The quest to reconstruct al-Qusur in 3D (1/2): aims, previous work and issues

After carrying out surface mapping between 2011 and 2018 (Humbert 2020), the French-Kuwaiti Archeological Mission in Failaka (MAFKF) decided during their 2023 campaign to create a 3D digital model (render) of the central part of al-Qusur’s settlement, focusing on three buildings and general views. Al-Qusur site, located on Kuwait’s Failaka Island, has virtually no preserved elevations and while there are some fifty-centimeter-high walls, the majority are just one or two courses high. In this situation, producing a model of what the site could have looked like was crucial for both the public and archaeologists.

The digital model seeks to depict the buildings and site in one of their probable phases using high probability criteria, either by deduction starting from the conserved parts of the site or after comparison with similar constructions or buildings from the same period and region. The process used to create the 3D model will be detailed in two blog posts. The current and first one will discuss the aims of the model, previous modelling attempts involving al-Qusur, and the limits of such an exercise.

Why create a 3D reconstruction of al-Qusur?

Al-Qusur, which was formerly a large settlement, is no longer easy for visitors to understand. This is because the structures have suffered from erosion over time, which has been accelerated by salinity and atmospheric conditions and has caused the parts of the buildings which were above ground to nearly completely disappear. The majority of the preserved elevations were ones which had been buried underground as this protected them from deterioration. As a result, the site appears only to exist in 2D, with traces of the former walls, making it very difficult to comprehend for people who are not used to this type of exercise (fig. 1).

Fig. 1. Aerial view of a part of the al-Qusur site in its current state (H. Al Mutairi, 2016). This shows the absence of any elevations and the fact that the site has a 2D-like aspect which hampers visitor understanding of the site.

Three-dimensional (3D) models have a didactic purpose as they make it possible to better understand a site as a whole for both archaeologists and the public. Indeed, archaeologists are used to looking at their site on two-dimensional (2D) models (plans, elevations and sections) and so might have trouble grasping an overall 3D perspective and could gain from locating their excavations within a whole settlement. Moreover, the public is usually composed of non-specialists who might have difficulties understanding technical drawings that can appear to be too ‘cold’ and too complex. Humans are accustomed to the ‘realistic’ aspect of a three-dimensional model, as we evolve in a 3D world that is very different from two-dimensional representations which are purely intellectual views. Thus, from a pedagogic and scientific standpoint, it was decided to create a 3D reconstruction of al-Qusur using the most recent hypotheses and findings as its basis.

Reconstructing the excavated site using 3D modelling allows the public to see what it could have looked like. However, it is important to note that it is necessary to create a global model for this and, in order to minimize calculation times (i.e. the amount of time the computer needs to spend creating the 3D images), the surroundings cannot be highly detailed. Also, even though the models for these types of reconstructions are based on archaeological facts, many details—such as the quantity and placement of openings, for example—are conjectural.

As well as being useful for the public, the 3D models developed for the MAFKF can be used to support research, in particular to validate working hypotheses. The modeling is primarily constrained by archaeological documentation, comparisons with other sites and the hypotheses used for previous reconstructions. While ethnographic comparisons can be useful, these only add to the range of possibilities, particularly technical ones. As a result, we used our experimental model to help test and validate a few of our hypotheses.

The idea behind this 3D reconstruction is not so much to exactly reproduce a past reality as to represent a verisimilar state of the site. Indeed, in the case of Failaka, the lack of iconographic and archeological evidence means that we are not able to create a perfectly realistic model. The purpose of the model is therefore to represent a state that presents the appearance of realism and gives an idea of what the site and its atmosphere could have been like. So, it is a purely hypothetical recreation of a state that possibly never existed and should not be mistaken for reality. For all these reasons, the render (3D model) cannot be considered to be an exact three-dimensional reconstruction of the buildings.

Previous models

In 2010, Erik Follain, an archeologist and architectural illustrator, created a 3D model of the monumental church A1 (fig. 2). This model was created using the documentation available at the time, which was rather restricted as the church’s remains had been recovered for preservation purposes. It was thus mostly based on the preliminary report written by B. Vincent, O. Callot, and J.-F. Salles (1991). However, the 1991 report was limited in its analysis, as few architectural comparisons with other churches in the area were made during this assessment. It also lacked observations about historical structures in the Gulf region and did not deeply examine the techniques and traditional construction methods which were used there. The 1989 plan, included in the 1991 report, was also incomplete (as most of the outer walls are missing from it) and simplified. E. Follain’s model proposed a suggestion for the church’s interior decorations, based on a stucco cross that had been discovered. At the same time, E. Follain proposed a reconstruction of a dwelling unit, the settlement and the island, although he was also confronted with a lack of accurate data for these models as well. In his proposition for the creation of a more complete model, E. Follain (Follain 2010) described his initial 3D model as crude as it lacked texture, and the materials and lighting were not coherent with the site’s topography. So, it was clear that the initial 3D model was limited and needed to be updated.

Fig. 2. First computer rendering of the monumental church A1 in al-Qusur (E. Follain, 2010).

In 2016, Catherine Raad, an archeologist and graphic designer, created a model of the church (fig. 3) using AutoCAD, a 2D architectural software that can be used to create 3D models from initial 2D plans, and Sketchup, a 3D modeling program which is known for its ease of use but which is rather limited for producing renders (Raad 2017). C. Raad proposed a model which was based on O. Callot’s surveys and all previous excavations and could be updated. However, Raad encountered the same limitations as Follain, that is to say the absence of existing elevations due to the fact that raw earth is a perishable material when used in such a challenging environment characterized by humidity, salinity, and strong precipitation. She also lacked the time to model other structures apart from the church and to render the landscape, making it necessary for our new model to recreate these. Moreover, the view of an architect-restorer about such a building is not the same as that of an archeologist, even though these views are complementary. Thanks to the latest discoveries and especially the progress of 3D rendering over the last few years, we were able to recreate more realistic renders for the new model. This meant it was possible to add complexity to the buildings through more details which could give a more verisimilar appearance to the model.

Fig. 3. Prior render of the monumental church A1 in al-Qusur (C. Raad, 2016).

Limits of the model

Even if the utmost care has been given to modeling historically and structurally realistic buildings, the inherent shortcomings of such an exercise must be kept in mind. For example, only a small portion of the site has been excavated so far meaning that the overall plan of the site is mostly based on surface surveys which have been completed using aerial photography. Even though this plan provides a good overall idea of the settlement, it is not as accurate and detailed as a proper excavation and therefore it is not possible to create a model which is 100% accurate using it as a starting point.

Also, even when excavations have been carried out they only show elevations limited to a maximum of 50 cm (and most of the time even less) as mudbricks are a perishable material, in particular in a saline and humid environment like Failaka Island. They melt easily in the presence of rain and turn back to their previous earthen state. Obviously, as the elevations are limited to 50 cm, all of the facades recreated in the models are extremely hypothetical even though they are based on the observation of local references. For instance, something as simple as the building’s windows leads to multiple unanswered questions such as: ‘How many windows were there?’, ‘What were their dimensions?’, and ‘Where were they located on the wall?’. Also, only a few poorly preserved elements of window roofing were discovered during the excavations and these are difficult to interpret.

The observation of local references is also complicated as there is a scarcity of similar buildings in the region. At present, only a handful of churches have been found in the area. Therefore, the classification of their structure is highly questionable even though this would be intellectually satisfying. On the same theme, even though the use of recent references for structural comparison is sound, this presents a bias as construction techniques might have evolved over the centuries.

The aesthetic bias

Quite often, when modelling a building, we tend to end up creating a perfect and well-finished object due to our own biases. Often this is because we are fascinated by symmetry and straight lines, mainly due to our artistic education and environment, and so we tend to project this attraction onto the buildings we are recreating. This bias makes us forget that at the time these buildings were built tools were not as precise as they are now, and the people who worked on the buildings might have only been average artisans or even low-qualified ones, resulting in miscalculations and asymmetry.

In the case of al-Qusur’s church, this is particularly evident when we compare the archaeological plan with the previous 3D models. As shown by the surveys carried out during the excavations, the southern wall presents several protrusions, while the whole building has sinuous walls and slightly misaligned irregular doorframes. However, these elements were not shown in the previous renders (figs. 2-3), which instead portray straight, orthogonal walls with no protrusions. Even though the thinking behind the creation of these designs is understandable, it can also be seen to have altered the aspect of the building, creating an idealistic image instead of one which strictly follows field-based information.

To follow a scientific approach as closely as possible, all these ‘irregularities’ were kept in the new model, even when their meaning was still a matter of debate. For example, the purpose of the protrusions on the southern wall are not clear (fig. 4). One of the explanations could be that they acted as buttresses to reinforce a weaker wall but another explanation could be that they played an ornamental role, hosted niches or benches, or supported roofs. Unfortunately, the answer cannot be confirmed without further discoveries as the elevations have all disappeared. But, it was decided to treat them as buttresses, giving the building a more realistic form and avoiding the aesthetic bias that would have led to a smooth and perfectly straight wall.

Fig. 4. Plan of the west monumental church, based on O. Callot’s drawings (vectorization by C. Raad). This portrays the church as it was discovered with walls which are not perfectly straight (the protrusions are highlighted in red).

Conclusion

The realization of a 3D render model of al-Qusur during MAFKF’s 2023 campaign was considered to be an important action both for its pedagogic value and its contribution to scientific research. 3D digital reconstructions are a useful tool for understanding a site, its buildings and their interconnectedness. They also provide an opportunity for archaeologists and architects to test structural and architectural hypotheses. However, it is important to take into account their limits, both when creating and when using them. As the MAFKF team were aware of these limitations in the case of al-Qusur, they attempted to mitigate their impact as much as possible.

The next paper (Coppé 2024) describes the methodology which was followed to recreate three buildings within the surrounding settlement, and the results which were obtained.

Bibliography

Bonnéric J., “Al‑Qusur, a Christian Monastery on Failaka Island, Kuwait”, An Archaeological and Historical Guide Book, National Council for Culture, Arts, and Letters of the State of Kuwait, Kuwait, 2016.

Bonnéric J., “Des églises monastiques de pèlerinage ?”, Bulletin d’études orientales, LXVIII | 2023, 95-119. 

Coppé E., “The quest to reconstruct al-Qusur in 3D (2/2): how did we create the model?”, Le carnet de la MAFKF. Recherches archéologiques franco-koweïtiennes de l’île de Faïlaka (Koweït), 14 octobre 2024, à paraître.

De Langhe K., “Early Christianity in Iraq and the Gulf: a view from the architectural remains”, In Actas del V Congreso Internacional de Arqueología del Oriente Próximo Antiguo, 2008, Vol. 1, p. 603–610.

Humbert J.,Cartographie du site d’al‑Qusur : le centre de l’établissement”, Le carnet de la MAFKF. Recherches archéologiques franco-koweïtiennes de l’île de Faïlaka (Koweït), 17 avril 2020. [En ligne] https://mafkf.hypotheses.org/2019

Follain E., Etat du Koweit – Ile de Failaka – Premiers essais pour une restitution en images de synthèse des vestiges archéologiques connus, 2010, unpublished report.

Vincent B., Callot O. & Salles J.-F., “L’église d’al-Qousour Failaka, Etat de Koweït. Rapport préliminaire sur une première campagne de fouille, Arabian Archaeology & Epigraphy 2, 1991, p. 145–181.

Raad C., “Entre enquête scientifique et imagination : produire une image tridimensionnelle de l’église d’al-Qusur (Faïlaka, Koweït)”, Le carnet de la MAFKF. Recherches archéologiques franco-koweïtiennes de l’île de Faïlaka (Koweït), 28 septembre 2017. [En ligne] https://mafkf.hypotheses.org/1680

Etienne Coppé

Architecte du patrimoine à l’Ifpo Beyrouth depuis juillet 2022, Etienne Coppé a travaillé dans divers pays méditerranéens. Il participe aux activités de recherche portant sur l'architecture, l'archéologie et le patrimoine, à travers des travaux de relevés architecturaux, d’analyses architecturales et de restitutions. Il conduit également des travaux de conservation et mise en valeur du patrimoine archéologique.

More Posts


OpenEdition vous propose de citer ce billet de la manière suivante :
Etienne Coppé (10 octobre 2024). The quest to reconstruct al-Qusur in 3D (1/2): aims, previous work and issues. Le carnet de la MAFKF. Consulté le 16 mars 2025 à l’adresse https://doi.org/10.58079/12g3o


Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

This site uses Akismet to reduce spam. Learn how your comment data is processed.