Discussion Points of the Remote Sensing Study and Integrated Analysis of the Archaeological Landscape of Rujm el-Hiri

Abstract

Remote sensing techniques provide crucial insights into ancient settlement patterns in various regions by uncovering previously unknown archaeological sites and clarifying the topological features of known ones. Meanwhile, in the northern part of the Southern Levant, megalithic structures remain largely underexplored with these methods. This study addresses this gap by analyzing the landscape around Rujm el-Hiri, one of the most prominent Southern Levantine megaliths dated to the Chalcolithic/Early Bronze Age, for the first time. We discuss the type and extent of the archaeological remains identified in satellite images within a broader context, focusing on the relationships between landscapes and these objects and the implications of their possible function. Our analysis of multi-year satellite imagery covering the 30 km region surrounding the Sea of Galilee reveals several distinct patterns: 40–90-m-wide circles and thick walls primarily constructed along streams, possibly as old as Rujm el-Hiri itself; later-period linear thin walls forming vast rectangular fields and flower-like clusters of ~ 20 m diameter round-shaped fences found in wet areas; tumuli, topologically linked to the linear walls and flower-like fences. Although tumuli share similar forms and likely construction techniques, their spatial distribution, connections to other archaeological features, and the statistical distribution in their sizes suggest that they might serve diverse functions. The objects and patterns identified may be used for further training neural networks to analyze their spatial properties and interrelationships. Most archaeological structures in the region were reused long after their original construction. This involved adding new features, building walls over older ones, and reshaping the landscape with new objects. Rujm el-Hiri is a prime example of such a complex sequence. Geomagnetic analysis shows that since the entire region has rotated over time, the Rujm el-Hiri’s location shifted from its original position for tens of meters for the thousands of years of the object’s existence, challenging theories of the alignment of its walls with astronomical bodies and raising questions regarding its possible identification as an observatory.

1. Introduction

Ancient megalithic structures, from Europe to Asia and the Americas, are found worldwide. These include dolmens, stone circles, and monumental enclosures, dating back at least as far as the 10th millennium BCE (e.g., Göbekli Tepe [1]). Examples range, among others, from the Neolithic megaliths of the British Islands to the first millennium BCE Nuragic civilization of Sardinia, e.g., [2,3]. While traditionally associated with ritualistic activities, these structures must have served various purposes, including sepulchral and non-sepulchral functions. Some, including the world-famous Stonehenge, appear to have astronomical alignments with celestial bodies and/or solar and lunar events, e.g., [4,5,6]. These ‘larger-than-life’ structures have been a focus of archaeological investigations and the public interest for centuries, yet many questions have remained unanswered.

The megalithic site of Rujm el-Hiri is one of the largest, best-known megalithic complexes in the Ancient Near East. It is located northeast of the Sea of Galilee (32°54′30.87″N, 35°48′3.89″E), as shown in Figure 1a. The aerial view of the site can be found in Figure 1b, and Figure 1c depicts its location along with the distance–altitude profile along the line drawn from the northern point of the Sea of Galilee towards Rujm el-Hiri). The Sea of Galilee, located in the Jordan Rift Valley along with the Dead Sea, is the second-lowest lake in the world. The northern part of the Sea of Galilee lies 210 m below sea level (Figure 1c). Rujm el-Hiri is situated 18 km northeast, in the middle of an 11.5 km-wide plateau, with a gradually increasing elevation ranging from 400 m to 616 m above sea level. This enigmatic archaeological object is often referred to as the “Great Stone Circles”, “Wheel of Ghosts” and even the “Levantine Stonehenge” because of its size, the precision of its ring structure, unique preservation, and much debated but still unclear function/s. The circles comprising Rujm el-Hiri and the tumulus in its center (Figure 1b) are usually dated to the Chalcolithic and Early Bronze Ages (ca. 4500–3600 BCE and 3600–2300 BCE, respectively), with possible later additions. The limited modern development of the area and its mainly agricultural character allowed for a remarkable preservation of the site, and other remains in the surrounding landscape, including isolated structures, larger settlements, and field and other wall systems.

The Sea of Galilee has historically been the main freshwater reservoir of the Levant and is fed, even nowadays, by numerous streams and small-scale lakes in the area, e.g., [7]. Paleo-climatological records suggest that during the beginning of the Holocene, the climate in the Mediterranean regions of the Levant was relatively stable, characterized by a wetter, warmer regime than the modern one, supporting rich fauna and deciduous (oak) trees growing, e.g., [8,9]. This period also witnessed the maximum development of the Mediterranean maquis, as indicated by marine pollen records from the coast of Israel. This was probably due to amelioration in climatic conditions, as corroborated by the continental pollen records from the Hula Valley and the surroundings of the Sea of Galilee, e.g., [10,11] and the northern coast of Israel [12], which suggests conditions favorable for early agriculturalist and pastoralist economies, which are usually reconstructed in the region during these periods, e.g., [13,14]. The peak in olive tree cultivation occurred between 7000 and 6500 years ago, marking a high point in socio-economic development during the same period. However, a shift toward drier conditions was identified in the mid-Holocene [15,16].

According to Develle et al. [8], the northern part of the Southern Levant was the first to experience these climate variations, which must have influenced settlement patterns and agricultural practices. This process of aridization peaked towards the beginning of the second Millennium BCE, and it has been suggested that this aridization should be linked to significant socio-cultural changes in early urban societies and even societal collapse across Western Asia and Egypt [15,16]. These climatic and environmental shifts could have triggered droughts, causing significant changes in the lifestyle of inhabitants across the entire region and were responsible, at least in part, for the decline of the culture behind the construction of Rujm el-Hiri and the surrounding megaliths, e.g., [17,18,19,20]. This information is essential for understanding the landscape features and the meaning of the archaeological remains.

Figure 1. Rujm el-Hiri. (a) Geographic location, (32°54′30.87″N, 35°48′3.89″E); (b) Aerial view, adapted from [21]; (c) Distance-height profile of the surrounding area relative to the northernmost point of the Sea of Galilee (vertical axis—altitude below/above sea level, in m; horizontal axis—the distance in km). The vertical line indicates the location of Rujm el-Hiri.

Figure 1. Rujm el-Hiri. (a) Geographic location, (32°54′30.87″N, 35°48′3.89″E); (b) Aerial view, adapted from [21]; (c) Distance-height profile of the surrounding area relative to the northernmost point of the Sea of Galilee (vertical axis—altitude below/above sea level, in m; horizontal axis—the distance in km). The vertical line indicates the location of Rujm el-Hiri.

Since its discovery in 1968, Rujm el-Hiri has captivated the imagination of many archaeologists and enthusiasts alike [22,23,24,25,26,27,28,29]. Comprised of a central cairn surrounded by several concentric circles made of basalt stones, the complex reached ca. 150 m in diameter and ca. 500 m in circumference. The concentric walls, preserved up to 2.5 m high and 3.5 m wide, were partially connected by a series of radial walls. Two oversized entrances were identified, one facing the northeast and the other to the southeast (Figure 1b). For more details, see a 3D aerial view of Rujm el-Hiri provided in [30].

Even though many years have passed since its discovery, and several excavations have been conducted at the site [24,28,31], this large monument, its chronology, and its function are still under debate. This is mainly because, much like other monuments across the old world, very few finds were recovered from within the structure during excavations (ibid; see discussion in [32]). The complex stratigraphy at the site presenting several episodes of use and reuse, deepens the arguments and several chronologies were suggested, ranging from the Neolithic through the Chalcolithic, Early Bronze Age, and Late Bronze Age, e.g., [33,34].

The functional purpose of Rujm el-Hiri has also been the subject of intense debates. While Epstein and Gutman [23] suggested that Rujm el-Hiri represents a fortified structure or defense enclosure, other explanations were given, ranging from an extravagant burial of a leader or essential persona to an observatory dedicated to astrologically based rituals [25]. Others suggested the Rujm was a regional gathering locale for otherwise scattered local tribes [24,34].

Whether Rujm el-Hiri was a ritual center or had a more ‘mundane’ meaning, it was part of a broader physical and cultural landscape. To fully understand Rujm el-Hiri’s purpose, it cannot be studied in isolation; it must be considered and explored within the broader context of the surrounding archaeological remains and landscape. That, in turn, requires an extensive survey that will allow the mapping and recording of these remains over a relatively wide area. Field surveying has been essential to archaeological research since the pioneering 19th-century exploration missions to the Holy Land, e.g., [35,36]. Indeed, these earliest surveys included parts of the Golan [37,38,39]. However, most surveys, including modern ones, were often partial or focused on specific chronological or cultural phases, e.g., [40,41]; see discussion in [42]. It was only with the initiation of the large national Golan Survey conducted by Hartal on behalf of the Israel Antiquities Authority that the wide extent of archaeological remains in the Golan was revealed in full [42,43]. However, even this large, ambitious project could not cover the entire region. While the Golan offers a remarkable archaeological record, it also presents critical issues when conducting an on-field foot survey. The closeness to the modern-day geopolitical borders, as well as the existence of minefields and other remnants of the troubled past of the region, represent crucial limiting factors for such an endeavor. These limiting factors have hindered the extent and quality of the archaeological record in the area and our understanding of Rujm el-Hiri and its role within the landscape.

Archaeological surveys are not only limited by modern-day (and natural) physical constraints. They are also inherently time-consuming and require skilled participants, accuracy, and precision. As such, they are usually also quite expensive. The use of remote sensing methods and their fusion with ‘traditional’ archaeological survey methods have the potential to alleviate these factors. Remote sensing substantially benefits archaeological surveys, particularly in complex locations with problematic field conditions, e.g., [44,45,46,47]. Moreover, remote sensing is geared towards examining large landscapes, providing high-resolution data that can reveal even subtle features that might be missed during ground surveys. Surprisingly, attempts to use remote sensing in the Golan have been minimal (as noted in [48], and the situation has not changed). The one published research was site-oriented, applying airborne thermal video radiometry (ATVR) for detecting subsurface remains at Tel Leviah [49]. In 1988, a geophysical survey was conducted at Rujm el-Hiri itself as preparation for the first excavation project at the site, but this was focused on the architectural remains themselves, aiming to clarify stratigraphical issues and assist in selecting excavation areas [24].

In this work, we present the first results of a new interdisciplinary project focusing on the study of Rujm el-Hiri and the landscape in which it existed. Archaeological objects identified in satellite images are classified into several types, systematically organized, and analyzed. We demonstrate the high potential of using remote sensing techniques to discover new archaeological sites and present newly discovered objects not previously recorded. Compared to geological information about the studied region, our analysis suggests that Rujm el-Hiri’s historical function was unlikely to involve astronomical observations. This conclusion is drawn from the fact that the entire region rotated over time, leading to no alignment of the original Rujm el-Hiri’s structural features with astronomical bodies, as proposed by several authors. Our study shows the importance of harnessing front-line technologies to facilitate ‘traditional’ archaeological survey methods.

2. Data and Methodology

We have analyzed remote sensing data from the Landsat/Copernicus Earth-observing satellite mission [50] provided by the U.S. Geological Survey [51] for the period before 2012. Further, we used imagery from two VHR Earth Observation Pleiades satellites operated by the French Centre National d’Etudes Spatiales (CNES, [52] and Airbus Defense and Space (the commercial provider of the VHR imagery of Israel and Palestine, [53]). For simplicity of checking the results, examples given below are obtained from Google Earth Pro (version 7.3.6.9796) using Maxar technologies [54] in cooperation with CNES/Airbus, which provides preprocessed images with 0.5 m resolution in a panchromatic band and 2 m resolution in multispectral bands.

Satellite imaging, mainly through Google Earth Pro and Google Earth Engine, is a valuable tool in archaeology. It improves the visualization and dissemination of scientific data and facilitates discoveries. Examples include the Nasca geoglyphs’ visibility and the Chinese pyramids’ orientations, e.g., [55,56]. Google Earth Pro, used in this work for illustration, can significantly save time and resources before fieldwork, estimating sizes and typical features of the objects under investigation. Modern remote sensing methods allow solving the problem of the “Nasca lines effect” when a landscape feature much larger than a human body is not interpreted by a human brain as a unit but rather as a local obstacle or a set of irregular features. In other words, poorly preserved archaeological objects of a size larger than 10–20 m can sometimes be recognized as patterns associated with human activity only from above.

Smaller-size objects can also be identified with remote sensing using different methods, including modern AI applications [57,58,59,60,61,62]. However, the initial study is always based on the expert evaluation of the landscape features. Since the Rujm el-Hiri’s area has never been studied using satellite images, in this work, we conduct an initial visual analysis of the area of interest to recognize the most typical landscape peculiarities associated with historical human activities, applying the method for restoring poorly visible archaeological objects described by Eppelbaum et al. [63] to the key sites. This method overlaps images taken over different years, seasons, and under varying light angles, which creates different shadows and enhances the visibility of the objects’ shapes that might otherwise remain hidden when using a single image. Analogous multi-temporal methods are used in archaeology to trace differences in the condition of sites, climate changes affecting archaeological objects, damage caused by human activities, and changes in vegetation, e.g., [64,65,66,67,68]. Cropmarks were also considered in the analysis, where available, for better recognition of the shape of objects of interest, e.g., [69,70]. Combined, it helps reveal poorly preserved or initially doubtfully identified objects. Each site was visually examined for the presence of distinct archaeological stratification, such as newer walls constructed atop older ones, identifiable by differences in color and shape. The typical sizes of tumuli identified in the study were measured manually in the best available high-resolution images, with an accuracy of 0.5 m.

Despite all the fame of Rujm el-Hiri, there have not yet been established references for how this megalithic structure and other associated archaeological objects in Golan Heights appear in satellite imagery. Therefore, the initial work on mapping and documenting the archaeological landscape should be carried out. Initial studies typically focus on identifying the most commonly occurring types of objects without analyzing specific sites in detail. Afterward, fieldwork will follow, combining on-site observations with satellite image analysis. These combined results will then be used to train neural networks, enabling the automatic identification of additional sites and the study of their characteristics.

In this regard, we take the first step in studying the Rujm el-Hiri-associated landscape. The most typical landscape patterns recognizable in satellite images and associated with Rujm el-Hiri have been recorded within the 30 km zone around the Sea of Galilee, with a particular focus on the ten km-radius zone encompassing the megalith. We have distinguished several types of archaeological objects and analyzed their spatial characteristics. We use geodynamic–paleomagnetic maps based on studies of Eppelbaum and Katz [71] and Eppelbaum et al. [72] to understand the possible impacts of natural forces on the region.

3. Geophysical Studies in the Region

Israel is located between 29° and 33° north of the equator. It is a subtropical region between the temperate and tropical zones, characterized mainly by semi-arid and arid climates [73]. Such a climate causes increased productivity and water-use efficiency due to higher CO₂, which tends to increase ground cover, counteracting the effects of higher temperatures [74] and climate change [75]. As a result of this effect, the soils of Israel are complex formations with variable physical properties, even within small areas [76]. The multi-layered and variable structure of the upper part of the geological sequence, e.g., [7] causes definite difficulties in conducting archaeo-geophysical studies [48].

Among the variety of applied geophysical methods used in studying Israeli archaeological sites, we can note the magnetic survey [77,78,79,80], paleomagnetic studies [81,82,83,84], seismics [85,86], near-surface temperature [80,87]; Ground-Penetrating Radar (GPR) [88,89,90], microgravity [80], self-potential [80,91], piezoelectric [92], resistivity [80,93], and remote sensing [49,63,94].

To analyze the region’s geological–archaeological properties in the Chalcolithic and Bronze Age, we refer to the comprehensive geophysical–geodynamic mapping that highlights the relationship between the revealed quasi-circular counterclockwise rotating mantle structure below the Eastern Mediterranean and different subsurface (surface) geological–geophysical features seen in Figure 2A,B, e.g., [71,72,95]. GPS data indicate that this counterclockwise rotation occurs at an average yearly velocity of 8–15 mm/year, e.g., [96]. Northern Israel is a part of the eastern segment of this giant rotating region. Based on the numerous paleomagnetic–radiometric and tectonic-structural data available for northern Israel (Figure 2B), several geodynamic-paleomagnetic maps (for the areas of Carmel, Atlit, Sea of Galilee and its surroundings, Hula, and Hermon) were constructed (see [71,72], and references therein). They unambiguously indicate the counterclockwise rotation of the region, including the Rujm el-Hiri site under study. Assuming the average rotation velocity of 10 mm per year and the age of the studied prehistoric site as 4000 years, we will receive an arc rotation of about 40 m. The corresponding direct movement can be estimated as 30 m to its today’s position.

Therefore, the extensive geophysical–geodynamic analysis (including GPS, paleomagnetic data, and tectonic reconstructions) suggests that the tectonic blocks of northern Israel have continuously moved along an elliptical trajectory in a counterclockwise direction [95,96]. This implies that the primary axes of Rujm el-Hiri have rotated over time along with the entire region, casting significant doubt on the popular theory that Rujm el-Hiri was an observatory. Studies such as Aveni and Mizrachi [26] have recalculated the sky map and aligned the directions of the solstices, equinoxes, Sirius, and other celestial bodies as they would have appeared in 2500–3500 BCE, matching them with the current orientation of Rujm el-Hiri’s symmetry axes and entrances. However, our analysis reveals that the alignment of the gateways and radial walls during that historical period was entirely different from today’s.

A general literature search indicates that no archaeo-geophysical methods (including Remote Sensing (RS)) have been applied in Rujm el-Hiri and its vicinity. One publication noted that GPR was used to identify the main chamber in the object’s center [24]. Therefore, the current work’s analysis of satellite images is the first geophysical application method in this area, and comprehensive geophysical-archaeological study in the future should enhance it.

4. Results of the Remote Sensing Study

4.1. The Structure of Rujm el-Hiri

Figure 3 shows Rujm al-Hiri observed from space in four different years and seasons (August 2013, April 2018, November 2021, and January 2022). The first lesson that one can learn from comparing these images is that different peculiarities of the archaeological object are better seen under different light (exposition angles) and other vegetation conditions. This aspect can be used to reconstruct the shape of poorly preserved sites since the visual information obtained about any differently highlighted object potentially allows for reconstructing its 3D view. Recently developed automatic techniques for 3D image reconstruction from multi-date satellite images use advanced algorithms that account for changes in lighting and seasonal variations [98,99,100,101,102,103]. These methods are rarely employed in archaeology but can potentially revolutionize the field (see [59,69] and references therein).

Given the vast area under study, which contains hundreds of thousands of archaeological sites and lacks previous landscape analyses using remote sensing, we have applied the various-illumination-based technique [63] to select a representative group of objects. This approach, which involves enhancing and overlaying multitemporal images of the sites, is more superficial than 3D reconstructions but helps visualize archaeological features that are difficult to discern in individual images from specific years. By extracting data on the most common landscape features, these key structures will later be used to train neural networks to analyze their typical parameters, frequency of occurrence, interrelationships, and more.

In the simplest case, this allows for restoring parts of the archaeological object covered by extensive vegetation or poorly seen from space in some seasons or because of insufficiently high spatial resolution, as illustrated in [63]. It means that “the most recent” does not always mean “the best” when applied to imaging archaeological objects, and sometimes older images with a poorer resolution give more information than those with a higher resolution.

For Rujm el-Hiri, this observation pertains to the smaller-scale circular walls, both external and internal, which are more clearly visible in the 2021 and 2018 images. Typically, archaeological remains that belong to a single unit (i.e., of the same construction phase) exhibit consistent colors across observational bands and possess similar spatial characteristics [63,104,105]. Examples are given below in Section 4.2 and Section 4.3. The images reveal that the smaller circular structures with fragile walls adjacent to the 10 m-wide outer walls do not conform to the outer design of Rujm el-Hiri, raising the possibility that they were added in a later period. This also applies to the narrow walls built upon the monumental walls of Rujm el-Hiri, as evidenced in both satellite images and aerial photographs (see details in Section 4.3 below). This was also identified in the excavations conducted at the site by Freikman [28], pp. 37–38, Figure 2.37.

Figure 3. Rujm el-Hiri site, as seen from space in different years and seasons. High-resolution images from Pleiades satellites processed by CNES/Airbus are provided by Google Earth Pro. Eye altitude is 460 m, tilt—zero.

Figure 3. Rujm el-Hiri site, as seen from space in different years and seasons. High-resolution images from Pleiades satellites processed by CNES/Airbus are provided by Google Earth Pro. Eye altitude is 460 m, tilt—zero.

4.2. Rujm el-Hiri’s Vicinity—Tracing the Critical Repeating Landscape Patterns

The upper panel of Figure 4 provides a broader view of the area surrounding Rujm el-Hiri. Structures, settlements, and field walls extend along the slopes of the Golan Heights, northeast of the Sea of Galilee, reaching up to 30–40 km from Rujm el-Hiri, both to the north and south. A similar landscape pattern is observed in some areas to the northwest and west of the Sea of Galilee, but with lesser preservation, as those areas were more intensively cultivated in the following periods. Except for modern roads and objects shown in Figure 4 in brown, the landscape represents a mixture of topological features characteristic of agriculture and farming, shown with yellow arrows. As shown in the bottom panel gallery, these have been divided into several distinct types, including linear-shaped walls, ‘flower-like’ fences, large round-shaped singular structures, and tumuli. In the following section, we will discuss these in detail.

4.2.1. Linear Walls

The area’s geological history is characterized by volcanic activity, resulting in thick basalt layers that make the terrain poorly acceptable for intensive agriculture. At the same time, the region’s fertile volcanic soil has historically created an attractive environment for grazing cattle and farming. The region’s inhabitants primarily bred sheep, goats, and later cattle, with a minimal number of pigs, e.g., [106,107]. They also occasionally consumed fish from local lakes and probably brought it from the Sea of Galilee, e.g., [16,108]. The primary feature identified is long, linear walls. Their lengths and the spaces between them vary with the landscape, ranging from 100 to 400 m in length and 30 to 100 m in width. Defining large rectangular spaces measuring tens of meters on each side, these linear walls likely served multiple purposes: they may have enclosed livestock, marked the boundaries of crop fields, etc.

Figure 5 provides an example of how these linear features were structured. The site shown in Figure 5 has been submerged beneath the Revaya reservoir for the past four decades. Historical satellite image analysis reveals that before the reservoir’s creation, a small inlet-outlet lake with a seasonally fluctuating water level existed in its place. The natural lake is visible in Landsat images from the spring of 1981, before the reservoir’s construction in 1983–1984. A series of artificial reservoirs, including the Revaya reservoir, were created in the region between 1970 and 2019 not through excavation but by constructing C-shaped dikes that blocked the outflow of naturally occurring lakes or depressions fed by small rivers or springs [109]. This technique partially prevents stream water from continuing to flow downhill and allows water accumulation in natural synclines. Over the last 20 years, the water levels in these small reservoirs, located east of the Sea of Galilee, have fluctuated significantly due to climate change, sometimes revealing exceptionally well-preserved archaeological remains. Figure 5a shows the reservoir with the maximal water level seen from above in 2018, while Figure 5b depicts its’ southern end while the water level is down in 2021. The site is 4.5 km southeast of Rujm el-Hiri and repeats all critical features in its vicinity but with better clarity.

Image analysis shows that the historical level of the seasonal body of water was significantly lower when the walls were constructed compared to the current state of the artificial reservoir. Figure 5b illustrates that the ancient lake level was shallow, even lower than what is observed during today’s driest periods. There are still walls and tumuli submerged underwater. Visual analysis of geological features suggests an extended period during which the lake maintained a second, quasi-stable water level before the dike’s construction. The irrigation canal in the eastern part of the lake was built later (in comparison with the period of the thick walls), as it runs over the older, more square-shaped enclosed areas. The canal terminates at this second stable level, marked by the dashed yellow line. Geophysical studies are necessary to accurately determine the periods when the original site and the canal were constructed based on the stratification of lake deposits.

The area is divided into rectangular sections that conform to the local topography, with sizes ranging from approximately 50 × 60 m to elongated sections of 35 × 200 m. Inside the sections, one can see large-scale tumuli, often connected by the walls, and smaller-scale signatures looking like small dots scattered inside several sections. Such signatures are either very small tumuli or remains of rooting sites of long-cultivated trees. The area exhibits clear signs of secondary use, a common feature in the region. The site’s good preservation allows us to estimate the walls’ length and original width. There are two types of walls seen in satellite images. They generally can be described as “thick” (i.e., at least 2 m wide) and “very thin” (the exact width of which is difficult to measure from above) at the edge of the satellite resolution. The measured width of the walls can sometimes appear greater than it is due to the scattering and post-depositional disturbance and dispersal of the building material. Preliminary results indicate that the thick walls may be up to 3.5 m wide, like the walls of Rujm el-Hiri. The thin walls often follow the thick walls, separating the area into smaller sections and creating passageways that were not part of the original design if one suggests that the thicker walls are older since thinner walls often go above or cross them, breaking the initial symmetry. Figure 5c, a close-up of the area indicated by the green rectangle in Figure 5b, illustrates the design of these passageways. A possibly older, thicker wall forms one side of a passageway, while the other side consists of a later, thinner wall that was apparently added to it. There are also passageways or waterway systems in the area created in the later period using two thin walls. Notably, thin walls are sometimes built on top of thick ones.

Figure 5d enlarges the area indicated by turquoise in Figure 5b. It shows a fragment of the lake’s eastern bank as it appeared in ancient times, featuring two passageways and four large tumuli numbered 1–4. The passageway near Tumulus 1 was constructed from ‘old’ walls and, according to the elevation landscape profile, could have initially been used as a waterway connected to the stream feeding the fields surrounding the lake and the ancient lake itself. The southern passageway, made from a combination of the old thick and later-period thin walls, could have served the same purpose during a later period when the lake level rose to the second stable level. Figure 5e,f are close-ups of Tumulus 1 and the rarely observed elongated Tumulus 4. The possible purposes of such objects will be discussed further in Section 4.2.4.

Figure 6 illustrates how the area’s inhabitants adapted to changing climate and environmental conditions to enhance farming efficiency. This site, which combines old thick walls and later-period thin walls, is located just 800 m from Rujm el-Hiri (see the “larger-scale walls” sign to the right of the oval in the upper panel of Figure 4). Topographic analysis indicates that the larger thick walls ran along a former stream branching from the Nachal Samach stream. Along with another former stream connected to the still-active Daliyot stream, this waterway fed the field of round-shaped fences, likely used for agricultural purposes (see Section 4.2.2 below). In Figure 4, the convergence of two former streams in the area of the flower-like wall cluster is depicted as landscape features resembling a light-colored fork, with the linear walls visible between its two prongs to the northeast of the oval. Analogous to Figure 5, the old thick walls often served as a foundation for thinner walls that divided the field into smaller sections. The space between the old and later-period walls likely became a passageway between newly built sections when the stream dried up. In many areas, the new walls overhang the old ones; in others, they are separate and indicate a field restructuring. The older walls are estimated to be 5 m wide, though their original width was probably about 3–4 m.

4.2.2. Flower-Like Fences

The round-shaped fencing method, deeply rooted in the historical context of the Eastern Mediterranean, has been influenced by various climatic, economic, and social factors over millennia. From the Late Neolithic to modern times, round-shaped fencing has played a critical role in livestock management and the overall organization of agricultural landscapes. In the Southern Levant, with its variable environments, these round constructions served multiple purposes, including controlling livestock movements, protecting crops from pests and wildlife, acting as windbreaks, and contributing to soil and water conservation. In contrast, some smaller ones might be used as dwellings covered by some light material. They may be called camp fields, platforms, or broad rooms in different studies, e.g., [110,111,112,113,114].

Meanwhile, some of the activities associated with small-scale, round-shaped structures could have been more effective in the rocky land of Golan Heights. As illustrated in Figure 4, round-shaped fences are relatively scarce compared to the vast areas covered with rectangular walls. They are usually smaller, typically ranging from 3 to 30 m in diameter. While the larger linear-shaped walls might be used primarily for livestock fencing (see Section 4.2.1), the smaller round-shaped fences, typically ranging from 3 to 30 m in diameter, could have been built for different purposes like agriculture and putting back the corral herds from the large rectangular fields for protection overnight. Their location implies another possibility. These ‘flower-like’ agglomerations are found near streams, suggesting they were built for purposes requiring ample water, likely agriculture. This is shown very clearly in Figure 4. Therefore, the difference in size and spatial distribution between the square areas delineated by the long walls and the smaller, round areas outlined by the round structures imply a different purpose or a different scale of activities (for example, a large field could be used for cereals, but a smaller one is needed for lentils such as chickpea and other species exploited during the period at hand).

Figure 7 shows six examples of typical flower-like clusters in the 30 km area around Rujm el-Hiri. Figure 7a is one more site covered mainly by water since the Dvash reservoir’s water level fluctuates like that of the Revaya reservoir. It is located 2.5 km from Rujm el-Hiri to the northwest. The image is from November 2021. The agglomeration of ~20 m-wide circular walls covers a small 90 m-wide hill, which is primarily underwater, resulting in excellent preservation of the walls. Parts of the site remained submerged even in 2021. An exciting feature is the presence of a singular small, ~7 m by 4 m, rectangular structure amidst the flower-like fences. The similarity in band color of the walls suggests that they might have been constructed during the same period as the circular walls.

Figure 7b shows a flower-like cluster stretching approximately 130 m along the Wadi Hafina stream. This site is 19 km northwest of Rujm el-Hiri. Many similar clusters are found along the stream.

Figure 7c depicts one of the numerous flower-like structures surrounding the Revaya reservoir, previously discussed and shown in Figure 5a. The semi-round-shaped fences are approximately 15–25 m in diameter. Like Figure 7a, this site features a rectangular structure within the fences. However, this structure is notably larger, measuring 37 m by 27 m, with at least one large room and two smaller rooms visible from above.

Another example of a large agglomeration of circular-shaped fences is shown in Figure 7d. It measures about 180 m in length and 130 m in width. The site is between the Nachal El Al stream and a vast wall, creating an approximately 80 m wide circular-shaped protected space with a large tumulus at the center, above which trees grow. It is worth noting that trees and bushes often grow above ancient sites because they retain water, which is crucial during hot summer. The site is located 4 km to the south of Rujm el-Hiri.

Another example is the flower-like cluster found on the sharp-edged hill by the Nachal Akbara stream (Figure 7e). The site is poorly preserved due to the hill’s loss of vegetation, leaving it primarily bald rock. Despite this, the site’s cells are more uniformly round than those in Figure 7b, with circular features measuring 10–20 m in diameter. Like objects shown in Figure 7a,c, some structures can be interpreted as houses: one is a square-like house measuring 9 m by 8 m, and the other is a round feature with an 11 m diameter.

Figure 7f illustrates the merging of flower-like agglomerations of typical 20 m-wide circular features into larger clusters connected by walls. This place is located 12 km to the north of Rujm el-Hiri. The clusters in the North become larger and more scattered because the area has more springs and streams. As a result, the clusters keep the circular form only occasionally, filling space between wide pastures.

Figure 8 provides examples of the evolution of flower-like clusters. The site in Figure 8a, located by Nahal Kanaf, 6.5 km southwest of Rujm el-Hiri, features a central area approximately 70 m wide. Figure 8b shows a similar layout but with clearly visible rectangular structures forming additional “leaves” around the flower’s center. The most prominent structure measures 14 m by 15 m. This site is located 14.3 km north of Rujm el-Hiri. Figure 8c depicts a site near Nahal Bazelet, just one kilometer north of Rujm el-Hiri. The center of this flower-like cluster contains a 19 m-wide tumulus, and the total diameter of the structure is about 60 m. The 32 m-long wall connects this agglomeration to a smaller, similar structure that seems poorly preserved or unfinished.

4.2.3. Large, Round-Shaped Singular Structures

Another distinct archaeological pattern identified through remote sensing is the tens-of-meters-wide round-shaped structures, comparable by size and shape with Rujm el-Hiri. Their conceptual similarity to Rujm el-Hiri makes their study crucial for understanding not only the use of the landscape but also the purpose of Rujm el-Hiri itself. Their preservation is typically poor, and there are numerous signatures of the secondary use of their material, building attachments to the internal and outer walls, or even roads constructed above them. Several examples are given in Figure 9 and Figure 10. Our team has recently analyzed the sizeable circular structure shown in Figure 9a using a combination of image reconstructions and neural network image recognition [63]. This site, located 13.5 km north of Rujm el-Hiri, features a double circle. The linear wall running above the double circle is recognized as a later-period noise. Still, the labyrinth- or spiral-shaped structure connected to the internal circle is built simultaneously with the walls.

Another large-scale circular object of the same type is located 750 m north of Rujm el-Hiri (see Figure 9b). It shows evidence of an internal wall, and analysis of images taken in different seasons reveals possible later-period, symmetrical attachments to the inner part of the outer wall, likely used as dwellings.

Two other round-shaped structures, shown in Figure 9c,d, are singular rings, possibly built as part of the same phase as the “thinner walls” discussed earlier and the houses attached to the larger-scale circle in Figure 9b. These rings are smaller and generally better preserved, although modern activities in the area have damaged them. It is possible they were constructed on top of older structures no longer visible.

The circle in Figure 9c is located 2 km east of Rujm el-Hiri, while the circle in Figure 9d is situated 3.2 km north. Both circles are divided into smaller sections by walls. The circle in Figure 9c is split into two parts: one filled with remains of buildings or tumuli and the other empty except for a linear wall crossing it.

Figure 10 shows three similar double-wall structures, each with an entrance facing southeast. The site depicted in Figure 10a is situated 3 km northeast of Rujm el-Hiri. The circle shown in Figure 10b is a newly discovered round-shaped object found near the one shown in Figure 9a, identified in the study by Eppelbaum et al. [63]. The circle in Figure 10c is located 13.5 km northwest of Rujm el-Hiri, in modern-day Syria, among farms. All these structures were readapted multiple times in later periods. As a result, two small circular features are connected to the main entrance of the outer wall in Figure 10a, along with other additions to the outer and inner walls. Similar later-period modifications distort the original shapes of the buildings shown in the other panels. In Figure 10b, these additions are mainly circles, while Figure 10c includes circles and tumuli.

The similarities in geometry, design, and measurements may suggest a connection between these sites, including Rujm el-Hiri, possibly implying that these structures might be related or signify a single phenomenon. In any case, their intricacy indicates a high level of sophistication of the society that built them.

4.2.4. Tumuli, Dolmens, Cairns or Roundhouses

The Levant is famous for its dolmen fields, cairns, and tumuli. These archaeological features are irregularly distributed worldwide and are typically dated from 4000 to 1000 BCE (2200–800 BCE in the Northern Levant). Dolmens are megalithic structures that form a chamber. They comprise two or more slab stones with a large flat capstone above them [115,116,117]. Dolmens generally range in size from small structures with capstones just a few meters in length to larger ones with capstones weighing several tons, creating large chambers. Various construction methods have been identified, including intricate engravings on the stones. Depending on their purpose, cairns can range from small, human-sized piles of stones used as land markers to larger mounds several meters in height and diameter, but they typically lack the internal space that dolmens have. Tumuli are earthen or stone mounds. They are generally larger than dolmens and cairns, often covering substantial areas with heights reaching several meters or more and diameters ranging from a few meters to over a hundred meters. Tumuli usually contain chambers within the mound, which can significantly add to their size. Tumuli may be constructed over dolmens, giving them a round shape when viewed from above [116]. A stone heap is another round-shaped feature visible from space, commonly found in mountainous areas with minimal soil cover. Initially, these stone heaps were one of the structure types mentioned above, but over time, they have compressed into a nearly homogeneous mass covering hills, distinguishable from above by color. They may also be collapsed structures on current hills or mountain slopes with poor soil cover. When all these objects have a diameter of less than 6 m, it is often impossible to differentiate between their types in satellite images. For simplicity, we will refer to all of them as tumuli. Although researchers admit the existence of various kinds of tumuli might have had different functions and/or meanings, this remains debated, mainly because some (if not most) excavated tumuli were found empty of indicative artefacts. This phenomenon is usually attributed to looting, as most researchers agree that tumuli functioned as graves (ibid).

Almost all figures shown above contain tumuli of various sizes, with a notable example depicted in Figure 5b. First, a large-scale tumuli chain runs along the lake’s west bank. Three are visible on the exposed lake bottom, two are located to the north (out of the frame of the figure), and one is visible in the water. Two others are near the southern bank, and presumably, two more are submerged. These tumuli are 20–25 m in diameter, with 60–80 m of space between them (from top to top). Other similar or smaller-scale tumuli are located within rectangular sections.

A noteworthy aspect of the tumuli identified is that many seem to be interconnected by long walls. This feature is often seen both in the Northern and Southern Levant, e.g., [116,118]. Additionally, some tumuli, such as the ones shown in Figure 5b, exhibit attached structures, which may be later-period additions but may also represent integral parts of some of the so-called ‘tumuli’ as suggested by the remote sensing analysis. Examples are shown in Figure 5e,f. In Figure 5e, where there is a rectangular structure, while in the elongated tumuli of Figure 5f, it appears as a series of houses.

Figure 11 shows an agglomeration of tumuli, part of the tumuli field north of Rujm el-Hiri depicted in Figure 4. The sizes of these tumuli range from 5 m to 21 m. The straight lines crossing the field are later-period walls. Image analysis indicates that all tumuli are linked by walls from plausibly the same period. Additionally, some tumuli have spherical courtyards around them, as highlighted in the zoomed-in box in Figure 11a.

Considering the earlier suggestion that 20 m-wide circular structures seen as cells from above are used in the wettest parts of the land for agricultural purposes, it brings an idea that some of the so-called ‘tumuli’ could be collapsed round-shaped buildings used as warehouses or dwellings rather than graves. Courtyards and fields attached to structures and houses are still common in the countryside. This is clearly illustrated in Figure 8c, which shows a large-scale tumulus in the very center of the flower-like structure. Therefore, it is possible that some of the tumuli, especially the larger ones seen on this site, represent other types of structures, such as warehouses, shelters, or dwellings.

This can be further exemplified by another landscape containing tumuli, located 0.7 km southwest of Rujm el-Hiri (see Figure 11b). Fences link all the medium-sized tumuli. Figure 11c shows a tumuli field within rectangular structures near the site, as shown in Figure 7e. Some tumuli are connected by walls and form chains, while others are isolated. A notable characteristic is the absence of other dwellings, which challenges the notion that all tumuli are solely graves. Some theories suggest that herding groups, part of a settled population in nearby regions, migrated seasonally for herding. Their settlements might have been located elsewhere, separate from the areas where they herded during winter or summer. In this context, the tumuli could have served social and territorial purposes, marking boundaries, pathways, and other significant features.

To demonstrate this with statistical analysis, let us calculate the number of tumuli of various sizes, constructing a histogram of their diameters. Figure 12 displays the distribution of tumuli sizes in three selected areas: (i) the tumuli field shown in Figure 11a, (ii) the field located northwest of the Revaya reservoir, centered at 32.878660° 35.762248° with a 1.5 km radius, (iii) the bottom of the Revaya reservoir depicted in Figure 5, and (iv) the tumuli field near Rujm el-Hiri, centered at 32.905833° 35.798311° with a 500 m radius. The black distribution, representing the frequency of tumuli of specific sizes across the entire dataset, is skewed to the left with a median diameter of 8 m, a mean diameter of 9.4 m, and a standard deviation of 3.6 m. It suggests at least three sub-distributions: one with the most common diameters around 7 m, the second with typical sizes between 11 and 13 m, and the third with 15–16 m diameters.

A visual analysis indicates that the large tumuli are located near the water, with the most prominent structures embedded in the flower-shaped formations, as shown in Figure 8c. The white histogram, representing the distribution of tumuli beneath the Revaya reservoir, highlights two peaks contributing to the overall distribution of larger structures.

In statistical terms, multiple peaks suggest different mechanisms influencing the process. Specifically, the histogram exhibits multimodality, where each peak may represent a distinct underlying population or category of tumuli. This implies that the distribution is not normal (unimodal) but rather a mixture of several distributions. The presence of multiple modes, or “humps”, in the data may indicate that the tumuli were built for various purposes or that different construction methods or cultural practices led to distinct size categories. The skewness of the distribution and the presence of outliers (i.e., very large tumuli) further suggest that while most tumuli fall within a common size range, exceptional cases may reflect unique historical, cultural, or environmental influences. Applying measures such as kurtosis (2.6 m) provides additional insights into the distribution’s shape, helping us understand the extent to which the data are clustered around the central values or spread across a broader range.

From the archaeological viewpoint, these statistical features suggest that tumuli found in different landscapes could have served diverse purposes, with burial being just one possibility. Understanding the multimodal nature of the distribution could help archaeologists infer the range of activities associated with these structures. It should be noted, however, that the actual sizes of the tumuli can only be confirmed through on-site examination. Since no statistical studies have been conducted on this topic, we begin with approximate estimates limited by the precision of remote sensing.

In sum, these results suggest that what seems to be a single phenomenon or “design” of ritual/burial objects may represent several phenomena, including mundane structures, complicating the determination of a particular function. Future remote sensing and field studies can help suggest further characterization of these remains for different purposes based on the surrounding landscape.

4.3. Mixed Types of Landscape Patterns; Older vs. Later Period Layers

Using the information obtained from analyzing various object types typical for the landscape near Rujm el-Hiri, one can attempt to interpret more complex structural combinations. A significant feature observed in satellite images is the multi-layer nature of archaeological objects, indicating that many structures were reused or damaged by subsequent use, sometimes with little regard to their original purpose.

Figure 13a encapsulates diverse elements: smaller and larger tumuli, a circular structure (not as old as those shown in Figure 9a,b and Figure 10), and field walls creating passageways between them. Walls link the largest tumuli, and one is even attached to a probable passageway. The structural differences between the largest tumuli and those scattered in the field suggest varying periods of construction or, alternately, different functions, which can only be validated via systematic archaeological excavations. Figure 13b is an example of the combined use of the old thick wall, suggesting an unfinished building of the Rujm el-Hiri size and attached later-period flower-like structure possibly used for agricultural purposes, with a large structure attached. The site is located 1.7 km south of Rujm el-Hiri.

Figure 13c shows a large double-wall circle almost filled with structures that do not belong to the period of its construction. Furthermore, thin lines across the object are from the third period, even later than the semi-circular structures inside.

Reusing the above archaeological objects was so common that finding a spot unmodified since the Chalcolithic or Early Bronze Age period is difficult. The task becomes incredibly challenging when later-period objects, such as roads used for ages by many generations, follow the old ones.

To illustrate how this pertains to Rujm el-Hiri, Figure 14a shows the landscape of an area located 2 km east of the circles discussed in [63] and shown in Figure 9a and Figure 10b. In this case, it is possible to differentiate between later-period thin walls and older-period thick ones. Although the former primarily follow the older walls, they diverge in some places, as seen if one follows their course.

This phenomenon allows training of the human eye and neural networks, suggesting that the same combination of older and later-period walls is seen in Rujm el-Hiri’s satellite and aerial images (e.g., Figure 1 and Figure 3), as shown in Figure 14b. Although following the old walls’ general shape, the lighter, thin walls run less accurately, which may be attributed to two known periods of the monument’s construction noted in [21,26].

It is conceivable that the space between the thin walls might have been used for secure passages. Some researchers [119,120,121,122] propose that Geshur was a contested region during the late Bronze Age. However, this is only a hypothesis since no direct evidence has been found to confirm this. Rujm el-Hiri could have functioned as a defensive structure in the context of the biblical land of Geshur in later periods.

According to the analysis, the outer small-scale thin circles, three indicated in Figure 14b, do not belong to the original construction either. Attaching additions to the entrances of older structures appears to have been widely employed, as evidenced by the objects shown in Figure 10. More studies, including on-site investigations, are required to understand the details and nature of these additions and the periods during which they were built upon Rujm el-Hiri.

5. Discussion

Remote sensing is a well-established tool in archaeology, with recent examples demonstrating its effectiveness in identifying and studying Levantine archaeological sites, e.g., [49,118]. However, the megalithic structures of the Southern Levant have not yet been extensively analyzed using these techniques. Our study represents the first step toward filling this significant gap in understanding the region’s exploitation in ancient times, particularly those influenced by one of the most famous Levantine megaliths, Rujm el-Hiri. The work shows how remote sensing may help identify new archaeological objects, study known ones, and understand links between landscape features and the meaning of different sites. Our findings raise many questions regarding the life-ways of the community that built and used Rujm el-Hiri. They can be answered only after a new comprehensive analysis of the corresponding structures and Rujm el-Hiri itself, and of course, only when integrated in systematic on-field archaeological excavations.

An analysis of remote sensing data indicates that most identified objects were constructed during two distinct periods. Rujm el-Hiri, the thick walls along the streams, and large-scale circles were likely built in the earlier period, predating the main tumuli construction period, alongside flower-like fences and thin linear walls. Given the paleoclimatic reconstructions discussed above (see Introduction), two main periods may be suggested as a possible timeframe: between 5000 and 4500 BCE and between 4000 and 2200 BCE. Sebag [110] finds that the thick fortification walls were built in the Early Bronze Age (3100–2000 BCE), which would adjust the possible construction date of Rujm el-Hiri to 3100–2200 BCE, based on the similarity of these objects observed in the satellite images. Further research, including excavations of selected features, including perhaps Rujm el-Hiri, is needed to resolve this timeline.

Although the remote sensing data analysis cannot reveal the exact dates of the construction of the objects under study, it indicates two consecutive episodes (excluding modern impacts). The first episode includes Rujm el-Hiri, associated objects, and features like thick walls and sizable circular structures. In contrast, the second episode overlaps the first and consists of thin walls, round-shaped fences, and dolmens or tumuli that appear coeval and interconnected. Previous studies, e.g., [115,116,123] suggest that most of the dolmens in the area were constructed during the Intermediate Bronze Age (2200–1550 BCE). Therefore, it is logical to indicate that the objects linked to the tumuli (walls and fences) belong to the same period. Regarding Rujm el-Hiri, it is supposed that monumental structures are built during the most favorable climate conditions combined with the highest development of the cultures. This evidence implies that Rujm el-Hiri and other similar large-scale round-shaped structures discussed in this study should be dated earlier than tumuli, either to the period between 5000 and 4500 BCE (during the peak of societal development in the stable-warm period) or to the period between 4000 and 2200 BCE, before the construction of the dolmens.

Future studies require conducting a comparative analysis of large-scale circular structures across the Levant to trace the origins of this tradition and understand the reasons behind the substantial investment in these monuments. For instance, a recently discovered ring in Egypt [124] is even larger than those identified in our previous study [63]. The labyrinthine internal structure of Rujm el-Hiri bears similarities to a recently uncovered Minoan-period structure in Crete, which features several circular walls connected by intersecting radial walls and a round building in the center [125]. This design is uncommon in Minoan civilization, and its purpose remains unclear. Notably, from an aerial view, the Crete structure displays two distinct rings. It is plausible that these structures were built around the same period and served similar functions, such as defense and/or other social or ritual purposes. This highlights the importance of examining Rujm el-Hiri’s purpose within a broader context and comparing discoveries with those from other regions.

There is another aspect rarely mentioned in the archaeological literature: the actual height of the Rujm el-Hiri site remains unknown, as reconstructions differ between excavators. We see only partial remains resulting from collapse and other post-depositional processes. Additionally, we lack information regarding the number and locations of the internal doors connecting the circles, whether they were scattered in a labyrinth manner or had a regular sequence. We know nothing about the eternal chambers/rooms or yards between the walls, although we can see the partition walls from above. As others have suggested, such rooms could be used for people gathering as storage rooms or as small corals. Some information can be obtained from geophysical methods, but only future excavations can answer the main question regarding the actual purpose of the monument.

Let us estimate the possible height of the Rujm el-Hiri walls. The current height of the object is from 2.5 m (walls) to ~4 m (tumulus in the center). However, given that (i) all massive buildings sink into the ground under their weight, e.g., [126], (ii) the soil level around high walls gradually increases because of natural processes, such as dust and alluvial mass accumulation and deposition of fragments of the fallen ceiling (if there was any) and the wall decay. The sink effect cannot give more subsidence than the soil thickness at the Golan Heights plateau, which is ~0.5 m [127], plus the base height (if there is any). Meanwhile, the sedimentation process around and inside left buildings can be significant because they represent a barrier for floods’ gravel and dust particles; as a result, old buildings become surrounded by more soil. It aligns with a detailed cross-section height profile of the landscape, which shows an elevation increase within a 100 m radius around Rujm el-Hiri. The area exhibits a 3 m height difference at its peak compared to the surrounding ground level. The natural landscape may determine some part of it, but a part is also defined by slowly accumulated soil. If one estimates the rate of accumulation of dust in Israel from all sources [128,129,130], which is thought to be at least 1 mm per year by the seashore and ~0.5 mm/year farther from the sea, an increase in the soil level for 4000 years may be ~2 m. Considering the slope and the erosion of soil accumulated around the outer wall, it can be estimated that the actual height of the walls at Rujm el-Hiri could be at least 0.5 m higher. More exact estimations of the actual volume of debris around the object need an on-site analysis of the volume of the collapsed material around the walls.

The other important aspect is discussing the purpose of tumuli in the area. The debate on structure, style, and function in archaeology, especially regarding architecture and landscapes, has a long-standing history, e.g., [131,132,133,134,135]. It is now widely recognized that assuming buildings, structures, or landscapes serve the same purpose based solely on their form may overlook the complex social, cultural, and environmental factors that influence their design and use. Simply equating form with function risks neglecting the nuanced roles these elements might play in society. The literature analysis [115,116,118,136] shows that the interpretation of artifacts found in dolmens and tumuli can vary significantly based on the quality and context of the findings. We propose that the similarity of the shapes of all tumuli does not mean the only purpose of using these structures. In the tested area, the objects that appear to be tumuli or even built with the same technique could be used as warehouses, shelters, or dwellings, in addition to their widely admitted purpose as burial locales. In particular, the connection of tumuli by walls has been noted in the literature, but without remote sensing, it was challenging to determine how often this occurs. Our analysis reveals that most tumuli are linked by structures believed to serve herding or agricultural purposes, with only a tiny number of tumuli standing independently. The statistical analysis confirms that the distribution of tumuli sizes suggests the existence of at least three types of tumuli, perhaps built for different purposes.

Large 15–20 m-wide tumuli are found alongside flower-like agglomerations in wet areas. They are often positioned at the center of the “flower” or serve as network nodes. Figure 8c and Figure 13b clearly show this, with the tumulus shape aligning with the surrounding leaf-like fields, providing access to the structure from all directions.

Tumuli are often connected by walls, as also noted by other researchers, providing quick access from two or more neighboring fields. Evidence of agricultural and farming activities is present in many sites, yet no structures other than tumuli are found. This suggests that some tumuli may have also served different purposes. The presence of circular window-like portholes and holes in the upper stones of certain dolmens supports this interpretation. The latter could have functioned as smoke vents if fires were lit inside dwellings or shelters. Our analysis also helps explain the presence of empty tumuli without signs of looting. This would be expected if they were used as warehouses or large silos for organic materials. Other tumuli filled with animal bones could have been used for food preparation or conservation. The use of tumuli as shelters or dwellings is also in agreement with the discovery of drawings (that could only have been created over extended periods), as well as atypical collective burials with heavily fragmented bones often found in unusual for ritual burials places and mixed with objects that typically exist in dwellings, e.g., [115,136]. Meanwhile, these ideas do not dismiss previous theories about the broader significance of tumuli, suggesting that, in many instances, tumuli could be later added to older features, reusing their stone material and referring to the old features as a sort of ‘memory of places’.

More combined archaeo-geological studies are required to check the ideas proposed in the research and answer the questions regarding the initial purpose of building Rujm el-Hiri. Undoubtedly, an examination of the remotely revealed archaeological targets in this area must be continued using advanced surface geophysical methods (Electric Resistivity Tomography (ERT), Ground-Penetrating Radar (GPR), and high-precision magnetic prospecting) and archaeological sequence excavation.

6. Conclusions

The key lessons that we have learned from the analysis of the landscape surrounding Rujm el-Hiri are the following:

• The landscape surrounding Rujm el-Hiri exhibits several recurring patterns: large-scale singular circles comparable in size to Rujm el-Hiri, thick walls analogous to those of Rujm el-Hiri, linear thin walls dividing the land into rectangular sections, flower-like clusters of small round-shaped fences, and tumuli.

  -

  Large ~40–90-m-wide circles are extremely rare and appear as old as Rujm el-Hiri. These typically consist of two concentric circles with spiral-like structures attached to the inner circle.

  -

  The three-meter-wide walls are primarily built along streams or near lakes. Based on their visual characteristics, they were likely constructed during the same period as Rujm el-Hiri.

  -

  The thin linear walls, constructed later, often overlay older landscape features. They form rectangular fields that were perhaps used as pastures.

  -

  The flower-like clusters of small ~20 m-wide circular walls or fences are less common than quasi-rectangular pastures. The fact that they are predominantly found in wet areas, typically near existing or dried-up streams, may suggest that these structures were used for agricultural purposes.

  -

  Based on satellite imagery analysis and structural examination, the tumuli appear to have been constructed during the same period as the linear walls and flower-like clusters. Our study suggests that while the tumuli exhibit similar designs, they may have served different purposes while sharing a common, circular form.

• Most archaeological structures in the region were reused long after their original construction. In some instances, this involved adding new layers or building walls over older ones, while in others, it damaged the original structures. Rujm el-Hiri is a prime example of such repurposing. The simplified architectural designs of later-period walls and attached structures may indicate that the original construction technologies were lost over time.
• The geological structure of the Rujm el-Hiri’s area has been shaped by the tectonic evolution of the region, leading to the rotation of blocks and, therefore, the migration of its location and the direction of the main entrance and the radial walls over time. The region’s integrated geophysical analysis (mainly GPS and paleomagnetic reconstructions) reveals that the Rujm el-Hiri site has rotated counterclockwise and shifted from its original location by tens of meters. This means that the current orientation of the radial walls and entrances was not the same as ~4000–2000 BCE, and the speculations that they were aligned with celestial bodies of the past are not supported. Therefore, Rujm el-Hiri was unlikely an observatory.
• Studies of the unmasked archaeological features should be continued using advanced remote sensing procedures and surface geophysical methods (e.g., Ground-Penetrating Radar, high-precise magnetics, and Electric Resistivity Tomography (ERT)) with further archaeological excavations.
• Our study reveals the objects besides Ryjm el-Hiri that have not been categorized before and illustrates how remote sensing can identify and contextualize archaeological landscapes by detecting previously unknown sites and revealing patterns that suggest complex relationships between structures and the natural environment. It aligns with ongoing efforts to use technology to uncover how ancient societies adapted to and transformed their surroundings, adding a crucial piece to how megalithic cultures developed in the Levant and beyond.
• This study’s implications extend beyond the Levant, inviting comparative studies with other megalithic structures and tumuli worldwide. The possible parallels between Rujm el-Hiri and other monumental Mediterranean structures of the same period suggest that a broader cultural or technological tradition may have influenced the design and construction of such sites across the region. The similarities in structure, such as concentric circles and radial walls, highlight the need for further interdisciplinary research that combines archaeological, geophysical, and paleoenvironmental data to understand these monuments’ origins and purposes better.