First archaeointensity results from Ecuador with rock magnetic analyses and 14C dates to constrain the geomagnetic field evolution in South America: Enhancing the knowledge of geomagnetic field intensity

Highlights

• This article is to determine the strength of the geomagnetic field.

• We recovered ~56 pieces of pottery from the highlands in northern Ecuador from 2 mounds and one excavation site.

• The radiometric dates were determined by means of ¹⁴C.

• The archaeomagnetic intensity was obtained by means of a modified Thellier-Coe protocol.

Abstract

Archaeomagnetic records from low-latitude sites are very scarce. Thus, the information about the geomagnetic field evolution in South America or Ecuador is sparse and more data are needed to constrain the field evolution in South America. In this study we present new absolute archaeointensity results from a set of ¹⁴C well-dated (i.e. calibrated ages) prehistoric archaeological sites in northern Ecuador (South America). Potsherd fragments were sampled during archaeological excavations of 2 earthen mounds in the vicinity of Atuntaqui and Otavalo cities, and pits at La Chimba site. The pottery covers 2 distinct archaeological periods defined as 1) Ecuadorian Formative period (700 BC – 1AD) and 2) the prehistoric Late period (1250–1505/1525 AD). Successful archaeointensity determinations were obtained from 56 sherds using the Thellier-Coe protocol and their rock magnetic properties.

The absolute palaeointensity determinations have yielded AI values of 43.6 ± 1.3 μT for Atuntaqui, 34.2 ± 1.4 μT for the Otavalo Mound 3 (lower charcoal lens), 39.5 ± 1.4 μT for Otavalo Mound 3 (upper charcoal lens) and a range of absolute archaeointensities from 7.0 ± 0.97 μT up to 42.2 ± 2.06 μT, with a mean average of 21.20 ± 1.15 μT for La Chimba. Archaeointensity (AI) determinations of the successful mean abolute archaeointensity for the early phase (700–440BC) excavation levels 27.8 ± 1.315 μT, middle levels phase (440–40BC), 15.07 ± 0.78 μT and the late levels phase (44BC-250AD) 25.52 ± 1.575 μT. These results are in good correlation with the data from Palpa in South Peru as well as a few data results from Colombia, South America.

Introduction

It has been known for several decades that archaeological artifacts can be reliable recorders of the Earth's magnetic field behavior in the past (e.g. Eighmy and Sternberg, 1990; Shibuya, 1980; Aidona et al., 2010, Aguilar-Reyes et al., 2013). The archaeological baked clays can generally offer a better temporal and spatial resolution than volcanic rocks, mainly due to their precise ¹⁴C radiometric dating, and due to the fact that they can be found all around the world without being limited to volcanic areas. The radiocarbon dating of the pieces of pottery of northern Ecuador highlands (Fig. 1) was carried out using charcoal remains, (e.g. maize, corn, organic food residues, Appendix A, Fig. 2), as well as pollen and volcanic ash, for Atuntaqui and Otavalo mounds (e.g. Athens and Osborn, 1974; Athens, 1999, 2003). The radiocarbon dating of La Chimba was based on Athens (1990) and Stahl and Athens (2001) where almost 40,000 specimens were recovered during excavations that started in 1972 at La Chimba site, Pichincha Province, Ecuador. A listing (including weights and counts) of all remains recovered from the 1989 investigations is provided in Athens (1990), including diagnostic ceramics. Identified macrobotanical remains at La Chimba include charred specimens of small maize cobs and kernels, two types of tubers (potatoes and oca), and seeds of quinoa, beans, and other unidentified types Potatoes, oca, and quinoa had not been previously identified in archaeological sites in the northern highlands (Athens, 1990). Fifteen radiocarbon determinations were obtained from TP-7 and appear to encompass the complete occupational history of La Chimba. Using absolute dates assigned to all levels based on a non-quantitatively determined best fit radiocarbon depth-age curve, it appears that initial occupation occurred sometime around 700 B.C. (2640 B.P. calibrated). Occupation was apparently continuous until the site was abandoned around A.C. 250 (1700 B.P. calibrated). The ages of the different test units are summarized in Table 1 and Table 2 of Athens (1990) which presents a generalized chronological scheme for the site (see Athens, 1978). La Chimba is at present the earliest known Ecuadorian ceramic period site north of Quito. Archaeomagnetic data, however, still remain sparse and very unevenly distributed in both time and space to well-describe and understand the pattern of the geomagnetic field's secular variation (SV) over millennial timescales on a global scale (Valet et al., 2008; Pavón-Carrasco et al., 2014a; Panovska et al., 2015). For instance, very few data come from the southern hemisphere and less than 3% of the archaeointensity entries in the GEOMAGIA database (Brown et al., 2015) come from sites with latitudes between −10 and 10°. In particular, archaeomagnetic data from South America are scarce, and in many cases, come from old studies that are characterized by scattering (Goguitchaichvili et al., 2019). Thus, in order to enhance, improve and contribute to the development of reliable global geomagnetic field models, it is necessary to improve the spatial global coverage with well-dated and high-quality reference data (e.g. Pavón-Carrasco et al., 2014b). In this paper, we present a new set of absolute archaeointensity determinations from pottery collections recovered from three archaeological sites located at the highlands of north central Ecuador. The new results were obtained with the modified Thellier-Coe protocol that implies that instead of measuring the NRM first, we preferred to apply a TRM before heating the sample in zero field (e.g. that Aitken et al., 1988, Valet et al., 1998, Herrero-Bervera and Valet, 2005, 2009; Herrero-Bervera et al., 2016), including partial thermoremanent magnetization (pTRM) checks that helps out to monitor possible magnetomineralogical alterations that can take place during experimental heating. The ages of the samples studied were determined through calibrated radiometric data and archaeological evidence based on diagnostic pottery. Only, a few previous available studies from Ecuador, data from neighboring Peru and Colombia covering the same time periods of our study are avaliable and were used for comparison with our new archaeointensity determinations. It is very important to point out that the archaeointensity results of this study will only be very useful and meaningful for the study of the evolution of the generation of the geomagnetic field at equatorial latitudes of ~11⁰ North and for a time windows of our analysis (i.e. 700 BC-1 AD, 1250–1505/1525 AD, 700–440BC, 440–40BC and 44BC-250AD) and within a 600–700 km circle (e.g. Tema and Knodopoulou, 2011) and centered at the northen ecuador sites correlatable to the Earth's core generated signals. Beyond this circle (i.e. radius of 600–700 km) any attempt to correlate to distant locations such as central America, Brazil, Argentina, Uruguay, Paraguay, Chile and Bolivia is meaningless due to the way the geomagnetic field is generated.