|►◄ Reverse Zone|
Complete List of Posts
Tue, 28 Jul 2009
The current version of LEED-ND is being put to a vote at the Congress for the New Urbanism, with voting to begin any day now. Essentially, with the LEED for Neighborhood Development Second Public Comment Draft having just closed its public comment period but before revisions are made based on those comment, the CNU, one of the 3 organizations developing the standard, has decided to cut to the chase and ask its members whether to start applying it. According to New Urban News, "Instead of calling for more discussion, LEED organizers want to put the Neighborhood Development program up for a yes-or-no vote." They quote New Urbanist Doug Farr, "the general consensus was that we wanted to get LEED-ND out on the street and get people able to use it."
LEED, Leadership in Energy and Environmental Design, is a certification for green buildings. Already in trouble because there is no statistical correlation between LEED levels and energy use, LEED buildings tend to be the big wasteful forms to begin with. However, even having actually efficient buildings does not good if that building is located somewhere that will force people to use a lot of energy on transportation. Hence, LEED-ND.
Although the latest draft has lots of good features, its flaws are great and have gotten even worse. These flaws go directly against CNU principles, but are hidden in the math.
Simply put, this document has a strong bias for high-rise concrete residential buildings over the types of forms and communities that CNU prefers. Look at the diversity index, on the surface one of the key principles that CNU stands for. What this standard does is use the Simpson Diversity Index, a number used in various fields (this name for it comes from biology). This is the wrong index to use. It is appropriate only when you assume that the underlying population has equal numbers of each species. So for instance if you categorize people by income and you get equal numbers of people with income above $200,000 and below, would you say you have achieved diversity of incomes? Of course not; the distribution that gives the highest value for the index is nothing like the distribution you should be trying to achieve. If you split that up into multiple categories, $200,000-$250,000, $250,000-$300,000, $300,000-$350,000 and so on, the highest value of the index will be achivable only by shutting out lower-income people completely.
This is what has happened to the Diversity score. Houses, duplexes, and townhouses are only 4 of the 20 categories of housing. The majority of the categories are hair-splitting distinctions between different types of high-rise apartment units. But the average person is not equally likely to live in an apartment compared to some kind on house or townhouse, and is definitely not five times more likely, the assumption that LEED-ND has built into this index. This must be made clear: the highest scores in this category are only available to those who build several high-rise buildings in their project, and are not available to those who build ground-oriented housing. You can get full points while having absolutely no houses or townhouses. Actually, they revised the definition of townhouse so that a townhouse can now be within a high-rise apartment building.
This, coupled with a scale where greater density always gives more points means that the higher-scoring projects will be predominantly high-rise. How else can you achieve over 63 dwelling units per net acre and over 3.0 floor area ratio for non-residential? By calculating based on dwelling units rather than population per acre, you are already biasing strongly for apartments and against families. You need 3 times as many apartment units to house the same population as a house or townhouse. Apartments get triple the points per population by giving them the same weight.
To be truly smart and green, we must house all of the population, not create areas where only the apartment-dwelling minority will live. Doing that just fuels more sprawl and emissions, by putting families, those with the highest demand for transportation, preferentially further away from transit. This particular prejudice about putting apartments near transit is further aggravated by making sure that a larger number of units (not of people) are closer to transit and to amenities, and by allowing a much lower density further from transit. In fact, projects without transit are not required to meet the stringent density standards at all. This will simply perpetuate having a large proportion of the population not having transit-supporting density. It ought to be the other way around. Areas without transit should be built to high enough density that they will support the addition of transit, whereas areas that already have transit do not need change. In the case of infill, the new project should be substantially similar to the existing community.
This is another one of the fatal flaws of the LEED-ND standard: there is very little attempt to adapt to the surrounding community and to improve it, despite now having a greater emphasis on infill and redevelopment. The standard is mostly free of context. It is perfectly happy to enter a tight-knit community of old townhouses with a walkable school having marginal enrolment, to demolish half and to build modernist highrises. You can lose one point, but only if you demolish buildings that have been registered as historic, without getting approval.
Adapting to the local community means analyzing what is good and what is not good about it, and preserving what is good and compensating for what is not good. So for instance, if an area has no housing for seniors, building that form is highly desirable. If it has the amenities for families (schools, parks, community centres, etc) but is losing families, then family housing is what you must build; project-level diversity is not as important as the community's ability to continue. If it is in the middle of a transit-served employment district, you must build employment. If it is not dense enough for transit, you must give it the required density. If most of the area is not served by walkable small-scale retail, then forget the scoring system that tells you need a large number of square feet all in one place and do what's right for the population that's already there.
On the energy efficiency front, LEED-ND is a bit of a dud. It only requires that large buildings be a few percentage points more efficient than ASHRAE 90.1-2007. Given how ineffient both residential and non-residential high-rises are, this doesn't even bring the energy use per square foot down to the level of the average single-detached house. Having separate standards for different housing forms is another way to give highrises a free pass. An Energy Star house has much lower energy intensity than the ASHRAE 90.1-2007 code. And, granted that following the code is not highly correlated with energy use to begin with, 5-10% beyond code and maximum points at 22-26% beyond code is a pretty timid target. Especially since federal law about to be passed will require energy reductions 30% below ASHRAE 90.1-2004 in 2010 and 50% below in 2014. Another major missing factor is embodied energy. Although it is mentioned in passing in the introduction to adaptive reuse, ignoring embodied energy or embodied GHG emissions is another way to ignore the relative environmental merits of different building forms. Concrete and steel used in building certain forms take a huge toll on the environment compared to wood frame construction. This factor is nowhere in LEED-ND, and is one of the many elephants in the room.