Vila Franca do Campo, São Miguel, Azores. 2015 Fieldwork Short Report.



In September 2015, the Armedea Project completed the first archaeological evaluation in Vila Franca do Campo, island of São Miguel, Azores (Portugal). The fieldwork was carried out by the Department of Archaeology of Durham University and the Centro de História d’Aquém e d’Além-Mar (CHAM), Universidade Nova de Lisboa – Universidade dos Açores, with the participation of the School of Environmental Sciences, University of East Anglia.

Vila Franca, the earliest capital of the Azores, was almost completely destroyed by a seismic-induced landslide on 22 October 1522 (Fructuoso 1981). Between 2,000 and 3,000 people were thought to have died in the event.

From 1967 to 1982, this site was investigated by Manuel de Sousa d’Oliveira, a Portuguese archaeologist who identified some remains of the ancient settlement. Sadly he never published his investigations. After this pioneering research, the Department of Volcanology at the University of the Azores produced a probabilistic model for the 1522 landslide, collecting evidence through coring and geological fieldwork. They estimated that the mass movement had an area of about 4.5km2 and a volume of 6.75x106m3 (Marques et al. 2009). This research, however, did not shed new light on the archaeology of the old settlement and what was known about the archaeology of the old Vila Franca do Campo was simply data extracted from interviews which Oliveira released to the local press (Bento 1989).

The 2015 archaeological excavation. The overall aims of the campaign were: (a) to understand the impact that the 1522 landslide had on the landscape of Vila Franca do Campo; (b) to evaluate the possible remains of the town under the landslide; (c) to investigate the archaeological evidence related to the aftermath of the 1522 disaster; (d) to study the geomorphological characteristics of the landslide; (e) to analyse the material culture yielded by our excavation; (f) to carry out a preliminary assessment of any material recovered during the previous excavations and now stored at the local museum in Vila Franca. 

Our fieldwork has involved the excavation of 17 trenches across the eastern part of the modern Vila Franca do Campo located within 5 different excavation areas (fig. 1), situated in the main zone to be known to be affected by the 1522 landslide (based on the reconstruction of Marques et al. 2009 and the observations by Oliveira in Bento 1989).

Fig. 1. Location of the excavation areas. The red triangle marks the exposed section where the 1522 landslide deposit is visible in the upper part of Vila Franca do Campo.
Fig. 1. Location of the excavation areas. The red triangle marks the exposed section where the 1522 landslide deposit is visible in the upper part of Vila Franca do Campo.

The 1522 landslide deposit was identified in all the 5 excavation areas and, in particular, within 8 of the excavated trenches (fig. 2). It has a mean thickness of around 1.5m and the pre-1522 buried soil lies at a depth of approximately -2.00m from the modern surface. Detailed sediment logging was undertaken to capture textural variations within the local deposits and for comparison between excavation areas located elsewhere in the town (fig. 3).

Fig. 2. Area 1, Trench 4. The l522 landslide deposit with boulders at its base. In the section, the buried soil is clearly visible at the bottom.
Fig. 2. Area 1, Trench 4. The l522 landslide deposit with boulders at its base. In the section, the buried soil is clearly visible at the bottom.
Screen Shot 2015-10-28 at 12.20.33
Fig. 3. Area 1, Trench 4. Clast fabric measurements were made for L2-L5 and L1-L5 were sampled for grain size and composition analysis. The surface of the deposit including L5 shows signs of anthropogenic modification. It is possible that the landslide deposit was thicker in this location. Drawing by M. Froude.

More specifically the logging was aimed at documenting the interaction of the landslide with the archaeological features, the nature of the landslide basal contact, whether vertical differences in grain size and sediment structure arose at each site, and any subsequent modification to the deposit by fluvial action. Observations are targeted at reconstructing the landslide dynamics and impact on the local landscape. There were distinct differences in the deposits logged in trenches in the town centre compared with a vertical exposure of the landslide on the peripheries of the debris fan. In the central areas, the deposit was massive in texture, vertically less heterogeneous with a near-horizontal sharp contact between the landslide and ash deposits from the Furnas C eruption (80 AD ± 120). Observations and clast fabric were diagnostic of deposition by a debris flow. Conversely, the deposit at the peripheries of the debris fan contained scours and structures indicative of fluctuations in the rheological behaviour of landslide to include  pulses with lower sediment concentration. Samples from the deposit are currently being analysed to assess grain size and composition, to ascertain the bulk characteristics of the flow at different locations and consider the physical impact of the landslide on the settlement of Vila Franca.

In one of the trenches  (T5), the excavation has documented the impact of the landslide deposit on the hydrology of the area, identifying the appearance of two new water streams on the new surface created by the mass movement (fig. 4).

Fig. 4. Area 2, Trench 5. The l522 landslide deposit (5015) eroded by later water channels infilled with rhythmic beds of gravels and sands (5005, 5020).
Fig. 4. Area 2, Trench 5. The l522 landslide deposit (5015) eroded by later water channels infilled with rhythmic beds of gravels and sands (5005, 5020).

Evidence of the pre-1522 occupation has been identified in the trenches located just outside the modern town, suggesting that this area was being cultivated on the fringes of the town in the late 15th-early 16th centuries (fig. 5). Here, animal bones and snails associated with fragments of bricks were found on the top of the buried pre-1522 surface. Archaeological contexts related to the pre-1522 settlement were also identified in one of the trenches (T17), located within the modern town.

Fig. 5. Area 2, Trench 7. The pre-l522 surface sealed by the landslide. Charcoal and occasional animal bones were found here, suggesting that this area was probably cultivated on the fringes of the town.

The same trench also confirmed that the reoccupation of those areas affected by the landslide was fast (fig. 6); the trench contained a metalworking furnace which produced fuel ash slags and a rubbish pit (fig. 7). This pit produced 44 sherds of pottery (1.3kg), including 16th century imported wares, such as a Tuscan maiolica sherd (Montelupo-type; fig. 8a), two Seville green-glazed large bowls (or ‘lebrillos’) and two other possible Seville Morisco Wares (see 

Fig. 6. Area 5, Trench 17, Section looking W. The arrow indicates the residual 1522 landslide deposit, cut by the rubbish pit 17015 on the left.
Fig. 6. Area 5, Trench 17, Section looking W. The arrow indicates the residual 1522 landslide deposit, cut by the rubbish pit 17015 on the left.
Fig. 7. Area 5, Trench 17. The rubbish pit 17015 under excavation. Note the small furnace 17012 on the left.

Finds. A total of 1,817 items were recovered during the 2015 excavations. They were washed, quantified and then deposited at the local Museum in Vila Franca. Most of the material is ceramics: pottery vessels (741 sherds, weighing 8.7kg) and also building materials, especially roof tile (502 sherds, 24.3kg); but animal bones (464 fragments), including a complete skeleton of a cat, metal items (55, mainly iron nails), shell, glass, mortar, slag and stone were also found.

Museum finds. Thanks to Dr Ricardo Rodrigues we were able to assess the material recovered by Sousa d’Oliveira in the 1960s-80s, which is now in store at the Municipal Museum of Vila Franca do Campo. A preliminary inspection of this material confirms that d’Oliveira must have reached the earliest levels of occupation of Vila Franca, given the presence of pottery clearly dated to the 15th century. At least two examples of Spanish lustrewares made in the Valencia region date to this century and they are both decorated with the briony pattern (fig. 8b, 8c). Further imports of the late 15th-early 16th century are the Seville wares, including ‘cuerda seca’ and blue-and-purple dishes. A further sherd from a Montelupo-type maiolica of the early 16th century was also found. Most of the material, however, are  redwares both of local production and also coming from mainland Portugal. The collection is an important one and its study would shed new light into Vila Franca’s history. Also human and animal skeletal remains were identified during this preliminary survey. 

Pottery composite
Fig. 8. A: Early 16th century Italian maiolica from Tuscany, from Trench 17 (context 17015); B, C: Valencian lustreware of the 15th century from d’Oliveira excavations

Future work. This preliminary research allowed us to open an exceptional window on the effects of the 1522 seismic event, but also on the reaction of the local population of the ancient Vila Franca. The ‘new’ Vila Franca do Campo was established along the western side of the landslide (for instance, the convent of São Francisco was completed immediately after the earthquake, in 1525), showing an extraordinary example of community resilience within a context of repeat destructive earthquakes; another destructive seismic event occurred shortly afterwards in 1536.

Future work to continue and develop this first season of archaeology will be planned in collaboration with the Municipality of Vila Franca in order to extend and enrich the small window already opened into the past through this pilot project. In addition, further studies on the material culture of Vila Franca will be jointly undertaken by the Durham University and the CHAM. Specific laboratory analyses are already under way in the Department of Archaeology of Durham, and they are mainly focused on grain size fraction analysis of the landslide deposit, OSL dating and phosphate analysis of the pre-1522 buried soil. 

Paolo Forlin1, Chris Gerrard1, Alejandra Gutiérrez1, André Teixeira2, José Bettencourt2, Melanie Freude3, Dave Petley3

1Department of Archaeology, Durham University (UK)

2 CHAM, Universidade Nova de Lisboa and Universidade dos Açores (Portugal)

3 School of Environmental Sciences, University of East Anglia (UK)


Bento, C. M. 1989, Escavações Arqueológicas em Vila Franca do Campo 1967-1982, São Miguel, Associação Arqueológica do Arquipélago dos Açores

Frutuoso, G. (1522-1591†). Livro Quarto das Saudades da Terra, in Saudades da Terra, G. Frutuoso (Ed.), 2a ed., Ponta Delgada, Instituto Cultural de Ponta Delgada, II, 1981

Marques R., Zêzere J.L., Gaspar J.L., Amaral P. 2009, Reconstituição e modelação probabilística da escoada detrítica de Vila Franca do Campo desencadeada pelo sismo de 22 de Outubro de 1522 (S. Miguel, Açores), Publicações da Associação Portuguesa de Geomorfólogos, Volume VI, APGEOM, Braga, 2009, p. 175-182

Acknowledgements: We would like to thank first everybody who took part to this fieldwork, in particular Peter Brown and Ed Treasure (Durham), Elena Fiorin (Barcelona), Cristóvão Fonseca, Thiago Silva and Cristelle Chouzenoux (CHAM). We are very grateful to everybody who supported our research in Vila Franca last September. In particular, we would like to mention the kind support of Dr Ricardo Rodrigues, mayor of Vila Franca, whose help and encouragement were instrumental in carrying the work to good fruition; N’Zinga Oliveira, member of CHAM; Prof. Margarida Machado, Universidade dos Açores; Mario and Lourdes Martins, who allowed us to excavate deep holes in the most beautiful garden in the island of São Miguel.  Without their help and that from many others, this research would not have been possible.

The VFDC 2015 fieldwork was supported by a Marie Curie Intra European Fellowship within the 7th European Community Framework Programme.

Screen Shot 2015-10-28 at 12.20.20

Leave a Reply

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

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

Google photo

You are commenting using your Google 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 )

Connecting to %s