Green Roof modeling for EAc1

Your question appears to assume that there are energy savings. Based on the research I have done and attempts to model green roofs I would say that - it depends.

The modeling of a living system from a thermal perspective is not at all simple. These are dynamic systems that change characteristics depending on the time of year and the weather. There will be times when they add to the conductive insulting value of the roof and times when they detract from the insulting value. In my experience a green roof will have virtually no direct thermal effect related to conductive losses.

One aspect which can be modeled to a limited extent is the effect of a green roof on the ambient roof temperature. Modeling reflectance/absorbtance can affect roof temperature and should have a measurable effect if you are bringing in outside air from the roof.

Given your small roof area I would model it similar to a white roof as this is as close as you can get in most models to a simulation of the effect.

I completely agree with Marcus on this. The research we have done on projects where we have a vegitative roof, even on 100% of the building, do not support any energy savings other than the actual roof insulation under the growing medium and usually above the membrane. The manufacturers all claim various levels of energy efficiency and then have very little guidance other than roof temperature.

First of all, please do not take my comments as someone that does not like green roofs. In fact, there are many very good reasons to have a vegetative roof, I just do not think energy efficiency is one of them, and do not feel attempts should be made to “payback” the roof based on energy savings. There are major benefits including connections to nature, views from adjacent buildings, heat island effect, delay of rainwater release, reduction in stormwater quantity and improvement in stormwater quality. All very good reasons to consider this technology.

There are two main claims for energy savings, conductive and evaporative. For conductive, just the ability to insulate a roof, the thermal conductivity of soil is directly related to moisture content…the more moisture, the more insulation, and at that it is not that great. Having just gone through a sustained drought in the Midwest, I would have a difficult time including any insulation contribution for a green roof in my calculations.

The evaporative gains are from evapotranspiration, or the ability of the plants to transpire water, and for the water to evaporate…evaporative cooling in a way. I have not seen evidence that this has a major effect on “cooling” a building, and if someone has that data, I would love to see it.

On one project, we wanted to test what the savings could be if the roof did not exist, in other words infinite insulation, and no solar gain. The savings this would represent were so small, it was not worth discussing past that point. This was done as part of a utility rebate program by a very well known energy modeling firm. They came to the same conclusion.

So, that is my reasoning, but am open to more information and data. But the real test will be when ASHRAE includes how to calculate the heating and cooling loads of a green roof in their handbooks!

We did not really do field research but relied on the research we could find. It was very limited. As Scott alluded to roof temperature data is really the only thing you can used to try and determine any energy savings. There are two effects here. On a large scale implementation of many, many green roofs would lower ambient temperatures related to the urban heat island effect and result in savings. On the individual roof scale LBNL and ORNL have done research on the roof temperature effects of more reflective roofing. The research we found showed that the temperature differences between the dark and light roofing materials is relatively small (smaller that we expected). I recall that there was some data from the Chicago City Hall green roof retrofit related to energy savings due to cooler roof temperatures.