We like the map too. In the spirit of critical examination, we couldn’t help but begin to explore its strengths and weaknesses. We have two concerns:
- As many of the map’s commenters have pointed out, the map is normalized by lot size, not by buildings’ floor space or usage. That means that small lots with heavily-used energy-intensive buildings (e.g., lower Manhattan’s skyscrapers) are shown in red for their high energy use, while lower-density lots (in, e.g., the outer boroughs) tend to be shown in green, indicating lower energy use. This is probably an artifact of limited data; it’s non-trivial, for example, to get access to the data you’d need to estimate the floor-space of every building in New York City. (It’s even harder to estimate how intensively buildings are used when you’re considering how to compare residential buildings to commercial buildings.) Nevertheless, it’s very hard to honestly compare how efficiently large, dense buildings use their energy compared to smaller buildings based on lot size alone.
- The map’s methodology, although sound, carries more limitations than the casual observer might realize. The researchers had energy usage data that had been aggregated to the ZIP code level, not data for individual buildings. They used statistical regression to estimate how much energy any given building is likely to consume. At a glance, though, you could be forgiven for mistakenly thinking that these data reflect actual measurements, not merely estimates. Ultimately, however, there’s no way of knowing from this map whether any given building is more or less energy efficient than another based on this map alone.
The map’s makers clarified their purpose in the comments section the New York Times’ coverage of the map:
First we would like to clarify that the energy consumption visualized on the map does not embody energy efficiency. The annual building energy consumption values portray what a particular building is estimated to consume, on average, given its size and building type. The actual building may consume more or less depending on a variety of factors.
Perhaps a building is LEED-certified, containing many energy efficiency measures. If one were to look at the energy bills of the building and compare them to the estimated values on the map, the energy consumption may be significantly less. On the other hand, maybe a particular building uses old inefficient systems to provide space heating and hot water. If one were to compare the energy bills of that building with the estimates on the map, the energy consumption may be significantly higher.
The main goal of the building energy map is to get the conversation started as to how much energy is being consumed and where.
We agree with the comments that perhaps normalizing by the square footage of the building would have provided a more interesting perspective. The map, however, would not look much different. In addition to the buildings in midtown being extremely large, they are also used for energy-intensive purposes. Offices and Stores are among the highest consumers of energy, according to our estimates.
What would stand out in a map of energy per building area however are hospitals and educational facilities. Also we’d like to point out that only total energy consumption is visualized on the map. The patterns of consumption change when one looks at one particular end use.