Since we are in an energy transition , in which we completely switch to renewable energy, the energy at building level will mainly have to come from solar energy. We assume that, the more local energy the better, and a ‘0 energy house’, including compensation for material energy, to be provided on an annual basis by the building itself. *
Then, particularly in the built environment, unimpeded access to solar radiation has become a precious commodity (or will be). (Or will be) With the consequence that the space that is shaded behind the house or the apartment should be kept free, it becomes useless in the energy equation, because it brings nothing more, in terms of solar energy production capacity (for energy, but also for material and food) The potential to generate energy is basically gone, and you can’t pass it on to the neighbors, because they have the same problem. Which, of course, at a time when we need to switch completely to renewable energy is of enormous importance. That is the link from the building performance to the higher scale level: the (maximum) potential (or energy space) of all buildings together in a renewable energy based society. Hence a little thought exercise how that might play out:
So there is land occupied, the land for construction, but also via the shaded space by that building, together what I would like to call, the Solar Space of a building. Statically speaking then. Of course shading is dynamic, it varies in angle and length and duration. And how exactly to apply that in practice still requires some study, but we assume for a moment that cumulatively all those part-time shaded areas behind that building, add up to a more or less comparable in time permanently shaded projection area of about the size of that building itself. So the accumulated and claimed sum of ‘solar space’ is, as an assumption, the building footprint itself plus a shadow area the size of the building with and height . (And by the way that is most critical in winter: that is when there is the most need for energy and when the shadows are the longest and the yields the lowest… )
For our latitude, however, you could roughly set the (average) loss angle at 45 degrees for the most productive months. In other words: For a building of x meters high, belongs , seen from the south, an area of x meters deep behind the highest point to energetically lost area. And must therefore belong to the own area of that building. Because in that area there is, say, no more potential for a building to provide its own energy.
So for example for a tower of 100 meters, a strip the width of the tower, of 100 meters deep. That is the area from which the solar energy is claimed, or stolen from the neighbors, if it is not part of one’s own land. For that reason alone, high-rise buildings are a no go area, and therefore not an option in the future, when legally there will be established a ‘right to sun’ . (There are more reasons why high-rise is not effective, except for developers, see here  for that. )
So apart from the impact per useful m2 of living space, see part 1, there is also energetically a number of m2 of land occupation per useful m2 of living space: the solar space. And of course this is directly related to the impact per m2, certainly when this energetic impact has to be provided for, or at least compensated for by solar energy: Then the impact per built m2 floor (in Operational and Embodied energy) directly affects the required m2 solar space to compensate or provide that per m2 floor. ( or also translated into m2 PV/m2 floor, see ) . or vice versa: the building site determines the maximum generated solar energy potential, and the operational demand and pay back of embodied impact should stay within that budget.
So if the basic calculation of the impact per m2 floor, is combined with the occupied solar space per m2 floor (i.e. the total occupation of land as solar receiver space) then a calculation of buildings, and standards, in a per m2 approach would make sense, because that m2 of building is then directly linked to the availability of energy producing space (cq. land), and thus building space is directly linked (and limited) to the impact in the big picture, on that overarching scale level.
We can link this back to the example in part 1: the terraced or stacked apartments: in the terraced apartments, an area equal to the rear wall (height times width) is part of the solar space, in a 45 degree projection. Two properties next to each other then has no influence, both have the same non-overlapping solar space. (theoretically, statically)
If we stack the two apartments its the height times 2, and also the length of the shadow x 2 , so net shadowed area remains the same. However then the total solar space is smaller (built area plus shaded area) . This seems favorable but if we continue to calculate we see that the available ‘solar production’ space is also (logically) smaller than in semi-detached houses (for the same double floor area). This remains true for other and higher configurations. The solar production space = equal the land (solar) occupancy space. Only, if the solar production space is smaller, it is also increasingly difficult to meet an energy and climate neutral criterion (within the own plot – the zero energy option, incl. possible compensation for Embodied Energy). So in fact every building will eventually need the same amount of land per m2 useful floor…. So this also results in an optimal density! Higher density makes no sense, is counterproductive.
In fact, depending on how effective (low energy demand and low embodied impact) we build**, the lower the demand for operational energy, and the more productive we can capture and use solar energy. The solar space, (or solar footprint) becomes less. Regardless of whether we use that solar energy passively or actively, the basis is then the available solar space. In other words, the plot size then determines what is possible, and depending on the materials used and the creativity of the designer, that can vary somewhat. And then without claiming solar or environmental space outside of it. (Energetically speaking, that is; this is still without a claim on land for the growth of materials themselves, or for the recovery of those supplies)
So much for this (basic) exploration of the use of the m2-normalization in buildings with respect to energy and material impact. It is clear that there is more than just a turnover per m2 , and especially if you would base regulations on that, which in fact aim at a higher scale level than that of 1 building. You would then have to work in a fairly fundamental way to prevent yourself from improving on a square millimeter and losing sight of the big picture. (Compare that with the step-by-step reduction of CO2 from products, while cumulatively we go far above national budget ceilings).
Starting from solar space could however provide an indicator that would allow you to steer at building level, in terms of m2 averages.
* energy-neutral = operational energy, climate-neutral is operational and embodied combined, see for instance here: http://www.ronaldrovers.com/circular-energy-neutral-climate-neutral-system-neutral/ and http://www.ronaldrovers.com/co2-rekenen-klimaatneutraal-jaar-x/
** Keep in mind that when stacking over greater heights, in any case a higher energy and material use per m2 useful floor area is inevitable, and thus the situation becomes even more unfavorable.
(And remember that the infrastructure in fact also has to be directly linked to this, charged per m2 of housing…both (new) energy infrastructure and road infrastructure)