Integrated Archaeological Prospection Studies in Mexico: A Review

Abstract

This chapter provides a retrospective of the work carried out by the Archaeological Prospection Laboratory, of the Institute of Anthropological Research, National Autonomous University of Mexico (IIA, UNAM). In 1983, geophysical surveys, combining different methods, were performed for the first time to study the San José Ixtapa archaeological site. Since, multi-technique geophysical surveys, integrating a wide range of complementary methods such as satellite remote sensing and soil analyses, have been performed at other sites in Mexico. To illustrate the development of the work carried out for this institution, the chapter syntheses the main achievement of some key case studies.

1 Introduction

The term Archaeological Prospection describes the employment of several techniques to discover buried sites or gather information on the location and extent of know ones. Barba (1984) has posited that to recover archaeological information in a minimally invasive manner, we must combine different methods in a sequence and apply each of them in an appropriate fashion to ensure both efficiency and success. This sequence aims to acquire the most complete range of information from an archaeological site by studying the context’s chemical and physical properties and assisting archaeologists in deciding the most suitable excavation strategy while saving time and money (Barba, 1990).

Following the experiences of other laboratories, but specially the integrated proposal of the Lerici Foundation in Italy, the IIA-UNAM’s Archaeological Prospection Laboratory was founded in 1983 to provide with new methods for archaeological research in Mexico. The initial methodological base has been further developed including new technology such as satellite imaging digital analysis, ground penetrating radar survey and spectrometric analysis of soil samples (Fig. 1).

Fig. 1

Methodology applied by the IIA-UNAM’s Archaeological Prospection Laboratory to study archaeological sites. (Barba, 1984, 1990; figure modified by Blancas, 2012)

2 Case Studies

Table 1 summarises the methods used at the five case studies: San José Ixtapa, Oztoyahualco (Teotihuacán), Santa Cruz Atizapán, and Tlajinga (Teotihuacán). These sites have been selected to provide an idea of the technique’s evolution over time, to emphasise the role of the lime as building material and chemical indicator, and the importance of encouraging the integration of methods.

Table 1 Summary of the techniques and data collection parameters used in the San José Ixtapa, Oztoyahualco, and Santa Cruz Atizapán sites, and the most important characteristics recorded with each technique

2.1 San José Ixtapa

In 1983, the first study integrating geophysical methods and chemical analysis of soil samples at an archaeological site in Mexico was carried out at San José Ixtapa. This is a Postclassic (AD 900–1521) site in the Temascalcingo Valley, Mexico State. It was first located by aerial photography due to the presence of clear soil spots and later verified by finding ceramic sherds covered by a mud and grass mixture (Limón, 1978; Limón & Barba, 1981). The evidence recovered from surface suggested the use of cinnabar and quicklime to produce mercury in ceramic pots sealed with a mud mixture. The motivation of the study was to test the methodological proposal and provide more lines of evidence to support the apparent production activities at the site (Barba, 1984).

Firstly, the entire terrain was explored to observe surface materials distribution. Afterwards, the topographic mapping of the site was undertaken by putting three grids over the areas of maximum material concentration (A, B and C). On top of them, all the surface materials were registered. To interpret results, environmental data, magnetometer surveys, surface artefacts distribution and chemical analysis of topsoil samples were taken into consideration. Simple, quick, reliable, low-cost chemical tests (carbonate and phosphate tests and pH determination) were performed in the lab.

Magnetometer survey, particularly from Grid B, indicated the presence of structures, stone retaining walls, and heating areas. Near these heating zones, the distribution of surface material showed the concentration of clay-covered potsherds. The combined results suggested that at this site liquid mercury was produced by heating cinnabar using quicklime inside small ovens made of two conjoined vessels sealed with mud mixed with straw (Barba & Herrera, 1986). The results of the magnetometer survey suggested hotspots where potential ovens could be located. The concentrations of mud-covered pottery fragments around Structure 2 suggested that in that area the ovens’ lids were broken (Ibid: 101).

The work at San Jose Ixtapa not only showed for the first time the benefits of studying archaeological sites with a minimally invasive methodology, but also the usefulness of integrating the ethnographic record to illustrate such production activities (Fig. 2). For example, until recent times, gambusinos (rural miners) of nearby communities produced mercury using this kind of ceramic pots (Barba & Herrera, 1986).

Fig. 2

Map of artefacts distribution and anomalies detected after the magnetometer survey in Grid B at San José Ixtapa. (Modified after Barba, 1984)

2.2 Oztoyahualco, Teotihuacán

In 1985, the Antigua Ciudad de Teotihuacán Project (PACT) appointed a minimally invasive survey to study the Oztoyahualco site. The site contains a Three Temples Courtyard and a residential area excavated in 1986–1988 (Manzanilla, 1993: 23; Barba, 1993: 47; Barba & Ortiz, 1993: 545). The objective was to identify possible archaeological features of interest as well as to explore the nature of clear soil marks identified in aerial photographs. The survey area (1 ha) was divided into the five sub-areas shown in Fig. 3.

Fig. 3

Aerial photography of the studied area from a helicopter. Observe the clear soil spots produced by calcium carbonate concentration during the dry season. (Manzanilla, 1993)

A topographic survey (readings every 4 m) revealed a low mound at Area B and another at Area E. During a walkover survey, abundant building materials (e.g. volcanic scoria, stucco, and flagstones) were recorded at the centre-south of Areas A and B, west of Area C, and throughout the entire Area E. There were also clusters of small tezontle (volcanic escoriaceous stone) fragments impregnated in carbonates, especially in the north-eastern part of Module B, in central C Module, in the Centre-South sector of D, and the entire eastern band of E (Ibid: 50). The magnetometer surveys conducted at these areas detected magnetic anomalies of high intensity possibly related to subsurface hearths and burnt structures. The high resistance values of the earth resistance survey using a mobile array of electrodes (Wenner-Alpha configuration) also detected the presence of structural remains (Ibid: 57). The topographic data helped to define the total extent of visible mounds (suggesting subsurface structures), with good correspondence with the results of the earth resistance survey.

At the areas with clear soil marks, targeted manual augering was carried out to obtain information about the depth of these deposits. A maximum depth of 0.8 m was reached. Soil samples were collected every 10 cm for chemical analysis. Soil samples (of the top 30 cm) were also taken at modules C, D and E at 4-m intervals for chemical analysis (Barba et al., 1991). A direct correlation was seen between the carbonate concentration values and the location of the clear soil marks (Fig. 3).

Overall, the combined interpretation of all data suggested that north-eastern of the Three Temples Courtyard and north-western of excavated residential complex lay another residential unit (HU2) and other structures (S1–S8) (Fig. 3). The presence of high values of phosphate concentration in Module E indicated an accumulation of garbage and other organic waste close to the Oztoyahualco residential unit. The collapse of structures points to an accumulation of construction debris. Among these materials, the disintegration of lime plaster may have caused the high carbonate concentration values in the soil, hence the clear colour of the soil marks. Considering the distribution of the surface findings, the high concentration of carbonates, phosphates and the geophysical data, it was evident the potential presence of a rectangular structure in Module E and was suggested as a promising location to target with an excavation. This was later excavated, appearing a residential complex with several rooms, patios, corridors with several levels of floors and walls, as can be seen in Fig. 4 in the red outline and in the aerial photography.

Fig. 4

Results of the different analyses carried out in Area D and E. The excavated area of the residential area of Oztoyahualco is outlined in red. (Figure modified from Barba & Ortiz, 1993)

2.3 Santa Cruz Atizapán

The Santa Cruz Atizapán site (AD 550–900) is located on the eastern bank of the Chignahuapan bog, in the Toluca valley, Mexico State, spanning approximately 1 km², encompassing 100 archaeological mounds. To build them, materials, such as sediments, wood, and volcanic stones were collected from surroundings. Many of these mounds were found near the shoreline of the bog, but the main structures were found on hillsides, ~10 m above the water surface (Fig. 5).

Fig. 5

Balloon aerial photography mosaic, overlapping digital terrain model in Santa Cruz Atizapán. (Blancas et al., 2017a)

The surveys were conducted during several field seasons. The magnetometer survey revealed the presence of volcanic stones used to build the mounds, which had domestic and ceremonial functions (Blancas et al., 2017b). It also revealed a circular building of 10 m of diameter buried alongside a bell-shaped mound within the settlement’s core and a straight stone paved road over 300 m long, and several small stone structures in a big artificial island (Fig. 6a) (Ibid: 44–139). GPR was used to determine the internal structure of the mounds, but due to high water saturation of sediments it suffered penetration problems.

Fig. 6

(a) Results of the magnetometer survey at Santa Cruz Atizapan showing three areas with important archaeological remains: small section of a large circular structure, straight stone-paved road, and an isle with a complex of archaeological mounds. (b) Distributions of protein residues, carbohydrates, carbonates, and phosphates concentrations expressed on arbitrary scales (Barba, 2007). Their enrichment correlates with the larger mounds and suggests that more intense activities were performed in these areas

Soil samples were taken every 100 m in the centre of the agricultural fields and selected mounds. Ten centimetres cores were also drilled to register the main stratigraphic changes caused by the building of the mounds (Terreros et al., 2017). In addition, semi-quantitative chemical tests were applied to 269 soil samples based on the Laboratory’s analysis protocol (Barba et al., 1991, 2009b). This study made it possible to obtain distribution maps of chemical residues and to relate: proteins to fish and insect consumption areas; carbohydrates to accumulation of decomposing cellulosic material around the mounds; carbonates to lime; and phosphate residues to areas inhabited by human groups. Figure 6b shows that the distributions of these residues are clearly associated with the area occupied by the pre-Hispanic mounds, especially in the area called “The Isle”, and in the areas related to waste and food consumption activities.

2.4 Tlajinga, Teotihuacán

Tlajinga is in the southern part of the large Teotihuacan complex, north-east of Mexico City, following the Avenue of the Dead, the city’s main pilgrimage route and its N-S axis, 1.5 km from the Citadel. Spanning roughly 1 km², it is set on a gentle slope of vertisol soil farmland. It receives water from the Patlachique range that spill to the north towards the San Lorenzo River (Fig. 7).

Fig. 7

Upper left, area of study location. Upper right, RGB satellite composite colour image of principal components (PC1, PC2, PC3), the shallow material of archaeological structures shows in orange colour, vigorous vegetation is related with green colour, and zones with greater moisture in blue colour. The soil samples (T4, T5, T6…) were analysed with FRX, DRX, micromorphology and FTIR. Below, magnetometer survey map of main archaeological structures (3:S4W1, 4:S4W1, 7:S3W1 …)

Arising from the work of Nichols (1988), Storey (1992) and Widmer and Storey (1993, 2012) in the Tlajinga quarter, an interdisciplinary research project was undertaken combining remote sensing, geophysical survey, soil physicochemical analysis, and excavations (Blancas et al., 2019).

Multi-spectral satellite imagery was helpful in identifying the locations of archaeological structures owing to the way their degraded building materials had combined with the surrounding soils and sediments. In the processing of this imagery, which included atmospheric correction by Chavez’s method (1996), the multispectral and panchromatic images were merged, giving a 41 cm’s spatial resolution, but with the same chromatic contrast of the multispectral images (Blancas, 2012). Using principal components analysis, anomalous patterns appearing with greater brightness and contrast were detected, which in some areas correlated with structures reported by the Teotihuacan Mapping Project (Millon, 1973) (Fig. 7). The distinctive reflectance was due to calcium carbonate from partially destroyed and buried structures that had mixed with the original soil (San Pablo Black Paleosoil (SPBP)) (Sánchez et al., 2013; Solleiro-Rebolledo et al., 2011). This study was extended by spectroradiometrically to determine the spectral signature of soil samples taken across the site. Those with high reflectance due to the presence of stucco or lime-based plaster were associated with collapsed archaeological structures.

The subsequent magnetometer surveys covered a total of ~6 ha and revealed landfills, heated surfaces, wall foundations, and other structures built mainly of volcanic stone. At the eastern part of the map (Fig. 7, right) there was an exceptionally large hill with a linear arrangement of magnetic dipoles related to the volcanic stones in walls. Other disordered dipoles were associated with debris. The magnetometer survey results from the 34: S3W1 complex showed a large platform and a other structural remains to the south (Fig. 7, bottom).

Besides, we collected reference materials to compare the geophysical results with the stuccos pf the Teotihuacan structures, caliches (natural carbonates) and soils, such as San Pablo Black Paleosoil (SPBP) upon which the Teotihuacan culture was developed (Fig. 8b).

Fig. 8

Reflectance and micromorphology results. (a) Spectroradiometric measurements. Micromorphology in crossed polars of (b) reference and (c) archaeological samples

Petrographically, the soil samples were taken from the central part of Tlajinga (Fig. 8, upper right) contain assorted materials, of volcanic origin and related to cultural activities. Cultural calcium carbonates with well classified volcanic glass sherds were identified. Those were added for the preparation of flooring, and walls are mainly made of volcanic materials, such as ferromagnesium, plagioclases, or volcanic glass in different amounts. Stucco and caliche are chemically similar since both have calcium carbonate (Fig. 8c). However, caliche contains eolithic calcium carbonate as well as volcanic clays and minerals, while stucco has as aggregate sorted volcanic glass shards and calcite micritic crystallisation (Barca et al., 2013; Barba et al. 2009a, b; Guillén, 2018).

X-ray fluorescence (XRF) for caliche and stucco revealed an elementary chemical composition with a larger calcium concentration in comparison to soil samples (T4, T5, T8, T14 and T15). Conversely, soil samples had larger quantities of elements, such as Si, K, Ti, Mn, and Fe, all volcanoclastic origin. The elementary chemical content of the San Pablo Black Paleosoil (SPBP) layer and selected soil samples was basically the same. This is explained by the volcanic origin of these soils (Fig. 9a). The chemical concentration of the caliche and the stucco had remarkable differences in Ti, Mn, and Fe if compared to the soil samples and the SPBP.

Fig. 9

Results of (a) XRF, (b) XRD and (c) FTIR analyses on soil and reference samples

Volcanic minerals found in the soil samples by micromorphology and X-ray diffraction (XRD), such as ferromagnesians, plagioclases and even glass, are in different proportions in the aggregates used for the preparation of the bases in Teotihuacan floors and walls. A striking difference between stucco and caliche, both calcium carbonates, is that caliche contains not only calcium carbonate, but also clays and minerals of volcanic origin. In contrast, stucco has sorted volcanic glass sherds as aggregates in the mixture (Barba et al., 2009a, b; Guillén, 2018). According to the results of this study, the calcium carbonate particles, mixed with the upper layer of the soil in this region of Tlajinga, modified and enriched the mineralogical content of the soils, and increased the reflectance of the upper soil horizon. Due to the intrinsic reflective properties of this mineral, mainly in the visible and infrared band, many of these anomalies come from the destruction of archaeological structures from the Teotihuacan period.

Finally, Fourier-transform infrared spectroscopy (FTIR) was applied to the stucco and caliche samples to determine whether the calcium carbonate was of geological or cultural origin (i.e. if the calcium carbonate crystal formation was related to an anthropogenic heating processes) (Chu et al., 2008). Looking at the three main infrared absorption peaks in the FTIR spectra in Fig. 9c, the caliche reference sample had a ν2/ν4 ratio of 3.01 indicating a natural geogenic origin. The stucco had a ratio of 5.86 that revealed a great atomic disorder and a larger probability of being of cultural origin. Soil T5, classified as of high reflectance, had a ν2/ν4 ratio of 5.72, consistent with a probable cultural origin. Soil T15 has a ν2/ν4 ratio of 4.23, which is intermediate and an indicator of a mixture of calcium carbonates both cultural and geogenic. The medium reflectance samples, T4 and T8, had more geogenic calcium carbonates, as most of the caliche could have been brought over from nearby natural sources. The T14 low reflectance sample was micromorphologically like the SPBP layer, with a ν2/ν4 value of 3.11, suggesting a clear geogenic origin.

3 Conclusions

This chapter has briefly reviewed the 40-year progress of the Archaeological Prospection Laboratory, of the National Autonomous University of Mexico’s Institute of Anthropological Research. The early experiences and observations have over time been enriched with new equipment and the integration of a wide variety of technologies that provide a more robust information for the interpretation of buried sites. The soil, as packing material of archaeological remains, preserves a great deal of information that deserves our attention. In the examples presented here the calcium carbonate particles are a remarkable cultural material that plays a key role in understanding the buried sites studied with satellite imagery and land-based techniques. This is important since the presented case studies are in volcanic environments, then most of the carbonates detected are from anthropic origin.

San Jose Ixtapa was the first case study to test the proposal of prospection techniques integration. For the first time, it was established the direct relationship between clear spots in terrain and high values in carbonates, in this case, as a by-product of past activities. In our second experience in Oztoyahualco, Teotihuacan, the clear spots in terrain revealed the existence of archaeological structures once covered by lime plasters and was confirmed by high topographic and earth resistance survey values. In Santa Cruz Atizapan, carbonates were almost absent but the distribution of some other chemical indicators revealed the areas of more intense occupation, while magnetic survey was successful to detect the presence of the volcanic stones used as building material and at the end, we had the active participation of earth resistivity and ground penetrating radar as verification techniques.

At Tlajinga, which is the most recent and successful application of our study methodology, the results show that calcium carbonate mixed with the upper layer of the soil modifies the soil’s mineralogical composition, increasing reflectance due to the properties of this mineral, which is visible in infrared light. Therefore, many of the anomalies in the processed satellite images come from the destruction of the Teotihuacan archaeological structures covered by lime plaster (Fig. 8). By gathering and analysing spectroradiometric, petrographic, chemical, and mineralogical data, we can assess the relationship between calcite in the soil’s upper horizon and reflectance in satellite images to establish their correspondence. The location and characteristics of the destroyed archaeological structures were verified indirectly using geophysical survey with magnetometer survey, earth resistance survey, and ground penetrating radar. Later, archaeological excavations have been conducted in certain selected zones to verify results.

The results of Tlajinga study are applicable to all of Teotihuacan’s spaces, since they shared the same building techniques, and were built over the same type of soil. Thus, most of the clear spots seen on the surface should be attributed to collapsed archaeological structures, especially spots on low hills. The origin of this observation in Teotihuacan was at Oztoyahualco in 1985, when for the first time these clear spots were seen on the ground surface and confirmed by chemical spot test. It can be concluded that the calcium carbonate particles in the soil are an inorganic compound that can be observed by remote sensors, verified by chemical tests as well as seen through a microscope, although thus far it has not been detected by geophysical measurements.

We have emphasised the role of carbonate particles because lime was an important material in Mesoamerica and on the other hand, it is quite common in literature to have reports of geophysical studies but is rather unusual to combine archaeo-geophysics and chemical data. Among more than 300 projects we have at present participated, we selected those cases where most of the techniques included in our methodological proposal were successfully applied and provided part of the information that at the end, produced a more integrated and better interpretation of the studied site.

In Mexico, there is just one Archaeological Prospection Laboratory (celebrating its 40 anniversary) full time devoted to applying geophysical techniques to the study of archaeological and cultural heritage sites. In addition, there are a couple of groups in research laboratories, that sometimes use geophysical techniques in cultural heritage projects. Occasionally, some teams from abroad come to Mexico to participate in some specific projects. There is just one commercial company in Mexico that provides this kind of services, the rest of the surveys have been performed by research laboratories. As a consequence, the great majority of the geophysical studies have been part of research work but in recent times some cases have involved federal government institutions and large infrastructure projects. By far, the most popular technique is magnetometer survey, but GPR is becoming more common every day. Less popular is the earth resistance survey.

Most of the studies have been carried on detecting and studying buried archaeological remains, but in recent times applications increasingly involve cultural heritage buildings diagnostic, sometimes to detect pre-Hispanic remains below colonial or modern buildings, some others, to define materials and construction techniques before historic buildings interventions.

For the future of archaeo-geophysics in Mexico, it will be necessary to increase funds to acquire geophysical equipment, as well as to promote the training of young geophysicist who want to be involved in cultural heritage studies.