blogged on the subject of selecting data sets (and when to consider having a custom dataset generated) and I recommend you give his post a read.

Since PHIUS+ 2015 is a climate specific standard, it’s all the more important to use the best available.  We all know that bad data is not exclusive to PHIUS (remember the Seattle weather debacle in early versions of the PHPP).

It’s incumbent on project teams to use science, reason and judgment in interpreting climate data sets. Being on the water, in the middle of a field or in the tarmac of an airport makes a difference.

In New York City, for example, we have an oddity: There are three dataset location choices.

A satellite photo of NYC with Central Park outlined. The climate date for the Park is substantially different than that for other parts of the city.

A satellite photo of NYC with Central Park outlined.

One is Central Park, and the PHIUS+ 2015 targets for that are substantially different than the others. But, counter to a Tweet calling into question the validity of the PHIUS+ NYC target numbers, they are different because the Central Park climate data is substantially different – probably due to vegetation countering the urban heat island effect. It has a dramatic and pretty fascinating effect on the microclimate, and the U.S. DOE has a nice read on the subject.

For project teams lucky enough to have access to multiple data sets for their location, by rational comparison, they should be able to make an intelligent decision to use a canned set or to have a custom set generated.

It also more important than ever that the PHIUS+ certifiers to examine the weather data provided by a project teams to see if the project team made a logical, rather then an easy selection of climate data.

In addition, we on the PHIUS Technical Committee will continue to collect and monitor data and will tweak certification protocols as we see the need. But, I remind all my fellow CPHCs that bad climate data sets are endemic in the industry and it is important that project teams make careful decisions and that they reach out to PHIUS staff to help when climate data sets just don’t seem right.

2 thoughts on “Climate Data and PHIUS+ 2015

  1. In a way this NY case points to a basic precision limit of the performance approach itself – there may often be microclimate differences between the weather stations and the project locations. There are modifiers in Meteonorm to account for some situations, but we just don’t have 10 or 30 years weather data for every address. So try not to fall into “climate data paranoia” either.

Leave a Reply

Your email address will not be published. Required fields are marked *