Detecting and Predicting Archaeological Sites Using Remote Sensing and Machine Learning – Application to the Saruq Al Hadid Site, Dubai, UAE

Abstract

In this paper, the feasibility of satellite remote sensing in detecting and predicting locations of buried objects in the archaeological site of Saruq al-Hadid, the United Arab Emirates (UAE) was investigated. Satellite borne Synthetic Aperture Radar (SAR) is proposed as main technology for this initial investigation. In fact, SAR is the only satellite-based technology able to detect buried artefacts from space and it is expected that fine resolution images of ALOS/PALSAR-2 (L-band SAR) would be able to detect large features (> 1 m) that might be buried in subsurface (< 2 m) under optimum conditions i.e., dry, and bare soil. SAR data were complemented with very-high resolution Worldview-3 multispectral images (0.31 m panchromatic, 1.24 m VNIR) to have a vis-ual assessment of the study area and its land cover features. An integrated approach through the application of advanced image processing techniques and geospatial analysis using machine learning was adopted to characterise the site while automating a process and investigate its ap-plicability. Results from SAR feature extraction and geospatial analyses showed detection of the already under excavation areas on the site and predicted new archaeological areas unexplored yet. The validation of these results was performed using previous archaeological works, geolog-ical and geomorphological field surveys. The modelling and prediction accuracies are expected to improve using the insertion of a neural network and backpropagation algorithms based on the performed cluster groups following more recent field surveys. The validated results can provide guidance for future on-site archaeological work. The pilot process developed in this work can, therefore, be applied to similar arid environments for the detection of archaeological features and guidance of on-site investigations

Full text

Citation: Ben-Romdhane, H.; Francis,
D.; Cherif, C.; Pavlopoulos, K.; Ghedira,
H.; Griffiths, S. Detecting and
Predicting Archaeological Sites Using
Remote Sensing and Machine
Learning—Application to the Saruq
Al-Hadid Site, Dubai, UAE. Geosciences
2023,13, 179. https://doi.org/
10.3390/geosciences13060179
Academic Editors: Jesus
Martinez-Frias and Deodato Tapete
Received: 9 May 2023
Revised: 10 June 2023
Accepted: 12 June 2023
Published: 15 June 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
geosciences
Article
Detecting and Predicting Archaeological Sites Using Remote
Sensing and Machine Learning—Application to the Saruq
Al-Hadid Site, Dubai, UAE
Haïfa Ben-Romdhane 1,2 , Diana Francis 1, * , Charfeddine Cherif 1, Kosmas Pavlopoulos 2, Hosni Ghedira 1,3
and Steven Griffiths 1
1Earth Sciences Department, Khalifa University, Abu Dhabi P.O. Box 127788, United Arab Emirates;
haifa.ben@sorbonne.ae (H.B.-R.); charfeddine.cherif@ku.ac.ae (C.C.); hosni.ghedira@mbzuai.ac.ae (H.G.);
steven.griffiths@ku.ac.ae (S.G.)
2Geography and Planning Department, Sorbonne University Abu Dhabi,
Abu Dhabi P.O. Box 38044, United Arab Emirates; kosmas.pavlopoulos@sorbonne.ae
3Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi 144534, United Arab Emirates
*Correspondence: diana.francis@ku.ac.ae
Abstract:
In this paper, the feasibility of satellite remote sensing in detecting and predicting locations
of buried objects in the archaeological site of Saruq Al-Hadid, United Arab Emirates (UAE) was
investigated. Satellite-borne synthetic aperture radar (SAR) is proposed as the main technology for
this initial investigation. In fact, SAR is the only satellite-based technology able to detect buried
artefacts from space, and it is expected that fine-resolution images of ALOS/PALSAR-2 (L-band
SAR) would be able to detect large features (>1 m) that might be buried in the subsurface (<2 m)
under optimum conditions, i.e., dry and bare soil. SAR data were complemented with very high-
resolution Worldview-3 multispectral images (0.31 m panchromatic, 1.24 m VNIR) to obtain a visual
assessment of the study area and its land cover features. An integrated approach, featuring the
application of advanced image processing techniques and geospatial analysis using machine learning,
was adopted to characterise the site while automating the process and investigating its applicability.
Results from SAR feature extraction and geospatial analyses showed detection of the areas on the
site that were already under excavation and predicted new, hitherto unexplored archaeological
areas. The validation of these results was performed using previous archaeological works as well
as geological and geomorphological field surveys. The modelling and prediction accuracies are
expected to improve with the insertion of a neural network and backpropagation algorithms based
on the performed cluster groups following more recent field surveys. The validated results can
provide guidance for future on-site archaeological work. The pilot process developed in this work
can therefore be applied to similar arid environments for the detection of archaeological features and
guidance of on-site investigations.
Keywords:
archaeology; remote sensing; geospatial analysis; machine learning; artificial intelligence;
SAR; PALSAR-2; Worldview-3; Saruq Al-Hadid; Dubai; United Arab Emirates
1. Introduction
Since launch, the advances realised by several remote sensors and technologies, such
as the potential of providing systematic data over large areas, have led to satellite remote
sensing being widely applied to various archaeological studies in several parts of the
world [1–5].
The application of imaging radar to archaeological research is gaining momentum [
1
,
6
–
9
].
These data often offer direct detection of archaeological sites and indirect detection of areas
of unknown human activity [
10
]. Remote sensing allows for regional surveys on a scale
that would not be possible from the ground. This reduces the effort and time spent on
Geosciences 2023,13, 179. https://doi.org/10.3390/geosciences13060179 https://www.mdpi.com/journal/geosciences

Geosciences 2023,13, 179 2 of 34
conventional archaeological reconnaissance and surveys and ameliorates the quality of
the intended outcomes. This is particularly valuable in areas that are inaccessible due to
difficult or hazardous terrains, such as deserts [11,12].
Research has been performed on the applicability of spaceborne imaging and radar
to archaeological surveys focused on geological investigation and vegetation assessment
and mapping [
7
]. While wetland archaeology has placed the focus on vegetation discrim-
ination [
13
,
14
], archaeological research in arid environments has emphasised geological
applications [
15
–
17
] and the detection of cultural heritage sites, shedding light on early
civilisations about which little is known through the use of direct methods such as surveys,
excavations, and dating techniques and/or indirect methods including ground-penetrating
radar and aerial photography [6,18].
A major problem in remote sensing studies of arid and semi-arid environments such as
the United Arab Emirates (UAE) is the deterioration of the image content by dust particles
or cloud cover [
19
]. The development of radar sensors and signal penetration abilities
throughout the atmosphere have precluded this difficulty [7].
Moreover, the application of AI machine learning and deep learning-based classifi-
cation of remote sensing data has advanced the detection and mapping of archaeological
features in arid environments, making the process more accurate and automated [20–22].
The study area (Saruq Al-Hadid site, Dubai, UAE) is contained within the Rub’ Al-
Khali (the Empty Quarter), a sand desert containing large areas covered by dry sand sheets,
sand dunes, and drift sand. Archaeological fieldwork and post-excavation analyses were
conducted by several authors e.g., [
23
–
28
]. However, a very limited number of studies
have approached the site using satellite remote sensing [29].
To the best of the authors’ knowledge, the current study is the first to use advanced
image processing and machine learning techniques for the detection, prediction, and
guidance of archaeology within the area of interest and the Rub’ Al-Khali desert since the
pioneering work by Blom et al. [
18
] which coupled radar remote sensing technologies with
traditional archaeology, giving insight into the ancient southern Arabian frankincense trade,
including the discovery of the “lost city of Ubar” in present-day Oman. This pilot study is
thus incentivised by the main objective of setting up a benchmark for the development of
national and regional remote sensing archaeology capabilities while automating the process
and investigating its potential errors and accuracy before generalising it to larger areas.
The approach proposed to achieve this objective is based on the hypothesis that the
artefacts discovered so far in the study area were produced on site, which might indicate
the possible presence of buried settlements in the surroundings of the Saruq Al-Hadid
site used by ancient indigenous workers. The methods used to implement this approach
are motivated by recent trends in archaeological prospection, remote sensing, artificial
intelligence (AI), and machine learning (ML). Results from the implementation of these
methods are presented and discussed. The outcomes are expected to be enhanced with
the integration of suitable archaeological training and testing data and with the broader
application to similar environments.
2. Materials and General Approach
2.1. Study Site
Saruq Al-Hadid (“Valley of iron”) archaeological site (24
◦
39
0
47
00
N, 55
◦
13
0
55
00
E) is
located within the northern extension of the Rub’ Al-Khali desert on the southern border
of Dubai, UAE (Figure 1a). It is characterised by active dune fields and considered one
of the most unique and important archaeological sites uncovered in the past few years
in the UAE [
25
]. In fact, it represents one of the main centres for copper smelting, manu-
facturing tools, and various utensils in the region since the beginning of the Second Iron
Age
(1270–800 BC) [27].
It is characterised by the presence of thousands of archaeological
artefacts spread over an area of >1 km
2
and buried within and beneath dunes up to 6 m
deep (Figure 1b,c).

Geosciences 2023,13, 179 3 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 3 of 34
2. Materials and General Approach
2.1. Study Site
Saruq Al-Hadid (“Valley of iron”) archaeological site (24°39′47″ N, 55°13′55″ E) is lo-
cated within the northern extension of the Rub’ Al-Khali desert on the southern border of
Dubai, UAE (Figure 1a). It is characterised by active dune fields and considered one of the
most unique and important archaeological sites uncovered in the past few years in the
UAE [25]. In fact, it represents one of the main centres for copper smelting, manufacturing
tools, and various utensils in the region since the beginning of the Second Iron Age (1270-
800 BC) [27]. It is characterised by the presence of thousands of archaeological artefacts
spre ad o ver an area of >1 km
2
and buried within and beneath dunes up to 6 m deep (Figure
1b,c).
(a)
(b)
(c)
Figure 1. (a) Left inset: Saruq Al-Hadid archaeological site (24°39′47″ N 55°13′55″ E). Middle inset:
Worldview-3 multispectral image (left scene—19 August 2019; central scene—19 November 2019;
right scene—9 January 2019). Right inset: ALOS-2/PALSAR2 (L-band) image (2015-05-
17T20:15:36Z). (b) Metalworking slag interspersed with metal artefacts, ceramics, and other cultural
material. (c) Slag artefacts on the dune surface extending to over 1 km
2
post-excavation.
2.2. Geological and Stratigraphic Seing
The lithology of the study area consists of (from youngest to oldest): linear longitu-
dinal dune landforms (mostly low dunes) and dune ridges, with large interdunes under-
lain by variably dolomitised sandstone, siltstone, and conglomerate of the Barzaman For-
mation (Miocene) or the Quaternary fluvial sandstones, conglomerates, and carbonate
Figure 1.
(
a
) Left inset: Saruq Al-Hadid archaeological site (24
◦
39
0
47
00
N 55
◦
13
0
55
00
E). Middle inset:
Worldview-3 multispectral image (left scene—19 August 2019; central scene—19 November 2019;
right scene—9 January 2019). Right inset: ALOS-2/PALSAR2 (L-band) image (2015-05-17T20:15:36Z).
(
b
) Metalworking slag interspersed with metal artefacts, ceramics, and other cultural material.
(c) Slag artefacts on the dune surface extending to over 1 km2post-excavation.
2.2. Geological and Stratigraphic Setting
The lithology of the study area consists of (from youngest to oldest): linear longitudinal
dune landforms (mostly low dunes) and dune ridges, with large interdunes underlain by
variably dolomitised sandstone, siltstone, and conglomerate of the Barzaman Formation
(Miocene) or the Quaternary fluvial sandstones, conglomerates, and carbonate sandstones
of the Hili Formation [
30
]. The dunes vary in height by up to ~20 m. Previous studies
e.g., [
31
] and field observations conducted in 2015 showed that between the two areas of
large Barchanoid ridges there are smaller dune ridges, aligned broadly east-west, which
cover the site compound and the land to the immediate east (Figure 2a,b).

Geosciences 2023,13, 179 4 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 4 of 34
sandstones of the Hili Formation [30]. The dunes vary in height by up to ~20 m. Previous
studies e.g., [31] and field observations conducted in 2015 showed that between the two
areas of large Barchanoid ridges there are smaller dune ridges, aligned broadly east-west,
which cover the site compound and the land to the immediate east (Figure 2a,b).
During the geological field survey, it was observed that the Barzaman Formation out-
lines the rockhead across most of the area and is locally well exposed in interdune areas.
Additionally, carbonate-rich Ghayathi Formation aeolianites were observed to form dis-
tinctive palaeodune ridges that trend southwest–northeast across the area, particularly to
the west and northwest. These observations align with statements made by other authors
[25,30,31], who determined deposits of up to 7 m of stratified layers well conserved be-
neath the protective crust of slag (Figure 2c). The comprehension of the geological and
stratigraphic seing of the study site will complement the geospatial analysis and inter-
pretation conducted in Section 3.3.
(a)
(b)
(c)
Figure 2. (a) The basic stratigraphy of the study area consisting of the geological bedrock of the
Barzaman Formation and the overlying longitudinal dune ridges. Prosopis cineraria trees, also known
as Ghaf, can be seen in the interdunal area. (b) The gypsum pavement which underlies the excava-
tion in the middle of the site has an elevation of approximately 100 m above sea level. This pavement
is exposed at several locations in the study area. (c) The general stratigraphic seing of (1) the surface
deposits with abundant evidence of metallurgical production debris and other artifacts and archae-
ological remains; (2) the thick series of intermient occupation/deposition activities represented by
diffuse material remains in a sandy matrix, including a variety of artefacts and slag; (3) a buried
Figure 2.
(
a
) The basic stratigraphy of the study area consisting of the geological bedrock of the
Barzaman Formation and the overlying longitudinal dune ridges. Prosopis cineraria trees, also known
as Ghaf, can be seen in the interdunal area. (
b
) The gypsum pavement which underlies the excavation
in the middle of the site has an elevation of approximately 100 m above sea level. This pavement
is exposed at several locations in the study area. (
c
) The general stratigraphic setting of (1) the
surface deposits with abundant evidence of metallurgical production debris and other artifacts
and archaeological remains; (2) the thick series of intermittent occupation/deposition activities
represented by diffuse material remains in a sandy matrix, including a variety of artefacts and slag;
(3) a buried dune horizon marked by a particularly dense concentration of artefacts laid down over
what may have been an informal ‘platform’ incorporating a high frequency of rough sandstone
blocks; and (4) a series of midden deposits characterised by a very high frequency of animal bone,
often forming concentrated layers that may reflect deflation.
During the geological field survey, it was observed that the Barzaman Formation
outlines the rockhead across most of the area and is locally well exposed in interdune
areas. Additionally, carbonate-rich Ghayathi Formation aeolianites were observed to form
distinctive palaeodune ridges that trend southwest–northeast across the area, particularly
to the west and northwest. These observations align with statements made by other
authors [
25
,
30
,
31
], who determined deposits of up to 7 m of stratified layers well conserved
beneath the protective crust of slag (Figure 2c). The comprehension of the geological

Geosciences 2023,13, 179 5 of 34
and stratigraphic setting of the study site will complement the geospatial analysis and
interpretation conducted in Section 3.3.
2.3. Datasets
2.3.1. Worldview-3
The Worldview-3 satellite sensor is a multispectral Earth-observing satellite that is
owned and operated by space technology firm Maxar Technologies Incorporated to collect,
in addition to the standard panchromatic and multispectral bands, eight-band short-wave
infrared (SWIR) and 12 CAVIS imagery. The Worldview-3 satellite provides 0.31 m panchro-
matic resolution, 1.24 m multispectral resolution, 3.7 m short-wave infrared resolution
(SWIR), and 30 m CAVIS resolution. The satellite has an average revisit time of <1 day
and can collect up to 680,000 km
2
per day [
32
]. Such high spatial and spectral resolutions
are anticipated to assist archaeological applications such as feature and elevation data
extraction and soil classification in a cost-effective manner. Three scenes from January,
August, and November 2019 were acquired to cover the study site (Figure 1a). Analysis
and interpretation of these imagery data are performed in Sections 2.4 and 3.1.
2.3.2. ALOS-2/PALSAR-2
The PALSAR-2 aboard the Advanced Land Observation Satellite 2 (ALOS-2) is a
synthetic aperture radar (SAR) operated by the Japanese Aerospace Exploration Agency.
It transmits and receives the L-band microwave characterised by deeper penetration of
the SAR signal in dry sand cover overlying potential buried structures [
33
]. PALSAR-2
demonstrated its usefulness in achieving high-resolution wide-swath width and image
quality (lower noise floor and range ambiguities) by expanding its transmission power
and bandwidth and adopting new technologies such as dual-beam receivers, complex
chirp modulations, and highly efficient data compression [
34
]. The Committee on Earth
Observation Satellites (CEOS) SAR format of PALSAR-2 stripmap high-sensitive mode full
polarisation Level 1.1 data were acquired on 17 May 2015 over the Saruq Al-Hadid site
(Figure 1a). These are complex numerical data on the slant range following compression
of the range and azimuth. As one-look data, they include phase information and are the
basis for later processing. In wide-area mode, image files are created for each scan. The
image is fully (quadratic) polarised and comes in single-look complex (SLC) data format.
The resolution is approximately 4.3 m (high-sensitive, range resolution 6.0 m/azimuth
resolution 4.3 m). Analysis and interpretation of these imagery data are performed in
Sections 2.4 and 3.2.
2.4. Approach
Satellite data and geographic information system (GIS) data from historic and recent
maps were blended for a comprehensive geospatial analysis of the site and its geographic
context to examine the current archaeological site in its historic context and provide predic-
tion and guidance for new potential sites in the UAE. Data blending involved integrating
information from multiple sources with different data characteristics, resolution, and limi-
tations. To ensure a maximum of compatibility and consistency among the used datasets,
techniques such as georeferencing and spatial alignment, data format standardisation,
colour balancing, resampling, coregistration, etc. were applied. Satellite image analysis re-
quired the acquisition, analysis, and interpretation of Worldview-3 and ALOS-2/PALSAR-2
data. Geospatial analysis required the use of AI using ML to automate the classification,
clustering, spatial pattern detection, and multivariate prediction of these data. Figure 3
summarises the approach followed in analysing and interpreting the original and trans-
formed multispectral and radar data. Image visualisation and processing were performed
using the Exelis
®
ENvironment for Visualising Images (ENVI) interface and Interactive
Data Language (IDL) version 5.3 (Exelis Visual Information Solutions, Boulder, CO, USA).
ENVI/IDL handled all pre- and post-processing work for the multispectral and SAR im-
ages. Figure 4summarises the approach undertaken during the geospatial analysis process.

Geosciences 2023,13, 179 6 of 34
Processing of the geospatial input data using AI and ML algorithms was performed fol-
lowing the steps shown in Figure 5. ESRI
®
ArcGIS Pro platform version 3.0.0 (Redlands,
CA; Environmental Systems Research Institute) supported these tasks by providing the
tools for the deep learning (DL) workflow, which involved data labelling and preparation,
training models and deploying them for inferencing, and disseminating results.
Geosciences 2023, 13, x FOR PEER REVIEW 6 of 34
and Interactive Data Language (IDL) version 5.3 (Exelis Visual Information Solutions,
Boulder, CO, USA). ENVI/IDL handled all pre- and post-processing work for the multi-
spectral and SAR images. Figure 4 summarises the approach undertaken during the geo-
spatial analysis process. Processing of the geospatial input data using AI and ML algo-
rithms was performed following the steps shown in Figure 5. ESRI
®
ArcGIS Pro platform
version 3.0.0 (Redlands, CA; Environmental Systems Research Institute) supported these
tasks by providing the tools for the deep learning (DL) workflow, which involved data
labelling and preparation, training models and deploying them for inferencing, and dis-
seminating results.
Figure 3. The adopted approach for analysing and interpreting the multispectral and radar data.
Number-indexed bands correspond to multispectral bands acquired in the different regions of the
electromagnetic spectrum: B2 (Blue (B): 450–510 nm), B3 (Green (G): 510–580 nm), B5 (Red (R): 630–
690 nm), B6 (Red Edge: 705–745 nm), and B7 (Near Infrared 1: 770–895 nm). Number-indexed prin-
cipal components (PC) were obtained using the principal component analysis (PCA). The PCs are
ranked using the variation in the data (largest to smallest), as elaborated in Section 3.1.1 and Ap-
pendix B. Band ratios, e.g., the normalised difference vegetation index (NDVI) and the normalised
high-frequency difference index (NHFD), were produced used spectral band transformation, as de-
tailed in Section 3.1.1 and Appendix C.
Figure 4. The adopted geospatial analysis process.
Figure 3.
The adopted approach for analysing and interpreting the multispectral and radar data.
Number-indexed bands correspond to multispectral bands acquired in the different regions of the
electromagnetic spectrum: B2 (Blue (B): 450–510 nm), B3 (Green (G): 510–580 nm), B5 (Red (R):
630–690 nm), B6 (Red Edge: 705–745 nm), and B7 (Near Infrared 1: 770–895 nm). Number-indexed
principal components (PC) were obtained using the principal component analysis (PCA). The PCs
are ranked using the variation in the data (largest to smallest), as elaborated in Section 3.1.1 and
Appendix B. Band ratios, e.g., the normalised difference vegetation index (NDVI) and the normalised
high-frequency difference index (NHFD), were produced used spectral band transformation, as
detailed in Section 3.1.1 and Appendix C.
Geosciences 2023, 13, x FOR PEER REVIEW 6 of 34
and Interactive Data Language (IDL) version 5.3 (Exelis Visual Information Solutions,
Boulder, CO, USA). ENVI/IDL handled all pre- and post-processing work for the multi-
spectral and SAR images. Figure 4 summarises the approach undertaken during the geo-
spatial analysis process. Processing of the geospatial input data using AI and ML algo-
rithms was performed following the steps shown in Figure 5. ESRI
®
ArcGIS Pro platform
version 3.0.0 (Redlands, CA; Environmental Systems Research Institute) supported these
tasks by providing the tools for the deep learning (DL) workflow, which involved data
labelling and preparation, training models and deploying them for inferencing, and dis-
seminating results.
Figure 3. The adopted approach for analysing and interpreting the multispectral and radar data.
Number-indexed bands correspond to multispectral bands acquired in the different regions of the
electromagnetic spectrum: B2 (Blue (B): 450–510 nm), B3 (Green (G): 510–580 nm), B5 (Red (R): 630–
690 nm), B6 (Red Edge: 705–745 nm), and B7 (Near Infrared 1: 770–895 nm). Number-indexed prin-
cipal components (PC) were obtained using the principal component analysis (PCA). The PCs are
ranked using the variation in the data (largest to smallest), as elaborated in Section 3.1.1 and Ap-
pendix B. Band ratios, e.g., the normalised difference vegetation index (NDVI) and the normalised
high-frequency difference index (NHFD), were produced used spectral band transformation, as de-
tailed in Section 3.1.1 and Appendix C.
Figure 4. The adopted geospatial analysis process.
Figure 4. The adopted geospatial analysis process.

Geosciences 2023,13, 179 7 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 7 of 34
Figure 5. Processing the geospatial data using the AI’s DL and ML algorithms.
3. Methods and Results
Results from the processing and analysis of the multispectral and radar data are pre-
sented and discussed first. Geospatial modelling and analysis are presented and discussed
later.
3.1. Spectral Analysis and Classification of Multispectral Data
Worldview-3 multispectral data were used to produce a very high-resolution mosaic
(true colour) of the Saruq Al-Hadid site and surrounding areas. Image visual analysis and
interpretation were performed after localisation of the site based on the produced ortho-
mosaic. Figure 6 shows a Worldview-3 red-green-blue (RGB) orthomosaic of the Saruq Al-
Hadid site located in the mobile dune fields of the northeastern edge of the Rub’ Al-Khali
desert, captured on 26 November 2019. The orthomosaic is colour-balanced within the
region tiles to minimise the colour differences due to side-to-side shading, contrast varia-
tions, time of day, sun angle, atmospheric conditions such as haze, and the use of multiple
cameras over multiple days among adjacent region tiles.
Previous and ongoing excavations at several of the Saruq Al-Hadid areas, as reported
by [26], were located on the multispectral images (Appendix A—Figure A1). Appendix
A—Figure A2a–c shows some remarkable spots and locations identified in the western
zone of observation of the site based on the produced orthomosaic. Appendix A—Figure
A3 shows some remarkable locations in the eastern zone of observation.
Figure 5. Processing the geospatial data using the AI’s DL and ML algorithms.
3. Methods and Results
Results from the processing and analysis of the multispectral and radar data are
presented and discussed first. Geospatial modelling and analysis are presented and
discussed later.
3.1. Spectral Analysis and Classification of Multispectral Data
Worldview-3 multispectral data were used to produce a very high-resolution mosaic
(true colour) of the Saruq Al-Hadid site and surrounding areas. Image visual analysis
and interpretation were performed after localisation of the site based on the produced
orthomosaic. Figure 6shows a Worldview-3 red-green-blue (RGB) orthomosaic of the
Saruq Al-Hadid site located in the mobile dune fields of the northeastern edge of the Rub’
Al-Khali desert, captured on 26 November 2019. The orthomosaic is colour-balanced within
the region tiles to minimise the colour differences due to side-to-side shading, contrast
variations, time of day, sun angle, atmospheric conditions such as haze, and the use of
multiple cameras over multiple days among adjacent region tiles.
Previous and ongoing excavations at several of the Saruq Al-Hadid areas, as reported
by [
26
], were located on the multispectral images (Appendix A—Figure A1). Appendix A—
Figure A2a–c shows some remarkable spots and locations identified in the western zone
of observation of the site based on the produced orthomosaic. Appendix A—Figure A3
shows some remarkable locations in the eastern zone of observation.
3.1.1. Feature Mark Detection Using Principal Component Analysis (PCA) and Spectral
Band Transformation
Prior to classifying the site thematically, principal component analysis (PCA) was
performed. PCA is one of the most conventional unsupervised feature extraction methods
which extracts features with the largest power [
35
]. PCA discards the components of data
with small variance, while components with small variance may have useful information
for discrimination between classes in the classification process. In other words, it helps in
finding a low-dimension set of axes that summarise data so that the informative compo-

Geosciences 2023,13, 179 8 of 34
nents for classification are extracted instead of being discarded. The produced principal
components (PC) are shown in Appendix B—Figure A4. Statistical investigation confirmed
that most information is contained in the first three PCs (Appendix B—Figure A5).
Geosciences 2023, 13, x FOR PEER REVIEW 8 of 34
(a) (b)
Figure 6. Worldview-3 colour-balanced RGB orthomosaic of the Saruq Al-Hadid site located in the
mobile dune fields of the northeastern edge of the Rub’ Al-Khali desert, captured on 26 November
2019. (a) Western zone of observation. (b) Eastern zone of observation.
3.1.1. Feature Mark Detection Using Principal Component Analysis (PCA) and Spectral
Band Transformation
Prior to classifying the site thematically, principal component analysis (PCA) was
performed. PCA is one of the most conventional unsupervised feature extraction methods
which extracts features with the largest power [35]. PCA discards the components of data
with small variance, while components with small variance may have useful information
for discrimination between classes in the classification process. In other words, it helps in
finding a low-dimension set of axes that summarise data so that the informative compo-
nents for classification are extracted instead of being discarded. The produced principal
components (PC) are shown in Appendix B—Figure A4. Statistical investigation con-
firmed that most information is contained in the first three PCs (Appendix B—Figure A5).
The site contains drought-tolerant floral species. Indeed, the site has Prosopis cineraria
trees, also known as Ghaf. Ghaf are particularly present within the interdunal area and
are used as biological indicators to support the geospatial analysis in delineating the geo-
morphological classes. At least 12 Ghaf trees are contained within the study area (Figure
6) and more are present within the surrounding area for which the geospatial modelling
is generalised. In addition, relatively denser vegetation can be observed through the
growth of shrubs across more than 17 species such as Rubiaceae, Solanaceae, Chenopodiaceae,
and Ochradenus arabicus and more than 17 herbs and grass species such as Neuradaceae,
Poaceae, and Cyperaceae, depending on the season (assuming the vegetation presence is
more suitable to the adopted analytical and modelling approaches). Therefore, spectral
band transformation was performed to produce indices such as the crop vegetation soil
index, the normalised difference vegetation index (NDVI), and the normalised high-fre-
quency difference index (NHFD) for archaeological prospection and detection of anoma-
lies related to buried archaeological structures [36] (Appendix C—Figure A6). Both NHFD
and NDVI bands provide valuable information for archaeological remote sensing, but
Figure 6.
Worldview-3 colour-balanced RGB orthomosaic of the Saruq Al-Hadid site located in the
mobile dune fields of the northeastern edge of the Rub’ Al-Khali desert, captured on 26 November
2019. (a) Western zone of observation. (b) Eastern zone of observation.
The site contains drought-tolerant floral species. Indeed, the site has Prosopis cineraria
trees, also known as Ghaf. Ghaf are particularly present within the interdunal area and
are used as biological indicators to support the geospatial analysis in delineating the geo-
morphological classes. At least 12 Ghaf trees are contained within the study area (Figure 6)
and more are present within the surrounding area for which the geospatial modelling
is generalised. In addition, relatively denser vegetation can be observed through the
growth of shrubs across more than 17 species such as Rubiaceae,Solanaceae,Chenopodiaceae,
and Ochradenus arabicus and more than 17 herbs and grass species such as Neuradaceae,
Poaceae, and Cyperaceae, depending on the season (assuming the vegetation presence is more
suitable to the adopted analytical and modelling approaches). Therefore, spectral band
transformation was performed to produce indices such as the crop vegetation soil index,
the normalised difference vegetation index (NDVI), and the normalised high-frequency
difference index (NHFD) for archaeological prospection and detection of anomalies re-
lated to buried archaeological structures [
36
] (Appendix C—Figure A6). Both NHFD and
NDVI bands provide valuable information for archaeological remote sensing, but they
serve different purposes. NHFD emphasises the detection of archaeological structures and
materials based on their spectral characteristics, while NDVI focuses on identifying areas of
interest based on vegetation patterns. Integrating the techniques diagrammed in Figure 3
is expected to enhance the effectiveness of the adopted approach. In fact, these techniques
were demonstrated to be effective in allowing researchers e.g., [
37
–
39
], to identify potential
sites and prioritise areas for ground-based investigations more efficiently.

Geosciences 2023,13, 179 9 of 34
3.1.2. Multimodal Analysis: Classification and Geocontextualisation
In the absence of user-defined training classes, unsupervised classification was per-
formed to cluster pixels in a dataset based on statistics using only the ISODATA algorithm.
ISODATA unsupervised classification calculates class means evenly distributed in the
data space, then iteratively clusters the remaining pixels using minimum distance tech-
niques [
40
]. Each iteration recalculates means and reclassifies pixels with respect to the
new means. This process continues until the number of pixels in each class changes by less
than the selected pixel change threshold or the maximum number of iterations is reached.
Classification aggregation was used to aggregate smaller adjacent class regions to a larger
region. Post-processing, including smoothing and aggregation, was performed to improve
the classification results. Aggregation is a useful post-classification clean-up process when
the classification output includes many small regions, which is true for the studied area.
Classification of the first three PC (PC1, PC2, and PC3) after enhancement and aggregation
was performed, including various classes (Figure 7a). Analysis and investigation of the
different results have led to the identification of two major categories of classes: “Water
classes”—Classes 1 to 3; and “Excavation classes”—Classes 20 to 22 (Figure 7b). This
unsupervised classification is complemented with multimodal (geostatistical, geophysical,
and ground truth) data, in addition to user expertise and familiarity with the study area for
more advanced contextual re-classification in Section 3.3.4.
Multimodal analysis and user expertise [
41
] were used for studying local relief gra-
dients in the context of geophysical field verification. Local relief gradients refer to the
variations in elevation or slope within a specific area. By applying multivariate statistical
techniques, insights into the spatial patterns and relationships between different relief
features are gained, which proved useful in archaeological investigations [
42
]. The multi-
variate statistical method applied to local relief gradients for field verification included the
following steps.
•Data collection
Data on local relief gradients within the study area were collected. This involved the
use of surveying equipment such as handheld Global Positioning System (GPS) receivers
(Garmin 62S GPS, Garmin International, Olathe, Kansas, USA—~3 m horizontal resolution
95% of the time) and data from the existing topographic grids set by the site archaeologists
e.g., [
43
], using total stations to measure elevation and slope at multiple points across
the landscape.
•Data pre-processing
This involved organising the data into a structured form, a matrix where each row
represents a sampling point and each column represents a variable (e.g., elevation, slope,
aspect). Additional qualitative attributes, such as vegetation cover or soil characteristics,
were also included where relevant.
•Exploratory data analysis
Before applying multivariate statistical techniques, data were explored to understand
their characteristics. This involved visualising the data using scatterplots and contour
maps to identify any outliers, trends, or patterns, e.g., dune patterns. Exploratory data
analysis helped in identifying potential relationships between variables, e.g., Ghaf trees
and dune ridges.

Geosciences 2023,13, 179 10 of 34
•Multivariate statistical analysis
The multivariate statistical technique applied to analyse local relief gradients was the
PCA and cluster analysis, i.e., ISODATA. These techniques helped to identify underlying
patterns (e.g., dune patterns), group similar areas (e.g., metal working slags), differentiate
between different relief types, and assess the spatial relationships between variables.
•Interpretation and field verification
Once the multivariate statistical analysis was performed, the results were interpreted
in the context of the research objectives, i.e., archaeological guidance. The identified
patterns and relationships could provide insights into the landscape’s geomorphological
processes, human activities, and site geological formation processes (Barzaman Formation
(Miocene) or the Quaternary fluvial sandstones, conglomerates, and carbonate sandstones
of the Hili Formation, etc.). Field verification was crucial to confirm the interpretations
made based on geostatistical analysis. However, this verification is limited by natural
spatial heterogeneity with diverse and dynamic relief features (e.g., sand dunes) and by
data collection challenges such as the inevitable discrepancy in resolution of the multisource
data (e.g., the difference among the remote sensing data and the field observations’ spatial
resolutions). This challenge is expected to be remedied when differential GPS is deployed
during future site visits.
•Integration with other data sources
Multivariate analysis on local relief gradients was combined with other geospatial data
sources such as multispectral and radar imagery, GIS data, and reported archaeological
artefact distributions and historical records. Integrating multiple data sets provides a more
comprehensive understanding of the landscape and facilitates the identification of potential
archaeological sites or features [44].
Despite the limitations stated above, by applying this integrative and multimodal anal-
ysis, better quantitative and spatial understandings of the landscape’s topographic charac-
teristics are gained. When combined with further geospatial analysis
(Sections 3.2 and 3.3),
these insights are expected to formulate more informed research strategies for further
archaeological investigations, thereby fulfilling the main objective of the study.
Geosciences 2023, 13, x FOR PEER REVIEW 10 of 34
• Multivariate statistical analysis
The multivariate statistical technique applied to analyse local relief gradients was the
PCA and cluster analysis, i.e., ISODATA. These techniques helped to identify underlying
paerns (e.g., dune paerns), group similar areas (e.g., metal working slags), differentiate
between different relief types, and assess the spatial relationships between variables.
• Interpretation and field verification
Once the multivariate statistical analysis was performed, the results were interpreted
in the context of the research objectives, i.e., archaeological guidance. The identified pat-
terns and relationships could provide insights into the landscape’s geomorphological pro-
cesses, human activities, and site geological formation processes (Barzaman Formation
(Miocene) or the Quaternary fluvial sandstones, conglomerates, and carbonate sandstones
of the Hili Formation, etc.). Field verification was crucial to confirm the interpretations
made based on geostatistical analysis. However, this verification is limited by natural spa-
tial heterogeneity with diverse and dynamic relief features (e.g., sand dunes) and by data
collection challenges such as the inevitable discrepancy in resolution of the multisource
data (e.g., the difference among the remote sensing data and the field observations’ spatial
resolutions). This challenge is expected to be remedied when differential GPS is deployed
during future site visits.
• Integration with other data sources
Multivariate analysis on local relief gradients was combined with other geospatial
data sources such as multispectral and radar imagery, GIS data, and reported archaeolog-
ical artefact distributions and historical records. Integrating multiple data sets provides a
more comprehensive understanding of the landscape and facilitates the identification of
potential archaeological sites or features [44].
Despite the limitations stated above, by applying this integrative and multimodal
analysis, beer quantitative and spatial understandings of the landscape’s topographic
characteristics are gained. When combined with further geospatial analysis (Sections 3.2
and 3.3), these insights are expected to formulate more informed research strategies for
further archaeological investigations, thereby fulfilling the main objective of the study.
(a)
Figure 7. Cont.

Geosciences 2023,13, 179 11 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 11 of 34
(b)
Figure 7. Thematic classification. (a) Unsupervised classification of PC1PC2PC3 (enhanced/aggre-
gated) followed by smoothing using smooth kernel size = 3 pixels. (b) Aggregation and class rela-
belling. Aggregate minimum size = 9 regions. Classes 1 to 3 = water classes. Classes 20 to 22 = exca-
vation classes.
3.2. Processing and Analysis of Synthetic Aperture Radar Data
3.2.1. Pre-processing for Data Calibration and Intensity Processing
ENVI-integrated SARscape tools were used for the SAR image visualisation, analysis,
and processing, such as digital elevation model (DEM) generation. Looks are the sub-im-
ages formed during SAR processing. Pre-processing for data calibration was performed
to obtain magnitude data. Intensity processing of SLC data (having a rectangular shape
of pixels) was performed. A multi-look operator was used with a window of m = 5 pixels
in row and n = 1 pixel in column, thus transforming the product into a more familiar geo-
metric visualisation. ENVI produces an intensity image of each input file with four range
looks and nine azimuth looks. The range resolution is 24.550900 m, and the azimuth res-
olution is 22.771800 m.
3.2.2. Image Speckle Filtering
The use of advanced image processing is central to the utilisation of any remotely
sensed data for archaeological studies. It is unrealistic to define features at the pixel scale
in undespeckled imagery; statistically based algorithms can accomplish this despeckling
with minimal loss of detail. SAR images are characterised by speckle. Speckle is a spatially
random multiplicative noise due to coherent superposition of multiple backscaer sources
within a SAR resolution element [7]. Image speckle filtering was conducted to correct the
speckle potential to corrupt the polarimetric observables (phase and intensity). A specific
procedure, Lee adaptive filter [45,46], was used to retrieve relevant polarimetric infor-
mation and to reduce the randomness of the acquired signals. Lee’s filter determines the
unspeckled intensity estimate that minimises the mean squared error (MMSE). This
MMSE filter is based on a linearised speckle model. The Lee filtering results are shown in
Appendix D—Figure A7a.
Figure 7.
Thematic classification. (
a
) Unsupervised classification of PC1PC2PC3 (en-
hanced/aggregated) followed by smoothing using smooth kernel size = 3 pixels. (
b
) Aggregation
and class relabelling. Aggregate minimum size = 9 regions. Classes 1 to 3 = water classes. Classes 20
to 22 = excavation classes.
3.2. Processing and Analysis of Synthetic Aperture Radar Data
3.2.1. Pre-Processing for Data Calibration and Intensity Processing
ENVI-integrated SARscape tools were used for the SAR image visualisation, analysis,
and processing, such as digital elevation model (DEM) generation. Looks are the sub-
images formed during SAR processing. Pre-processing for data calibration was performed
to obtain magnitude data. Intensity processing of SLC data (having a rectangular shape
of pixels) was performed. A multi-look operator was used with a window of m = 5 pixels
in row and n = 1 pixel in column, thus transforming the product into a more familiar
geometric visualisation. ENVI produces an intensity image of each input file with four
range looks and nine azimuth looks. The range resolution is 24.550900 m, and the azimuth
resolution is 22.771800 m.
3.2.2. Image Speckle Filtering
The use of advanced image processing is central to the utilisation of any remotely
sensed data for archaeological studies. It is unrealistic to define features at the pixel scale
in undespeckled imagery; statistically based algorithms can accomplish this despeckling
with minimal loss of detail. SAR images are characterised by speckle. Speckle is a spatially
random multiplicative noise due to coherent superposition of multiple backscatter sources
within a SAR resolution element [
7
]. Image speckle filtering was conducted to correct
the speckle potential to corrupt the polarimetric observables (phase and intensity). A
specific procedure, Lee adaptive filter [
45
,
46
], was used to retrieve relevant polarimetric
information and to reduce the randomness of the acquired signals. Lee’s filter determines
the unspeckled intensity estimate that minimises the mean squared error (MMSE). This
MMSE filter is based on a linearised speckle model. The Lee filtering results are shown in
Appendix D—Figure A7a.
3.2.3. Digital Elevation Model Extraction and Geocoding
SAR systems measure the intensity and phase of the transmitted radar pulses follow-
ing their reflection (backscatter) from the Earth’s surface. The data are recorded in a 2D

Geosciences 2023,13, 179 12 of 34
coordinate system (slant-range geometry) onto which 3D objects on the Earth’s surface
are projected. Geocoding of the SAR image was performed, i.e., 2D SAR coordinates
were associated to 3D coordinates in given horizontal and vertical datums. Furthermore,
radiometric calibration was conducted to make the radar intensity independent of the ac-
quisition geometry and of the SAR processor. Co-registration was applied to the multiband
sensor data layers. Devolution of mixed pixels into ground and vegetation components
was also performed. Geocoding and radiometric calibration of the SAR data were per-
formed using the global digital surface model (DSM) (horizontal resolution 1 arcsec) by
the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) (30 m
ground resolution and 5 m elevation accuracy) (Appendix D—Figure A7b). The produced
DEM (30 m resolution) is used to better understand the environmental context and to create
the slope map of the study site.
3.2.4. Creating the Slope Map from the Digital Elevation Model
Given the resulting 30 m DEM (Appendix D—Figure A8), the slope is calculated
in degrees (
◦
) in the easting and northing directions (Appendix D—Figure A9a). The
output product is a floating complex datum where the real and imaginary parts are two
floating format files containing the easting and northing slope components, respectively
(Appendix D—Figure A9b,c, respectively). The slope and elevation contribute as input data
in predictive modelling as part of the subsequent geospatial analysis to assist in locating
potential archaeological features.
3.2.5. Single-Date Feature Extraction
Single-date features, based on first-order statistics, can be derived from SAR intensity
data. Depending on the targeted product, these features facilitate detection and extrac-
tion of structures, which can be additionally used for segmentation and/or classification
purposes. In desert regions, SAR’s properties, such as the transmissivity of dry sand to mi-
crowave wavelengths, the sensitivity of radar to roughness, and micro relief, are exploited
for archaeological prospection [
12
,
33
]. The feature extraction uses one polarisation at a time
(Exelis, 2016). The extracted features are represented using RGB composite showing three
different values on the same image, assigning each one to a different channel (Figure 8).
The standard deviation index (Std) is assigned to the red channel, the minimum index
(Min) to the green channel, and the gradient index (Grad) (maximum absolute variation
between consecutive acquisition dates) to the blue channel. Features of higher backscatter
in comparison to the surrounding sand are present in both the western and eastern zones.
As explained in Section 3.1.2, during the field survey the types of sediments were identified
using lithostratigraphic data of previous archaeological reports. Areas of archaeological
artefacts and remains are differentiated from areas covered by modern constructions. A
comparison of these intermediate results with the optical imagery identifies the high rela-
tive backscatter features of the manmade road that separates both zones and the excavation
camps. The backscatter properties of artificial linear objects (roads, excavation camps)
and of known natural linear objects (sand dune ridges) are similar. This is in conformity
with results communicated by other works such as the feature extraction research for
archaeological application conducted by Stewart et al. [
33
] in the North Sinai Desert. In the
literature e.g., [
47
–
49
], features of low relative backscatter possibly correspond to archaeo-
logical structures such as buried valleys, geologic structures, and possible prehistoric age
occupation sites. Analysis of SAR data is complemented by geospatial analysis to further
assess this prediction.

Geosciences 2023,13, 179 13 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 13 of 34
Figure 8. RGB composite is made of the standard deviation, minimum, and gradient indices pro-
duced during the single-date feature extraction of the area visualised in Figure 6. The vertical align-
ment of bright areas corresponds to the road that separates the western zone from the eastern zone
of the study area. The area encircled in yellow represents an approximate zonation of previous and
ongoing excavations.
3.2.6. Post-processing of Synthetic Aperture Radar Data
The main objective of this task was to make archaeological features and sites more
visible in imagery to facilitate the generalisation of the approach to similar environments
for the detection of archaeological sites while simplifying and reducing field surveys and
efforts. The main archaeological features that are expected to be detected within the stud-
ied area and similar environments are sludges and metal crafts remains of ~15 m × 15 m
surface appearance, wells stone-built orifices of ~1 m diameter, and remains of stone-built
structures of irregular ~2 m × 3 m surface with undefined paerns [25] (Weeks et al., 2017).
The selected high-sensitive fully polarised SAR data of ~4.3 m resolution (6 m Rg × 4.3 m
Az) and 50~70 km swath are well suited to revealing such archaeological features and
structures [50,51].
In order to achieve this objective, different combinations of the different polarised
bands (HH, HV, VH and VV) [50] (Figure 9) and of the different extracted features (Section
3.2.5) were examined with reference to the content displayed in appendix A—Figure A1
to visually explore the site and its surrounding environment. The best combination is
shown in Figure 10, when L-band imagery composite was produced from the three polar-
ised bands (HH, VV, and HH) displaying the majority of the locations explored in Section
3.1. Using this band combination, aempts were made to extend the search area to show
new potential archaeological sites to help in guiding the excavation in Saruq Al-Hadid.
There are, however, great impracticalities in utilising this “trained eye” approach, the
main issue being that not all eyes are equally trained. Moreover, an image may or may not
reveal site locations acceptably depending upon the peculiarities of the monitoring dis-
play. Brightness is a relative term and is unquantified in the “trained eye” approach.
Consequently, geospatial analyses of the multispectral and SAR data were performed
using AI and self-supervised ML techniques for the classification and multivariate clus-
tering of the data in order to understand the spatial distribution of the different elements
extracted by the GIS process. Results are described and discussed in the next section.
Figure 8.
RGB composite is made of the standard deviation, minimum, and gradient indices produced
during the single-date feature extraction of the area visualised in Figure 6. The vertical alignment
of bright areas corresponds to the road that separates the western zone from the eastern zone of
the study area. The area encircled in yellow represents an approximate zonation of previous and
ongoing excavations.
3.2.6. Post-Processing of Synthetic Aperture Radar Data
The main objective of this task was to make archaeological features and sites more
visible in imagery to facilitate the generalisation of the approach to similar environments
for the detection of archaeological sites while simplifying and reducing field surveys and
efforts. The main archaeological features that are expected to be detected within the studied
area and similar environments are sludges and metal crafts remains of ~15 m
×
15 m surface
appearance, wells stone-built orifices of ~1 m diameter, and remains of stone-built structures
of irregular ~2 m
×
3 m surface with undefined patterns [
25
] (Weeks et al., 2017). The
selected high-sensitive fully polarised SAR data of ~4.3 m resolution (6 m Rg
×
4.3 m Az)
and 50~70 km swath are well suited to revealing such archaeological features and
structures [50,51].
In order to achieve this objective, different combinations of the different polarised
bands (HH, HV, VH and VV) [
50
] (Figure 9) and of the different extracted features
(Section 3.2.5) were examined with reference to the content displayed in appendix A—
Figure A1 to visually explore the site and its surrounding environment. The best combi-
nation is shown in Figure 10, when L-band imagery composite was produced from the
three polarised bands (HH, VV, and HH) displaying the majority of the locations explored
in Section 3.1. Using this band combination, attempts were made to extend the search
area to show new potential archaeological sites to help in guiding the excavation in Saruq
Al-Hadid. There are, however, great impracticalities in utilising this “trained eye” approach,
the main issue being that not all eyes are equally trained. Moreover, an image may or may
not reveal site locations acceptably depending upon the peculiarities of the monitoring
display. Brightness is a relative term and is unquantified in the “trained eye” approach.

Geosciences 2023,13, 179 14 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 14 of 34
Figure 9. The different post-processed polarised bands. (a) HH. (b) HV. (c) VH. (d) VV.
Figure 10. RGB composite is made of the HH-VV-HH band combination. Area encircled in yellow
represents previous and ongoing excavations. Area encircled in red represents areas with no infor-
mation provided to date on excavation works.
3.3. Geospatial Analysis of Multispectral and Radar Data Using AI and ML
The main objective of this analysis was to create spatial statistics to describe and
model the spatial distribution, paerns, processes, and relationships among the various
spatial elements of the archaeological site. The results and findings are intended to be
generalised to a larger area to predict and guide new potential archaeological sites and
activities. The observations collected over Saruq Al-Hadid were clustered based on
Figure 9. The different post-processed polarised bands. (a) HH. (b) HV. (c) VH. (d) VV.
Geosciences 2023, 13, x FOR PEER REVIEW 14 of 34
Figure 9. The different post-processed polarised bands. (a) HH. (b) HV. (c) VH. (d) VV.
Figure 10. RGB composite is made of the HH-VV-HH band combination. Area encircled in yellow
represents previous and ongoing excavations. Area encircled in red represents areas with no infor-
mation provided to date on excavation works.
3.3. Geospatial Analysis of Multispectral and Radar Data Using AI and ML
The main objective of this analysis was to create spatial statistics to describe and
model the spatial distribution, paerns, processes, and relationships among the various
spatial elements of the archaeological site. The results and findings are intended to be
generalised to a larger area to predict and guide new potential archaeological sites and
activities. The observations collected over Saruq Al-Hadid were clustered based on
Figure 10.
RGB composite is made of the HH-VV-HH band combination. Area encircled in yel-
low represents previous and ongoing excavations. Area encircled in red represents areas with no
information provided to date on excavation works.
Consequently, geospatial analyses of the multispectral and SAR data were performed
using AI and self-supervised ML techniques for the classification and multivariate clus-
tering of the data in order to understand the spatial distribution of the different elements
extracted by the GIS process. Results are described and discussed in the next section.
3.3. Geospatial Analysis of Multispectral and Radar Data Using AI and ML
The main objective of this analysis was to create spatial statistics to describe and model
the spatial distribution, patterns, processes, and relationships among the various spatial

Geosciences 2023,13, 179 15 of 34
elements of the archaeological site. The results and findings are intended to be generalised
to a larger area to predict and guide new potential archaeological sites and activities. The
observations collected over Saruq Al-Hadid were clustered based on similarities of values.
Then, they were used to generate a model using PCA. Five geospatial analysis procedures
were performed to create the model pattern as described in Figure 4.
3.3.1. Input GIS Data Extraction
To study the entities that characterise the study site, spatial objects such as topography,
slope, soil, and hydrography were extracted. These objects represent the model’s inputs
and are defined by their spatial variation.
•Elevation
To understand the topography and improve the visualisation of the terrain, a triangu-
lated irregular network (TIN) with three-dimensional coordinates x, y, and z was extracted
from the DEM generated in Section 3.2.3. Appendix E—Figure A10 shows an elevation
map of the study area. The elevation within the Saruq Al-Hadid site and surroundings was
found to vary between 80 and 130 m.
•Slope
Weeks et al. [
25
] reported excavations faced with challenges mainly due to the instabil-
ity of the sand matrix in the study site. Excavations have therefore proceeded by stepping
or terracing trenches, and only the trenches located towards the centre have been excavated
to the basal gypsum layer that underlies all deposits in the central sector of the site. It
could thus be deduced that the landscape slope plays a crucial role in the stability of the
dune flanks which, due to aeolian processes (e.g., erosion, transportation, and deposition
of sediment by the wind), expose or bury the archaeological features. Due to its importance
in the historical establishment of the habitat, the classification of the slope was carried out
from the DEM of the site. The study area was found to be characterised by a moderate
slope varying between 1 and 15◦(Appendix E—Figure A11).
•Natural resource assessment
Assessing the presence of water resources and features is essential to any studied
environment. Following the unsupervised classification which led to the detection of water
classes in Section 3.1.2, traces of the historic hydrographic network were mapped. The
process of the hydrographic network extraction from the DEM pixels was automated based
on the flow routing model [
52
]. In this method, pixels are centred on the DEM grid points,
and each pixel discharges into one of its eight neighbours: the one located in the direction of
the steepest descent. The hydrographic network was extracted using ArcGIS Pro hydrology
tools (ESRI 2022), and the process was automated to extract streams from the 30 m DEM
using ArcGIS model builder (ESRI 2022). Results from the hydrographic network mapping
show a stream order of types 2 and 3, i.e., immediate tributaries of the main stem (type 2)
and tributaries emptying into the type 2 (type 3) [53] (Appendix E—Figure A12).
•Dune Pattern Detection Using DL and Convolutional Neural Networks
The shape variation of the dunes is often due to climatic events. It can be an indication
of an archaeological trace if an immediate change of shape is detected in a series of homo-
geneous dunes [
54
]. An AI method for the detection of the shape of the dunes was applied
using convolutional neural networks (CNN) to classify the acquired multispectral data
(Figure 11). DL and CNNs have been exploited in SAR-guided archaeological research to
extract valuable information from images and aid in the field, as reviewed by Argyrou and
Agapiou [
55
]. In this study, CNNs were used for semantic segmentation classification of
dunes in areas where urban features are absent. The assignment of every pixel in the 30 cm
multispectral data is based on the categorisation “dune” or “not dune”. For dune recogni-
tion, the CNN architecture used is U-Net to perform semantic segmentation that not only
requires discrimination at pixel level but also a mechanism to project the discriminative

Geosciences 2023,13, 179 16 of 34
features learnt at different stages of the encoder onto the pixel space [
56
]. The model was
created using Keras, a DL application programming interface (API) written in Python run-
ning on an NVIDIA A40 graphics processing unit (GPU) installed on a high-performance
computing (HPC) system. The dataset used for this modelling is composed of 120 RGB
images of dune samples and semantic labels over the UAE desert; 96 images (80% of the
whole dataset) were used to train the model and 24 images (20%) were used to test it. The
loss function is set to binary cross entropy, which is appropriate for binary classification
problems with a sigmoid activation function. Additionally, the model is configured to
track the BinaryAccuracy metric, which calculates the accuracy of the binary classification
predictions. The results were evaluated by experimenting with the model according to
several numbers of epochs, but after 30 epochs the validation accuracy, with 0.16 loss
and 0.95 binary accuracy, was not improving. Based on the dune pattern estimation, the
detection of dune variation within the study site could be automated (Figure 11).
Geosciences 2023, 13, x FOR PEER REVIEW 16 of 34
Python running on an NVIDIA A40 graphics processing unit (GPU) installed on a high-
performance computing (HPC) system. The dataset used for this modelling is composed
of 120 RGB images of dune samples and semantic labels over the UAE desert; 96 images
(80% of the whole dataset) were used to train the model and 24 images (20%) were used
to test it. The loss function is set to binary cross entropy, which is appropriate for binary
classification problems with a sigmoid activation function. Additionally, the model is con-
figured to track the BinaryAccuracy metric, which calculates the accuracy of the binary
classification predictions. The results were evaluated by experimenting with the model
according to several numbers of epochs, but after 30 epochs the validation accuracy, with
0.16 loss and 0.95 binary accuracy, was not improving. Based on the dune paern estima-
tion, the detection of dune variation within the study site could be automated (Figure 11).
Figure 11. Dune paern detection using convolutional neural networks (CNN) in the Saruq Al-
Hadid site and surroundings. (a) Input image. (b) Feature extraction and paern classification. (c)
Dune paern map. (d) Automation of dune paern detection.
3.3.2. Geoprocessing: Extracting, Summarising, and Aggregating Geospatial Data
Data points extracted from geoprocessing generated a large dataset (Appendix F—
Figure A13). Using geoprocessing, millions of points were generated, such as points ex-
tracted from hydrographic networks, within CNN-classified dunes, etc. These points con-
tain information about each pixel (i.e., soil, elevation, location within or outside hydro-
graphic streams, dune areas, etc.) to be used for clustering. There were undetected pat-
terns within this dataset with no pre-existing labels and with a minimum of human
Figure 11.
Dune pattern detection using convolutional neural networks (CNN) in the Saruq Al-Hadid
site and surroundings. (
a
) Input image. (
b
) Feature extraction and pattern classification. (
c
) Dune
pattern map. (d) Automation of dune pattern detection.
3.3.2. Geoprocessing: Extracting, Summarising, and Aggregating Geospatial Data
Data points extracted from geoprocessing generated a large dataset (Appendix F—
Figure A13). Using geoprocessing, millions of points were generated, such as points
extracted from hydrographic networks, within CNN-classified dunes, etc. These points
contain information about each pixel (i.e., soil, elevation, location within or outside hydro-

Geosciences 2023,13, 179 17 of 34
graphic streams, dune areas, etc.) to be used for clustering. There were undetected patterns
within this dataset with no pre-existing labels and with a minimum of human supervi-
sion. To overcome these issues, cluster analysis using the K-means++ cluster unsupervised
ML algorithm [
57
] was performed. To determine the optimal number of clusters in the
geospatial dataset, the elbow method technique was used. It serves as a useful heuristic
to guide the selection process and provide insights into the structure of the data [
58
]. The
chosen clustering algorithm, K-means, was applied to the dataset. K-means partitions the
data into a specified number of clusters (K), ranging from a minimum value to a maxi-
mum value. For each value of the number of clusters, the within-cluster sum of squares
(WCSS) clustering quality metric, also known as the inertia, is calculated. It measures
the compactness or coherence of the data points within each cluster. A line plot is then
created where the x-axis represents the number of clusters and the y-axis represents the
WCSS (Appendix F—Figure A14a). The plot is then examined, and the point where the
decrease in the WCSS begins to level off, forming an “elbow” shape, is identified. The
elbow point 5 was selected to represent the number of clusters that provides a good balance
between maximising the similarity within clusters and minimising the similarity between
clusters (Appendix F—Figure A14). Adding more clusters may not provide substantial
improvement of the WCSS, while excessively increasing the number of K clusters can lead
to overfitting and reduced interpretability.
3.3.3. Geostatistical Analysis
The main purpose of this step was to calculate the covariances between the different
mined variables, i.e., unsupervised classes, radar polarisations, elevation, slope, hydrologic
resources, and dune pattern. The covariances were calculated, and the main components
were extracted using PCA. The order of PCs was defined based on the calculation of the
probability of the independent variables (IV). After decomposition using Kernel PCA, a
non-linear dimensionality reduction through the use of kernels [
59
] of the ML library for
the Python programming language scikit-learn, probability 0.369 and 0.193 defined the
main PCs (Table 1). The other PCs were considered secondary.
Table 1.
Covariance of independent variables calculated using the ML Kernel PCA decomposition
algorithm.
1 2 3 4 5 6 7 8 9 10 11 12
Probability of IV 0.369 0.193 0.108 0.074 0.062 0.049 0.041 0.025 0.023 0.019 0.017 0.013
3.3.4. Reclassification Using Multimodal Data and User Expertise
Throughout the study area made of the soil formations described in Section 2.2, there
are occasional small tufts of vegetation, and to the east, south, and west of the compound
there are very occasional Ghaf trees (Figures 2a and 6). In terms of visible reflectance:
(1) at canopy level, reflectance results from various factors, such as vegetation chemical
properties, leaf morphology, canopy structure, and tree sizes [
60
–
62
]; and (2) at soil level, it
is a differentiating characteristic for many classes and is an essential part of the definitions
for both surface and subsurface diagnostics. Main factors influencing the reflectance for bare
soils are roughness and texture, organic matter content, and moisture conditions [63,64].
Reclassification of the study site was conducted using the fusion of multispectral, band
ratio (e.g., soil and vegetation bands), GIS (e.g., slope), derived geostatistical, lithological,
and hydrographic data, along with user expertise and familiarity with the site’s geomor-
phological context and setting. This multimodal analysis led to the distinction between
three main different geomorphological assemblages.
1.
Geomorphological assemblage (I) is represented by classes 4 to 10 of the unsupervised
classification presented in Section 3.1 (Figure 7a), which correspond to areas related
to manmade constructions’ generated shadows such as areas of light intercepted
and blocked by tents and buildings; and to vegetation such as Ghaf trees and dune

Geosciences 2023,13, 179 18 of 34
ridges. This geomorphological assemblage consists of sand dune formations with
shaded ridges facing north and containing relatively more humidity than the general
surroundings, in addition to classes 1 to 3 that were identified as water classes in
Section 3.1 (Figure 7b). The multimodal data reclassification regrouped these features
into “Geo assemblage I” (Figure 12).
Geosciences 2023, 13, x FOR PEER REVIEW 19 of 34
Figure 12. Reclassification of the study site, based on its landscape context and multimodal data,
into three main geomorphological assemblages. The area of field survey and verification contained
the reported main excavation sectors and surroundings. Based on the field surveys, three site loca-
tions were elected for future archaeological investigation: A (55°14′11.616″ E 24°39′52.58″ N), B
(55°14′18.603″ E 24°39′51.519″ N), and C (55°14′28.123″ E 24°39′51.499″ N). To the best knowledge of
the authors as of the current study time frame, there are no reports on excavation works within the
area contained in the eastern zone of the study area.
3.3.5. Paern Modelling
The penultimate step was the paern model definition. Given the different statistical
distributions of the input elements, PCA and gaussian modelling were chosen. This re-
sulted in a nonlinear model for the precision of the selection of groups. To solve the non-
linear model, the gaussian radial basis function (RBF) kernel was used [67].
This allowed us to iterate the process several times in order to optimise the paern
modelling. The unsupervised ML technique using Kernel PCA was able to detect the ex-
cavation areas that had already been reported. The paern modelling was employed to
predict areas of potential archaeological significance, including previously reported exca-
vation sectors and ground-surveyed areas (Figure 13a) as well as areas contained in the
eastern zone which were identified in the previous steps of the integrated process (Figures
A3 and 13b).
Figure 12.
Reclassification of the study site, based on its landscape context and multimodal data,
into three main geomorphological assemblages. The area of field survey and verification contained
the reported main excavation sectors and surroundings. Based on the field surveys, three site
locations were elected for future archaeological investigation: A (55
◦
14
0
11.616
00
E 24
◦
39
0
52.58
00
N), B
(55
◦
14
0
18.603
00
E 24
◦
39
0
51.519
00
N), and C (55
◦
14
0
28.123
00
E 24
◦
39
0
51.499
00
N). To the best knowledge
of the authors as of the current study time frame, there are no reports on excavation works within the
area contained in the eastern zone of the study area.
2.
Geomorphological assemblage (II) is represented by classes 10 to 15 of the unsuper-
vised classification provided in Section 3.1 (Figure 7a), corresponding to gypsum
pavement at the base of the excavated sequences (Figure 2b), sporadic drought-
tolerant vegetation in interdune areas such as shrubs and bushes, sand veneer, low
dunes, and the exposed formations at the subsurface; the geological substratum of

Geosciences 2023,13, 179 19 of 34
the Barzaman and/or Hili sandstones to siltstones formations. After reclassification,
these structures were identified as “Geo assemblage II” (Figure 12).
3.
Geomorphological assemblage (III) is represented by classes 16 to 20 of the unsuper-
vised classification in Section 3.1 (Figure 7a), corresponding on the ground to metal
working slags. Indeed, these features were detected during the field survey’s direct
observations and autopsies and reported by previous geophysical surveys as the three
main archaeological excavation sectors, i.e.: (1) the three-year Saruq Al-Hadid Ar-
chaeological Research Project (SHARP)’s excavation trenches reported by Cable [
43
];
(2) the excavation sector ongoing since 2019 that was reported by Weeks et al. [
26
];
and (3) the previous excavation sector reported by Weeks et al. [
25
] (Figure A1). This
assemblage also contains the longitudinal sand dune ridges facing southward, in
addition to the excavation’s areas reported in Section 3.1 (Figure 7b) and artificial
infrastructures made of metal or concrete. These geomorphological constituents were
reclassified into “Geo assemblage III” (Figure 12).
This multimodal reclassification was verified based on reports that correspond tem-
porally with the SAR data acquisition period (e.g., Cable (2015)), with the geological and
geomorphological autopsies conducted in October, November, and December 2015, and
with the multispectral data acquisition period, e.g., the pre- and post-archaeological in-
vestigations of Weeks et al. [
25
], Stepanov et al. [
65
], and Valente et al. [
66
] (Figure 12).
Figure 12 shows the reported main excavation sectors extending as part of the field survey
and verification. Based on observations of features of potential archaeological importance
during the field surveys (e.g., metal working slags), three site locations (A, B, and C) were
elected for future archaeological investigation (Figure 12). As of the current pilot study
time frame and to the best knowledge of the authors, there are no reports on excavation
works within the area contained in the eastern zone identified using the developed research
process. Further investigation on this area was conducted using the ML pattern modelling
developed in Sections 3.3.5 and 3.3.6.
Based on the geological and geomorphological user expertise and the familiarity
gained with the study site, multimodal reclassification works best for metal working slags,
archaeological remains including artifacts, and rock blocks in dune ridges, especially in
Geo-assemblage III. As reported in Section 3.1.2, misclassifications were due to differences
in object scale and data resolution; fine-scale variations in relief features and subtle patterns
that cannot be captured by this approach.
3.3.5. Pattern Modelling
The penultimate step was the pattern model definition. Given the different statistical
distributions of the input elements, PCA and gaussian modelling were chosen. This resulted
in a nonlinear model for the precision of the selection of groups. To solve the nonlinear
model, the gaussian radial basis function (RBF) kernel was used [67].
This allowed us to iterate the process several times in order to optimise the pattern
modelling. The unsupervised ML technique using Kernel PCA was able to detect the
excavation areas that had already been reported. The pattern modelling was employed
to predict areas of potential archaeological significance, including previously reported
excavation sectors and ground-surveyed areas (Figure 13a) as well as areas contained in
the eastern zone which were identified in the previous steps of the integrated process
(Figures A3 and 13b).
Recalling the main objective of this study, which is to enhance archaeological research
in Saruq Al-Hadid desert and similar environments through, among other techniques, the
application of a series of successive clustering algorithms (K-means++ and kernel PCA),
the three main excavation sectors reported in the literature were retrieved. Results showed:
(1) larger variance values (>0.041) for these archaeological sites in comparison to the other
locations within the study site, with the largest variance (0.369) corresponding to the area
that was elected, using remote sensing data and throughout the different steps of the
process, to be of potential archaeological importance; and (2) lower variances (<0.019) or

Geosciences 2023,13, 179 20 of 34
rejection from the model corresponding to dune areas or recent constructions identified in
situ or remotely. This proves that the model has worked fairly well for the area of interest
and has the potential to provide further insights into the spatial patterns and relationships
between different relief features, thus aiding in archaeological investigations. Recent site
visits and verification are required and planned to validate these findings as soon as access
to the site is granted by the governing authority.
Geosciences 2023, 13, x FOR PEER REVIEW 20 of 34
Recalling the main objective of this study, which is to enhance archaeological re-
search in Saruq Al-Hadid desert and similar environments through, among other tech-
niques, the application of a series of successive clustering algorithms (K-means++ and ker-
nel PCA), the three main excavation sectors reported in the literature were retrieved. Re-
sults showed: (1) larger variance values (>0.041) for these archaeological sites in compari-
son to the other locations within the study site, with the largest variance (0.369) corre-
sponding to the area that was elected, using remote sensing data and throughout the dif-
ferent steps of the process, to be of potential archaeological importance; and (2) lower
variances (<0.019) or rejection from the model corresponding to dune areas or recent con-
structions identified in situ or remotely. This proves that the model has worked fairly well
for the area of interest and has the potential to provide further insights into the spatial
paerns and relationships between different relief features, thus aiding in archaeological
investigations. Recent site visits and verification are required and planned to validate
these findings as soon as access to the site is granted by the governing authority.
Figure 13. Cluster decomposition and geospatial distribution in the Saruq Al-Hadid site. Through-
out the study area, dunes and recent constructions have low variances (<0.0190). (a) Reported exca-
vations areas and field-elected site locations for archaeological investigation (A, B and C) showed
higher values (>0.0250). (b) The remote-sensing elected area (eastern zone) was clustered using the
highest variance (0.3690).
3.3.6. Paern Prediction and Application to Similar Environments
The final step was to predict and guide archaeological activities and research through
the identification of potential areas for further on-site investigation. This was achieved
based on the ML and DL techniques. With a more precise decomposition of clusters based
Figure 13.
Cluster decomposition and geospatial distribution in the Saruq Al-Hadid site. Throughout
the study area, dunes and recent constructions have low variances (<0.0190). (
a
) Reported excavations
areas and field-elected site locations for archaeological investigation (A, B and C) showed higher
values (>0.0250). (
b
) The remote-sensing elected area (eastern zone) was clustered using the highest
variance (0.3690).
3.3.6. Pattern Prediction and Application to Similar Environments
The final step was to predict and guide archaeological activities and research through
the identification of potential areas for further on-site investigation. This was achieved
based on the ML and DL techniques. With a more precise decomposition of clusters based
on a mathematical representation of geospatial distribution, areas of reported archaeo-
logical significance were identified. Areas of potential significance were predicted. The
prediction of these potential areas is in conformity with the reclassification performed in
Section 3.3.4 (“Geo assemblage III” (Figure 12)). This modelling, based on the succession of
the unsupervised ML analysis steps, can be applied to similar environments for the extrac-
tion of geographic information. Transfer learning, which involves leveraging pre-trained
models on similar tasks or datasets, can help overcome the scarcity of labelled archaeo-
logical data in desert environments. Additionally, data augmentation techniques, such as
image rotation, scaling, etc., can increase the diversity of training data and improve the

Geosciences 2023,13, 179 21 of 34
generalisability of the developed ML model. Additionally, it is expected that data collected
during future geophysical field and ground-truthing surveys will help to further validate
the model-predicted areas. The pattern can then be inferred to a larger extent in the same
environment (Figure 14) and the approach can be applied to similar arid environments,
although with Saruq Al-Hadid being a pilot study for the testing of the efficiency of the
developed approach, limitations were faced due to several factors such as the lack of more
up-to-date in situ data and the heterogeneity in the resolution of the multisource data.
Previous studies show that best results are obtained through the use of a combination of
remotely sensed data sources, i.e., multispectral, thermal, SAR, ground-penetrating radar
(GPR), as supported by field research [37,68–73].
Geosciences 2023, 13, x FOR PEER REVIEW 21 of 34
on a mathematical representation of geospatial distribution, areas of reported archaeolog-
ical significance were identified. Areas of potential significance were predicted. The pre-
diction of these potential areas is in conformity with the reclassification performed in Sec-
tion 3.3.4 (“Geo assemblage III” (Figure 12)). This modelling, based on the succession of
the unsupervised ML analysis steps, can be applied to similar environments for the ex-
traction of geographic information. Transfer learning, which involves leveraging pre-
trained models on similar tasks or datasets, can help overcome the scarcity of labelled
archaeological data in desert environments. Additionally, data augmentation techniques,
such as image rotation, scaling, etc., can increase the diversity of training data and im-
prove the generalisability of the developed ML model. Additionally, it is expected that
data collected during future geophysical field and ground-truthing surveys will help to
further validate the model-predicted areas. The paern can then be inferred to a larger
extent in the same environment (Figure 14) and the approach can be applied to similar
arid environments, although with Saruq Al-Hadid being a pilot study for the testing of
the efficiency of the developed approach, limitations were faced due to several factors
such as the lack of more up-to-date in situ data and the heterogeneity in the resolution of
the multisource data. Previous studies show that best results are obtained through the use
of a combination of remotely sensed data sources, i.e., multispectral, thermal, SAR,
ground-penetrating radar (GPR), as supported by field research [37,68–73].
Figure 14. Extending the search area geographically to guide the archaeological activities and re-
search provided that more recent data are collected for further validation and improvement of the
modelling and prediction accuracies. (a) Area encircled in yellow is at the convergence of previous
and ongoing excavations. (b) Area encircled in red is predicted to contain archaeological features,
as elected by the integrated remote sensing and geospatial research approach. This area is consid-
ered the focus of future investigations.
4. Discussion and Conclusions
Arid regions, such as the Saruq Al-Hadid site, offer a challenging environment for
the detection of historical monuments using ground-visual or physical methods due to
challenging abiotic factors such as high sand dunes. These factors, however, make these
Figure 14.
Extending the search area geographically to guide the archaeological activities and research
provided that more recent data are collected for further validation and improvement of the modelling
and prediction accuracies. (
a
) Area encircled in yellow is at the convergence of previous and ongoing
excavations. (
b
) Area encircled in red is predicted to contain archaeological features, as elected by
the integrated remote sensing and geospatial research approach. This area is considered the focus of
future investigations.
4. Discussion and Conclusions
Arid regions, such as the Saruq Al-Hadid site, offer a challenging environment for
the detection of historical monuments using ground-visual or physical methods due to
challenging abiotic factors such as high sand dunes. These factors, however, make these
environments some of the most promising sites for satellite SAR penetration and appli-
cations in archaeology. These applications were demonstrated in the literature to have
successfully guided archaeological works and research [
12
,
33
,
47
–
49
]. SAR’s potential has
been further evidenced when integrated with ML- and DL-based classification of multitem-
poral data [
20
–
22
]. Advanced image processing techniques and multimodal data analysis
using ML and DL were integrated within the geological context to develop an automated
process and investigate its accuracy before generalising it to larger areas. Results from
SAR feature extraction and geospatial analyses allowed for detection of the areas of the
site that were already under excavation as well as further geomorphological assemblages.

Geosciences 2023,13, 179 22 of 34
The geospatial modelling highlighted anomalies to predict areas of potential archaeological
value. Potential archaeological areas were predicted with the implementation of a more
precise decomposition of clusters based on a mathematical representation of geospatial
distribution. The validation of these results was performed using reported archaeological
research findings that temporally coincide with the used datasets, e.g., [
43
,
65
,
66
], as well
as geological and geomorphological autopsies, along with direct observations conducted
during the field surveys.
While the field verification is valuable [
16
], it was dependent on the scale and reso-
lution of the data; fine-scale variations in relief features and subtle patterns might have
been missed. In addition, it was contingent on contextual factors; local relief gradients
are influenced by a variety of factors, including distinct geomorphological and geological
processes, land use and cover (such as sparse vegetation and limited surface water), and
human activities. The complex interplay of these factors potentially limits the accurate
guidance of the archaeological research.
These limitations, also discussed in the literature, e.g., [
74
], highlight the need for
further validation and integration of the developed process with recent archaeological
knowledge; especially in desert landscapes, which often present limited spectral signatures
and complex surface interactions [75,76].
In fact, desert environments often exhibit limited spectral variability, especially when
dominated by sand and sparse vegetation, as is the case in Saruq Al-Hadid and the larger
geographic area. This limited range of spectral signatures poses challenges for ML al-
gorithms that rely on distinct spectral patterns for classification or feature detection [
22
].
Discriminating between different archaeological features or materials may therefore become
more difficult in these environments.
Moreover, they can have complex surface interactions due to the presence of sand
dunes, rocky outcrops, and various surface materials. These surface interactions can result
in complex spectral mixtures and scattering effects, making it challenging to accurately
interpret remote sensing data (Figure 14). This complexity can affect the performance of
ML algorithms, which may struggle to differentiate between different surface materials or
archaeological features.
The modelling and prediction accuracies are expected to improve using the insertion of
a neural network and backpropagation algorithms based on the performed cluster groups
following more recent field surveys. The validated results can provide guidance for future
on-site archaeological work. The study’s findings are also essential for reliable mapping of
paleo-drainage systems within the study area, which is characterised by low topographic
variations [
77
]. The applicability and efficiency of the developed process are anticipated to
improve following imminent, more recent field surveys and further validation efforts using
multitemporal data. The analysis was carried out with the expectation that the techniques
described would be adjusted to better fit the requirements of archaeological research in
larger areas and similar environments.
This research is recommended for long-term, multimodal, and multitemporal inves-
tigations into the prehistoric landscape of the study site. An integrated workflow would
combine ML and DL techniques with automated feature detection in multisensory, multi-
temporal, remotely sensed data and suitable archaeological training data and knowledge
in the interpretation process. Such an innovative process is expected to more accurately
generate and validate detections of hitherto unidentified archaeological objects and sites in
the surroundings of the study site and similar environments. This would contribute to the
creation of reliable, labelled archaeological training datasets, thus guiding archaeological
studies through the identification of significant areas, the prediction of potential site loca-
tions, and the formulation of more informed research strategies for further investigations
while cutting down on extensive and costly ground-based sensing. In addition, mapping
watersheds of continental scale could be enabled, thus assisting in the reconstruction of the
paleo-hydrology of desert regions in other areas around the world.

Geosciences 2023,13, 179 23 of 34
Furthermore, local and regional archaeological and geophysical databases can be
expanded through the large spatial and temporal coverages offered from space, the unique
resolutions of recent optical remote sensors, the penetrating capabilities of remote sensing,
and AI.
Author Contributions:
Conceptualisation, H.B.-R. and H.G.; data curation, H.B.-R. and C.C.; formal
analysis, H.B.-R., D.F. and C.C.; funding acquisition, D.F. and S.G.; investigation, H.B.-R. and C.C.;
methodology, H.B.-R. and C.C.; visualization and software, H.B.-R. and C.C.; interpretation and
contextualisation, H.B.-R. and K.P.; project administration, S.G.; resources, D.F.; supervision, D.F.;
writing—original draft, H.B.-R.; review and editing, D.F. and H.B.-R. All authors have read and
agreed to the published version of the manuscript.
Funding: This research was funded by Khalifa University, grant number ‘8474000305’.
Data Availability Statement:
The data presented in this study are available on request from the
corresponding author, Diana Francis.
Acknowledgments:
The authors would like to thank the ALOS data distribution team of the PASCO
Corporation for supporting with the ALOS-2/PALSAR-2 data acquisition and the data distribution
team at Bayanat AI for facilitating the Worldview-3 data acquisition.
Conflicts of Interest: The authors declare no conflict of interest.
Appendix A
Geosciences 2023, 13, x FOR PEER REVIEW 24 of 34
Appendix A
Figure A1. Main previous and ongoing excavation sectors at the western zone of the Saruq Al-Hadid
site as per Weeks et al. (2019). Worldview-3 RGB composite of the western zone of observation. * A
three-year programme of archaeological fieldwork and post-excavation analysis in 2017.
(a)
Figure A1.
Main previous and ongoing excavation sectors at the western zone of the Saruq Al-Hadid
site as per Weeks et al. (2019). Worldview-3 RGB composite of the western zone of observation. * A
three-year programme of archaeological fieldwork and post-excavation analysis in 2017.

Geosciences 2023,13, 179 24 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 24 of 34
Appendix A
Figure A1. Main previous and ongoing excavation sectors at the western zone of the Saruq Al-Hadid
site as per Weeks et al. (2019). Worldview-3 RGB composite of the western zone of observation. * A
three-year programme of archaeological fieldwork and post-excavation analysis in 2017.
(a)
Geosciences 2023, 13, x FOR PEER REVIEW 25 of 34
(b)
(c)
Figure A2. (a–c). Some remarkable spots and locations on the site based on the produced orthomo-
saics—western zone of observation. Projection: UTM, Zone 40 North. Map: 321634.89 E, 2729207.78
N Meters. LL: 24°40′0.76″ N, 55°14′14.40″ E.
Figure A2. Cont.

Geosciences 2023,13, 179 25 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 25 of 34
(b)
(c)
Figure A2. (a–c). Some remarkable spots and locations on the site based on the produced orthomo-
saics—western zone of observation. Projection: UTM, Zone 40 North. Map: 321634.89 E, 2729207.78
N Meters. LL: 24°40′0.76″ N, 55°14′14.40″ E.
Figure A2.
(
a
–
c
). Some remarkable spots and locations on the site based on the produced
orthomosaics—western zone of observation. Projection: UTM, Zone 40 North. Map: 321634.89 E,
2729207.78 N Meters. LL: 24◦4000.7600 N, 55◦14014.40” E.
Geosciences 2023, 13, x FOR PEER REVIEW 26 of 34
Figure A3. Some remarkable spots and locations on site based on the produced orthomosaics—east-
ern zone of observation. Projection: UTM, Zone 40 North. Map: 321634.89 E, 2729207.78 N Meters.
LL: 24°40′0.76″ N, 55°14′14.40″ E.
Appendix B
Figure A4. Principal components.
Figure A3.
Some remarkable spots and locations on site based on the produced orthomosaics—
eastern zone of observation. Projection: UTM, Zone 40 North. Map: 321634.89 E, 2729207.78 N Meters.
LL: 24◦4000.7600 N, 55◦14014.400 0 E.

Geosciences 2023,13, 179 26 of 34
Appendix B
Geosciences 2023, 13, x FOR PEER REVIEW 26 of 34
Figure A3. Some remarkable spots and locations on site based on the produced orthomosaics—east-
ern zone of observation. Projection: UTM, Zone 40 North. Map: 321634.89 E, 2729207.78 N Meters.
LL: 24°40′0.76″ N, 55°14′14.40″ E.
Appendix B
Figure A4. Principal components.
Figure A4. Principal components.
Geosciences 2023, 13, x FOR PEER REVIEW 27 of 34
(a) (b) (c)
Figure A5. Statistical analysis of the principal components. (a) Resulting statistics; (b) basic statistics
for each PC band; (c) eigenvalues plot.
Appendix C
Figure A6. Geocoding and radiometric calibration using the global DSM (horizontal resolution 1
arcsec) by the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM).
Figure A5.
Statistical analysis of the principal components. (
a
) Resulting statistics; (
b
) basic statistics
for each PC band; (c) eigenvalues plot.

Geosciences 2023,13, 179 27 of 34
Appendix C
Geosciences 2023, 13, x FOR PEER REVIEW 27 of 34
(a) (b) (c)
Figure A5. Statistical analysis of the principal components. (a) Resulting statistics; (b) basic statistics
for each PC band; (c) eigenvalues plot.
Appendix C
Figure A6. Geocoding and radiometric calibration using the global DSM (horizontal resolution 1
arcsec) by the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM).
Figure A6.
Geocoding and radiometric calibration using the global DSM (horizontal resolution
1 arcsec) by the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM).
Appendix D
Geosciences 2023, 13, x FOR PEER REVIEW 28 of 34
Appendix D
Figure A7. (a) Complex data multi-looking (HH, HV, VH, VV), speckle-filtered using Lee filtering
(3 × 3). (b) Geocoding and radiometric calibration using the global DSM (horizontal resolution 1
arcsec) by the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) (30 m
ground resolution and 5 m elevation accuracy). Saruq Al-Hadid study site is delineated in red.
Figure A8. The produced digital elevation model (DEM) over Saruq Al-Hadid site.
Figure A7.
(
a
) Complex data multi-looking (HH, HV, VH, VV), speckle-filtered using Lee filtering
(3
×
3). (
b
) Geocoding and radiometric calibration using the global DSM (horizontal resolution
1 arcsec) by the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) (30 m
ground resolution and 5 m elevation accuracy). Saruq Al-Hadid study site is delineated in red.

Geosciences 2023,13, 179 28 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 28 of 34
Appendix D
Figure A7. (a) Complex data multi-looking (HH, HV, VH, VV), speckle-filtered using Lee filtering
(3 × 3). (b) Geocoding and radiometric calibration using the global DSM (horizontal resolution 1
arcsec) by the ALOS-Panchromatic Remote-sensing Instrument for Stereo Mapping (PRISM) (30 m
ground resolution and 5 m elevation accuracy). Saruq Al-Hadid study site is delineated in red.
Figure A8. The produced digital elevation model (DEM) over Saruq Al-Hadid site.
Figure A8. The produced digital elevation model (DEM) over Saruq Al-Hadid site.
Geosciences 2023, 13, x FOR PEER REVIEW 29 of 34
Figure A9. (a) DEM slope. (b) S2N. (c) W2E.
Appendix E
Figure A10. A triangulated irregular network of the Saruq Al-Hadid site (in red) and surroundings.
Figure A9. (a) DEM slope. (b) S2N. (c) W2E.

Geosciences 2023,13, 179 29 of 34
Appendix E
Geosciences 2023, 13, x FOR PEER REVIEW 29 of 34
Figure A9. (a) DEM slope. (b) S2N. (c) W2E.
Appendix E
Figure A10. A triangulated irregular network of the Saruq Al-Hadid site (in red) and surroundings.
Figure A10.
A triangulated irregular network of the Saruq Al-Hadid site (in red) and surroundings.
Geosciences 2023, 13, x FOR PEER REVIEW 30 of 34
Figure A11. Slope map classification of the Saruq Al-Hadid site (in red) and surroundings.
Figure A12. Hydrographic network map of the Saruq Al-Hadid site (in red) and surroundings.
Figure A11. Slope map classification of the Saruq Al-Hadid site (in red) and surroundings.

Geosciences 2023,13, 179 30 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 30 of 34
Figure A11. Slope map classification of the Saruq Al-Hadid site (in red) and surroundings.
Figure A12. Hydrographic network map of the Saruq Al-Hadid site (in red) and surroundings.
Figure A12. Hydrographic network map of the Saruq Al-Hadid site (in red) and surroundings.
Appendix F
Geosciences 2023, 13, x FOR PEER REVIEW 31 of 34
Appendix F
Figure A13. Geoprocessing: extracting data from radar and optical satellite images. (a) Radar com-
position. (b) High-volume scaering. (c) Unsupervised classes. (d) Classification centroids.
Figure A14. Geoprocessing: cluster analysis using the K-means++ Cluster Algorithm. (a) Cluster
groups detection. The relationship between the number of clusters and the WCSS is graphed. The
number of clusters where the change in WCSS begins to level off is then selected (elbow method).
(b) Data aggregation.
Figure A13.
Geoprocessing: extracting data from radar and optical satellite images. (
a
) Radar
composition. (b) High-volume scattering. (c) Unsupervised classes. (d) Classification centroids.

Geosciences 2023,13, 179 31 of 34
Geosciences 2023, 13, x FOR PEER REVIEW 31 of 34
Appendix F
Figure A13. Geoprocessing: extracting data from radar and optical satellite images. (a) Radar com-
position. (b) High-volume scaering. (c) Unsupervised classes. (d) Classification centroids.
Figure A14. Geoprocessing: cluster analysis using the K-means++ Cluster Algorithm. (a) Cluster
groups detection. The relationship between the number of clusters and the WCSS is graphed. The
number of clusters where the change in WCSS begins to level off is then selected (elbow method).
(b) Data aggregation.
Figure A14.
Geoprocessing: cluster analysis using the K-means++ Cluster Algorithm. (
a
) Cluster
groups detection. The relationship between the number of clusters and the WCSS is graphed. The
number of clusters where the change in WCSS begins to level off is then selected (elbow method).
(b) Data aggregation.
References
1. Parcak, S. Archaeology from Space: How the Future Shapes Our Past; Henry Holt and Company: New York, NY, USA, 2019.
2.
Alexakis, D.; Sarris, A.; Astaras, T.; Albanakis, K. Detection of Neolithic settlements in Thessaly (Greece) through multispectral
and hyperspectral satellite imagery. Sensors 2009,9, 1167–1187. [CrossRef] [PubMed]
3.
Alexakis, D.; Sarris, A.; Astaras, T.; Albanakis, K. Integrated GIS, remote sensing and geomorphologic approaches for the
reconstruction of the landscape habitation of Thessaly during the neolithic period. J. Archaeol. Sci. 2011,38, 89–100. [CrossRef]
4. Altaweel, M. The use of ASTER satellite imagery in archaeological contexts. Archaeol. Prospect. 2005,12, 151–166. [CrossRef]
5.
Agapiou, A.; Lysandrou, V. Remote sensing archaeology: Tracking and mapping evolution in European scientific literature from
1999 to 2015. J. Archaeol. Sci. Rep. 2015,4, 192–200. [CrossRef]
6.
Chen, F.; Lasaponara, R.; Masini, N. An overview of satellite synthetic aperture radar remote sensing in archaeology: From site
detection to monitoring. J. Cult. Herit. 2017,23, 5–11. [CrossRef]
7.
Henderson, F.M.; Lewis, A.J. (Eds.) Principles and applications of imaging radar. In Manual of Remote Sensing; John Wiley and
Sons, Inc.: Hoboken, NJ, USA, 1998; Volume 2, ISBN 0-471-29406-3.
8. Comer, D.C.; Harrower, M.J. Mapping Archaeological Landscapes from Space; Springer Science & Business Media: Berlin, Germany,
2013; Volume 5, ISBN 1-4614-6074-3.
9.
Lasaponara, R.; Masini, N.; Rizzo, E.; Orefici, G. New discoveries in the Piramide Naranjada in Cahuachi (Peru) using satellite,
Ground Probing Radar and magnetic investigations. J. Archaeol. Sci. 2011,38, 2031–2039. [CrossRef]
10. Philibert, R. Archaeology from Space: Advanced Satellite Imagery Through the Work of Sarah Parcak. Spectrum 2017,6, 6.
11.
Lasaponara, R.; Masini, N. Satellite Synthetic Aperture Radar in Archaeology and Cultural Landscape: An Overview; Wiley Online
Library: Hoboken, NJ, USA, 2013; ISBN 1075-2196.
12.
Stewart, C.; Montanaro, R.; Sala, M.; Riccardi, P. Feature extraction in the North Sinai desert using spaceborne synthetic aperture
radar: Potential archaeological applications. Remote Sens. 2016,8, 825. [CrossRef]
13.
Cox, C. Satellite imagery, aerial photography and wetland archaeology: An interim report on an application of remote sensing to
wetland archaeology: The pilot study in Cumbria, England. World Archaeol. 1992,24, 249–267. [CrossRef]
14.
Sever, T.L.; Irwin, D.E. Landscape archaeology: Remote-sensing investigation of the ancient Maya in the Peten rainforest of
northern Guatemala. Anc. Mesoam. 2003,14, 113–122. [CrossRef]
15.
Hu, N.K.; Li, X. Historical ruins of remote sensing archaeology in arid desertified environment, northwestern China. In Proceedings
of the IOP Conference Series: Earth and Environmental Science; IOP Publishing: Bristol, UK, 2017; Volume 57, p. 012028.
16.
Breeze, P.S.; Drake, N.A.; Groucutt, H.S.; Parton, A.; Jennings, R.P.; White, T.S.; Clark-Balzan, L.; Shipton, C.; Scerri, E.M.;
Stimpson, C.M. Remote sensing and GIS techniques for reconstructing Arabian palaeohydrology and identifying archaeological
sites. Quat. Int. 2015,382, 98–119. [CrossRef]
17.
Deroin, J.-P.; Téreygeol, F.; Heckes, J. Evaluation of very high to medium resolution multispectral satellite imagery for geoarchae-
ology in arid regions–Case study from Jabali, Yemen. J. Archaeol. Sci. 2011,38, 101–114. [CrossRef]
18.
Blom, R.; Zairins, J.; Clapp, N.; Hedges, G. Space Technology and the Discovery of the Lost City of Ubar; IEEE: Piscataway, NJ, USA,
1997; Volume 1, pp. 19–28.

Geosciences 2023,13, 179 32 of 34
19.
Sultan, M.; Sturchio, N.; Al Sefry, S.; Milewski, A.; Becker, R.; Nasr, I.; Sagintayev, Z. Geochemical, isotopic, and remote sensing
constraints on the origin and evolution of the Rub Al Khali aquifer system, Arabian Peninsula. J. Hydrol.
2008
,356, 70–83.
[CrossRef]
20.
Orengo, H.A.; Conesa, F.C.; Garcia-Molsosa, A.; Lobo, A.; Green, A.S.; Madella, M.; Petrie, C.A. Automated detection of
archaeological mounds using machine-learning classification of multisensor and multitemporal satellite data. Proc. Natl. Acad.
Sci. USA 2020,117, 18240–18250. [CrossRef] [PubMed]
21.
Garcia-Molsosa, A.; Orengo, H.A.; Lawrence, D.; Philip, G.; Hopper, K.; Petrie, C.A. Potential of deep learning segmentation for
the extraction of archaeological features from historical map series. Archaeol. Prospect. 2021,28, 187–199. [CrossRef] [PubMed]
22.
Bachagha, N.; Elnashar, A.; Tababi, M.; Souei, F.; Xu, W. The Use of Machine Learning and Satellite Imagery to Detect Roman
Fortified Sites: The Case Study of Blad Talh (Tunisia Section). Appl. Sci. 2023,13, 2613. [CrossRef]
23.
Soriano, I.; Perea, A.; Escanilla, N.; Rodrigo, F.C.; Al Ali, Y.Y.A.; Karim, M.B.R.; Zein, H. Goldwork technology at the Arabian
Peninsula. First data from Saruq al Hadid Iron Age site (Dubai, United Arab Emirates). J. Archaeol. Sci. Rep.
2018
,22, 1–10.
[CrossRef]
24.
Herrmann, J.T.; Casana, J.; Qandil, H.S. A sequence of inland desert settlement in the Oman peninsula: 2008–2009 excavations at
Saruq al-Hadid, Dubai, UAE. Arab. Archaeol. Epigr. 2012,23, 50–69. [CrossRef]
25.
Weeks, L.; Cable, C.; Franke, K.; Newton, C.; Karacic, S.; Roberts, J.; Stepanov, I.; David-Cuny, H.; Price, D.; Bukhash, R.M. Recent
archaeological research at Saruq al-Hadid, Dubai, UAE. Arab. Archaeol. Epigr. 2017,28, 31–60. [CrossRef]
26.
Weeks, L.; Cable, C.M.; Franke, K.A.; Karacic, S.; Newton, C.; Roberts, J.; Stepanov, I.; McRae, I.K.; Moore, M.W.; David-Cuny, H.
Saruq al-Hadid: A persistent temporary place in late prehistoric Arabia. World Archaeol. 2019,51, 157–182. [CrossRef]
27.
Weeks, L.; Cable, C.M.; Karacic, S.; Franke, K.A.; Price, D.M.; Newton, C.; Roberts, J.; Al Ali, Y.Y.; Boraik, M.; Zein, H. Dating
persistent short-term human activity in a complex depositional environment: Late Prehistoric Occupation at Saruq al-Hadid,
Dubai. Radiocarbon 2019,61, 1041–1075. [CrossRef]
28.
Karacic, S.; Weeks, L.; Cable, C.; Méry, S.; al-Ali, Y.; Boraik, M.; Zein, H.; Glascock, M.D.; MacDonald, B.L. Integrating a complex
late prehistoric settlement system: Neutron activation analysis of pottery use and exchange at Saruq al-Hadid, United Arab
Emirates. J. Archaeol. Sci. Rep. 2018,22, 21–31. [CrossRef]
29.
El Rai, M.C.; Al-Saad, M.; Aburaed, N.; Al Mansoori, S.; Al-Ahmad, H.; Marshall, S. Automatic detection of potential buried
archaeological sites in Saruq Al-Hadid, United Arab Emirates. In Proceedings of the Remote Sensing Technologies and Applica-
tions in Urban Environments V, Online, 21–25 September 2020; International Society for Optics and Photonics: Bellingham, WA,
USA, 2020; Volume 11535, p. 115350G.
30.
Farrant, A.R.; Price, S.J.; Arkley, S.L.B.; Finlayson, A.; Thomas, R.J.; Leslie, A. Geology of the Al Lisaili 1: 100 000 Map Sheet, 100-6,
United Arab Emirates; British Geological Survey: Nottingham, UK, 2012; ISBN 0-85272-716-X.
31.
Herrmann, J.T. Three-Dimensional Mapping of Archaeological and Sedimentary Deposits with Ground-penetrating Radar at
Saruq al-Hadid, Dubai, United Arab Emirates. Archaeol. Prospect. 2013,20, 189–203. [CrossRef]
32.
Longbotham, N.; Pacifici, F.; Malitz, S.; Baugh, W.; Camps-Valls, G. Measuring the spatial and spectral performance of WorldView-
3. In Hyperspectral Imaging and Sounding of the Environment; Optical Society of America: Washington, DC, USA, 2015; p. HW3B.2.
33.
Stewart, C.; Lasaponara, R.; Schiavon, G. ALOS PALSAR analysis of the archaeological site of Pelusium. Archaeol. Prospect.
2013
,
20, 109–116. [CrossRef]
34.
Kankaku, Y.; Sagisaka, M.; Suzuki, S. PALSAR-2 launch and early orbit status. In Proceedings of the 2014 IEEE Geoscience and
Remote Sensing Symposium, Quebec City, QC, Canada, 13–18 July 2014; pp. 3410–3412.
35. Bro, R.; Smilde, A.K. Principal component analysis. Anal. Methods 2014,6, 2812–2831. [CrossRef]
36.
Calleja, J.F.; Pagés, O.R.; Díaz-Álvarez, N.; Peón, J.; Gutiérrez, N.; Martín-Hernández, E.; Relea, A.C.; Melendi, D.R.; Álvarez, P.F.
Detection of buried archaeological remains with the combined use of satellite multispectral data and UAV data. Int. J. Appl. Earth
Obs. Geoinf. 2018,73, 555–573. [CrossRef]
37. Wiseman, J.; El-Baz, F. Remote Sensing in Archaeology; Springer: Berlin/Heidelberg, Germany, 2007.
38.
Lasaponara, R.; Masini, N. Satellite remote sensing in archaeology: Past, present and future perspectives. J. Archaeol. Sci.
2011
,9,
1995–2002. [CrossRef]
39.
Tapete, D.; Cigna, F. Detection of archaeological looting from space: Methods, achievements and challenges. Remote Sens.
2019
,
11, 2389. [CrossRef]
40.
Ball, G.H.; Hall, D.J. ISODATA, a Novel Method of Data Analysis and Pattern Classification; Stanford Research Inst.: Menlo Park, CA,
USA, 1965.
41.
Adediran, A.O.; Parcharidis, I.; Poscolieri, M.; Pavlopoulos, K. Computer-assisted discrimination of morphological units
on north-central Crete (Greece) by applying multivariate statistics to local relief gradients. Geomorphology
2004
,58, 357–370.
[CrossRef]
42. Orlando, P.; Villa, B. de Remote sensing applications in archaeology. Archeol. E Calc. 2011,22, 147–168.
43. Cable, C. SHARP S1 Excavation; University of New England: Armidale, NSW, Australia, 2015.
44.
Orengo, H.A.; Garcia, A.; Conesa, F.C.; Green, A.; Singh, R.N.; Petrie, C.A. Combining TanDEM-X with multi-temporal, multi-
source satellite data for the reconstruction of the Bronze Age landscapes of the Indus Civilisation. In Proceedings of the
IGARSS 2019-2019 IEEE International Geoscience and Remote Sensing Symposium, Yokohama, Japan, 28 July–2 August 2019; pp.
4581–4584.

Geosciences 2023,13, 179 33 of 34
45. Lee, J.-S. Refined filtering of image noise using local statistics. Comput. Graph. Image Process. 1981,15, 380–389. [CrossRef]
46.
Lee, J.-S.; Jurkevich, L.; Dewaele, P.; Wambacq, P.; Oosterlinck, A. Speckle filtering of synthetic aperture radar images: A review.
Remote Sens. Rev. 1994,8, 313–340. [CrossRef]
47.
Holcomb, D.W. Imaging radar and archaeological survey: An example from the Gobi Desert of Southern Mongolia. J. Field
Archaeol. 2001,28, 131–141.
48.
McHugh, W.P.; Breed, C.S.; Schabers, G.G.; McCauley, J.F.; Szabo, B.J. Acheulian sites along the “radar rivers” southern Egyptian
Sahara. J. Field Archaeol. 1988,15, 361–379.
49.
McCauley, J.F.; Schaber, G.G.; Breed, C.S.; Grolier, M.J.; Haynes, C.V.; Issawi, B.; Elachi, C.; Blom, R. Subsurface valleys and
geoarcheology of the eastern Sahara revealed by shuttle radar. Science 1982,218, 1004–1020. [CrossRef]
50.
Chapman, B.D.; Comer, D.C.; Isla, J.A.; Silverman, H. The measurement by airborne synthetic aperture radar (SAR) of disturbance
within the Nasca world heritage site. Conserv. Manag. Archaeol. Sites 2015,17, 270–286. [CrossRef]
51.
Comer, D.C.; Blom, R.G. Detection and Identification of Archaeological Sites and Features Using Synthetic Aperture Radar (SAR) Data
Collected from Airborne Platforms; Springer: Berlin/Heidelberg, Germany, 2007; ISBN 0-387-44453-X.
52.
Stanislawski, L.V.; Shavers, E.J.; Wang, S.; Jiang, Z.; Usery, E.L.; Moak, E.; Duffy, A.; Schott, J. Extensibility of U-Net neural
network model for hydrographic feature extraction and implications for hydrologic modeling. Remote Sens.
2021
,13, 2368.
[CrossRef]
53.
Hack, J.T. Studies of Longitudinal Stream Profiles in Virginia and Maryland; US Government Printing Office: Washington, DC, USA,
1957; Volume 294.
54. Knight, J.; Burningham, H. Sand dune morphodynamics and prehistoric human occupation in NW Ireland. Geol. Soc. Am. Spec.
Pap. 2011,476, 81–92.
55.
Argyrou, A.; Agapiou, A. A Review of Artificial Intelligence and Remote Sensing for Archaeological Research. Remote Sens.
2022
,
14, 6000. [CrossRef]
56.
Ronneberger, O.; Fischer, P.; Brox, T. U-net: Convolutional networks for biomedical image segmentation. In Proceedings of
the Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich,
Germany, 5–9 October 2015; Part III 18. Springer: Berlin/Heidelberg, Germany, 2015; pp. 234–241.
57.
Min, Z.; Kai-fei, D. Improved Research to K-means Initial Cluster Centers. In Proceedings of the 2015 Ninth International
Conference on Frontier of Computer Science and Technology, Dalian, China, 26–28 August 2015; pp. 349–353.
58. Hastie, T.; Tibshirani, R.; Friedman, J.H.; Friedman, J.H. The Elements of Statistical Learning: Data Mining, Inference, and Prediction;
Springer: Berlin/Heidelberg, Germany, 2009; Volume 2.
59.
Schölkopf, B.; Smola, A.; Müller, K.-R. Kernel principal component analysis. In Proceedings of the Artificial Neural Networks—
ICANN’97: 7th International Conference, Lausanne, Switzerland, 8–10 October 1997; Springer: Berlin/Heidelberg, Germany,
2005; pp. 583–588.
60. Colwell, J.E. Vegetation canopy reflectance. Remote Sens. Environ. 1974,3, 175–183. [CrossRef]
61.
Strahler, A.H. Vegetation canopy reflectance modeling—Recent developments and remote sensing perspectives. Remote Sens. Rev.
1997,15, 179–194. [CrossRef]
62.
Lin, Y.; Tian, Q.; Qiao, B.; Wu, Y.; Zuo, X.; Xie, Y.; Lian, Y. A Synthetic Angle Normalization Model of Vegetation Canopy
Reflectance for Geostationary Satellite Remote Sensing Data. Agriculture 2022,12, 1658. [CrossRef]
63. Ben-Dor, E.; Irons, J.R.; Epema, G.F. Soil reflectance. Remote Sens. Earth Sci. Man. Remote Sens. 1999,3, 111–188.
64.
Heiden, U.; d’Angelo, P.; Schwind, P.; Karlshöfer, P.; Müller, R.; Zepp, S.; Wiesmeier, M.; Reinartz, P. Soil Reflectance Composites—
Improved Thresholding and Performance Evaluation. Remote Sens. 2022,14, 4526. [CrossRef]
65.
Stepanov, I.; Weeks, L.; Franke, K.; Rodemann, T.; Salvemini, F.; Cable, C.; Al Ali, Y.; Radwan, M.B.; Zein, H.; Grave, P. Scrapping
ritual: Iron Age metal recycling at the site of Saruq al-Hadid (UAE). J. Archaeol. Sci. 2019,101, 72–88. [CrossRef]
66.
Valente, T.; Contreras Rodrigo, F.; Mahmud, A.; Boraik Radwan Karim, M.; Al Mansoori, M.S.; Zein, H. Five seasons of excavations
in Areas 2A and G of Saruq al-Hadid (Dubai, UAE): Iron Age II evidences of copper production, workshop area and ceremonial
activities. Isimu 2020,23, 169–195. [CrossRef]
67.
Chang, Y.-W.; Hsieh, C.-J.; Chang, K.-W.; Ringgaard, M.; Lin, C.-J. Training and testing low-degree polynomial data mappings via
linear SVM. J. Mach. Learn. Res. 2010,11, 1471–1490.
68.
Clark, C.D.; Garrod, S.M.; Pearson, M.P. Landscape archaeology and remote sensing in southern Madagascar. Int. J. Remote Sens.
1998,19, 1461–1477. [CrossRef]
69.
Maktav, D.; Crow, J.; Kolay, C.; Yegen, B.; Onoz, B.; Sunar, F.; Coskun, G.; Karadogan, H.; Cakan, M.; Akar, I. Integration of remote
sensing and GIS for archaeological investigations. Int. J. Remote Sens. 2009,30, 1663–1673. [CrossRef]
70.
Agapiou, A.; Hadjimitsis, D.G.; Alexakis, D.D. Development of an image-based method for the detection of archaeological buried
relics using multi-temporal satellite imagery. Int. J. Remote Sens. 2013,34, 5979–5996. [CrossRef]
71.
Agapiou, A.; Lysandrou, V.; Sarris, A.; Papadopoulos, N.; Hadjimitsis, D.G. Fusion of satellite multispectral images based on
ground-penetrating radar (GPR) data for the investigation of buried concealed archaeological remains. Geosciences
2017
,7, 40.
[CrossRef]
72.
Parcak, S. Moving from space-based to ground-based solutions in remote sensing for archaeological heritage: A case study from
Egypt. Remote Sens. 2017,9, 1297. [CrossRef]

Geosciences 2023,13, 179 34 of 34
73.
Wheatley, D.; Gillings, M. Spatial Technology and Archaeology: The Archaeological Applications of GIS; CRC Press: Boca Raton, FL,
USA, 2013.
74. Brooke, C. Thermal imaging for the archaeological investigation of historic buildings. Remote Sens. 2018,10, 1401. [CrossRef]
75.
Gxokwe, S.; Dube, T.; Mazvimavi, D. Multispectral remote sensing of wetlands in semi-arid and arid areas: A review on
applications, challenges and possible future research directions. Remote Sens. 2020,12, 4190. [CrossRef]
76.
Govender, T.; Dube, T.; Shoko, C. Remote sensing of land use-land cover change and climate variability on hydrological processes
in Sub-Saharan Africa: Key scientific strides and challenges. Geocarto Int. 2022,37, 10925–10949. [CrossRef]
77.
Ghoneim, E. Optimum groundwater locations in the northern United Arab Emirates. Int. J. Remote Sens.
2008
,29, 5879–5906.
[CrossRef]
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.