Introduction

Stone relics carry unique historical, artistic, and scientific value and are the common heritage of all mankind. However, large open-air stone relics have been exposed to natural environments for a long time and have experienced severe weathering and erosion, leading to complex damage. For example, the Leshan Giant Buddha (LGB), the world’s largest stone statue of the seated Maitreya Buddha, has experienced tremendous damage under the dual effects of natural weathering and anthropogenic influences, with increasing damage due to weathering (Fig. 1). In addition, air pollution can lead to surface darkening, acidification, and crust formation, whereas biological factors, such as colonization by algae, mosses, and lichens, can exacerbate surface degradation and promote moisture retention. Owing to long-term natural weathering, immovable stony relics are susceptible to reduced rock strength due to increased porosity1. During the weathering process, calcite, feldspar, and other components in rock react, reducing the degree of cementation and increasing the number of pores and cracks2. These complex interactions between environmental factors and the porous structure of stone require specialized research and conservation strategies. Unlike indoor or sheltered heritage sites, open-air artifacts need to be considered for direct and prolonged exposure to harsh weather conditions. The unique environment of the LGB highlights the need for an integrated approach that addresses both macroscopic and microscopic weathering processes to ensure effective conservation and long-term stability.

Fig. 1: Statue of the LGB.
figure 1

a Photographs of LGB. b–e The degradation of the LGB.

Full size image

In recent years, an increasing number of scholars have begun to carry out conservation research on the cultural heritage of sandstone relics3,4,5,6,7. The depositional and tectonic features of sandstones lay a material foundation for various weathering characteristics and degrees8. Meng et al.9 proposed an innovative method that combines terahertz spectroscopy and ultrasonic velocity and uses least squares support vector machines to classify and predict the weathering levels of large stone relics efficiently. Yang et al.10 developed an intelligent evaluation method that integrates hyperspectral imaging and machine learning techniques for classifying and quantifying the weathering stages of sandstone in humid environments. Sandstones that are dominated by silica and calcium carbonate are more complex in composition after weathering11. Salt crystallization is one of the important factors that causes weathering of sandstone grottoes12,13,14,15. The prevention and control of condensation damage is especially important for tomb-like sandstone stony cultural heritage relics through onsite monitoring, indoor tests, and software simulations16. Shi et al.17 proposed a method for rating condensation damage and provided corresponding preventive measures. The microscopic pore structure of sandstone is more damaged by the combined action of freezing, thawing, and loading18. The interaction of rock dissolution and drying is the main factor that causes changes in the microstructure and mineral composition of rocks, which ultimately leads to the decay of their physical and mechanical properties19. On the basis of hyperspectral images, Li et al.20 established a prediction model and a deterioration identification model to provide a new method for investigating the cultural deterioration of stone relics. Hyperspectral methods can also be used to evaluate the status of salt-induced weathering on the surfaces of sandstone artifacts21. Chen et al.22 used spectral indices and an intelligent model to evaluate various degradation patterns of stone carvings in humid regions of southern China. Porosity is the main factor controlling the durability of cultural heritage objects made from stone23. Nondestructive testing technology provides an effective means to assess the deterioration status of cultural heritage objects. This method not only reduces the cost of assessment but also plays an important role in ensuring the protection of cultural heritage objects24. The capillary water absorption process can be observed via the high-density resistivity method, and the deterioration characteristics of the stone surface can be analyzed25. The weathering rates of historic sandstone buildings in semiarid environments are controlled by the age of the buildings as well as the orientation of the studied sandstone wall sections26. Dissolution weathering of quartz sandstone is a fundamental process that increases rock porosity and reduces erosion resistance through the intergranular dissolution of quartz27. Tetsuya Waragai28 proposed an equation that can estimate the maximum depth of voids in a sandstone block, and the coefficients of the equation can help in predicting the degree of deterioration of stone cultural heritage sites such as Angkor Wat. The final weathering of rocks depends on a complex combination of three parameters: environmental conditions, salt compositions, and rock physical properties29. The above studies provide an extremely important reference and foundation for the conservation of sandstone cultural heritage.

Because large stone relics are mostly located in open-air settings, natural environmental factors have a greater impact, but because of the variation in rocks in different regions, establishing unified, quantitative standards for classifying the degree of rock weathering is difficult. Traditional engineering geology classifies the degree of rock weathering by using more qualitative evaluation methods of engineering geology, such as the rock structure, rock color, mineral composition, rock fragmentation, ease of excavation, and other aspects of comprehensive analyses to determine the degree of weathering, given the weathering grade. The quantitative evaluation mostly uses some indicators of physical and mechanical properties30, which is not very helpful for research on rock weathering mechanisms. The current research on large stone relics is focused mostly on local or small-scale analyses and lacks a systematic and comprehensive analysis method. The microscopic characteristics of rocks and compositional changes in rocks are effective methods used to identify the weathering of lithic artifacts31,32. Therefore, a systematic and scientific study of weathering variations and a quantitative analysis method are necessary.

Most of the studies on the LGB have focused on preliminary explorations of the nature of the bedrock, whereas relatively few studies have been conducted on its weathering characteristics. Earlier studies divided the sandstone of the LGB into three sections33, and the weathering depths of the top, middle, and bottom layers were initially determined to vary from 6–8 cm, 5–6 cm, and 8–9 cm, respectively34. A stratigraphic investigation revealed that the bedrock of the LGB has spatial differences in terms of mineral composition and structural configuration35. Tian36 concluded that the typical vertical sequence of the LGB consists of two sets of rocks with significantly different grain sizes. The middle to lower part consists of fine- to medium-grained feldspathic quartz sandstone with very large oblique laminations, which are associated with inland sand genesis. The upper part consists of siltstone with sandy grain laminations and is the result of intermittent sheet-flow deposition. Huang et al.37 analyzed the effects of bacterial and algal symbionts on the weathering of the LGB and reported that there was severe damage to the LGB caused by biological weathering. Sun et al.38 reported that carbon quantum dots prepared with Ginkgo biloba as a precursor could be used for tracing stone cultural heritage objects through laboratory tests and traced the source of seepage on the chest of the LGB. These studies provide an important basis for the conservation of the LGB. Rock characteristics and the environment in which they are located are the main factors affecting weathering. Differences in the properties of the strata cause the weathering of the LGB in different strata to differ. LGB is strongly affected by chemical weathering, which is often closely related to the loss of rock cement, soluble salt transportation, and various chemical reactions. Chemical weathering not only leads to changes in the mineral and chemical compositions of rocks but also affects the pore characteristics and microstructures. However, the LGB shows significant spatial variations in the degree of weathering due to its large volume, and systematic and quantitative studies on its weathering variations are still lacking. Targeted studies are urgently needed to conserve the LGB, especially to analyze the weathering mechanisms of its different strata.

This study aims to establish a systematic and scientific method for the study and quantitative analysis of weathering variations. The stratigraphy of large open-air stone relics is identified through field surveys and thin-section examinations, the natures and characteristics of different layers are analyzed, and their macroscopic and microscopic characteristics are compared to systematically analyze the variations in weathering. Weathering indicators were established using the test results obtained for the mineral compositions and elemental contents to quantify the weathering variations among layers. This study provides a systematic and quantifiable analysis method to evaluate the weathering variations of large open-air stone relics, which in turn provides a scientific basis for their conservation and restoration.

Materials and methods

Study area

The LGB, also known as the Lingyun Big Buddha, is a seated statue of the Maitreya Buddha that is chiseled from the riverfront cliffs of Qixia Peak, Lingyun Mountain, on the southern bank of the Minjiang River. The LGB is located in Leshan City, Sichuan Province (Fig. 2), at the confluence of the Min River, Qingyi River, and Dadu River, so it has long been in a humid environment33,39. The LGB, carved during the Tang Dynasty (713–803 CE), is a monumental stone statue. Standing 71 m tall, it is the largest stone Buddha in the world. In 1982, the LGB was listed as a protected unit as a national key cultural relic by the Chinese State Council, and in December 1996, the Emei Mountain LGB was approved by the United Nations Educational, Scientific, and Cultural Organization as a “World Cultural and Natural Heritage” site.

Fig. 2
figure 2

Map of the study area.

Full size image

The LGB is located in a subtropical monsoon climate zone with high humidity and abundant rainfall. According to the years of meteorological data from the Leshan Meteorological Station, the average temperature of Leshan city has been 17.2 °C for many years, with the highest temperature of 38.1 °C and the lowest temperature of −4.3 °C. The average annual rainfall is 1384.8 mm, with a maximum rainfall of 248.2 mm per day. Rainfall can reach up to 104.9 mm during heavy rainfall, and rainfall is mostly concentrated in the months of May and September, which accounts for 78.5% of the annual rainfall. The average relative humidity is 81% per year, and the average evaporation is 617.1 mm per year. These conditions greatly contributed to the weathering processes observed on the surface of the Buddha, including water seepage and biological diseases40,41,42.

Stratigraphy

To explore the main causes of weathering in the LGB, in this study, samples were collected from several parts of the right rock wall of the LGB from the bottom up for systematic identification and analysis.

Thin-section microscopic examination of the sandstones of the LGB revealed obvious differences in the rock structure of the sandstone strata, which can be divided into two main categories: a coarse–medium sandy structure or fine–medium sandy structure, and a powder-fine sandy structure or medium–fine sandy structure (see Fig. 3). Different parts of the rock assemblage show significant differences and can be divided into two parts with the knee of the LGB as the boundary: the knee part consists mainly of powder-fine clastic quartz sandstone with a high content of heterogeneous bases; the upper part (except for the head and middle and lower parts) is mainly composed of clastic quartz fine sandstone, fine-meso-sandstone, and meso-fine sandstone; and the lower part consists mainly of clastic quartz fine sandstone and meso-fine sandstone. This study focuses on the upper part of the LGB and provides insight into the weathering differences in the area above its abdomen.

Fig. 3: Thin-section identification results of LGB sandstones.
figure 3

a Coarse- to medium-grained clastic quartz sandstone and b powdery–fine-grained clastic quartz sandstone.

Full size image

On the basis of the field investigation, this study divides the area above the belly of the LGB into 10 layers from bottom to top on the basis of the natural geographic conditions, rock structure, stratigraphy, and microstructure of the rocks. The lithological characteristics and basic descriptions of each layer are shown in Fig. 4. The rock compositions mainly include quartz, feldspar, carbonatite clasts, sandstone clasts, mudstone clasts, flint clasts, kyanite clasts, quartzite clasts, and ejecta clasts, and small amounts of chlorite and mica can also be observed. The pore types consist mainly of intergranular pores and intragranular soluble pores, and the pore fillings are mainly composed of a heterogeneous clay matrix and calcareous and ferruginous cement.

Fig. 4
figure 4

Lithologic column of LGB formation.

Full size image

Stratigraphic sampling

In geotechnical engineering, sampling is a critical step to ensure the representativeness and reliability of data. In this study, sampling locations were designed on the basis of the spatial distribution of weathering features observed in the LGB. Through a combination of visual inspection and non-destructive testing techniques, sampling was carried out in the area of the right cliff wall of the LGB with obvious weathering features (Fig. 5). To ensure the accuracy of the results, the samples compared between different layers were surface (0–2 cm) samples. Samples were taken from both different strata and different depths to capture various degrees of weathering for macro- and microanalysis. All sampling was conducted under strict cultural heritage protection guidelines to minimize the impact on the LGB. In accordance with the standard test method for engineering rock masses43, the samples were processed into cylinders.

Fig. 5
figure 5

Sample of a stratum in the Leshan Giant Buddha.

Full size image

Test methods

To ensure the reliability and reproducibility of the experimental results, all tests in this study were conducted with at least three independent repetitions under identical conditions. The average values of the results were used for analysis, and standard deviations were calculated to assess the variability. The observed fluctuations in the data were within acceptable ranges and consistent with the inherent heterogeneity of porous stone materials, which is typical in weathering studies. The test methods used for the study are as follows:

  1. (1)

    X-ray diffraction (XRD; Model: X’ Pert Pro MPD) was used to identify the minerals from different strata of the LGB and to analyze their mineral contents.

  2. (2)

    X-ray fluorescence spectroscopy (XRF; Model: PW2403) was used to analyze the elemental content of the sandstone.

  3. (3)

    Pore size distribution tests were carried out via mercury-in-pressure (MIP; Model: PoreMaster60) tests.

    The sandstone permeability magnitude reflects its ability to allow fluids to pass through and is closely related to sandstone porosity. Darcy’s law for one-dimensional steady seepage of gases was used to determine and calculate the permeabilities of sandstones in different layers4,44,45,46. Gas permeability is absolute permeability, which is a property of the rock itself and depends on the pore structure of the rock. During a test, air is usually used as the medium in the rock core being tested, so the absolute permeability of the core is also known as the air permeability.

    $$Q=-\,k\cdot\, A\,\cdot \,\frac{{dP}}{{dx}}$$
    (1)

    where \(Q\) is the seepage volume (flow rate), \(k\) is the permeability, \(A\) is the cross-sectional area, \({dP}\) is the pressure difference, and \({dx}\) is the flow path length. The unit “mD” represents millidarcy, which is a standard unit of permeability in geosciences and petrophysics.

  4. (4)

    An overburden porosity–permeability meter (Model: PoroPDP-200) was used to determine the porosity and permeability of samples from different layers.

  5. (5)

    The types and contents of easily soluble salts in different layers of the LGB were determined via chemical analysis of water. A Research Ion Chromatograph (Model: ICS-2500) was used.

  6. (6)

    Scanning electron microscopy (Model: SU-1500) was used to study the microstructures of the sandstone samples obtained from various strata of the LGB.

Characteristics of the spatial variations in lithology

Mineral compositions and elemental contents of the strata

The mineral composition of a rock influences its physical and chemical properties, which, in combination with factors such as texture, porosity, and cementation, contribute to its overall strength. The XRD test results (Fig. 6a) reveal that the main constituents of the sandstone strata of the LGB are quartz, feldspar, and calcite, with small amounts of clay minerals. Among them, quartz occupies the largest proportion, with a total content ranging from 50% to 79%, so it can be assumed that quartz has a decisive influence on the properties of the sandstone strata in the LGB. Feldspar is second only to quartz in this sandstone, fluctuating from 14% to 36%. The calcite content ranges from 4% to 18%, whereas the clay mineral content is low, at only 3% to 5%, indicating that the sandstone in the LGB strata has a low cement content and is weakly cemented and easily weathered.

Fig. 6: XRD test results of rock samples of the LGB.
figure 6

a Different sampling points, and b differentiation.

Full size image

As shown in Fig. 6b, by analyzing the average values of the mineral contents of the samples between the different layers, the variations in the quartz content are highly consistent with the stratification boundaries, but there are also some variations within the thicker layers. The quartz contents in layers 1, 2, and 3 are relatively low, whereas the quartz content in layer 4 significantly increases and remains high in the remaining layers. Moreover, the feldspar and calcite contents in layers 1, 2, and 3 are high and decrease with increasing quartz content. These findings indicate that the changes in quartz content are negatively correlated with the changes in feldspar and calcite contents. Therefore, it can be inferred that the sandstone strata of the LGB use quartz as the skeleton, whereas feldspar, calcite, and clay minerals are used as fillers.

The elemental content of a rock has an important influence on its mineral composition and physicochemical properties, which in turn affect the weathering rate and erosion resistance of the rock. The results of the XRF test (Fig. 7a) reveal that the SiO2 content of the sandstone strata in the LGB is the highest, accounting for approximately 50%, followed by those of CaO and Al2O3, which account for ~20% and 15%, respectively. The contents of Fe2O3, MgO, Na2O, and K2O are low, accounting for ~2%, and the sum of the remaining elements is less than 1%. This indicates that SiO2 dominates the strata of the LGB sandstones, further confirming the results of the previous mineralogical composition analysis that revealed that quartz is dominant because it is composed mainly of SiO2. The high SiO2 content indicates that the rock skeleton is mainly composed of quartz, whereas the CaO content indicates the importance of calcite (CaCO3) in the sandstones. The 15% Al2O3 content is closely related to the presence of feldspar and clay minerals. and clay minerals, indicating that the sandstone strata of the LGB underwent different degrees of chemical weathering and sedimentation.

Fig. 7: XRF test results of a rock sample of the LGB.
figure 7

a Different sampling points, and b differentiation.

Full size image

In the strata from layers 1 to 10, the contents of Al2O3, Fe2O3, MgO, Na2O and K2O do not change much, whereas the contents of CaO and SiO2 change significantly, and the two show opposite trends: the higher the content of CaO is, the lower the content of SiO2 (Fig. 7b). This trend is consistent with the change rule of quartz and calcite in the XRD test, and the changes in the CaO and SiO2 contents are highly consistent with the stratification boundaries, indicating that the stratification of the LGB strata sandstone is reasonable. Therefore, the SiO2 and CaO contents reflect the weathering intensities of different strata to some extent. Since SiO2 does not easily weather, the sandstone strata of the LGB are mainly exfoliated and weathered by materials other than SiO2 during the weathering process, so the rocks with higher SiO2 contents are weathered to a greater extent, whereas the CaO content is relatively low.

Stratigraphic pore distribution

The pore distribution characteristics of rocks significantly affect their permeability, water storage capacity, and mechanical properties, and these factors play a key role in weathering and stability studies. To investigate the characteristics of the pore size distributions in the sandstone strata of the LGB, MIP tests were carried out, and the results are shown in Fig. 8. The pores of the LGB sandstone can be categorized into small pores (diameter < 2 μm), medium pores (2 μm < diameter < 50 μm) and large pores (50 μm < diameter < 1000 μm). The results show that mesopores are the main pore type in the sandstone strata of the LGB, and more mesopores are present in all layers, which is consistent with the low cement content and weak cementation of the LGB and is one of the important reasons for its easy weathering.

Fig. 8: Pore characteristics of different strata in the LGB.
figure 8

a Pore distributions of layers 1–5, b pore distributions of layers 6–10, c cumulative curves of the apertures of layers 1–5, and d cumulative curves of the apertures of layers 6–10.

Full size image

The pore sizes and distribution densities varied significantly among the different layers. The pore sizes of layers 5–8 were relatively large and densely distributed, and the density of mesopores larger than 10 μm was relatively high, providing good channels for water infiltration. The other layers have smaller pore apertures and relatively low densities of mesopores larger than 10 μm. Although macropores are developed in all the layers, layers 5–8 have more macropores and larger pore diameters. The high-density distribution of mesopores typically increases the surface area of rock, increasing its susceptibility to chemical and physical weathering. Specifically, the entry of water into the rock interior through micropores may lead to salt loss and chemical reactions, accelerate the weathering process, increase the pore size of existing pores, and create new pores. In addition, the high-density distribution of micropores enhances the contact area between the rock and the weathering medium, further accelerating the weathering rate and increasing susceptibility to disintegration and erosion in the natural environment.

Stratigraphic porosity and permeability

Figure 9 shows that there are large differences in porosity and permeability between the sandstones of different layers, and they are positively correlated. Both the porosity and permeability significantly fluctuate with increasing layer height. The trends of porosity and permeability are consistent in most of the layers; i.e., when the porosity increases, the permeability also increases, and vice versa.

Fig. 9
figure 9

Curves showing the variations in the porosity and permeability coefficients of sandstone in different layers.

Full size image

Specifically, the porosities and permeabilities gradually increase in layers 1 to 6, with slower growth rates in layers 1 to 3 and significant increases after layer 4. There is a slight decrease in layer 7, followed by a maximum in layer 8, where the porosity reaches a maximum value of 22.6% and the permeability reaches a maximum value of 10.39 mD. This suggests that the sandstone in layer 8 has high fluid storage and transport capacities. However, in layer 9, both the porosity and permeability decreased significantly, indicating that the sandstone in this layer is relatively dense and has low storage and transport capacities.

Overall, there are differences in the pore structures and fluid transport capacities of the sandstones in different layers, which provide conditions for the generation of weathering differences. These differences not only affect the physical properties of the rock but also have important effects on the weathering process and its stability in the natural environment.

Stratigraphic soluble salt content

The contents of labile salts in rocks have an important influence on their weathering processes and durability. The dissolution and recrystallization of salts may lead to the destruction of the rock structure, thus accelerating the onset of weathering and erosion. The test results shown in Fig. 10 reveal that the distributions of the soluble salt contents in layers 1 to 10 of the LGB strata are not uniform. Among the cations, the Ca²+ and Na+ contents are the greatest, generally fluctuating between 1 and 50 mg/100 g, whereas the K+ and Mg²+ contents are generally <1 mg/100 g. The anion contents vary considerably, with the NO3− and Cl− contents dominating in layers 1–3, and their contents are much higher than those in the other layers. Above layer 4, the ion content decreases significantly until it is generally less than 5 mg/100 g in layer 6.

Fig. 10: Ion contents and conductivities in the sandstone of the LGB.
figure 10

a Ion contents; b conductivities.

Full size image

The information from the site investigation shows that layers 4 to 8 and 10 are covered by vegetation, especially layers 6, 7, 8, and 10, which are better covered by vegetation. This finding indicates that the vegetation on the cliff surface is closely related to the distribution of soluble salts. In areas covered by vegetation, the soluble salt contents are relatively low, whereas they are relatively high in uncovered areas. This is related to the pore sizes and pore distributions of the sandstones in different layers.

The conductivity of the leachate of the cliff core samples indirectly reflects the salt content of the cores, and the higher the salt content is, the greater the concentration and conductivity of the leachate. The experimental results in Fig. 10b show that the conductivity values of the leachate are consistent with the trend of the ion concentrations, and the conductivities of layers 1 to 3 are significantly higher than those of the other layers, with a maximum value of more than 2000 μs/cm, whereas the conductivities of layers 4 to 8 and 10, which are covered by vegetation, decrease significantly and generally remain below 200 μs/cm.

Stratigraphic microstructure

To reveal the microstructural differences between the different sandstone strata of the LGB, in this study, the rock samples from each stratum were analyzed via SEM, with a focus on comparing the grain arrangement, surface roughness, and porosity. A qualitative description was developed in terms of surface morphology, grain structural integrity, cleavage and surface deposits. This analysis contributes to an in-depth understanding of the variations in the natural weathering process of each stratum, thus indicating the durability and structural deterioration characteristics of each layer.

Figure 11a, b show the loose and inhomogeneous particle distribution, large intergranular gaps and broken edges, suggesting that the samples have undergone significant weathering, especially the effects of physical and chemical weathering. In contrast, the samples in Fig. 11c, d have tightly arranged particles and small crevices, indicating less weathering and more intact surface material.

Fig. 11: SEM images of the LGB sandstone.
figure 11

a Layer 1, rock sample Y1; b Layer 2, rock sample Y6; c Layer 3, rock sample Y11; d Layer 4, rock sample Y12; e Layer 5, rock sample Y13; f Layer 6, rock sample Y18; g Layer 7, rock sample Y20; h Layer 8, rock sample Y21; i Layer 9, rock sample Y25; and j Layer 10, rock sample Y27.

Full size image

Figure 11e, f clearly reveal weathering characteristics with loose particle distributions and rough surfaces, suggesting that the weathering process leads to a significant increase in physical decomposition. The morphology of the grains in Fig. 11g, h shows a fractured structure and pronounced porosity, reflecting strong physical weathering, which may be related to moisture erosion and temperature fluctuations.

Finally, the samples in Fig. 11i, j contain uniformly arranged particles and large contact surfaces with minimal gaps, indicating that these samples are in a less weathered state. The dense particle structure of these materials results in greater weathering resistance, probably due to the presence of weathering-resistant minerals or less exposure to external environmental changes.

The combined observations reveal that the degree of weathering of rock samples from different strata significantly varies. The more severely weathered strata (e.g., Fig. 11e, g) exhibit obvious granular decomposition and surface roughness, whereas the less weathered samples (e.g., Fig. 11i, j) present tighter granular structures. These differences not only reflect the environmental conditions of each stratum but also provide a reference for the study of the durability of stone relics.

Weathering characteristics at different depths in the same strata

To avoid interference from severely weathered layers and soluble salt accumulation layers, the fourth layer was chosen to study weathering differences in sandstones at different depths. Interval depths of 0–2 cm, 2–4 cm, and 4–6 cm were analyzed. Samples in the middle of each interval were subjected to XRF testing to determine their chemical composition, whereas MIP testing was used to characterize the pore structure.

The chemical composition analysis (Fig. 12a) revealed that the amount of oxides varied considerably with depth. Notably, the content of calcium oxides increased significantly with depth, suggesting a relatively high concentration of calcite or related calcium-bearing minerals in the deeper horizons. This trend suggests that weathering diminishes with depth. In contrast, the SiO2 content decreases slightly with depth.

Fig. 12: Test results of LGB sandstone at different depths.
figure 12

a XRF test and b MIP test.

Full size image

Pore structure analysis (Fig. 12b) further illustrates the effects of weathering processes on the properties of sandstone at different depths. The pore size‒density function curves show a large variation in pore distribution, with a greater proportion of larger pores in the shallower horizons (0–2 cm) than in the deeper horizons. This may be related to more intense weathering near the surface, which results in larger pores and weaker structures. In contrast, deeper strata (4–6 cm) feature a greater proportion of smaller pores, reflecting the relative integrity of the sandstone strata, which has been minimally affected by aggressive weathering.

These findings emphasize the depth-dependent nature of sandstone weathering, with significant changes in both chemical and physical properties occurring at the most superficial levels. The interplay between changes in mineral composition and pore structure highlights the complexity of weathering mechanisms.

Discussion

Mineral compositions and elemental contents

As determined from the characterizations of the components and structures of the rocks, the high feldspar and rock debris contents of the rocks that form the LGB, as well as the properties of weak cementation and well-developed pores, make it easy for the rock to weather and flake off layer by layer. The bedrock skeleton of the LGB is mainly composed of two parts: first, quartz chalk particles, whose SiO2 content accounts for ~50% and is the most dominant mineral component; second, the cement is composed of calcium, aluminum, iron, sodium, potassium and other trace elements, which bind the quartz particles together.

In the shallow samples, owing to long-term exposure, the content of easily soluble components such as calcium, potassium, and sodium decreased significantly, leading to mineral depletion and structural weakening, and the increase in porosity of the surface rocks increased the water‒salt migration capacity, which exacerbated the deterioration process. In contrast, the loss of these components in the deeper samples was lower than that in the deeper samples, and the minerals remained relatively intact with significantly lower weathering. In addition, the crystallization pressure due to salt migration and crystallization is another important driver of weathering in shallow layers, especially during wet-dry cycles, which is more pronounced in the destruction of microporous structures.

Differences in the mineral composition of different sandstone layers not only determine the resistance to weathering but are also closely related to changes in the interlayer pore structure and saltwater behavior. These variations reveal the intrinsic driving force of sandstone weathering and the mechanisms of its interaction with the external environment, and further study of these mechanisms is important for the development of conservation strategies.

Since quartz is very stable physically and chemically and montmorillonite is usually used as a product in chemical reactions, the chemical reactions of water with rock mainly involve reactions between feldspar, calcite, and CO2 in the groundmass, a process that takes place with the participation of water (Fig. 13). This reaction further aggravates the weathering process of the rocks. The reaction equation was as follows:

Fig. 13
figure 13

Schematic diagram of the water-rock reactions in the LGB.

Full size image

Calcite:

$${\rm{CaC}}{{\rm{O}}}_{3}+{\rm{C}}{{\rm{O}}}_{2}+{{\rm{H}}}_{2}{\rm{O}}={{\rm{Ca}}}^{2+}+2{\rm{HC}}{{\rm{O}}}_{3-}$$
(2)

Albite:

$$2{\rm{NaAlSi}}{{\rm{O}}}_{8}({\rm{s}})+2{\rm{C}}{{\rm{O}}}_{2}+11{{\rm{H}}}_{2}{\rm{O}}=2{{\rm{Na}}}^{+}+2{\rm{HC}}{{\rm{O}}}_{3-}+{{\rm{Al}}}_{2}{\rm{Si}}{{\rm{O}}}_{5}{({\rm{OH}})}_{4}({\rm{s}})+4{{\rm{H}}}_{4}{\rm{Si}}{{\rm{O}}}_{4}$$
(3)

The high porosity and permeability of the LGB sandstones make them susceptible to atmospheric precipitation and groundwater to form aquifers. The presence of this water leads to hydrolysis and cement loss, which reduces the strength of the sandstone and causes spalling and loss of quartz grains47. Therefore, the index of colloid loss of the SiO2/(CaO+Al2O3 + Fe2O3) LGB sandstone strata can be defined as (a). A higher value indicates greater colloid loss, and more sandstones are subjected to dissolution and erosion by water. Moreover, K and Ca have a weak affinity for O in rocks, often in the form of free ions, and these elements are prone to migration during weathering35. Therefore, Al2O3/(K2O+CaO) can be used as an indicator of the degree of chemical weathering of sandstone (b), and the higher its value is, the greater the degree of chemical weathering. These two indicators can effectively reflect the weathering and deterioration characteristics of LGB sandstone.

The different depth indices (a) and (b) for the fourth stratum were calculated, as shown in Fig. 14. Both indices decrease with depth, indicating that sandstones experience more severe weathering at the surface (0–2 cm and 2–4 cm) than in deeper layers (4–6 cm). The index of loss of cementitious material (a) indicates that leaching is more pronounced at the surface, where it is directly affected by rainfall, for example. Similarly, the chemical weathering intensity index (b) decreases significantly with depth, indicating that the surface layer is subjected to more intense chemical reactions such as hydrolysis and ion migration than are relatively stable deeper layers.

Fig. 14
figure 14

Curves showing changes in cementation loss and chemical weathering strength of sandstones at different depths.

Full size image

Sandstones at depths of 4–6 cm are considered representative of unweathered sandstones and can be used as benchmarks for comparison. This assumption is based on the stability of the physical and chemical parameters observed at this depth, indicating minimal exposure to external weathering media. In addition, deeper samples typically retain their original mineral composition and pore structure, providing a reference for assessing changes due to surface weathering. This approach effectively quantifies the degradation of sandstones due to natural weathering processes and highlights the need for targeted conservation strategies for heavily weathered surface layers.

Figure 15 shows the changes in the cementitious material loss indicator (a) and chemical weathering intensity indicator (b) in different strata sandstones of the LGB, and their respective baselines (red and blue dashed lines) represent the values of unweathered sandstones. The differences in Fig. 15 are due to natural variability in the rock samples, a common challenge in such studies. These differences reflect general trends rather than precise measurements.

Fig. 15
figure 15

Curves showing the changes in the loss of cementation and chemical weathering strength of sandstone in the LGB.

Full size image

The indicator of cementitious material loss (a) shows that the loss of cementitious material in the upper body of the LGB tends to increase and then decrease, but the loss of cementitious material in its chest and abdomen (layers 4–8) is still high. It reaches a maximum value of 1.7 in layer 8, which shows that the loss of cemented material in these layers is more serious, and the pore aperture inside the sandstone is larger and densely distributed, which indicates that the degree of weathering is significantly -3. This pattern suggests that the sandstones in the chest and abdomen of the LGB experience the most serious dissolution of cemented material, which may be due to greater exposure to water infiltration and dissolution processes. The lower layers, although closer to the bottom, appear to be less affected by these processes, possibly due to reduced water ingress.

A similar trend was observed for the chemical weathering strength indicator (b), which varied considerably between layers, with fluctuations ranging from 0.5 to 0.8. Layers 5, 6, 7, 8, and 10 exhibited stronger chemical weathering, whereas layers 1, 3, and 4 were relatively weaker. This fluctuation amplitude reflects the significant difference in chemical weathering between different layers, indicating that sandstones have differences in composition and structural properties between different layers, which affects their weathering process.

In the uppermost strata, the values of both indicators return to levels closer to the baseline, indicating relatively low weathering intensity. This may be due to better drainage measures at the head of the LGB and the protective effect of decorative materials on the rock surface.

The deviation of the two indices from the baseline values highlights stratigraphic differences in weathering intensity driven by physical and chemical processes. The thoracic and abdominal parts of the LGB appear to be the most susceptible to weathering, possibly due to their ___location, which may be subject to greater water retention and weathering. Previous studies have shown that the chest and abdomen of LGB develop severe hollowing disease, which provides excellent conditions for water retention48. These results emphasize the importance of targeted conservation measures for the strata with the most pronounced weathering degradation.

Pores and microstructures

The degree of rock weathering is positively correlated with porosity and pore size49. The weathering of the same rock can be characterized with the help of the porosity and pore distribution50,51. As the weathering of the bedrock of the LGB strata intensified, the MIP test results revealed a gradual increase in porosity and pore size, which is consistent with the findings of Zhou et al.’s study on the Yungang grottoes52, indicating that the proportion of large pores increased with increasing weathering depth and that the mechanical and hydrological properties subsequently decreased. However, the crystallization of soluble salts on bedrock surfaces may also fill pores, thus reducing the average pore size and porosity53.

Different pore size distributions between different layers lead to different water permeabilities, which results in weathering differences between different layers. In layers 5–8, the pore sizes are relatively large, and the proportions of large pores are high, which provide good channels for water infiltration, resulting in higher water permeability, increased water-rock interactions, and greater weathering in these layers. In contrast, layers 1–4 have reduced proportions of large pores, poor water permeabilities, and weak water-rock interactions, resulting in a relatively slow weathering process.

Differences in sandstone properties between different layers contribute to the seepage characteristics of the sandstone in the LGB. Five to eight layers of sandstone have porosities and permeabilities higher than average, forming a sol-filtration layer that enhances seepage. Associated with the seepage damage to the main body of the Buddha, these layers correspond to the chest and shoulder of the Buddha, especially the chest, which further collects seepage due to the void between the restoration material and the main body, leading to hollowing and cracking.

The analysis of porosity reveals that there are significant differences in the degree of physical weathering in different strata of the LGB. Porosity, as an important physical property, directly affects the sensitivity of sandstones to environmental pressures, such as water intrusion and wet-dry cycles, thus accelerating the weathering process. In the strata of the chest and abdomen of the LGB (layers 5–8), higher porosity is associated with greater loss of ground colluvium, which leads to increased physical weathering. This is manifested in the increased connectivity of pores and microfractures in these strata, which promotes water infiltration and ionic leaching, both of which are detrimental to rock integrity. These findings emphasize the key role of porosity as a determinant of physical weathering intensity and highlight the need to integrate physical parameters and chemical indices for a comprehensive assessment of weathering mechanisms.

Soluble salt contents

In the weathering of stone relics, the destructive effect of soluble salts on cultural relics is very obvious, especially their repeated crystallization and dissolution of immovable cultural relics, which causes damage that is difficult to repair. Soluble salt damage to stone relics often involves two processes. First, soluble salts dissolve with the migration of water flow, which results in the loss of some of the rock material, dissolution of the cement, pore development, and loosening of the rock particles, which is known as the leaching effect. In the other process, soluble salts accumulate where water evaporates, resulting in salt crystallization, which in turn causes phenomena such as rock pulverization, surface spalling, and internal swelling and cracking.

In terms of the distributions of the soluble salt contents and conductivities, the soluble salt contents of layers 1–3 are significantly greater than those of the other layers, indicating that these layers are prone to accumulate nitrate and chloride salts. Although the salt contents are high, owing to the local humid climate, salts are usually dissolved in groundwater in ionic form, and the degree of crystallization is not significant. This contrasts with the destructive effect of salt crystallization on artifacts in arid environments, reflecting the complexity of weathering mechanisms.

The LGB abdomen consists of 1–3 layers, while the chest consists of 4–6 layers. The chest rocks have coarse grains and well-developed intergranular pores, allowing water to easily flow through, which reduces soluble salt content with minimal fluctuation. In contrast, the soluble salt content in layers 1–3 of the abdomen is significantly higher and more variable, indicating less developed pore space and greater ion retention. Layers with smaller pores show weak percolation, leading to higher ion concentrations, while larger pores in other layers promote percolation, reducing ion concentrations. This also supports the growth of vegetation in layers 6–8, and 10. In unvegetated layers 1–3 and 9, soluble salt content increases in the middle and decreases at the edges, indicating that the junctions are areas of salt loss. This further validates the interlayer division and suggests that salts in the bedrock center are less prone to loss or enrichment by water evaporation. Future studies should focus on layers with dense vegetation to better understand and control the seepage process.

Preventive measures and conservation strategies

To mitigate weathering damage to the LGB, the following preventive measures are recommended based on the findings of this study and previous research: enhance drainage systems and apply water-repellent coatings to reduce water infiltration; regularly clean surfaces and use eco-friendly biocides to control biological growth54,55; establish long-term environmental monitoring to track temperature, humidity, and pollution levels; apply microbial-induced carbonate precipitation to strengthen the sandstone while preserving its appearance56,57; and use desalination treatments, like poulticing, to remove harmful soluble salts and stabilize the stone’s pore structure58,59.

Conclusion

This study systematically analyzes the weathering variations of the LGB from both macro- and microscopic perspectives and provides a quantifiable methodology for studying the weathering of large open-air stone relics. Through stratified sampling and multidisciplinary testing of the strata of the LGB, the following conclusions were drawn:

  1. 1.

    The LGB is affected mainly by chemical weathering, and there are significant differences in the degree of weathering of each layer. The indicators of cement loss and chemical weathering intensity indicate that the weathering of layers 5–10 is more severe than that of the other layers, especially in terms of the loss of cement at the sandstone surface, which leads to structural damage and increased weathering.

  2. 2.

    The Cl contents of the different layers indicate that the water contents of the sandstones in the different layers vary significantly. Layers 5 and 6 have relatively high Cl contents, which indicates that their water contents are relatively high, seepage damage is severe, and the degree of weathering is relatively high.

  3. 3.

    The pore sizes and distributions of the sandstones in different layers differ significantly, affecting the water permeability and degree of weathering of each layer. Layers 5–8 have high proportions of large pores and dense distributions, which provide good channels for water seepage and aggravate the loss of colloids and pore expansion, thus resulting in high degrees of weathering.

  4. 4.

    The different concentrations and conductivities of the soluble salt ions in each layer lead to differences in weathering. Layers 1–3 have relatively high soluble salt contents and conductivities, indicating that these layers are prone to salt accumulation. In the three-river confluence environment where the LGB is located, salt crystallization is not significant during the rainy season, but salt crystallization may accelerate sandstone weathering during the dry season.

These conclusions provide a scientific basis for the study of the weathering mechanism of the LGB and its conservation, as well as a reference for weathering studies of other large open-air stone relics.