Monument Future

209

DECAY OF PRE-COLUMBIAN PHALLIC STONE MONOLITHS

Christopher Pötzl¹, Ali D. Öcal², Thomas Cramer³, Siegfried Siegesmund¹

IN: SIEGESMUND, S. & MIDDENDORF, B. (EDS.): MONUMENT FUTURE: DECAY AND CONSERVATION OF STONE.

– PROCEEDINGS OF THE 14TH INTERNATIONAL CONGRESS ON THE DETERIORATION AND CONSERVATION OF STONE –

VOLUME I AND VOLUME II. MITTELDEUTSCHER VERLAG 2020.

1 Geoscience Centre of the Georg August University Göttingen, Germany

2 Department of Anthropology of the National University of Colombia, Bogota, Colombia

3 Department of Geosciences of the National University of Colombia, Bogota, Colombia

Abstract

The pre-Columbian astronomical and meteorological observatory of El Infiernito or ‘Little Hell’ is situated within the Monquirá Archaeological Park, near the colonial town of Villa de Leyva in the Colombian Boyacá-Department. Several burial mounds and a variety of phallic-shaped monoliths, made of local sandstone, bear witness to the culture and unique masonry of the ancient Muisca culture. The El Infiernito ranks among the most important historical and cultural monuments in Colombia. The monoliths were erected nearly two millennia ago and show severe damages today that developed over the time since its construction. Depending on their exposition, various weathering phenomena like differential back-weathering, scaling, cracking and discoloration can be observed. In order to identify the main damage causes and understand their relation with the stone properties, in-situ investigations, in the form of lithology and damage mapping, as well as laboratory work regarding petrography, petrophysical properties and weathering behavior of four sandstones were done. The research demonstrates the differential weathering behavior of the analyzed sandstones and identifies thermal and thermohygric expansion as the main weathering agents.

Keywords: Colombia, El Infiernito, monoliths, sandstone decay, damage assessment, weathering processes

El Infiernito: Historical Background

The stone monoliths at El Infiernito (Fig. 1) have been the subject of much attention ever since the earliest Spanish missionaries, because of their claimed associations with controversial Muisca rituals and alleged riotous ceremonies (Simon 1981). The arrangement of 26 finely carved and equidistant cylindrical pillars in two parallel rows led Silva Celis (1981) to the conclusion that the archaeological 210site had been a sacred field of astronomical and meteorological observation. The natives had measured the movements of the sun, and thus, solstices and equinoxes were calculated and periods of rain and drought were established, which regulated the production activities of the natives (Silva Celis 1981). A dolmen tomb found in 2006 together with 54 phallic-formed sandstone monoliths and indications of feasting events, suggest that El Infiernito was a central place where inter-community activities were carried out.

Figure 1: General view of the archaeological park.

In this study, preliminary results of an investigation on the main characteristics of the deterioration processes that affect the monoliths of the Monquirá archaeological park El Infiernito are presented. This work will serve as a base for later conservation and restoration activities.

Decay Phenomena

The exposed monoliths present significant problems due to the environmental conditions (Öcal et al. 2009) and show diverse, partly exposure-specific forms of weathering (Fig. 2). Approximately 200 monoliths and stone columns were used to visually assess and characterize the weathering behavior.

Figure 2: Decay phenomena on the monoliths. a) biological colonization and sanding, b) crust formation and salt efflorescence, c) alveolization, d) scaling.

The tropical climate at the ~2000 m a. s. l. El Infiernito area is characterized by two rainy and two dry seasons. The rainiest months are May and October with up to 200 mm, whereas January is the driest month with 40 mm followed by July with 90 mm. In addition to the rain, in the mostly semiarid microclimate, air humidity plays an important role as a water supplier. The relative air humidity fluctuates between 70–85 % depending on the season. Direct water absorption during rain events or capillary suction of surface and soil water, control the humidity regime. The monthly average temperatures are above 20 °C throughout the year, with lowest mean temperatures of 10 °C and highest mean temperatures of 22 °C and can rise to over 30 °C. Especially during the long dry and cloudless seasons, the temperatures on the monoliths’ surface can reach high values.

The monoliths show a large variety of different damage types, such as relief, back-weathering, bursting, rounding, covings and alveolization, which predominantly occurs. Neighboring areas of the stone surface weather back to different degrees, and create a moving relief, e. g. by weathering parallel to the layer or by the loss of components. Often honeycomb-like structures are formed, which are typical for alveolar weathering (Fig. 2c). The causes can be of a very different nature. These can include salt weathering and wind erosion, as well as structural and lithological peculiarities.

The deposition of dirt and dust particles in the pore space leads to a gradual compaction of the rock surface, and may favor biological growth on some monolith surfaces. Gray to brownish crusts (Fig. 2b) can be found on the stone surface as coherent layers, millimeter to centimeters thick. Visually, the limit between the natural patina and the damage pattern crust cannot always be clearly determined. Biological colonization by green algae, lichens and higher plants was observed on all monoliths (Fig. 2a). Loosely adhering whitish salt efflorescence is an indication of the build-up of harmful salts enriched in the stone (Fig. 2b).

Sanding, flaking, peeling (Fig. 2a, d), crumbling and exfoliation as well as transitional forms are evident. In addition to the relief formation, sanding 211is the second most common form of weathering with clearly varying intensities. Like sanding, scaling and flaking from plate-like areas parallel to the surface can lead to total loss. These phenomena are strongly connected to salt crystallization and expansional behavior, as may be fostered by strong sunshine during the day and followed by cold nights. Stress on the rocks’ fabric leads to the formation of cracks. However, the most noticeable damages to the stone surfaces are large, surface-parallel scales. Sometimes, nearly catastrophic damage events may occur on the exposed monoliths, like on Monolith V-0163, which was mapped in 2007 (Fig. 3 and 4), and was considerably damaged with a huge loss of material 12 years later (Fig. 3). In this example, material inhomogeneities like nearly invisible mica- and clay-layers inside the rock may have been an important primary damage factor. Scaling visibly occurs in connection with increased salt pollution, especially in the splash water area. Exposure-related scale formations are clearly recognizable. While on some monoliths only discoloration is observed, others show heavy break outs and material loss (Fig. 3).

Figure 3: Monolith V-0163. Critical condition and strong deterioration between 2007 and 2019.

Figure 4: Damage mapping of Monolith V-0163 in 2007.

Rock material and its characterization

As possible source rocks for the monoliths, four comparable sandstones (S1–S4) were taken in the immediate vicinity of the archaeological park. They likely correspond to four sedimentary geological formations consisting of the Arcabuco, Ritoque, Paja, and Churuvita, which are Late Jurassic to Cretaceous in age. The formations are described in more detail by Etayo-Serna (1968), Patarroyo (2008) and Renzoni (1983). Field observations point towards a higher utilization of S2 and S3 as possible source rocks for the monoliths.

S1 is a very fine-grained, gray sandstone, which appears heterogenous due to centimeter long white bands and dark gray to black colored lithic fragments (Fig. 5a). It shows sublitharenitic composition and a matrix (with argillaceous and organic components) supported fabric (Fig. 6a). Monocrystalline, angular quartz grains with various degrees of undulose extinction make > 75 % of the rock. 212Feldspar is occasionally visible (< 5 %). Quartz and feldspar show average grain sizes of < 0.2 mm. Mostly rounded, sometimes bent and elongated, chert fragments are greater in size and make up about 15 % of the rock. About 2 % phyllosilicates are found, preferably in the white bands of the sandstone. From the petrographical point of view, S1 could belong to the Ritoque Formation, which crops out near Villa de Leyva.

S2 is a whitish and reddish mottled fine-grained quartz arenite (Fig. 5b). Like in S3 and S4, the homogenous fabric is grain supported (bound by silica cement). Angular quartz grains (> 95 %) are partly polycrystalline and between 0.1–0.01 mm in size. About 30 % of the quartz grains are coated by iron oxides (Fig. 6b). Less than 5 % lithic components of very fine grained quartz and argillaceous material, as well as minor amounts (< 1 %) of mica can be found. A quartz arenite layer of the normally silty to pelitic Paja Formation, which actually crops out at the park, could be the possible source of this sandstone.

S3 shows macroscopically and microscopically strong similarities with S2. It has a whitish and reddish speckled appearance (like S2), but additionally contains white and red bands (similar to S1), which depicts a layering (Fig. 5c). The amount of very fine grained lithic fragments is higher than in S2 (> 5 %), the amount of iron oxides appears to be lower (Fig. 6c). Due to the strong similarities to S2, it can be assumed that S3 also originates from the Paja Formation.

S4 is a very fine-grained and very light colored, whitish-grey (with tiny yellowish spots) sandstone of sublitharenitic to quartz arenitic composition (Fig. 5d). The homogenous fabric is grain supported and very porous (Fig. 6d). The sandstone consists of > 90 % subrounded to subangular, polycrystalline quartz, with minor undulose extinction. The average grain size is about 0.01 mm. Less than 5 % fine grained lithic fragments (consisting of almost exclusively quartz), 2 % of opaques and < 1 % of mica can be found. A possible origin could be a very fine grained subgroup of the Arcabuco Formation, cropping out near Villa de Leyva. Although the Churuvita Formation crops out further to the southeast of the park, it cannot be ruled out as a possible source material.

Figure 5: Macroscopic photographs of a) S1, b) S2, c) S3 and d) S4 (size of the photos: 4.5 cm × 4.5 cm).

Figure 6: Thin section photomicrographs of a) S1, b) S2, c) S3, d) S4 in plane-polarized light and 2.5x objective. The blue resin visualizes the differences in pore space.

Petrophysical properties and weathering behavior

The petrophysical properties and weathering behavior of the sandstones were determined in accordance to the German industrial norms parallel (X-direction) and perpendicular (Z-direction) to the bedding. For further information see Siegesmund and Dürrast (2011). Table 1 gives an overview of the laboratory results.

In general, the sandstones can be divided into 213group one (S1, S3) and group two (S2, S4) of similar petrophysical properties and weathering behavior. Group two is characterized by higher porosities (> 20 vol%) and bulk densities around 2.0 g/cm³. Group two also shows unimodal pore radii distributions with large ratios of capillary pores (> 93 %) and mean pore radii between 0.9 and 5.7 µm. Group one (S1, S3) on the other hand, shows mean pore radii of 0.3–0.5 µm and contains considerable amounts of micropores (< 48 %). Up to nine times higher capillary water absorption (< 9 kg/m2√h) and lower resistance against water vapor diffusion in group two indicate better pore connection than in the sandstones S1 and S3 of group one (Tab. 1). The saturation coefficient S, as an indication of the materials’ frost resistance (Hirschwald 1912), classifies S1, S3 and S4 as frost resistant and S2 as uncertain.

Table 1: Petrophysical properties and weathering behavior of the four analyzed sandstones.

Regarding the moisture expansion, both groups show distinct differences when subjected to water (Tab. 1). While group two (S2, S4) shows barely any expansion (< 0.06 mm/m), the sandstones of group one (S1, S3) expand up to 0.2 mm/m. However, the hydric expansion values of the investigated sandstones can be generally considered low when compared to other sandstones (Siegesmund and Dürrast 2011).

The response to temperature changes is high in all investigated sandstone samples. Under dry conditions the sandstones expand about 0.8 mm/m. This expansion is increased up to 1.1 mm/m (S1, S3) when heating under wet conditions (Tab. 1). The average thermal expansion coefficients α range between 11 and 13 × 10^–6 K^–1 and can be considered very high. These high values, however, are not surprising, since quartz characteristically possesses a high thermal expansion coefficient and the investigated sandstones show high compositional maturity. The residual strains are considerably low under dry conditions and do not exceed 0.2mm/m under wet conditions.

The strength of the rock material is proportional to their dynamic modulus of elasticity (Young’s Modulus), which can be determined via the ultrasonic velocity. The ultrasonic velocity with up to 3.8 km/s as well as the Young’s Modulus with up to 32 GPa are significantly higher in group 1, meaning S1 and S3 possess a higher rock strength. In comparison, group two (S2, S4) shows ultrasonic velocities of 2.5 km/s and Young’s Moduli of up to 13 GPa.

Under laboratory conditions the sandstones show moderate resistance towards salt attack. After 30 cycles of salt bursting test (standard DIN EN 12370), the sandstones show a distinct darkening of their color tone and suffer minor material loss in the form of flaking and scaling (Fig. 7). Only S1 shows a constant loss of material in the form of bursting of edges and smaller components (Fig. 7a).

Figure 7: Discoloration and material loss of sample cubes S1–S4 after 30 cycles of salt bursting test.

Summary

The studied monoliths show diverse, partly exposure-specific weathering forms. Under laboratory 214conditions, the negative effects of thermal and thermohygric expansion were identified and are in accordance with the field observations. Water did show to increase the negative effect of thermal expansion on the investigated sandstones. Also its function as a transport medium has to be taken into account. The in-situ investigation showed severe damages associated with salt crystallization, which could be reproduced under laboratory conditions to a lesser extent. Possible protective measures for the future could be temporary roof constructions during seasons of high solar radiation and temperature or a protective planting of trees, which would be a permanent, but more natural solution. Nevertheless, individual monoliths should be considered for conservational treatment due to the high degree of weathering. Desalination and repairs with stone replacement mortars as well as measures to minimize the future salt input into the stone can be crucial. Regarding the origin of the monolith source material, the data from the in-situ and petrographic investigations, point towards a sandstone layer in the Paja Formation as origin of the sandstones S2 and S3. Therefore, further investigation should concentrate on the residual sandstone boulders covering the Paja Formation in the vicinity of the park.

Acknowledgements

We would like thank M. Silva Celis, director of the archaeological park Monquirá, for supporting our work as well as C. Gross for helping with the petrographical descriptions.

References

Etayo-Serna, F., 1968. Sinopsis Estratigrafica de la region de Villa de Leiva y zonas proximas. Boletín de Geología 8.

Hirschwald J (1912) Die Prüfung der natürlichen Bausteine auf ihre Wetterbeständigkeit. W. Ernst & Sohn, Berlin.

Öcal, A. D.; T. Cramer; S. Siegesmund, 2009. Caracterización de agentes del deterioro de los monolitos de piedra arenisca del Infiernito – Colombia, en 2do. Congreso Argentino y 1ro. Latinoamericano de Arqueometría 6–8 de julio de 2007, Tulio Palacios et al. – eds. – Buenos Aires: Comisión Nacional de Energía Atómica – CNEA, 2009, pp. 413–419.

Patarroyo Gama, P., 2008. La Formacion Ritoque en la zona de Vélez (Santander-Colombia). Geología colombiana 33, 109–110.

Renzoni, G., Rosas, H., Etayo-Serna, F., 1983. Mapa Geológico de la Plancha 171, Duitama, esc. 1:100.000. Ingeominas, Bogotá.

Siegesmund, S. and Dürrast, H. (2011), Physical and mechanical properties of rocks. In Siegesmund, S. and Snethlage, R. (Eds.), Stone in architecture, Springer, pp. 97–225.

Silva Celis, E. (1983) Descubrimiento Arqueológico en Villa de Leiva. Boletín de Antropología 5 (17–19), pp. 235–250.

Simón, P. 1981. Noticias Historiales de las Conquistas de Tierra Firme en las Indias Occidentales; Biblioteca del Banco Popular, Bogotá.

215

EVALUATION OF EFFECTIVE COMPRESSION STRENGTH OF NATURAL STONE BY DRILLING RESISTANCE MEASUREMENTS

Stephan Pfefferkorn¹, Christoph Franzen²

IN: SIEGESMUND, S. & MIDDENDORF, B. (EDS.): MONUMENT FUTURE: DECAY AND CONSERVATION OF STONE.

– PROCEEDINGS OF THE 14TH INTERNATIONAL CONGRESS ON THE DETERIORATION AND CONSERVATION OF STONE –

VOLUME I AND VOLUME II. MITTELDEUTSCHER VERLAG 2020.

1 HTW Dresden, FB Bauingenieurwesen, Friedrich-List-Platz 1, 01069 Dresden, Germany

2 Institut für Diagnostik und Konservierung an Denkmalen in Sachsen und Sachsen-Anhalt e. V. (IDK), Schlossplatz 1, 01067 Dresden, Germany

Abstract

The knowledge of strength parameters for material is necessary for the appraisal of the bearing behaviour of historic buildings. Usually these are determined by suitable check of drilling cores.

The columns under the vault in the ground floor of the historical city hall in Oederan (Germany, Saxony) are strongly loaded according to continuance statics. For the judgement their load-carrying capacity the pressure resistance should be determined. On account of the statics a cross section reduction was not admitted by taking drilling core. Besides, the columns from Hilbersdorfer Porphyrtuff showed considerable decomposition phenomena in its surfaces. The aim of the investigations was to find out the depth of the structure loosening and to estimate the pressure resistance of the uninjured rest cross section.

These questions could be answered with the drilling resistance measurement method. Besides, drillings with small diameter and a steady driving are introduced into the material. The measuring value of the drilling resistance arises from the contact force which is necessary for the realisation of the given driving. It can be given for every tenth millimetre of the drilling depth, so that a depth profile of the drilling resistance arises.

Between compressive strength and the drilling resistance a linear correlation exists. This relation could be determined on the basis of test specimens which were cut of simultaneous removal material. Drilling resistance as well as compressive strength were checked at these cubes and were confronted. Afterwards three drilling opposition measurements were executed per column. From the measuring profiles obtained could be read the depth of the structure decay as well as the drilling resistance of the unweathered core cross section be determined. With these values the compression strength was estimated under use of the before provided correlation function.