Enhanced interpretation of magnetic survey data using artificial neural networks: a case study from Butrint, southern Albania

Abstract

The classical city of Butrint in southern Albania embodies over three millennia of settlement history. A Roman colony wasestablished sometime after 31BC, whichled to the expansion ofthe city southwards onto a low-lying floodplain where settlement continued wellinto the late antique period. In this paper we describe the results of a detailed magnetometry survey undertaken to investigate Roman settlement upon the floodplain.The study included the use of multilayer perceptron neuralnetworks to further process the magnetic data and derive estimates of feature burial depths, allowing a three-dimensional reconstruction of buried subsurface remains to be made. The neural network approach potentially offers several advantages in terms of efficiency and flexibility over more conventional data inversion techniques.The paper demonstrates how this can lead to an enhanced interpretation of magnetic survey data, which when combined with other geoarchaeological data can provide a clearer picture of settlement evolution within the context of landscape change.The value of this processing technique is also evident within the context of cultural resource management strategies, which potentially restrict more intrusive methods of investigation.

Detecting Roman land boundaries in aerial photographs using Radon transforms

“…2). The greater extent of these building has been recorded by past geophysical surveys (see [4,15]), while a small section of road recorded within the southwest portion of the settlement confirms this alignment.…”

Human–climate interactions in the central Mediterranean region during the last millennia: The laminated record of Lake Butrint (Albania)

“…2000 BC), when slope-derived 'terra rossa' in Konispol Cave (headwaters of Pavllo river) are interpreted as a result of incipient grazing, shepherding and deforestation favored by warmer conditions (Ellwood et al, 1997). The Butrint Peninsula was intermittently occupied during the Bronze Age and the Archaic period (8 th century BC) as a hilltop refuge, and a fortified trading post was established by the 6 th century BC (Bescoby et al, 2004) (Table 2). By the 4 th century BC, Butrint became a Hellenistic port and the city was expanded (Hodges, 2013).…”

“…A century later (after 31 BC) 'Buthrotum' was designated as a colony for Roman veterans, and the city duplicated its surface towards the Vivary Channel. It also expanded to over 2.5 ha to the Vrina Plain (Bescoby et al, 2004), with the construction of a new aqueduct and bridge across the Vivari Channel (late 1 st century AD) Wilson, 2013). Image analysis of historical aerial photographs revealed a complex farmland divisions in the Vrina Plain during Roman times, responding to a centuriated pattern.…”

Deep learning based automated analysis of archaeo‐geophysical images

“…Schmidt (), discusses the advanced processing of vector data for the automatic extraction of magnetometer anomalies. The first use of an ANN that targets the processing of magnetic data and the estimation of feature parameters for inversion purposes was proposed by Bescoby, Cawley and Chroston (). Indeed, the approaches on simulating GPR signals and parameter estimation for modeling using machine learning techniques is discussed by Giannakis, Giannopoulos, and Warren ().…”