Some Thoughts on 3D Models, or Putting 3D Models to Research Use

I have an unease with ‘historical reconstructions’, 3D modelling and VR/AR of historical remains that centres on questions of research (see here). This is not to say such reconstructions are ‘bad’, but they seem to be an end-in-themselves. The reconstruction is a way of visualising the total understanding of ‘X’ (whether a building, street, city, etc.) and that is that. No further questions.

However, this reduces the reconstruction to nothing more than a pretty spreadsheet. Yes, a very pretty spreadsheet, but an expensive catalogue of archaeological remains (like this). Instead, I see the ‘tool’ side of models as their primary purpose. Models can be used as a tool to critically evaluate the ancient remains in a way other forms of research do not. As an example, a friend posted this NY Times article on mapping the shadows of buildings in NYC. It is a fascinating read about the effects of the skyline in everyday choices, movements, and, most important for NYC, property values.

‘Sunlight and shadow shape the character and rhythm of New York’s public spaces. They have the power to control the flow of foot traffic on our city streets and decide which plazas hum with activity and commerce and which stay barren and desolate. And probably most noticeably, they have the power to change the rent.’ (Bui and White).

Air and sunlight are, for NYC, commodities that shape the value of a property. I was immediately drawn to Roman law. Air and sunlight were legal requirements that restricted extensions to buildings (Dig. 7.1.30; 39.2.25; 8.2.11) and Vitruvius discuss the placement of certain rooms based on seasonal sunlight patterns (De Arch. 6.3.11; 6.6.6). The importance of air and sunlight were recognised by the Romans and this importance worked its way into law. This process from lived experience to legal requirement can be seen in terms of the commodification of sunlight similar to NYC (although, not in the same capitalist sense).

In my own research, I have looked at some of the social processes (all related to sounds) that shaped the way the ancient urban environment was constructed. Sunlight is an example of a natural process (non-built) that shapes social interaction in urban space. It is that point, the non-built aspect of sunlight in relation to physical buildings that allows for models to be a critical tool. I can stand in Ostia and track the movement of the sun’s path, but by modelling the building and sun’s path, I can question building heights influence on street space, or requirement of internal lighting for room usage, or even hypothesise relative values based on direct sunlight in comparison with other buildings. The list goes on the more you think about (as I have found out…). So, with the legal requirement of light and the archaeological remains of Ostia (as well as a question from a friend), I went fishing for sun path models to apply to Roman buildings.

It did not take long to find Andrew Marsh’s blog, which has a web-based app for 3D sun path analysis. As the sun’s path is based on the geographical location of the space in question, the app uses Google earth to pull coordinates (top left in the screenshot below). There is a map of the daylight length (bottom left) and a moveable 3D model of a multifunctional urban space (right). The 3D is centred on the building in red and set on a compass so, you can orient it, if you like (I set it roughly to the orientation of the site so, I did not confuse myself).

screen-shot-2017-01-11-at-9-56-22-am

Sun-path for the Portico di Pio IX in Ostia on 12 Jan. at 10:30am (from http://andrewmarsh.com/apps/staging/sunpath3d.html)

I selected the Portico di Pio IX on the north card maximus in Ostia (see here, here and a forthcoming chapter here) and tomorrows date (12 Jan.) at 10:30am. The day length is relatively narrow and the suns path is rather low in the sky, running from 120∘ (SE) – 240∘ (SW). This creates long shadows to the west in the morning appearing between 8:10-20, direct sunlight down the street at midday, and long shadows to the east in the evening until sunset around 16:30-40. This forms the basic outline of sunlight throughout the day (at this point in the yearly cycle).

Now, what are missing in this scenario are the buildings along the cardo maximus. The mixed-use urban space in the 3D model is instructive, but only if you know the buildings on the street (this is where models become more than pretty pictures). Now, the cardo is an open space (8 m wide and 130 m long) with two identical buildings on either side (rows of 8 shops with a portico in front). At midday (12:00), the sun would be reaching its peak, almost due south of the street. Two things come to mind; 1) this indicates that the temple (Capitolium) in the forum would receive direct sunlight to its facade and steps, while 2) casting a long shadow down the cardo. This feature would not change throughout the year, as at the summer solstice and winter solstice the sun’s position, at midday, remains in the southern region of the sky.

I also found a dynamic daylight analysis app on Marsh’s blog, which allows one to simulate daylight in a simple room (initial release so just a square room). So, I built a shop in the Portico di Pio IX and set a window to the size of the front door (as there is no ‘door’ in the model).

Screen Shot 2017-01-10 at 5.02.30 PM.png
Dynamic daylight model of a shop (Portico di Pio IX) (from http://andrewmarsh.com/apps/staging/daylight-box.html)

The shops in the Portico di Pio IX had two levels (a mezzanine space above the ground floor); the model is of the ground floor area, under the mezzanine. Again, the app allows you to mess with the orientation and you can adjust all the elements as needed. The result, shown above, is as expected. A high daylight factor (DF) in the area around the shutters with the work plane height set at 1 m. The DF range is between 2.7% and 23.9% with an average of 5.7%. The DF is the ratio of light inside to light outside the room (a calculation used in architectural design. There are slight changes to the DF ratios as the work plane changes (avg increase to 6.8% at 0.05 m from the floor) and at a work plane height of 2.135 m the DF avg drops to 4%.

As an example, I placed the work plane height at 0.85 m, which corresponds to several bar counter heights in Ostia (see Hermansen 1982). The result is interesting, or at least allows for some comment. At 3 m into the shop, the DF is in the range of 6-8%, suggesting that natural light would light a counter (as counters were under 3 m in length).

Screen Shot 2017-01-11 at 11.25.28 AM.png
Dynamic daylight in shop (Portico di Pio IX) with a work plane height of 0.85 m (highlighted in red; bar counter height) and DF contour grid in 3D. (From http://andrewmarsh.com/apps/staging/daylight-box.html)

What is missing, which will change this model, is the portico space in front of the shop. As seen in the sun-path, there will be little direct sunlight into the shops along the Portico di Pio IX. In particular, the shops on the westside will get early morning sunlight and the westside, evening sunlight. The portico will block much of the light at these times, however in the midday sun, it will provide needed shade (as anyone who has worked in Italy in August can attest). What this model does suggest is that choices, such as counter height, could be dictated by natural light, and importantly, the heating capacity of natural light. Although, in the case of the Caseggiato del Termopolio (1.2.5; the most well-known bar in Ostia) it’s counter and space orientation will never receive direct sunlight. What we might begin to see is the material remains of the ‘commodification’ (for lack of a better term) of sunlight within an ancient context.

Finally, it is worth returning to the start of this post. 3D models and reconstructions are useful and helpful for historical research. The discussion here has benefited directly from such modelling techniques, although, as stated, it was the need to move past the image, fly-through, or even simply trying to document every element within the model that drove my interest. We need to push the utility of 3D models and reconstructions for critical questioning of our own academic assumptions and start to use them for addressing new modes of enquiry and topics of interest, rather than as an end in-themselves.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s