Experimental Raw Earth Building for Passive Cooling: A Case Study for Agricultural Application in a Mediterranean Climate

Abstract

Residential and agricultural buildings must prioritize environmental sustainability, employing locally sourced, bio/geologically sustainable materials, and reversible construction methods. Hence, adobe construction and earth-based building methods are experiencing a comeback. This article describes the hygrothermal performances of a real scale agricultural building prototype, in real field conditions, built and designed to be energy-efficient, environmentally friendly, and well-suited for the hot, dry climates typical of the Mediterranean region during summer. The building prototype is a small modular two room construction, one room based on wood (for control purpose) and the other one on raw earth. The experimental set up highlights the passive cooling and humidity regulation potential provided by raw earth and adobe brick technology in agricultural buildings used for fruit and vegetable storage. Such passive cooling alternatives in the Mediterranean climate could reduce the need for energy-intensive and environmentally impactful cold storage rooms.

1. Introduction

The growing demand for localized food supplies requires a comprehensive approach that includes diversifying agricultural production, establishing local food supply chains, and implementing broader localized food systems. This shift in paradigm introduces new requirements in the realm of food infrastructure, covering agricultural production, processing, packaging, logistics, and the retailing of local food products. The sustainable management of such infrastructure, whether located in urban or rural areas, poses a significant societal challenge intricately tied to contemporary imperatives related to ecological, environmental, and energy transitions. Notably, a substantial portion of land artificialization, leading to a decrease in available surface area per capita, is occurring on agricultural lands. This trend is particularly pronounced in France, where approximately 80% of artificial land utilization takes place on agricultural territories [1,2]. Since the 1970s, urban planning documents have shielded these agricultural lands from extensive development, with only constructions directly associated with farming activities being legally authorized within these agricultural zones [3,4].

Nonetheless, recent research findings reveal a significant intrusion of urbanization into farmland because of agricultural development. This intrusion involves the expansion of infrastructure, residential communities, and commercial areas into initially rural landscapes, such as forests and agricultural regions, without a coherent urban planning framework. Notably, a staggering 10 million square meters of agricultural structures emerge annually, accounting for over 35% of the total non-residential building area [5]. This escalating construction trend can be attributed to various factors, including advancements in agricultural technology, the increasing influence of health and environmental regulations, the relocation of landlocked farms to densely populated regions, the permissive interpretation of rules allowing unchecked utilization of agricultural building rights, and the growing proliferation of constructions disguised as agricultural projects but serving alternative purposes. To curb this widespread urban sprawl, national legislation, implemented since the early 2000s, has aimed to restrict building possibilities within agricultural zones [6].

However, a new type of agriculture that is more sustainable and benefits from short supply chains is emerging as one of the preferred outlets for local organic production. These short supply chains, characterized by a direct link between producers and consumers, belong to the broad field of alternative agri-food systems developed in response to the negative environmental and socio-economic effects of conventional systems [7]. This new type of agriculture has different needs, with smaller fields and a requirement for more temporary food production storage [8,9]. Therefore, a method to accommodate this growing demand for food infrastructure while preserving precious agricultural expanses from further urbanization must be found.

In this context, sustainable agricultural storage facilities were designed. The developed solution (i) allows for self-construction, thus providing a cost-effective and potentially participatory approach; (ii) enables disassembly and later reuse of the building, thereby simplifying the process of obtaining building permits, especially on public lands; and (iii) incorporates eco-design principles.

In the efforts to eco-design fruit and vegetable storage buildings, dual challenges were faced. The first aspect involves minimizing the environmental impact during the construction phase, emphasizing the crucial choice of utilizing locally sourced and organic materials. In a second step, it is necessary to minimize the environmental impact of the building during the operational phase as a fruit and vegetable storage area and, consequently, minimize the potential energy input required to manage temperature and humidity in the buildings.

The thermal and moisture performance of building materials plays a pivotal role in ensuring the stability of indoor environments, particularly in fruit/legume storage facilities. The Mediterranean climate, characterized by hot, dry summers and mild, humid winters, demand materials with high thermal inertia and moisture buffering capacity to mitigate temperature fluctuations and condensation risks. Studies have shown that lightweight concrete masonry and improved façade systems can significantly reduce energy consumption while maintaining thermal stability in such climates [10]. Moreover, vernacular architectural strategies—such as thick stone walls, small openings, and shaded courtyards—demonstrate effective passive cooling and humidity control, offering valuable insights for contemporary storage design [11]. Earth-based materials—such as rammed earth and adobe—have garnered renewed interest due to their low embodied energy, high recyclability, and ability to regulate indoor humidity through hygroscopic behavior. Studies have shown that such materials can maintain stable indoor temperatures and reduce relative humidity swings, making them particularly suitable for preserving fruits and legumes [12,13].

One class of materials that addresses these two eco-design challenges is raw earth. Indeed, raw earth in construction is making a comeback, primarily due to its widespread availability, ease of production, non-toxic nature, recyclability, potential for local production, and consequently, reduced energy and environmental costs. Additionally, this material exhibits outstanding thermo-hygrometric performance, a characteristic that has been harnessed and documented for centuries [14,15], allowing the construction of buildings that maintain a certain coolness in the summer and warmth in the winter. Lower temperatures are the key to maintaining the quality of most post-harvest vegetables. Cold reduces respiration and transpiration and limits the development of physiological and parasitic diseases, as well as most internal changes. Also, air is a mixture of water vapor and gases (nitrogen, oxygen, CO₂); water vapor content (ambient hygrometry) has a major impact on vegetable preservation. For vegetables with a high susceptibility to wilting, the optimum humidity is around 90%. For certain vegetables, it is advisable to limit humidity (regular aeration of the storage room): Alliaceae, squash, etc. The table shows some of the fruits and vegetables produced in the south of France during the summer. Storage conditions of less than seven days are indicated in

Table 1. Optimal short-term storage conditions for some fruits and vegetables produced in the south of France [16].

Vegetables or Fruits	Recommended Temperature for Storage Less than 7 Days	Optimum Relative Hygrometry (RH)
Tomatoes	13–18 °C	>80%
Zucchinis	8–12 °C	>80%
Eggplant	8–12 °C	>80%
Peppers	8–12 °C	>80%
Cucumbers	8–12 °C	>80%
Green beans	8–12 °C	>80%
Canary melon	13–18 °C	>80%
Watermelon	8–12 °C	>80%
Squash, pumpkin	13–18 °C	>60–75%
Strawberries	2–7 °C	>80%

.

Thus, this article focuses on experimentations with new modes of sustainable food buildings based on raw earth but also on reversibility, modularity, and self-construction. These experiments have been performed in a real environment, in collaboration with small market-garden farms (with limited storage volumes needs). The buildings are located directly on a functional farm and used by the local market gardener during the study. The experimental site is located on a rural area of southern France, in the commune of Lodève, (Figure 1a), where the local Mediterranean climate is characterized by hot, dry summers and mild, wet winters (Csa according Köppen–Geiger classification [17]), with 2668 h of sunshine per year and 629 mm of annual rainfall, mostly during winter months [18]. The modular low tech small size building with raw earth walls and floor that will be described in the first part is shown in Figure 1b–d. The raw earth and brick fabrication and characterization will also be described in the first part.

In the second part, the study will focus on: (i) exploring the hygrothermal performance of the constructed buildings under real environmental conditions; and (ii) analyzing the impact of raw earth walls and floors on indoor temperature and relative humidity (RH), with particular attention to amplitude and phase shift effects. Such an approach could reduce reliance on energy-intensive cold storage, particularly in the Mediterranean region during the summer months.

2. Raw Earth Building Fabrication

Traditional construction materials that are heavy, such as raw earth, are known for providing coolness in the summer due to their high volumetric mass, which imparts inertia into interior spaces. Thermal and hygric regulation in the building studied in the present work depends on the raw materials and the construction technique.

2.1. Raw Earth Preparation

The present preparation aims to produce earth bricks to regulate temperature and humidity as effectively as possible and to favor inertia. Thus, a high volumetric mass in the range of 1600 kg.m⁻³ is recommended. To obtain such a volumetric mass, the following proportions were considered:

• One volume of clay-rich raw soil mixed with water. This soil was excavated from the Montpellier region at a depth of 40–50 cm. It allows a very low environmental footprint. The proportion of water represents 35% of the raw soil volume.
• One volume of sand, traditionally used as masonry sand intended to produce concrete, mortar, and coatings with a particle size of 0–5 mm. This sand increases the volumetric mass of the mixture.
• One volume of hemp straw (approx. 2 cm length) for mechanical and thermal consideration.

The combination is pre-mixed within a large concrete mixer before being foot treaded for approximately 20 min.

2.2. Earth Brick Preparation

The prepared earth mixture is used to produce adobe bricks. A wooden mold is produced and tricked into the water before filling. The mold is deposited on a smooth plastic support; then, the bricks are molded by hand in a wooden mold (the molds are filled by adding successive earth clods). To ensure quick drying, the bricks are demolded the same day directly after filling. Also, the bricks were air-dried for 3 weeks under outdoor conditions (during June) without a fan. For rapid, uniform drying, the bricks are regularly turned over. The complete protocol for brick preparation is available on the project website and on YouTube [19,20]. Given the constraints of maintaining reasonables cost and keeping the walls thin relative to the size of the buildings, bricks with a thickness of 12 cm were considered. These bricks were obtained by using appropriately sized molds. Once dried, weighing the 12 × 10 × 25 cm³ bricks allowed us to estimate a density of approximately 1560 kg.m⁻³. Figure 2 (left) shows pictures of the brick preparation process in a wooden mold and an example of a dry complete brick (right) [20].

2.3. Building Design

The design of the buildings is based on demountable modules that can be combined in different ways according to the requirements and whose identical structure is adaptable to different uses. The building is designed to be easily dismantled to meet local planning constraints and reduce land artificialization. In this study, the notion of dismantling is ensured by (i) the absence of structural works, such as deep foundations, (ii) the steel and wood-frame structures are demountable and reusable, and (iii) the raw earth bricks are prefabricated on site and are assembled in a way ensuring dismantling and reuse. In this project, particular care was taken to favor low-tech, locally sourced materials and to minimize the carbon impact. The choice of materials and self-building also aim to reduce construction costs and give the project owner greater autonomy in the design and construction process. All-structural elements are designed with replicability and the principles of Open Science in mind: all building instruction are freely available on different media (documentation, YouTube) [19,20]. The final typical structure can be seen in Figure 3. It is made of two storage modules (2.50 m × 2.38 m wide and 2.30 m high each), without windows and with a door for each module. The entire assembly of the two modules is covered with wooden cladding. For practical reasons, a terrasse has been added. Finally, a green roof covers the entire structure.

The first storage module, called the ‘tool storage module’ or ‘wooden room’ is dedicated to the storage of agricultural equipment and tools. This module has no specific requirements regarding thermal performance. This module is mainly based on the wooden cladding on the walls (walls A, F, E) and OSB on the floor. The second module, called the ‘vegetable storage module’ or ‘earthen room’ is the one of interest for its hygrothermal performance and is dedicated to the storage of fruits and vegetables. This second module is based on raw earth bricks for the walls (B, C, D, and G) and floor and is oriented to the north to prevent too much solar irradiation in summer. Note that the dividing wall (G) between the two modules is made of earth bricks on the earthen room side and made of OSB on the wooden room side.

To build this structure, a wall and partition framework made of wood is installed, as seen in Figure 4c. On the outside of the wall wooden framework, a thin vapor barrier membrane (a Solitex Mento 3000) rain screen is installed to prevent the effects of climatic elements. Then, as shown in Figure 4b, long battens are placed directly on the thin barrier membrane (on the external part of the frame), screwed directly on the wooden framework. These battens, with a cross-section of 45 mm × 45 mm help maintain an air gap between the interior wall and the wooden cladding, reducing solar gains to limit overheating in the summer. Fixed at an angle of 45°, they can dissipate heat provided by the sun through a “chimney effect”. Moreover, the battens contribute to the bracing of the frames. On top of these battens, wooden cladding has been installed vertically. To minimize costs, rough-sawn Douglas fir boards have been installed. As shown in Figure 4d, the Douglas fir board is 22 mm thick and 150 mm long.

Finally, the earth bricks are then inserted into the wooden framework of the building to create an earthen envelope in the structure. An example of the integration of the bricks into the wood frame walls is illustrated in Figure 4a. The vertical assembly of the bricks was carried out using a very lean earth mortar to allow for adjustments to dimension and flatness defects while permitting easy removal later. The mortar used to assemble the bricks and fill the joints is made of a fine and very lean mixture (3 volumes of sand to 1 volume of sifted earth through a 4 mm sieve). A natural linen mesh was fixed in front of the bricks on the frame, and the entire assembly was covered with an earth plaster to perfect the waterproofing. The final cross-section of raw earth B, C, and D walls can be seen in Figure 4d. Walls A, E, and F are the same as in Figure 4d but without raw earth. For the floors of the two modules, a metal base was designed to support heavy loads and facilitate disassembly and reassembly. On the floor, in the earthen room, the bricks are arranged within the floor framework, and the joints are filled with lean mortar. In the wooden room, a simple OSB is installed. The roof of the building is composed of many layers. From the top to the internal side of the roof, there is a mixture of earth and volcanic rock (pozzolan), vapor barrier, semi-firm limestone, a plastic film, and OSB.

3. Raw Earth Brick Composition and Thermal Properties

3.1. Earth Brick Composition

A complete analysis of the bricks was performed using three different complementary techniques: (i) the mineralogical composition was determined using the Bruker D2 PHASER diffractometer, (ii) the chemical composition was analyzed using the Bruker S2 RANGE X-ray Fluorescence (XRF) analyzer, and (iii) thermogravimetric analysis were performed using the NETZSCH STA 449 F5 Jupiter instrument from NETZSCH Gerätebau GmbH, Selb, Germany.

3.1.1. Mineralogical Composition Analysis

The Bruker D2 PHASER diffractometer from Bruker AXS GmbH, Karlsruhe, Germany, was operated in continuous scanning mode and utilizes Cu Kα radiation with a wavelength of 0.15406 nm. The X-ray diffraction (XRD) analysis parameters included a 2θ range of 10–70, an increment of 0.02°, and an acquisition time of 0.15 s per step. A seen in the Figure 5, the X-ray diffraction pattern of the mixture samples, thanks to MAUD software (version 2.998), can reasonably be refined using 8 main phases: calcite (49.1%) (calcium carbonate (CaCO₃)), quartz (13.7%) (silicon dioxide (SiO₂)), Al-tobermorite (10.8%) (aluminium-substituted calcium silicate hydrate ((Ca,Al)₅Si₆O₁₈·5H₂O)), forsterite (6.9%) (magnesium silicate (Mg₂SiO₄)), kaolinite (8.7%) (aluminum phyllosilicate (Al₂Si₂O₅(OH)₄) [21]), muscovite (9.9%) (white mica (KAl₂(AlSi₃O₁₀)(OH)₂)) [21], traces of albite (sodium feldspar (NaAlSi₃O₈) potentially containing traces of Ca, K, and Mg [22]), and traces of tobermorite (hydrated calcium silicate (Ca₅Si₆O₁₈·5H₂O)). The results presented here are based on quantitative phase analysis performed via Rietveld refinement, yielding agreement factors of R_wp = 6.87% and R_B = 5.14%. The high quality of the fit (Goodness of Fit = 1.78) underscores the reliability and quantitative accuracy of our phase analysis.

3.1.2. Chemical Composition Analysis

To verify the consistency of the bulk composition as determined by XRD, identify elements not detected by XRD (amorphous phases, trace constituents), and refine the interpretation of the mineralogical data, chemical composition analysis of the earthen bricks was conducted. The composition analysis presented in

Table 2. Chemical composition of the earth bricks.

Element	wt.%
CaO	43.26
SiO2	36.08
Al2O3	9.48
Fe2O3	4.61
K2O	1.68
SO3	0.90
TiO2	0.84
MnO	0.11
BaO	0.06
La2O3	0.06
SrO	0.05
Loss on ignition	0.23

highlights significant proportions of key elements, with calcium oxide (CaO) predominating at 43.26%, silicon dioxide (SiO₂) at 36.08%, aluminum oxide (Al₂O₃) at 9.48%, and iron oxide (Fe₂O₃) accounting for 4.61% of the total composition. Characterization of this chemical composition constitutes an essential first step toward understanding the material according to the requirements of the ASTM D2487 standard. This analysis allows identification of the dominant mineralogical nature of the soil and guides the interpretation of its physical properties, particularly its cohesion, strength, and chemical reactivity. The presence of these major oxides suggests a matrix rich in silico-calcareous and clay minerals, which is entirely consistent with our XRD measurements. Indeed, we found large quantities of calcite—frequently associated with silico-calcareous matrices in natural materials, even though it is not a silicate—alongside typical silico-calcareous phases, such as Al-tobermorite and tobermorite, as well as significant amounts of clay minerals, such as kaolinite. Other minerals detected by XRD, such as quartz, forsterite, muscovite, and albite, contribute to the overall mineralogical composition but do not specifically characterize the silico-calcareous or clay matrix. In addition, the presence of Fe₂O₃ not explained by the main XRD phases suggests the existence of minor phases or impurities not identified by diffraction. These matrices, rich in silico-calcareous and clay minerals, are well-suited and commonly encountered in civil engineering and construction applications.

3.1.3. Thermogravimetric Analysis

Thermogravimetric analysis (TGA) and derivative thermogravimetry (DTG) measurements serve to complete the material characterization by revealing amorphous phases, residual water, or impurities not detected by XRD and chemical composition analysis, while providing valuable information on its thermal behavior and stability. These measurements were performed using the NETZSCH STA 449 F5 Jupiter instrument from NETZSCH Gerätebau GmbH, Selb, Germany, on samples weighing between 60 and 80 mg, contained within an alumina pan. The temperature was increased at a rate of 10 °C/min from 20 to 900 °C under an Ar-flowing atmosphere (flow rate: 50 mL/min). It is important to note that the mixture was dried in an oven at 105 °C for 24 h before being milled to a fine powder. This step is essential for eliminating absorbed water.

Figure 6 presents the thermogravimetric analysis (TGA) and derivative thermogravimetry (DTG) profiles for the brick specimens. The mass loss observed between 100 and 120 °C is attributed to the elimination of free water via dehydration and evaporation. In the temperature range from 120 to 140 °C, the DTG peak reflects both the continued evaporation of free water and the dehydration of calcium silicate hydrate (CSH) and calcium aluminum silicate hydrate (CSAH). The high CaO and SiO₂ contents, as revealed by our chemical analysis, are consistent with the formation of calcium silicate hydrates (C-S-H or CASH) under conditions, such as pozzolanic or hydration reactions. Furthermore, our mineralogical analyses (XRD) confirm the presence of Al-tobermorite and tobermorite, both of which are calcium silicate hydrates closely resembling the classical C-S-H phases found in cements. At 295 °C, the observed mass loss may be attributed to the decomposition of hydroxides, such as goethite (FeO(OH)). Although XRD did not detect the presence of goethite, the presence of Fe₂O₃ in the chemical composition (4.61% of the total) suggests that iron is present as an oxide or hydroxide, which could account for this mass loss. Between 500 and 560 °C, the minor mass loss can be readily ascribed to the dehydration of clay minerals, specifically kaolinite in our sample [23,24]. Additionally, a discrete endothermic peak observed around 575 °C is attributed to the allotropic transformation of α-quartz into β-quartz [25]. The endothermic peaks between 670 and 830 °C, characterized by a broad range in the DSC curve, correspond to the decomposition of calcite, which is the predominant phase in our material [26].

3.2. Earth Brick Thermal Properties

To assess the thermal performance of the bricks—specifically their thermal conductivity and heat capacity—a sample measuring 19 × 19 × 4 cm³ was prepared by joining two sections cut from a single earth brick. Thermal characterization was performed using a NETZSCH HFM 446 Lambda heat flow meter from NETZSCH Gerätebau GmbH, Selb, Germany (Figure 7).Thermal conductivity and heat capacity were measured according to the ASTM C518 [27] standard. Thermal conductivity was determined at a temperature of 20 °C with an upper plate at 30 °C and a lower one at 10 °C. The specific heat capacity was measured at 20 °C with a heating rate of 1 K/min from 10 °C to 30 °C.

The measured thermal properties of the earth brick were 0.54 W.m⁻¹.K⁻¹ for the thermal conductivity and about 1200 J.kg⁻¹.K⁻¹ for the specific heat capacity (see

Table 3. Thermal properties of the earth brick (measured) and OSB (found in the literature).

Material	Thermal Conductivity [W.m−1.K−1]	Specific Heat Capacity [J.kg−1.K−1]	Thermal Diffusivity [m2. s−1]
Earth brick	0.54 ± 0.01	1199 ± 36	0.29 × 10−6
OSB Wood	0.129 [17]	1552 [18]	0.17 × 10−6

). These values are consistent with what can be found in the scientific literature, where earth blocks show thermal conductivities ranging from 0.19 to 1.35 W.m⁻¹.K⁻¹ and specific heat capacities ranging from 648 to 1832 J.kg⁻¹.K⁻¹, depending on their composition and constructive technique [28]. These thermal properties ensure a good thermal diffusivity of 0.29 mm².s⁻¹ that increases the total thermal inertia of the building [29,30,31]. For comparison, the thermal properties of the OSB wood are also reported in Table 3.

4. Results and Discussion

4.1. Temperature and Relative Humidity Evolution

The built prototype building is instrumented with the aim of carrying out an experimental campaign during summer 2023, for a period of 24 days (between 26 June and 20 July 2023). The sensors used are BME680 from Bosh Sensortec, Reutlingen, Germany allowing measurement of the temperature and relative humidity (RH). These sensors’ operating range are from −40 to +85 °C for temperature and 0 to 100% for RH. Accuracy regarding temperature and RH measurements are ±1 °C (for T between 0 and 65 °C) and ±3% (for RH between 20% and 80%), respectively. This type of low-cost sensor was deliberately chosen since it is easy to find and easy to link to any modern processor to find codes and libraries. A total of 5 sensors were installed: two sensors were installed in the control area (wooden room); two sensors were installed in the vegetable storage area (earthen room) and one sensor outside the prototype. The acquisition card was an Adafruit Huzzah ESP8266 an ‘all-in-one’ ESP8266 WiFi development board with built-in USB and battery charging. A 4G Huawei modem was installed next to the Huzzah ESP8266 to create a local Wi-Fi network. Thanks to this 4G Modem, the data were sent every 10 min from the ESP8266 to the Adafruit IO cloud and to the ‘EPF’ cloud. It should be noted here that the measurements were taken without vegetables inside the storage rooms.

Thanks to this experimental setup, the relative humidity (RH) and temperature data were recorded outside the building and in each module. Figure 8 presents the raw data obtained for the measured external temperatures (blue curve), the temperatures inside the wooden room (green curve), and the temperature in the earthen room (orange curve). A preliminary qualitative analysis allowed us to quickly observe that the raw earth vegetable storage room exhibited a thermal phase shift of approximately 6–8 h and that this thermal wave was significantly attenuated by approximately 60–80%. It is also possible to see a temperature delay of a few hours in the wooden room and a certain attenuation, lower than that of the earthen room. It can be noted that the average daily temperature was approximately the same for the outdoors and for the earthen room, while the average temperature was much higher in the wooden room, see

Table 4. Synthetic data regarding temperature and relative humidity in the building rooms and outdoors.

Criteria	Temperature [°C]			Relative Humidity [%]
	Outdoors	Wooden Room	Earthen Room	Outdoors	Wooden Room	Earthen Room
mean	26.9	29.0	26.4	54.0	46.0	57.5
std	5.4	3.6	2.4	14.0	5.4	3.3
min	15.4	21.1	20.2	20.0	31.3	48.5
max	39.0	38.3	31.8	81.6	58.1	67.0
max − min	23.6	17.2	11.6	61.6	26.8	18.5

.

Figure 9 presents the raw data of relative humidity outdoors (blue curve), in the earthen room (orange curve), and in the wooden room (green curve). The daily cycle can easily be observed in outdoor conditions, with an increase and decrease in the relative humidity. The wooden and earthen rooms constituting the building show distinct behavior. The earthen room has a well-controlled hygrometry, with slight variations that are much less pronounced than those observed outdoors. There is also a significant phase shifting (time between two successive relative humidity peaks), often larger than 12 h. The relative humidity in the rooms decreases when the one outside increases, and vice versa. The humidity in the wooden room (green curve) is systematically lower than that of the earthen room (orange curve) and lies within the lower average of the relative humidity outside.

Table 4 presents the summary statistics of temperatures and relative humidities recorded during the measurement period. The average outdoor temperature was 26.9 °C (SD = 5.4 °C), with values ranging from 15.4 °C to 39.0 °C. These conditions are consistent with a Mediterranean summer without heatwave episodes. The wooden room exhibited a higher mean temperature of 29.0 °C (SD = 3.6 °C; range: 21.1 °C to 38.3 °C), while the earthen room demonstrated better thermal performance, with a lower average temperature of 26.4 °C (SD = 2.4 °C; range: 20.5 °C to 32.1 °C), notably 0.5 °C below the outdoor average. This improved performance is mainly attributed to its substantial thermal inertia, as indicated by the lower standard deviation and the reduced diurnal temperature variation (ΔT = 11.6 °C). The Shapiro–Wilk test revealed that the temperature data for all conditions deviate significantly from normality (outdoor: W = 0.9747, p < 0.001; earthen room: W = 0.9815, p < 0.001; wooden room: W = 0.9919, p < 0.001). Consequently, non-parametric Wilcoxon signed-rank tests were performed to compare paired measurements. The results show statistically significant differences between all temperature conditions: (i) outdoors vs. earthen room: W = 2,179,433, p < 0.001, (ii) wooden room vs. outdoors: W = 1,155,775, p < 0.001, and (iii) earthen room vs. wooden room: W = 121,292, p < 0.001. For relative humidity, similar trends were observed. The Wilcoxon signed-rank tests indicated significant differences across all room pairings: (i) outdoors vs. earthen room: W = 1,793,926, p < 0.001, (ii) wooden room vs. outdoors: W = 969,981, p < 0.001, and (iii) earthen room vs. wooden room: W = 0, p < 0.001. These results confirm that the wooden room environments exhibit distinct thermal and hygrometric behaviors, with the earthen room offering more stable conditions for fruit and vegetable storage.

There was also a noticeable natural regulation of the relative humidity (RH) in the earthen room. Indeed, even though the average RH value of the earthen room (57.5%) was close to the average RH value outdoors (54%), the global standard deviation (3.3%) and the difference between minimum and maximum RH values (18.5%) were by far much smaller than the outdoor values (14% and 61.6%, respectively). This finding confirms the capacity of the earthen material to uptake/release moisture, this has been already reported in several studies [32,33,34].

4.2. Temperature Attenuation and Phase Shifting

To go deeper into quantitative analysis, the exact time interval between the outdoor peak temperature and the indoor peak temperature for each day were analyzed. This time difference, or phase shift, indicates how long it takes for the heat wave to move through the walls and ceiling. The temperature phase shift was determined by measuring the time interval between the peaks outdoors and indoors. This phase shift analysis was conducted on both modules, and the results are displayed in Figure 10a and

Table 5. Thermal phase shift, indoor temperature, and temperature difference between indoor and outdoor ambiences during the 19 investigated days.

Criteria	Earthen Room			Wooden Room
	Shift Between Peak [h]	Temperature [°C]	Shift Between Peak [h]	Shift Between Peak [h]	Temperature [°C]	Thermal Buffering [°C]
mean	6.62	27.8	5.9	4.00	33.3	0.49
std	1.00	2.0	1.4	0.86	2.5	1.12
min	5.33	22.7	3.2	2.57	26.6	0.00
max	9.17	31.8	8.8	5.67	38.3	2.94

. For the earthen room, it can be seen in Table 5 that the phase shift values ranged from 5.33 to 9.17 h, with an average of approximately 6.6 h, as illustrated by the orange histogram and orange star on Figure 10a. Regarding the behavior of the wooden room, as expected, the phase shift was lower; it ranged between 2.57 h and 5.67 h, with an average around 4 h, see the green histogram and star on Figure 10a.

The values of phase shifting observed in the case of the earthen room (0.55 h/cm) are relatively higher compared to those found by El Fgaier [31] in the case of two different bricks (0.33 h/cm and 0.25 h/cm for optimal thicknesses of 30 and 40 cm, respectively). This may be due to the lower thermal diffusivity found in the present work.

Also, the temperature difference between the outdoor and the indoor maximum peaks were examined in both modules for each day. This difference is called the peak temperature attenuation and is shown for both modules in Figure 10b. As expected, the wooden room presents a reduced temperature peak attenuation compared to the earthen one. Indeed, the earthen room presents an average attenuation of 6 °C compared to 0.5 °C for the wooden room.

4.3. Relative Humidity Stabilization and Phase Shifting

Figure 11 demonstrates that the indoor relative humidity follows the evolution of the outdoor one, with a time shift. As for the thermal phase shift analysis, the phase shift of relative humidity was determined by measuring the time interval between the peaks outdoor and indoors in the case of both rooms. Out of the 19 days of measurement (excluding days with incomplete data), the maximum humidity was observed around 4:36 a.m. for the outdoor area, 8:18 a.m. for the wooden room, and 1:30 p.m. for the earthen room.

Figure 11 illustrates the phase shift of relative humidity and presents much more dispersed data compared to temperature, particularly for the earthen room. For the wooden room, the phase shift ranged from 0.6 h to 8.6 h, with an average of 3.7 h and a standard deviation of 2.1 h, see

Table 6. Relative humidity phase shift, indoor relative humidity, and relative humidity difference between indoor and outdoor ambiences during the 19 investigated days.

Criteria	Earthen Room		Wooden Room
	Shift Between Peak [h]	Relative Humidity (Peak) [%]	Shift Between Peak [h]	Relative Humidity (Peak) [%]
mean	8.8	60.7	3.7	51.5
std	3.7	2.62	2.1	4.21
min	2.5	57.6	0.6	40.9
max	16.4	70.0	8.6	58.0

.

For the earthen room, the phase shift ranged from 2.5 h to 16.4 h, with an average of 8.8 h and a standard deviation of 3.7 h due to the ability of earth bricks to absorb and release water vapor. These disparities between the two rooms may be due to the ability of the earth bricks to moderate variations in ambient humidity. This behavior may reflect the high moisture buffer value (MBV) of these earth bricks, especially as these bricks contain clay (kaolinite). It has been shown in the literature that the presence and nature of the clay minerals will have a great influence on the moisture behavior of the earth bricks [34].

4.4. Discussion

It was observed that temperature and relative humidity were better regulated in the earthen room compared to the wooden room. Additionally, it was noted that the earthen room allows for more significant phase shifts in temperature and relative humidity than the wooden one.

The understanding of the thermal performance of our earth material relies on a detailed analysis of its composition and hygrothermal properties. It is essential here to link the mineralogical and chemical formulation of the earth to its measured thermal characteristics. Our raw earth brick exhibits a thermal conductivity of 0.54 W/(m·K), a specific heat capacity of 1200 J/(kg·K), a thermal diffusivity of 0.29 mm²/s, and a density of approximately 1560 kg/m³. These values place our brick in the upper range of high-inertia materials, a result directly attributable to the intrinsic nature of the earth’s components. Indeed, as previously discussed, the chemical composition of the earth reveals high contents of calcium oxide (CaO) and silica (SiO₂), at 43.26% and 36.08%, respectively. These elements are responsible for the formation of minerals, such as calcite (CaCO₃), quartz (SiO₂), as well as hydrated calcium silicates, like tobermorite and Al-tobermorite, which were identified by X-ray diffraction. These minerals are recognized for their density and stability, properties that promote a high thermal storage capacity. The presence of clay minerals, particularly kaolinite, further enhances this specific heat capacity, while also contributing to humidity regulation—a complementary factor for indoor thermal comfort [35,36]. The high density of the brick, in the order of 1560 kg/m³, is another determining factor. Indeed, the volumetric heat capacity, which results from the product of the density and the specific heat capacity, reaches approximately 1,872,000 J/(m³·K). This high value means that the brick can absorb and release significant amounts of heat without causing sharp variations in the indoor temperature. This thermal inertia is particularly sought after in bioclimatic buildings, as it helps to smooth out external temperature fluctuations and maintain a stable indoor climate, even in the face of significant daily or seasonal thermal variations.

The measured thermal conductivity, although moderate, remains higher than that of pure insulating materials, but this is largely compensated for by the material’s high thermal inertia. This combination of properties explains the low observed thermal diffusivity, which indicates that heat propagates slowly through the wall. This phenomenon is essential for thermal wave shift—the delay between the moment when heat reaches the outer surface of the wall and when it manifests on the inner surface. Thus, the earth brick studied here offers an excellent compromise between insulation and thermal inertia, distinguishing it markedly from lighter materials, such as OSB panels, which were used alone in the second, wooden test room.

In addition to hygrothermal performance, it is essential to analyze the mechanical properties and durability of our material. The mineralogy identified by XRD (kaolinite, muscovite, residual illite, tobermorite, Al-tobermorite, and forsterite) directly elucidates both the mechanical and hygrothermal characteristics. Unlike swelling clays, such as smectites (charge = 0.3–0.8), the minerals present in our mixture—notably muscovite (high charge) and kaolinite (neutral charge)—do not exhibit significant interlayer expansion capacity. This absence of swelling reduces the risk of humidity-induced cracking but also limits intercrystalline water retention [37]. However, kaolinite contributes to passive hygrometric regulation within the earthen room by absorbing and releasing ambient moisture without causing structural alterations. Simultaneously, the high contents of silica (36.08% SiO₂), alumina (9.23% Al₂O₃), and calcium (43.26% CaO) promote pozzolanic reactions, leading to the formation of calcium silicate/aluminate hydrates (CSH, CSAH), such as tobermorite and Al-tobermorite. These phases, detected by XRD, enhance mechanical cohesion and durability. Forsterite (Mg₂SiO₄) plays a complementary key role: during the partial hydration of the clay, it promotes the development of magnesium-rich binders, improving long-term strength and stability.

To summarize, the high calcite and quartz content in our bricks provide enhanced mechanical strength, making them suitable for load-bearing and durable building applications. The presence of Al-tobermorite and tobermorite, both hydrated calcium silicates, contributes to improved water resistance and durability. Muscovite and forsterite further reinforce the material’s resistance to cracking and high temperatures, while the kaolinite content imparts moderate plasticity and moldability, which are beneficial during shaping when the material is combined with water [38]. Overall, this composite material delivers balanced construction performance, combining mechanical robustness, thermal insulation, and effective moisture management, although its plasticity is lower than that of conventional clay due to the predominance of non-clay minerals.

5. Conclusions

Two experimental modules—each measuring 2.50 m in length, 2.38 m in width, and 2.30 m in height—were constructed in the south of France (Mediterranean climate), one using raw earth bricks (earthen room) and the other using wood (wooden room). The hygrothermal performance of both modules, specifically phase shift and thermal attenuation, was monitored over a four-week period during the summer of 2023. This study demonstrated that dense earth bricks (approximately 1560 kg/m³), produced from local and low-cost materials, offer a promising solution for the short-term storage of fruits and vegetables during the summer in Mediterranean conditions. The earth brick walls exhibited transmission thermal inertia comparable to, or slightly better than, values reported in the literature. Notably, the amplitude of indoor temperature fluctuations was reduced by about 6 °C, and a mean time lag of 6.6 h (0.55 h/cm) was observed. Additionally, the raw earth walls effectively regulated and stabilized the indoor relative humidity, maintaining typical values between 50% and 65%. These temperature and humidity conditions are suitable for the temporary storage of certain fruit and vegetable varieties, and for others, they closely approach optimal storage requirements. As a result, the need for active cooling is reduced, highlighting the potential of raw earth construction for passive, energy-efficient food preservation in warm climates.

This study demonstrates that raw earth construction offers a practical, low tech, low cost, low-energy, and environmentally friendly solution for short-term agricultural storage in hot, dry Mediterranean climates. The adobe walls provide effective passive cooling and humidity regulation, without the need for expensive and non-sustainable refrigeration. The construction method is simple, uses locally available materials, and is well-suited for self-building and modular, reversible structures, making it highly replicable, adaptable, and scalable for different agricultural contexts and storage needs (whether for small-scale fruit and vegetable storage, larger agricultural warehouses, or even other uses, such as processing or packaging areas). The reversibility and potential for disassembly further facilitate adaptation to changing requirements or relocation. We believe our results obtained in the Mediterranean context could be largely transferable to other hot and arid regions, particularly effective in climates characterized by large diurnal temperature swings, low relative humidity, and high daytime temperatures, like in extremely dry regions (while in more humid areas (tropical regions), the earth’s ability to absorb and release moisture may be limited).

From a policy perspective, these results suggest that regional building codes and rural development strategies should recognize and support earth-based construction as a viable, sustainable alternative for agricultural infrastructure. Facilitating authorization procedures for reversible, eco-designed buildings and providing technical guidelines or incentives could accelerate the adoption of passive cooling solutions, contributing to both climate adaptation and the preservation of agricultural land.