Evaluating Earth Construction Techniques on Ancient Architecture: An Exploratory Use of Empirical Field Tests in West Asia (Tell Zurghul/Nigin)

Abstract

The research empirically evaluates ancient earth construction techniques through the analysis of archaeological adobe samples from Tell Zurghul/Nigin, south-eastern Iraq, dating from the mid-5th to mid-3rd millennium BCE. Simple, non-standardised empirical field tests were employed to obtain preliminary material characterisations, valuable for pilot assessments and gaining further significance when compared with quantitative analytical results. Their application evaluates the functionality of these tests while integrating archaeological insights with material science, underscoring the importance of multidisciplinary collaboration in earthen heritage conservation. Sixteen samples—fifteen archaeological and one modern—were analysed to assess raw material composition, grain size, clay behaviour, organic content, cohesion in wet and dry states, and surface adhesion. Results demonstrate notable homogeneity in material composition across the time span, primarily fine sands with minimal clay or silt. This suggests favourable drainage, minimal shrinkage, and reduced cracking but limited cohesion, implying a potential need for stabilisers such as plastic clays or fibres in construction. These findings inform conservative strategies for the preservation and restoration of earthen structures at the site.

1. Introduction

1.1. Significance and Challenges of Cultural Heritage Conservation

Cultural heritage is the legacy of tangible and intangible assets inherited from the past that a group or society deems worthy of preservation for future generations. With its artistic, historical, and economic significance, it is a vital part of human identity and represents the shared narrative that connects past, present, and future [1,2,3,4].

Tangible cultural heritage encompasses physical artifacts, such as archaeological sites, monuments, historic buildings, and works of art. These are concrete manifestations of a community’s history, creativity, and identity, and their preservation is essential to maintain a link with the past. Intangible cultural heritage, on the other hand, refers to the traditions, practices, expressions, knowledge, and skills that are passed down from generation to generation, representing living cultural elements that are deeply rooted in communities and essential for maintaining their identity and sense of continuity [3,5,6,7,8,9].

The need to safeguard cultural heritage, considered as the act of preserving this invaluable legacy, has become increasingly crucial as threats such as natural disasters and climate change, human-induced damage, lack of resources and capacity, and insufficient awareness and engagement continue to endanger these irreplaceable resources. Its preservation is, nevertheless, recognised as a global imperative, as evidenced by its inclusion as a specific target in the United Nations’ Sustainable Development Goals. Nevertheless, significant challenges remain, particularly in mitigating the deterioration and loss of both tangible and intangible elements under complex social, economic, and environmental pressures [10,11].

Recent research has emphasised the importance of analysing conservation strategies not only during site management but also immediately following excavation, considering financial constraints, preservation approaches, and locally available materials [12] (pp. 38–45). Such analyses are particularly crucial for archaeological earthen architecture, where rapid decay upon exposure to atmospheric and biological agents requires informed, site-specific interventions.

1.2. Archaeological Earthen Architecture and Its Vulnerabilities

Archaeological earthen architecture, a subset of tangible heritage, includes structures built using techniques such as pisé, cob, and adobe, which employ unfired natural earth mixtures composed of clay, sand, and other minerals left to air-dry. These techniques are traditionally and still used in specific world areas, generically comprised between the Tropic of Cancer and the Tropic of Capricorn, in regions where clay is abundant and the climatic conditions are favourable to their continuous use [13], as in southern Iraq, used as a case-study area within West Asia. The structures are normally surface-plastered and protected from precipitation by the roof, and they require continuous interventions (re-plastering of the surfaces, replacement of some decayed structural elements, etc.) for full functionality.

With archaeological earthen architecture, the reference is to ancient buildings recovered during archaeological excavations. They are no longer in use, in most cases the roofs originally protecting the walls from weathering have collapsed, and they have lost their surface plasters. At the time of the discovery, the structures are therefore directly exposed to atmospheric and biological agents without further protection [14]. The conservation of these kinds of structures is challenging, due to their tendency to decay within a brief time once exposed to the atmospheric agents (rain, snow, and wind). Other factors may participate in the rapid decay of earthen architecture, e.g., moisture, groundwater flooding, clay salinisation, the wet–dry cycle, as well as roots, animals, fungi, and human activities (see, e.g., [15,16,17,18]).

The analysis of archaeological earthen architecture in view of conservative strategies demands a multidisciplinary approach, involving archaeologists, geologists, engineers, and architects, to integrate knowledge of historical construction techniques with quantitative assessments of material properties.

Detailed historical research is indispensable to understand the experience of the building, its transformations over time, and the original construction methods. These qualitative data are essential to correctly interpret the degradation mechanisms and phenomena affecting the structure. Complementarily, quantitative analysis—including simple, non-standardised empirical field tests and laboratory-based examination of chemical, mineralogical, hydric, and mechanical properties of archaeological earthen structures and their distinct components (e.g., adobe), as well as monitoring and structural modelling—are essential to define any defects in the construction and to plan targeted interventions. The integration of these two types of data—qualitative and quantitative—is critical, as neither approach alone would be sufficient to fully understand the conditions of an archaeological building [19] (pp. 148–149), [20,21,22,23].

1.3. Preliminary Empirical Field Assessments and Their Role in Conservation

The primary aim of this study is to address the limited empirical knowledge on the properties of archaeological adobe structures at Tell Zurghul/Nigin by applying simple, non-standardised empirical field tests.

Field tests are low-cost, on-site assessments providing preliminary information useful to gain awareness of material properties and vulnerabilities [24,25,26,27]. To characterise the materials, field tests—including visual inspection, colour description, smell, hand manipulation, pastille, bottle, and peeling tests—are performed directly on-site to evaluate material composition, organic content, grain proportion, clay behaviour, and cohesion in wet and dry states [28,29,30,31,32,33,34]. Adapted from architectural practice, these tests allow us to assess ancient construction quality, make synchronic and diachronic comparisons, and perform preliminary evaluations of adobe replicas, which are commonly used in conservation interventions to restore earthen structures [35,36,37]. The comparison between archaeological adobe and purpose-made replicas constitutes an essential methodological step for validating raw material selection criteria, evaluating mechanical and compositional compatibility between ancient and modern adobe, and calibrating the interpretive reliability of field-testing procedures as preliminary diagnostic tools within conservation science.

Field tests are therefore used as a source of information to carry out a basic, descriptive, and approximate evaluation of construction materials, which may be conducted on site by archaeologists, including those without formal conservation training. However, it is essential to emphasise that these tests are preliminary and complementary to laboratory analyses, and they cannot replace thorough assessments and interventions by trained conservation specialists, whose presence is essential for definitive conservation planning [38,39,40]. This study proposes a pragmatic and replicable approach based primarily on empirical field tests, applied to the analysis of earthen architecture at Tell Zurghul/Nigin. By combining the results of these in situ tests with the broader framework of existing analytical data [41], the study contributes to the characterisation of ancient adobe technologies, supports immediate site-specific conservation decisions, and allows for qualitative comparison across different construction phases. This is the first application of such a combined approach across a stratified archaeological sequence at Tell Zurghul/Nigin, spanning the mid-fifth to mid-third millennia BCE, enabling refined characterisation of adobe technologies, informing immediate site-specific conservation decisions, and supporting diachronic comparisons of construction practices. It also offers insights into practical conservation challenges, methodological strategies, and the balance between efficiency, cost-effectiveness, and scientific rigour.

2. Materials and Methods

2.1. Material Dataset and Sampling Context

A total of 16 samples (15 archaeological and 1 modern adobe samples) have been included in the analysis (Figure 1;

Table 1. Archaeological adobe, with indication of provenance, archaeological phasing, period, and typology of the sample.

ID	Archaeological Context	Archaeological Dating	Typology
43	Mound B, Step V (W.389)	Uruk period	Adobe
46	Mound B, Step V (W.389)	Uruk period	Adobe
48	Mound B, AP 6 (SU 861)	Late Ubaid period	Adobe
51	Mound B, AP 4 (W.382)	Late Ubaid period	Adobe
53	Mound B, Step V (W.389)	Uruk period	Adobe
54	Mound B, AP 5 (W.386)	Late Ubaid period	Adobe
61	Mound B, AP 2b (W.362)	Late Ubaid period	Adobe
63	Mound B, AP 2b (W.362)	Late Ubaid period	Adobe
64	Mound B, AP 3a (W.701)	Late Ubaid period	Adobe
65	\	Late Ubaid period	Adobe
66	Mound B, AP 3a (W.701)	Late Ubaid period	Adobe
68	Mound B, AP 2c (W.361)	Late Ubaid period	Adobe
69	Mound B, AP 3a (W.701)	Late Ubaid period	Adobe
79	Mound A (W.458)	Mid-3rd Mill. BCE	Adobe
SG.22.E.302/9	Mound A (W.458)	Mid-3rd Mill. BCE	Earthen plaster

). All the samples come from the Nasiriyah region, in south-eastern Iraq (Dhi Qar Governorate). The archaeological samples have been collected at the site of Tell Zurghul/Nigin, in the course of excavation by an Italian archaeological expedition from Sapienza Università di Roma (MAIN: Missione Archeologica Italiana a Tell Zurghul/Nigin) and exported with the permission of the SBAH (State Board of Antiquities and Heritage) in Baghdad. The modern sample comes from a local adobe-making company in Nasiriyah.

The archaeological site of Tell Zurghul (31°22′36.06″ N; 46°29′36.24″ E), located approximately 7 km south-east of the ancient Sumerian city of Lagash, is an approximately 70 ha settlement known in cuneiform inscription as the ancient Sumerian city of Nigin, which was continuously occupied from the mid-5th to the end of the 3rd millennium BCE. The site is characterised by two extensive mounds located in the central part of the settlement and, respectively, called Mound A and Mound B (Figure 2). Mound A corresponds to the main occupational phases of the ancient city assigned to the 3rd millennium BCE, and it has been investigated in several operations (namely, Area A, D, and E; see [42,43]); Mound B was mainly occupied in the prehistoric phases of the site (mid-5th and late 4th millennium BCE), and it has been investigated in one excavation area named Area B [44,45].

A total of 13 samples come from Area B, and they pertain to various stratified architectural phases investigated both in an open-air operation on top of the mound [architectural phases—henceforth AP—2b, 2c, and 3a; Figure 3a], and in a step trench on the northern side of the mound [AP 4, 5, 6—all pertaining to the Ubaid prehistoric occupation of the mound—and Step V, dating to the late 4th millennium BCE; Figure 3b].

A total of 2 samples come from W.458 in Area E, a massive structure dating to the mid-3rd millennium BCE, belonging to one adobe of the wall and a portion of the preserved plaster applied to the surface of W.458 (Figure 3c).

Selected samples from the site were previously subjected to laboratory-based chemical, mineralogical, petrographic, mechanical, and hydric analyses [41], which provide baseline quantitative information. While detailed results are presented elsewhere, these data allow preliminary cross-validation and evaluation of the empirical field test methods presented here. The samples show a high degree of homogenisation in terms of chemical and mineralogical composition, with the use of local clay sources. Differences are evidenced in petrographic analysis, with the identification of at least two formulas based on varying proportions of clay matrix and aggregate content, one characterised by a clay-rich matrix, and another one characterised by an unbalanced matrix-to-aggregate ratio with high aggregate content. Samples exhibit a generalised weakness in terms of surface hardness, and in hydric tests, they show a high degree of hydrophilicity with a severe tendency towards weathering decay.

2.2. Methodology

Earthen structures are mostly composed of inorganic solid elements [28,46], such as stones (grain diameter 20–200 mm), gravel (2–20 mm), sand (0.06–2 mm), silt (0.002–0.06 mm), and clay (<0.002 mm). Specifically, the different proportions and the specific distribution of sand, silt, and clay determine the structure of the soil and therefore its physical and mechanical properties.

The experiments described below focus on simple, low-cost, non-specialist field tests, which were selected not only for their practicality but also for their potential to provide preliminary, descriptive information on the suitability of selected clay materials for building purposes, and which can be performed directly on-site [28,29,30,31].

Empirical tests used in this analysis are as follows:

• Visual inspection: the dry earth is examined with the naked eye, and the grain size is observed. The presence of gravels and sands visible to the naked eye is noted [28] (p. 39), [29] (p. 4).
• Colour description: colour properties can provide information about sediment composition, and they are relevant for registering visual properties of archaeological adobe to be considered in the restoration phase. Colour is registered through the Munsell Soil Color Chart book, a standardised system based on three components, such as hue (a specific colour), value (lightness), and chroma (intensity).
• Tactile examination: dry soil can be described as “sandy” when it is coarse and abrasive to the touch; “silty” when it is fine and easy to pulverise to the touch; “clayey” when it is hard to break down, fine, and sticky [29] (p. 9), [30] (p. 179).
• Olfactory examination: the humidified earth is smelled with the aim of assessing the presence of organic elements. The smell of mould and humus is an indication of organic presence [29] (p. 9), [30] (p. 179).
• Handwashing test: dry sample is placed between the hands and runs through the water: the harder it is to remove the soil by washing the hands, the more clayey it is [29] (p. 9), [30] (p. 179).
• “Pad” or dry strength test: the plastic-state soil is moulded into pastilles within a metal ring. A linear shrinkage is then observed after drying. The soil presents good qualities if the sample has a shrinkage of less than 1 mm and if it is difficult to reduce the tablet to powder [28] (p. 43), [29] (p. 4).
• Sedimentation test: a quantity (100 g) of weighed soil is inserted into a water bottle (1 L) to evaluate the percentages of solid elements. Measurements are taken with a ruler at given intervals during one day of sedimentation. The heavier materials (crushed stone and sand) settle to the bottom, while the finer, lighter materials (clay and silt) remain on the surface [28] (p. 38).
• Peeling or Scotch Tape test: the test is performed to assess surface cohesion, and to provide an indication of the surface deterioration of adobe. For this, 2.6 × 3.1 paper labels (approximately 8 cm²) with double-sided tape were prepared and weighed (unladen weight). The paper labels were attached to a flat sample face (both surface and core) with homogenous pressure (up to six pressures, adhesion time: 0.30 min). The labels were stripped off (tear-off angle of approximately 90°) and re-weighed to quantify the amount of material remaining adhered to the tape. Two to three measurements were taken on each sample, depending on their nature and preservation [31].

For each test, data were systematically recorded using a standardised form (Figure 4), allowing both qualitative evaluation and cross-comparison across archaeological layers and with modern adobe replicas.

3. Results

The field tests encompassed visual (including colour), tactile, olfactory, manual washing, pad, sedimentation, and surface cohesion analyses. Their objective is to ascertain the physical–mechanical properties of the 16 analysed samples from the Nasiriyah region of Iraq, together with their stability, and composition, and with particular emphasis on organic content, plasticity, and behaviour during desiccation (

Table 2. Results of visual, tactile, olfactory, and handwashing testing.

ID	Visual	Tactile	Odour/Washing
43	Compact, traces of organic/vegetal inclusions	Fine and slightly abrasive, easy to crumble	Earthy and dusty smell when dry, fungus scent when wet. Sticky, difficult washing.
48	Compact, traces of organic/vegetal inclusions	Fine and slightly coarse and abrasive, easy to crumble	Earthy and dusty smell when dry, fungus scent when wet. Easy to wash.
51	Compact, possible charcoal	Fine, hard to crumble	Earthy and dusty smell when dry, fungus scent when wet. Slightly sticky, medium washing.
53	Very compact, clayey	Fine, abrasive, hard to crumble	Earthy and dusty smell when dry, fungus scent when wet. Slightly sticky, medium washing.
54	Compact, clayey	Soft, hard to crumble	Earthy and dusty smell when dry, fungus scent when wet. Sticky, difficult washing.
61	Compact, clayey	Coarse, easy to crumble	Earthy and dusty smell when dry, fungus scent when wet. Slightly sticky, easy washing.
63	Compact, possible gypsum traces?	Coarse, easy to crumble	Earthy and dusty smell when dry, fungus scent when wet. Slightly sticky, medium washing.
64	Compact, traces of organic/vegetal inclusions	Fine and soft to the touch, crumbles with difficulty	Earthy and dusty smell when dry, fungus scent when wet. Easy to wash.
65	Friable on fractures, compact elsewhere, charcoal traces	Fine, abrasive, crumbles easily	Earthy and dusty smell when dry, fungus scent when wet. Earthy, sandy, easy washing.
66	Very compact, clayey	Fine, hard to crumble	Earthy and dusty smell when dry, fungus scent when wet. Slightly sticky, possible silt.
68	Very compact, clayey	Coarse, hard to break	Earthy and dusty smell when dry, fungus scent when wet. Sticky, difficult washing.
69	Very compact, clayey	Fine, easy to crumble	Earthy and dusty smell when dry, fungus scent when wet. Not sticky, possible silt, easy washing.
79	Compact, traces of organic/vegetal inclusions	Fine and slightly abrasive, easy to crumble	Earthy and dusty smell when dry, fungus scent when wet. Medium easy to wash.
302/9	Friable, mostly sandy traces of organic/vegetal inclusions	Slightly coarse and abrasive, easy to crumble	Odourless, easy to wash
MOD	Friable sandy soil, very small white mineral inclusions	Fine and slightly coarse texture, easy to crumble	Odourless, moderate washing difficulty

).

Visually, all the samples appear very compact, consisting of a predominantly clayey matrix, with traces of plant inclusions (straw) no longer visible. Sample no. 48 is friable in correspondence with the fractures, while it is very compact on the rest of the surface. Samples nos. 51 and 65 show the presence of possible black organic material such as charcoal, while sample no. 63 presents white mineral inclusions, possibly gypsum. Unlike the archaeological adobe, plaster sample no. 302/9 appears friable and is predominantly sandy in composition, with impressions of vegetal inclusions that have decomposed and are no longer visible. The modern adobe sample (MOD) appears to be composed of friable sandy soil, containing small white mineral inclusions and lacking any organic inclusions. The dry soil appears uniformly coloured and consistently falls within the 10YR6/3 range (pale brown) for all samples, including the modern adobe sample, whereas sample 302/9 is approximately within the 10YR7/3 range (very pale brown; Table 2).

The visual inspection test represents a fundamental step in the preliminary assessment of earthen materials [33,47,48,49,50]. Through macroscopic observation of colour, surface texture, and visible inclusions, it is possible to infer key compositional and mechanical properties of adobe, such as shrinkage potential, particle uniformity, and organic content. In particular, the detection of cracks or shrinkage patterns after drying has been recognised as an indicator of soil plasticity and suitability for construction. The consistent pale-brown colour range (10YR6/3) observed across the Tell Zurghul/Nigin samples supports the hypothesis of a homogeneous raw material procurement strategy throughout the site’s occupational phases.

Moreover, colour analysis holds a dual significance in earthen heritage research: beyond its compositional implications, it also contributes to chromatic compatibility assessments for conservation [33,49]. Even though advanced tools such as colorimeters can provide quantitative chromatic parameters [41], the use of Munsell Soil Color Chart book already enables reliable discrimination among material types and depositional contexts [48]. This preliminary classification can thus guide subsequent laboratory analyses and ensure the chromatic coherence of modern replicas used in conservation interventions.

All the samples appear slightly abrasive to the touch (except for nos. 54 and 64, which are soft), indicating a considerable quantity of sand, particularly in samples nos. 61, 63, and 68, which are also coarse and highly abrasive. The modern sample is friable and slightly coarse. The major difference identified by tactile examination concerns the ease of the samples breaking down: half crumble easily, while the other half resist, with no. 68 exhibiting exceptional resistance (Table 2). Such tactile properties reflect the balance between sand, silt, and clay content, which strongly influences plasticity, cohesion, and mechanical behaviour of earthen construction materials [51,52,53]. Samples that are harder to crumble likely have higher clay content and cohesion, providing greater structural stability, whereas easily crumbling samples indicate lower cohesion and higher sand fraction, affecting workability but potentially reducing shrinkage and cracking [54,55].

These observations confirm that simple tactile tests, when interpreted in combination with sedimentation and shrinkage measurements, provide meaningful first-order assessments of adobe suitability for construction and conservation purposes, even in field conditions where laboratory analyses are unavailable [51,56].

Prior to conducting the other tests, the samples are crushed using a mortar and then reconstituted by gradually adding water until the desired consistency (plastic or viscous) is attained. Ideally, the tests should be performed in both soil states, but due to the limited amount of material available, the test is conducted at the viscous state only.

Based on the odour, the samples exhibit an earthy and dusty smell when dry, whereas a humus-like scent becomes evident when the soil is wet. The olfactory analysis thereby confirmed the presence of organic components within the archaeological adobe samples, as previously indicated by visual examination, given that the moistened earth of these samples emits a scent of mould and humus. This observation aligns with previous studies, where the presence of a distinct smell in wet soil has been correlated with the existence of organic matter [47]. In the context of earthen construction, such organic content is generally considered undesirable, as it can compromise the mechanical stability and long-term durability of adobe materials. Conversely, sample no. 302/9 is odourless, suggesting an absence of preserved organic components in the plaster sample, although such components were originally present in the plaster composition, as evidenced by direct visual observation. The modern sample is odourless as well, in both the dry and the liquid state, confirming the absence of organic inclusions (Table 2). Therefore, the lack of odour in both the plaster and modern adobe sample suggests a more stable inorganic composition, consistent with materials intentionally selected to ensure better performance and reduced biodegradation.

The handwashing test is carried out to assess the quantity (in proportion) of clay within the samples. Samples number 43, 53, 54, and 68 are those that leave the greatest amount of clay residue on the hands, making it particularly difficult to remove the soil by washing them. Regarding the modern sample, it exhibits a notable presence of fine to coarse sand particles and presents a moderate washing difficulty, along with an oily consistency that distinguishes it from all the archaeological samples (Table 2).

According to previous experimental and applied studies on earthen construction [47,51,56], handwashing and other tactile tests provide a rapid, low-tech indication of the clay-to-sand ratio and correlate reasonably well with laboratory analyses. The difficulty of rinsing one’s hands is widely recognised as an indicator of high clay content and reduced coarse fraction, consistent with field-test guidelines. Such empirical approaches are particularly valuable in field conditions where laboratory facilities are unavailable, offering first-order insights into soil cohesion, plasticity, workability, and suitability for adobe construction.

Nevertheless, tactile or hand-based tests may overestimate clay content compared to quantitative sedimentation or granulometry methods [52,53], highlighting the importance of complementary measurements to validate empirical assessments. Recent research has shown that the proportion of clay critically influences not only plasticity but also drying behaviour and mechanical resistance of adobe blocks [55], supporting the relevance of the handwashing test as a preliminary screening tool in both archaeological and conservation contexts.

For the dry strength (or “pad”) test, the plastic-state soil was moulded into 6 cm diameter pastilles using a metal ring. All samples were analysed except for numbers 43, 64, and 302/9 due to insufficient material. Most samples showed no noticeable shrinkage during 24 h of drying, indicating good construction qualities, whereas sample no. 54 exhibited a non-linear shrinkage of approximately 7 mm, suggesting limited suitability for building purposes (Figure 5). The modern sample displayed low cohesion in the dry state, consistent with its friable texture, and no shrinkage (

Table 3. Results of the shrinkage, sedimentation, and peeling tests.

ID	Shrinkage (Pad Test)	Sedimentation	Peeling Test (mg/cm2)	Notes
43	\	\	a = 2.25; b = 2.16; c = 2.21 1
48	\	\	a = 5.7; b = 3.48; c = 0.87
51	No shrinkage (6 → 6 cm)	No foam	a = 7.63; b = 3.82; c = 0.4	Similar to 63 but more cohesive
53	No shrinkage	Moderate foam	a = 1.15; b = 2.76; c = 3.48	Balanced clay–silt
54	Shrinkage (6 → 5.28 cm)	\	a = 2.05; b = 4.55; c = 5.67	High plasticity, low stability
61	No shrinkage (6 → 6 cm)	No foam	a = 3.53; b = 1.91; c = 3.78	Coarse and workable
63	No shrinkage (6 → 6 cm)	No foam	a = 1.61; b = 1; c = 1.73	Possible mineral inclusions
64	\	\	a = 4.93; b = 3.53
65	No shrinkage (6 → 6 cm)	Foam with black dots	a = 4; b = 4.12; c = 3.58	Good workability, organic traces
66	No shrinkage (6 → 6 cm)	Moderate foam	a = 2.92; b = 5.61; c = 4.8	Good plastic balance
68	No shrinkage (6 → 6 cm)	No foam	a = 13.28; b = 1.4; c = 2.41	Highly clayey, strong cohesion
69	No shrinkage (6 → 6 cm)	Moderate foam	a = 4.13; b = 2.1; c = 5.61	Likely silt presence
79	\	\	a = 3.68; b = 7.57; c = 4.67
302/9	\	\	a = 0.57; b = 0.47; c = 0.1
MOD	No shrinkage (6 → 6 cm)	No foam	a = 6.15; b = 14.55; c = 5.06	Presence of fine sand grains; a very thin (1 mm) yellow layer composed of clay and lime observable during sedimentation test

¹ Two to three measurements per sample (a, b, and c). Values are reported in mg/cm² and indicate the amount of material (in mg) adhering to 1 cm² of paper labels with double-sided tape for each measurement.

). Shrinkage patterns in these pad tests provide practical insight into potential cracking and deformation, particularly in high-clay soils [57,58]. Monitoring crack formation and shrinkage in small blocks or pastilles is a well-established method for assessing the structural behaviour of earthen materials [48,59,60,61,62]. Such observations can guide conservation strategies, highlighting cases where stabilisers or fibres might be needed to mitigate shrinkage and improve durability [63,64].

The sedimentation test was conducted using 100 g of dry soil placed in a bottle filled with water up to 8 cm. Sedimentation was observed over a 24 h period, with measurements taken at specified intervals (1 min, 5 min, 10 min, 30 min, 1 h, 2 h, 6 h, 12 h, and 24 h). Due to the requirement of approximately 100 g of powdered material, the test was performed only on samples nos. 46, 48, 51, 53, 61, 63, 65, 66, 68, 69, and MOD, owing to insufficient material for the other samples. This method allows for a preliminary estimation of clay, silt, and sand proportions, by shaking the sample in water and observing the settling layers over time [33,34,65].

All the bottles exhibit rapid sedimentation times, with low visibility observed within the first ten minutes of the experiment. However, after thirty minutes, visibility improve significantly. In all archaeological samples, the water became clarified in less than two hours, indicating a rapid sedimentation rate and the absence of colloidal or expansive clays. The modern sample is the only one requiring more than two hours to achieve water clarification (approximately three hours).

After sedimentation, samples nos. 46, 53, 61, 65, 66, and 69 present a layer of fine grains of sand (2 cm), and a layer of foam on the surface (ca 2 mm; Figure 6). The foam observed in the samples possibly indicates the presence of organic materials. Notably, sample number 65 exhibits the greatest volume of foam, measuring approximately 3 mm, and contains several small black inclusions, possibly traces of charcoal from carbonised organic residues, which are absent from the other samples. Of interest, black organic inclusions were identified visually in both samples 51 and 65. However, while the presence of these small black organic inclusions is confirmed for sample 65 by the sedimentation test, this is not the case for sample 51, where such inclusions were not detected by the bottle test. Samples nos. 48, 51, 63, 68, and MOD do not exhibit any foam on the water surface after sedimentation, implying that the soil is well settled and lacks unstable organic components.

The dominant fraction in all samples is fine sand, accounting for 95–100% of the total sediment. Samples 51, 61, and MOD display a thin layer (approximately 1 mm) of silts and clays, comprising about 4.76% of the sediment. No coarse sands or gravel were identified in any of the samples (Table 3). These findings reinforce the sandy nature of the archaeological adobe, which supports both structural stability and rapid drainage. Sedimentation analysis provides crucial information on grain size distribution, which is a key factor in assessing soil suitability for adobe or rammed-earth construction, although it should be complemented with other field and laboratory tests to confirm suitability [48,66].

The surface cohesion of the samples is measured in mg/cm², based on the amount of material adhering to paper labels with double-sided tape. This test method evaluates the surface cohesion properties by pressing adhesive tape onto the sample and then measuring the weight of detached material per unit area [67,68,69]. The resulting “decohesion index” is inversely proportional to the cohesion forces of the grains.

The archaeological adobe generally exhibits medium surface cohesiveness, with material loss ranging from 2 to 6 mg/cm², except for some outliers. In particular, samples numbers 46, 68, and MOD demonstrate the poorest surface cohesiveness, whereas the plaster sample number 302/9 exhibits the highest surface cohesiveness, reflecting its role as an external surface cladding rather than an internal component like adobe (Figure 7; Table 3). Differences in surface adhesion indicate functional variation between adobe and plasters and can provide insights into the effectiveness of consolidation strategies in conservation practice. However, it is noted that the peeling test has sometimes been over-applied without sufficient standardisation, potentially leading to non-comparable or biased results [69].

4. Discussion

A total of sixteen samples (fifteen archaeological and one modern) were analysed using a series of visual, tactile, olfactory, manual washing, sedimentation, drying shrinkage, and surface cohesion tests to evaluate their suitability as construction materials.

The archaeological samples were examined in order to preliminarily assess the physical–mechanical properties of the materials, to facilitate comparisons between the different architectural phases, dating from the mid-5th to the mid-3rd millennium BCE, and to assess the suitability of the raw materials historically used for contemporary restoration projects. The archaeological samples from Tell Zurghul/Nigin exhibit a high degree of homogeneity across all the architectural phases and periods under investigation. This consistency was already evident from the results of mineralogical and chemical analyses conducted on the same samples [41]. Our field test results confirm that earthen architecture at the site consistently employs comparable raw materials in terms of suitability for building purposes.

In terms of physical–mechanical properties and composition, several samples yielded significant findings relevant to the assessment. Sample no. 65 had a sandy–clayey composition, exhibited no shrinkage, contained evidence of organic constituents, and demonstrated favourable workability, making it highly suitable for stable structural mixtures. Sample no. 68 displayed high cohesion and plasticity, encountered difficulties during washing, and produced no discernible foam, indicating its appropriateness for highly argillaceous mixtures. Samples nos. 53, 66, and 69 exhibited a balanced clay–silt composition, generated moderate foam levels, and showed no shrinkage, suggesting versatility for conservation or general applications. Samples nos. 63 and 51 were characterised by a coarser composition and absence of foam, thereby making them suitable for uses requiring resistance to shrinkage. Finally, sample no. 54 demonstrated high plasticity and significant shrinkage, classifying it as an unstable material, though it remains of interest for deformability testing. These observations align with studies highlighting the influence of clay-to-sand ratio on block plasticity, shrinkage, and resistance to cracking [57,58,63].

The sedimentation test results confirm that the analysed archaeological adobe predominantly consist of fine sands, with only minimal silty or clayey material present in a few samples. Sedimentation analysis provides a rough estimate of sand, silt, and clay proportions [33,65], which are crucial for assessing soil suitability for rammed-earth and adobe construction. This contrasts with the handwashing test, which suggested a higher clay proportion in samples 43, 53, 54, and 68; however, this may reflect the tactile sensation of clay during washing, while the actual clay quantity in the sediment was insufficient to settle. Tactile tests provide rapid, low-tech indicators of cohesion and workability [47,55,56], yet they may overestimate clay content compared to sedimentation measurements, emphasising the importance of complementary approaches.

The modern sample, originating from an adobe-making company in Nasiriyah, exhibits a friable and moderately coarse matrix with notable presence of fine to coarse sand particles, confirmed through the handwashing and visual tests. The oily nature of the sample may suggest the presence of organic or clay-bound compounds, although this was not corroborated by the smell test, which revealed no significant organic odour. The dry strength test indicates low cohesion in the dry state, consistent with the friable texture. Sedimentation revealed predominance of fine sand with a thin clay–lime layer, suggesting implications for water resistance and particle cohesion. The absence of foam further suggests a lack of surfactant-like substances or organic binders. These characteristics render it broadly comparable with archaeological adobe, though careful evaluation is necessary for restoration, considering mechanical and compositional compatibility.

The surface cohesion (peeling) tests show variable adhesion among archaeological adobe and between adobe and plasters. Archaeological adobe generally exhibits medium cohesion, while the plaster sample shows the highest surface adhesion. The “decohesion index” obtained from these tests has been widely used to evaluate surface cohesion and consolidation efficiency in conservation practice [67,68,69]. Differences observed among samples provide functional insights relevant for restoration strategies.

The study is necessarily constrained by the relatively limited number of archaeological samples analysed (n = 15), complemented by a single modern analogue. These limitations reflect material scarcity and ethical considerations regarding destructive sampling. The empirical field tests employed are inherently semi-quantitative and can be affected by preservation state and inter-operator variability. Consequently, results should be considered preliminary indicators rather than definitive characterisations, pending more robust laboratory analyses.

Nevertheless, by acknowledging these constraints explicitly, the present work establishes a transparent methodological baseline and underscores the potential of integrating low-cost, rapid field assessments with advanced analytical methods. Future research should expand the dataset, incorporate systematic laboratory investigations, and extend the application of this approach to other archaeological sites, refining reliability and contributing to broader discussions on earthen construction technologies, regional variability, and conservation strategies.

5. Conclusions

In conclusion, the results of the field tests presented here demonstrate that such methods provide a rapid, low-cost, and replicable approach for assessing the physical–mechanical properties of earthen construction materials. Field tests are particularly useful in contexts where laboratory analyses are constrained due to destructive sampling or limited access to material [70]. The archaeological adobe from Tell Zurghul/Nigin exhibit remarkable compositional consistency and structural stability across different chronological phases, confirming the reliability of historically used raw materials.

While modern analogues offer potential for restoration work, their mechanical and compositional properties must be carefully evaluated to ensure compatibility with the archaeological materials. Consideration of soil geochemistry and potential organic inclusions can be critical for evaluating such compatibility [71].

The development of consistent methodologies for characterising earthen materials using non-destructive field tests is crucial for bridging the gap between empirical observations and laboratory analyses, enabling archaeologists and conservators to obtain comparable data across sites [72]. By integrating these diverse field-testing approaches with continued laboratory research, it becomes possible to advance understanding of earthen materials and support their broader acceptance and application in sustainable conservation and construction practices.

The field tests should therefore be interpreted as preliminary screening tools, providing actionable first-order information to guide conservation decisions, without replacing laboratory-based analyses when these are feasible. By combining empirical field assessments with technical expertise, archaeologists can actively contribute to the evaluation of material properties while minimising the impact on fragile archaeological structures.

Finally, greater awareness of technical properties by archaeologists, in collaboration with conservation and materials specialists, fosters informed and sensitive strategies for preserving archaeological earthen architecture. This integrated approach ensures interventions that are technically sound, historically faithful, and culturally respectful, while maintaining the holistic perspective essential for safeguarding these heritage structures.