Extraction of Deterioration and Analysis of Vegetation Impact Effects on the South Palace Wall of Weiyang Palace

Extraction of Deterioration and Analysis of Vegetation Impact Effects on the South Palace Wall of Weiyang Palace | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Extraction of Deterioration and Analysis of Vegetation Impact Effects on the South Palace Wall of Weiyang Palace Sheng Gao, Liang Tao, Fulong Chen, Xiaochen Zhou, Pilong Shi, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4568335/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 18 Oct, 2024 Read the published version in npj Heritage Science → Version 1 posted 12 You are reading this latest preprint version Abstract Weiyang Palace, as the royal palace of the Western Han Dynasty, is a part of the Silk Roads: the Routes Network of Chang'an-Tianshan Corridor on the World Heritage list. The south palace wall of Weiyang Palace is a well-preserved earthen site within the palace, but it is undergoing continuous deterioration due to the influence of vegetation and external environmental factors. This study pioneers the integration of high-resolution three-dimensional LiDAR scanning with multi-source data analysis, including unprecedented on-site botanical surveys, to elucidate the nuanced impacts of different vegetation types on the structural integrity of the south palace wall. Through contour line analysis and facade grid analysis, we extracted the deterioration locations of typical sections of the earthen sites. And we classified the overlying vegetation types on the wall using an object-oriented classification algorithm. Our findings reveal a complex interaction between vegetation and earthen structures: paper mulberry exhibits protective qualities against erosion, while ziziphus jujuba significantly exacerbates structural vulnerabilities by inducing cracks. Methods employed in this study for extracting earthen site deterioration and combining multi-source spatial data analysis can serve as a technical application model for monitoring and analyzing the driving forces of surface earthen sites along the entire Silk Road network, thereby better guiding the conservation of earthen sites. Weiyang Palace earthen heritage LiDAR vegetation spatial technology Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 1 Introduction Cultural heritage represents a vital asset to humanity, bearing the weight of history and playing a crucial role in sustainable development by substantially contributing to the socio-economic enhancement of local communities [ 1 – 3 ]. However, due to climatic conditions and human activities, both cultural heritage and its surrounding environment are increasingly facing deterioration [ 4 , 5 ]. To address this, the United Nations has proposed the 2030 Agenda for Sustainable Development Goals (SDGs), which specifically mentions the need to further protect and safeguard cultural heritage in SDG 11.4 [ 6 , 7 ]. Cultural heritage encompasses a variety of categories, including but not limited to ancient ruins, stone carvings, and cultural landscapes. Each type of cultural heritage has its unique characteristics, presenting technical challenges in the preservation of various kinds of heritage [ 4 ]. Consequently, there is a need for universally applicable technical methods tailored to specific types of heritage, which can serve as benchmarks for heritage management practices. [ 8 , 9 ]. Earthen heritage constitutes a significant category in the World Heritage list, accounting for 10% of the total entries [ 10 ]. During the Central Plains Dynasties in China, rammed earth technology was widely used in large constructions like the Great Wall [ 11 ]. Most earthen sites, having been exposed to natural environmental factors such as wind, rain, and temperature variations over long periods, are progressively deteriorating [ 5 ]. Numerous studies have attempted to reveal the impacts of wind erosion and rain wash on these sites and to identify effective protective measures [ 12 , 13 ]. However, constrained by technological limitations, previous studies primarily concentrated on small-scale extraction of deteriorated areas in earthen heritage, primarily employing manual methods [ 14 – 16 ]. Such an approach, being labor-intensive and seldom supplemented by follow-up surveys, proves detrimental to the sustainable development of heritage sites. In recent years, the advancement of spatial technology and remote sensing has made it possible to automatically and repeatedly monitor and protect cultural heritage [ 17 – 21 ]. Among these technologies, LiDAR stands out as it can non-contact acquire information about earthen sites using lasers, and its penetrative ability minimizes environmental interference during operation [ 22 , 23 ]. Freeland et al. [ 24 ] combined LiDAR and automated feature extraction techniques to extract the earthworks in the Kingdom of Tonga; Wang et al. [ 25 ] extract ancient city wall by deep learning from LiDAR data. Despite these advancements, the primary application of LiDAR has been in recording and identifying earthen sites, with less emphasis on detecting and analyzing deterioration. In addition, previous research scenarios on earthen sites have also been inadequate. Most earthen sites are located in arid or semi-arid areas, where they are directly exposed to environmental factors, with research focusing on the impacts of wind and rain [ 5 , 10 , 26 – 29 ]. However, some earthen heritage lies in semi-humid and semi-dry regions, such as Xi'an, the eastern starting point of the Silk Road [ 30 ]. In these areas, there are earthen sites where vegetation can grow. While it is generally accepted that vegetation mitigates the effects of environmental factors, the direct impact of the vegetation itself has been insufficiently explored [ 31 , 32 ]. Moreno et al. [ 33 ] suggest that the impact of vegetation on earthen sites varies with the type of vegetation, being either positive or negative. Based on these research gaps, this study employs the south palace wall of Weiyang Palace as a case study, proposing a spatial analysis method that integrates the locations of wall deterioration with the types of overlying vegetation to assess the impact of vegetation on rammed earth walls. This method utilizes LiDAR data to evaluate wall collapses and on-site data collection to document wall cracks, alongside optical image data for vegetation classification. By correlating deterioration locations with vegetation distribution, this research elucidates the varying impacts of different vegetation types on the southern palace wall, informing future preservation strategies. 2 Materials and data 2.1 The south palace wall of Weiyang Palace Weiyang Palace, once the royal palace of the Western Han Dynasty, is situated in Xi'an, Shaanxi Province, China, with geographic coordinates at 34°18′16″N and 108°51′26″E, covering a total area of 6.1 square kilometers (Fig. 1 ). Recognized as the eastern starting point of the Silk Road, it was designated a part of the Silk Roads: the Routes Network of Chang'an-Tianshan Corridor on the World Heritage list in 2014. The layout of the Weiyang Palace is characterized by palace walls, city walls, and archaeological remains of buildings. To date, archaeologists have identified 10 palace wall sites, with the south palace wall, particularly its western section, being relatively well-preserved. The south palace wall, constructed using the rammed earth technique, is exposed to various environmental factors such as humidity, temperature fluctuations, salinization, wind erosion, and rain wash, compounded by human activities. These elements contribute to widespread deterioration, including cracking, powdering, erosion, and collapses. Vegetation on the wall also plays a significant role, with the canopy offering some protection against wind and rain, while the roots can destabilize the structure. Consequently, extracting the wall deterioration and understanding the interaction between vegetation and the wall are critically important for the conservation of the south palace wall. After conducting field surveys, we discovered that the south palace wall of Weiyang Palace hosts various types of vegetation, including paper mulberry, ziziphus jujuba, bromus, paulownia tomentosa, ailanthus altissima, and robinia pseudoacacia. Paper mulberry and ziziphus jujuba are the predominant species, whereas the others occur in significantly smaller quantities. Therefore, this study primarily focuses on the impact of paper mulberry and ziziphus jujuba on the deterioration of the wall. Additionally, the types of deterioration observed are numerous; this study categorizes them into small-scale deterioration, such as cracks, and large-scale deterioration, such as collapses. 2.2 Data description We utilized the DJI M300 drone to design flight paths and captured a series of overlapping optical images, achieving a lateral overlap rate of 70% and a longitudinal overlap rate of 50%. These optical images were registered and stitched together using DJI TERRA software, producing an orthophoto of the southern palace wall with a resolution of 0.17 m (Fig. 2 a). Additionally, the drone was equipped with a Livox LiDAR sensor to conduct low-altitude flights over the wall. By controlling the drone to direct laser pulses from both the top and the sides onto the wall, we obtained detailed information about the wall, especially its vertical facades (Fig. 2 b). The scale of the cracks in the wall is too minute for existing spatial technologies to effectively extract. Therefore, data regarding these cracks were obtained through field surveys: we measured the location, length, and width of the cracks on the vertical facades of the wall on-site, and recorded the growth of root systems within these cracks. Subsequently, we digitized these records into spatial data with location and attribute information. For large-scale deterioration, such as collapses, we utilized LiDAR point cloud to pinpoint the locations of the collapses and verified the accuracy using on-site recorded positions and photographs of the collapses. 3 Methods This section introduces a combined analysis method using optical image and LiDAR data to study the impact of vegetation on ancient sites (Fig. 3 ). We initiated by digitizing the wall's three-dimensional (3D) structure, followed by delineating the cracks at their corresponding locations on the wall model. Additionally, we employed the innovative application of contour line analysis and facade grid analysis to identify the collapse locations of the wall. For the optical image data captured by the drone, we conducted multi-scale segmentation to create image objects, and then calculated the internal texture features of each object. These texture features, along with the original image bands, were fed into a random forest classifier for classification, resulting in a classified map of the vegetation types covering the south palace wall. Finally, by integrating the results from the extraction of deterioration locations and the classification of vegetation, we conducted an analysis of how different vegetation types on the wall influence its deterioration. 3.1 Wall deterioration detection Firstly, we applied top-down ground filtering methods to the LiDAR point cloud data to remove the vegetation point cloud, retaining only the ground point data. In this process, we employed a cloth simulation filter to separate the vegetation point cloud from the point cloud of the wall itself. The cloth simulation filter works by simulating the physical process of cloth draping over an object to separate ground and non-ground points [ 34 ]. This process starts by inverting the point cloud, and then determining the final shape of the cloth through analyzing the interaction between the nodes of the cloth and the corresponding LiDAR points, with the final shape of the cloth representing the ground points. Once the ground points were obtained, we created a multipatch feature in ArcGIS software based on this ground point data, resulting in a spatially referenced 3D model. We then mapped the field-recorded locations, lengths, and trends of wall cracks onto this model, completing the spatial digitization of the wall cracks. To digitally identify collapse locations on the wall, we generated contour lines on the 3D model. These contour lines were segmented at approximately 2-meter intervals horizontally to form a series of curved segments. We calculated the curvature of these segments and categorized them by different colors based on their curvature, which was determined by dividing the length of each contour segment by the straight-line distance between its endpoints. In the color-graded map of contour line curvature, areas of the wall collapse were identified as regions where high-curvature contour lines clustered and concaved towards the inside of the wall. Following this principle, we recorded the identified collapse locations on the wall. In the initial top-down ground point filtering process, we considered the ground point cloud data as representing the wall itself, and used contour line analysis to extract collapses observed in the vertical direction. However, collapses such as partial indentations of the wall facade were not observable from a vertical perspective. To address this issue, Cai et al. [ 35 ] proposed a method involving the horizontal placement of the facade, followed by the use of existing ground filtering algorithms to separate the wall surface and protrusions. In this study, we manually separated the point cloud data containing the wall facade from original point cloud data, rotated it 90 degrees to horizontally place the facade, and then applied the cloth simulation filter to obtain the point cloud data of the wall facade. Based on this facade point cloud data, we located areas of local indentation, thus identifying collapse locations on the wall from a horizontal perspective. 3.2 Vegetation classification We employed an object-oriented approach for vegetation classification, a rapidly emerging technique in Geographic Information System (GIS) and archaeology, which provides a more accurate representation of geographical objects compared to single pixels [ 36 , 37 ]. Initially, the drone's optical images were subjected to multi-scale segmentation in eCognition software. The results of this segmentation were influenced by the scale of segmentation, shape, and compactness parameters. Based on preliminary research and subsequent experiments, we set the segmentation scale to 50, the shape parameter to 0.5, and the compactness to 0.8 [ 38 ]. After obtaining image objects, using only the original red, green, and blue bands of the drone's image as classification features was insufficient. Therefore, we calculated the internal texture features of these objects as supplementary features to improve the classification results. We constructed a gray-level co-occurrence matrix (GLCM) for the pixels inside the image objects. GLCM is a tabulation of the frequency of different combinations of pixel gray levels [ 39 ]. Statistical measures derived from the GLCM, such as mean, dissimilarity, second-order moment, contrast, correlation, variance, homogeneity, and entropy, were used to effectively enhance image classification [ 40 ]. Finally, we constructed a random forest classification model using these features to obtain the vegetation classification results. The random forest model achieves robust classification by combining multiple weak classifiers. These weak classifiers were built using decision tree algorithms, with the Gini index serving as the criterion for constructing the decision tree nodes [ 41 ]. The Gini index is calculated as follows: $$Gini=1-{\sum }_{i=0}^{I}{{p}_{i}}^{2}$$ 1 Where \(Gini\) is the Gini index, \(I\) represents the number of classes, \(i\) denotes the class index, and \({p}_{i}\) is the probability of occurrence of class \(i\) . A smaller Gini index indicates a higher purity of the sample set. Therefore, in constructing each tree node within the random forest classifier, the feature that results in the largest decrease in the Gini index is chosen for splitting, continuing until a leaf node is reached. 3.3 Analysis of the impact effects of vegetation on wall deterioration We analyze the overlapping digital data products of vegetation types, cracks, and collapses on the wall to establish their spatial interrelationships. This analysis involves conducting spatial statistical evaluations to digitally represent and quantitatively assess how vegetation growth and its types are spatially correlated with the deterioration patterns of the wall. This approach enables us to understand how different types of vegetation correlate with and potentially contribute to collapses of the wall. Additionally, based on field survey data, which includes measurements of the lengths and widths of cracks as well as the conditions of root system growth within them, we further identified the impact of vegetation on wall cracks through statistical analysis. 4 Result 4.1 Extraction of Wall Deterioration The south palace wall of Weiyang Palace is divided into eastern and western sections, separated by a road. According to the 3D model (Fig. 4 a), the model of the eastern section of the wall is uneven and almost lacks the wall's original and regular shape. Upon field inspection, it was observed that the eastern section suffered significant damage (Fig. 4 b). Additionally, the presence of large trees in this area impeded the penetration of the LiDAR signal. In contrast, the 3D model of the western section more clearly depicts the wall’s overall structure (Fig. 4 c). Consequently, our subsequent research and analysis focused on the western section. We established the starting point of the western section of the wall as the origin, with westward direction serving as the positive direction for positioning wall deterioration. Based on the cracks recorded during field surveys, we marked these cracks on the corresponding locations of the wall model. Through field surveys and analyses between orthophoto and LiDAR point cloud data, we found that the western section of the wall experienced significant collapses from 130 meters to 740 meters, where the collapsed earth formed slopes. The vegetation growth on these slopes was notably faster than on the rammed earth walls, leading to the wall's sides from 130 meters to 740 meters being obscured by large paper mulberry trees. These large trees not only made field sampling challenging but also significantly obstructed the LiDAR signal. For these reasons, the extraction of collapses in this study excluded the part from 130 meters to 740 meters. After excluding the area from 130 meters to 750 meters, we divided the remaining areas into three parts for ease of display. Part 1 is from 50 meters to 130meters; Part 2 is from 750 meters to 1050 meters; Part 3 is from 1200 meters to 1300 meters. On the facade of the western section of the wall, we generated contour lines at 0.2-meter elevation intervals and segmented these lines at a horizontal distance of 2 meters, subsequently calculating the curvature of these segments and applying color grading based on their curvature. In the color-graded curvature map of the contour lines, the areas of wall collapse are indicated by the clustering of high-curvature contour lines that concave inward towards the wall. Based on this criterion, we identified the areas of collapse on the wall. Additionally, the protruding paper mulberry trees on the side of the wall caused the contour lines to first bulge outward and then concave inward. This necessitated a comparison with the LiDAR point cloud data to manually exclude any misidentified collapse areas. Ultimately, using the contour line method, we identified a total of 29 collapse locations., as illustrated in Fig. 5 . Based on the 3D model, we manually extracted the point cloud data of the facade of the wall, and after rotating it by 90 degrees, we re-applied the cloth simulation filter. Utilizing the filtered point cloud data, we constructed a grid model and visually interpreted areas of local indentation, identifying them as horizontal collapse locations on the wall. In total, we additionally found 10 collapse locations (Fig. 6 ). Through the analysis of contour lines and the facade grid, we identified a total of 39 collapse locations on the wall. We physically photographed and recorded 20 instances of wall collapse on-site. Upon detailed comparison with the digitally identified collapse locations, 18 out of the 20 recorded instances matched those detected by digital methods (Fig. 7 ). The detection accuracy for the collapse of Weiyang Palace's south palace wall was determined to be 90%. 4.2 Vegetation classification To simplify the research, we limited our classification targets to ziziphus jujuba, paper mulberry, and 'others' categories. The 'others' category includes bare earth and less frequent vegetation types like bromus. After performing multi-scale segmentation, we obtained 179,202 image objects, delineating 1,075 patches as paper mulberry samples, 336 as ziziphus jujuba samples, and 186 as samples of the 'others' category. We divided the sample set into a training set and a test set in 7:3 ratio and evaluated the classification model on the test set. We used precision, recall and f1-score as metrics (Table 1 ). The final results of the vegetation types covering the wall are as presented in Fig. 8 . The final classification ratios for paper mulberry, ziziphus jujuba, and 'other' were 0.78, 0.07, and 0.14, respectively. Paper mulberry and ziziphus jujuba are in a competitive state on the wall, with paper mulberry occupying a dominant position. Table 1 Accuracy of vegetation classification. Category Precision Recall F1-score Paper mulberry 90.2% 97.5% 93.8% Ziziphus jujuba 82.1% 67.6% 74.2% Others 84.4% 71.7% 77.6% 4.3 Analysis of the impact effects of vegetation on wall deterioration Combining the collapse locations extracted in Section 4.1 with the vegetation classification results from Section 4.2 for an overlay analysis, we found that among the identified collapses, 75.9% were located under paper mulberry, 10.3% under ziziphus jujuba, and 13.8% were in areas without vegetation coverage. Comparing this with the overall proportions of vegetation types on the wall facade (paper mulberry: ziziphus jujuba: others = 0.78: 0.07: 0.14), it appears that overlying vegetation type does not directly influence the probability of collapse occurrence. We conducted an overlay analysis of the wall cracks' attributes in conjunction with the types of vegetation covering the wall. Figure 9 displays the locations, lengths, and widths of the cracks on the wall and the types of vegetation above them. It is evident that the cracks without vegetation above them are significantly larger in both length and width than those with vegetation, and these cracks are primarily concentrated in the 750 to 800 meters area. Based on our field investigations, we discovered that this particular section had undergone artificial treatment, resulting in the absence of vegetation growth above it. Consequently, due to the erosion caused by rainwater, this area experienced a significant number of large cracks. In the other cases, cracks under paper mulberry averaged 0.95 m in length and 3.37 cm in width, while those under ziziphus jujuba averaged 1.17 m in length and 1.41 cm in width, indicating that cracks under paper mulberry tend to be wider, whereas those under ziziphus jujuba are longer. Additionally, we noticed that 32% of the cracks had ziziphus jujuba growing above them, a proportion much higher than its occurrence on the wall (7%), suggesting a higher likelihood of crack formation in areas with ziziphus jujuba. During field investigations, we discovered that 62% of the cracks had vegetation roots growing directly within them; of these, 37% were from paper mulberry and 63% from Ziziphus jujuba. This distribution suggests that Ziziphus jujuba plays a significant role in the formation of wall cracks. Further statistical analysis reveals that, compared to conditions without roots, the roots of paper Mulberry and ziziphus jujuba simultaneously increase the average length and width of cracks. Specifically, paper mulberry roots significantly enhance the width, while ziziphus jujuba roots notably increase the length (Table 2 ). Table 2 Average lengths and widths of cracks under different roots conditions Roots conditions Average width (cm) Average length (m) Presence of paper mulberry roots 3.68 1.01 Presence of ziziphus jujuba roots 1.98 1.25 Absence of Roots 1.33 0.88 The analysis above indicates that the occurrence of collapses on the wall is not influenced by the type of vegetation, suggesting that vegetation type is not a significant factor in these large-scale phenomena. In contrast, the development of cracks in the wall is highly dependent on the type of vegetation present. Paper mulberry is associated with wider cracks, whereas ziziphus jujuba is linked to longer cracks, with areas hosting ziziphus jujuba being particularly susceptible to crack formation. Importantly, despite the adverse effects of Paper Mulberry on crack severity, it still offers better protection than exposing the wall directly to external environmental conditions. 5 Discussion 5.1 Advancement and limitation Spatial technology is increasingly being applied in the field of cultural heritage [ 20 , 42 ]. However, its use is often limited to digitalizing and demonstrating archaeological sites [ 43 ], with only a minority of studies employing intelligent algorithms to extract useful information for heritage monitoring and protection from spatial data [ 44 , 45 ]. The deterioration of earthen sites results from a combination of multiple factors [ 46 ]. Therefore, alongside using spatial technology to monitor these sites, it is crucial to investigate the underlying forces driving their deterioration to inform effective protection strategies. Regrettably, research in these areas tends to be conducted in isolation: studies utilizing spatial technology typically concentrate on data acquisition and site monitoring, whereas those analyzing deterioration mechanisms focus primarily on these processes [ 47 , 48 ], often using small samples in laboratory settings, which do not provide a holistic understanding of the sites or practical conservation solutions. This study integrated multi-source spatial data to monitor and analyze the deterioration and driving forces of the Han Dynasty Weiyang Palace's south palace wall. Using the cloth simulation filter, especially after the 90-degree rotation, we developed three-dimensional data of the wall itself and portrayed its deterioration through a series of spatial analysis methods. These methods are significantly relevant for data processing and information extraction for earthen sites like the wall. By combining intelligent algorithms with high-resolution drone image, we classified the overlying vegetation on the wall. Through overlay and statistical analysis, we identified the impact of the vegetation on the deterioration of the wall. The approach of integrating multi-source data for overlay analysis in this study can be extended to similar research on earthen sites. Beyond the impact of vegetation, sensors carried by drones can infer temperature, humidity and so on [ 49 , 50 ], which can be overlaid with the extracted locations of wall deterioration for further analysis, identifying driving forces and formulating scientific conservation recommendations. Our research has developed a method for analyzing the driving forces affecting earthen sites using spatial technology, but it has some limitations. With the advancement of deep learning algorithms [ 51 ], it has become possible to automatically extract wall deterioration from LiDAR point cloud data or 3D models. However, at this stage, our study only includes the case of Weiyang Palace's south palace wall, which is insufficient to build an adequate sample set for deep learning. In reality, the Silk Road is home to a multitude of earthen sites. Incorporating these sites into comprehensive research could significantly enhance the potential for automated and accurate identification of site deterioration. Moreover, during our research process, we found that the scale of the cracks in the wall is too small for detection by the data from airborne platforms, so our data on cracks was manually recorded. In winter, when vegetation leaves wither, close-range photogrammetric techniques can be used to identify cracks [ 52 ]. The identified cracks can be digitally mapped onto the surface of the wall, allowing for direct calculation of their lengths and widths. This is also a direction for our future research. 5.2 Suggestions for the protection of the south palace wall The deterioration of earthen sites in natural environments is complex, making it challenging to completely prevent deterioration [ 53 ]. Our experimental data shows that the south palace wall of Weiyang Palace is currently undergoing gradual deterioration, with collapses occurring regardless of whether vegetation is present or not. Therefore, it is necessary to use appropriate materials to reinforce the structure of the wall to prevent further extensive collapses [ 54 ]. As for the cracks, ziziphus jujuba significantly contributes to the formation of cracks in the wall, while paper mulberry, although slightly exacerbating the cracks, offers better protection than exposing the wall directly to external environmental conditions (Fig. 10 ). On the wall, paper mulberry and ziziphus jujuba are in competition, with paper mulberry in a dominant position. Considering these factors, we suggest trying chemical methods to eradicate ziziphus jujuba to prevent the formation of more cracks in the wall. Until an appropriate man-made shelter is created for the wall, the paper mulberry trees on it should be preserved to avoid direct erosion by wind and rain. 6 Conclusion This study has offered comprehensive insights into the deterioration of the south palace wall of Weiyang Palace. By integrating multi-source spatial data, including LiDAR and high-resolution drone image, we have successfully identified the deterioration along the wall and analyzed the impact effects of vegetation on wall deterioration. Utilizing innovative methods such as contour line analysis and facade grid analysis, we effectively pinpointed the locations of deterioration along the wall. These techniques enabled a detailed spatial representation of the wall's condition, demonstrating the value of advanced spatial analysis in cultural heritage preservation. And our research employed an object-oriented approach, integrating texture features to classify the types of overlying vegetation on the wall. This classification played a crucial role in understanding the interaction between vegetation and the wall's deterioration. Finally, we discovered that paper mulberry and ziziphus jujuba compete for dominance on the wall and their impact on the wall's structure is distinct. Paper mulberry, despite slightly exacerbating the cracks, offers a level of protection against direct environmental exposure, whereas ziziphus jujuba contributes significantly to the formation of cracks. The study underscores the complexity of conserving earthen sites, especially those continuously exposed to natural elements and human interventions. It is clear that a one-size-fits-all approach is not feasible for the preservation of such earthen sites. Our research provides a case study on the detailed analysis of the driving forces behind wall deterioration using spatial technology and intelligent algorithms. This approach can guide more precise conservation strategies for earthen sites. In conclusion, our study not only advances the understanding of biotic factors affecting earthen heritage deterioration but also sets a precedent for the application of integrated spatial and botanical analysis in the conservation of cultural heritage worldwide. Declarations Acknowledgements We would like to express our sincere gratitude to the Xi'an Academy of Conservation and Archaeology for their generous support and funding. Authors’ contributions Conceptualization, L.T. and F.C.; methodology, S.G., F.C. and L.T.; software, S.G., P.S and Z.X; formal analysis, S.G. and F.C.; investigation, S.G., W.L., Y.L, H.L and C.C.; resources, X.Y. and W.L.; data curation, S.G. and X.Z; writing—original draft preparation, S.G.; writing—review and editing, S.G. L,T and F.C.; visualization, S.G., W.Z., and M.Z.; supervision, F.C. and L.T.; project administration, F.C. and L.T.; funding acquisition, F.C. and L.T. All authors have read and agreed to the published version of the manuscript. Funding This work was jointly support by the National Natural Science Foundation of China (NSFC) (grant no. 42271327) and 'Silk Road: Digital Preservation of Cultural Heritage Sites in the Xi'an Section of the Chang'an-Tianshan Corridor Network' project of the Xi'an Academy of Conservation and Archaeology. Availability of data and materials Data sharing is not applicable to this article as no datasets were generated during the current study. Ethics approval and consent to participate Not applicable. Competing interests The authors declare that they have no competing interests. References Bowitz E, Ibenholt K. Economic impacts of cultural heritage – Research and perspectives. Journal of Cultural Heritage. 2009;10(1):1-8. Gravagnuolo A, Micheletti S, Bosone M. A Participatory Approach for “Circular” Adaptive Reuse of Cultural Heritage. Building a Heritage Community in Salerno, Italy. Sustainability. 2021;13(9):4812. Wang C, Fulong C, Wei Z, Huafen Y, Di WJNRSB. Sequential PSInSAR approach for the deformation monitoring of the Nanjing Ming Dynasty City Wall. National Remote Sensing Bulletin. 2022;25(12):2381-95. Wang X, Li H, Wang Y, Zhao X. Assessing climate risk related to precipitation on cultural heritage at the provincial level in China. Science of The Total Environment. 2022;835:155489. Shao M, Li L, Wang S, Wang E, Li Z. Deterioration mechanisms of building materials of Jiaohe ruins in China. Journal of Cultural Heritage. 2013;14(1):38-44. Petti L, Trillo C, Makore BN. Cultural Heritage and Sustainable Development Targets: A Possible Harmonisation? Insights from the European Perspective. Sustainability. 2020;12(3):926. Guo H, Chen F, Tang Y, Ding Y, Chen M, Zhou W, et al. Progress toward the sustainable development of world cultural heritage sites facing land-cover changes. The Innovation. 2023;4(5). Richards J, Wang Y, Orr S, Viles H. Finding Common Ground between United Kingdom Based and Chinese Approaches to Earthen Heritage Conservation. Sustainability. 2018;10(9):3086. Gutiérrez-Carrillo ML, Arizzi A. How to deal with the conservation of the archaeological remains of earthen defensive architecture: the case of Southeast Spain. Archaeological and Anthropological Sciences. 2021;13(8):131. Richards J, Zhao G, Zhang H, Viles H. A controlled field experiment to investigate the deterioration of earthen heritage by wind and rain. Heritage Science. 2019;7(1):1-13. Xie L, Wang D, Zhao H, Gao J, Gallo T. Architectural energetics for rammed-earth compaction in the context of Neolithic to early Bronze Age urban sites in Middle Yellow River Valley, China. Journal of Archaeological Science. 2021;126:105303. Rainer L. Deterioration and pathology of earthen architecture. Terra Literature Review. 2008:45. Fodde E, Khan MS. Affordable Monsoon Rain Mitigation Measures in the World Heritage Site of Moenjodaro, Pakistan. International Journal of Architectural Heritage. 2013;11(2):161-73. Parisi F, Asprone D, Fenu L, Prota A. Experimental characterization of Italian composite adobe bricks reinforced with straw fibers. Composite Structures. 2015;122:300-7. Du Y, Chen W, Cui K, Zhang K. Study on Damage Assessment of Earthen Sites of the Ming Great Wall in Qinghai Province Based on Fuzzy-AHP and AHP-TOPSIS. International Journal of Architectural Heritage. 2019;14(6):903-16. Zhang Y, Ye WM, Chen B, Chen YG, Ye B. Desiccation of NaCl-contaminated soil of earthen heritages in the Site of Yar City, northwest China. Applied Clay Science. 2016;124-125:1-10. Chen F, Lasaponara R, Masini N. An overview of satellite synthetic aperture radar remote sensing in archaeology: From site detection to monitoring. Journal of Cultural Heritage. 2017;23:5-11. Luo L, Wang X, Guo H, Lasaponara R, Zong X, Masini N, et al. Airborne and spaceborne remote sensing for archaeological and cultural heritage applications: A review of the century (1907–2017). Remote Sensing of Environment. 2019;232:111280. Remondino F. Heritage Recording and 3D Modeling with Photogrammetry and 3D Scanning. Remote Sensing. 2011;3(6):1104-38. Campiani A, Lingle A, Lercari N. Spatial analysis and heritage conservation: Leveraging 3-D data and GIS for monitoring earthen architecture. Journal of Cultural Heritage. 2019;39:166-76. Chen F, Guo H, Ma P, Tang Y, Wu F, Zhu M, et al. Sustainable development of World Cultural Heritage sites in China estimated from optical and SAR remotely sensed data. Remote Sensing of Environment. 2023;298:113838. Mallet C, Bretar F. Full-waveform topographic lidar: State-of-the-art. ISPRS Journal of Photogrammetry and Remote Sensing. 2009;64(1):1-16. Comer DC, Comer JA, Dumitru IA, Ayres WS, Levin MJ, Seikel KA, et al. Airborne LiDAR Reveals a Vast Archaeological Landscape at the Nan Madol World Heritage Site. Remote Sensing. 2019;11(18):2152. Freeland T, Heung B, Burley DV, Clark G, Knudby A. Automated feature extraction for prospection and analysis of monumental earthworks from aerial LiDAR in the Kingdom of Tonga. Journal of Archaeological Science. 2016;69:64-74. Wang S, Hu Q, Wang S, Ai M, Zhao P. Archaeological site segmentation of ancient city walls based on deep learning and LiDAR remote sensing. Journal of Cultural Heritage. 2024;66:117-31. Guo Q, Wang Y, Chen W, Pei Q, Sun M, Yang S, et al. Key Issues and Research Progress on the Deterioration Processes and Protection Technology of Earthen Sites under Multi-Field Coupling. Coatings. 2022;12(11):1677. Du Y, Chen W, Cui K, Gong S, Pu T, Fu X. A Model Characterizing Deterioration at Earthen Sites of the Ming Great Wall in Qinghai Province, China. Soil Mechanics and Foundation Engineering. 2017;53(6):426-34. Richards J, Mayaud J, Zhan H, Wu F, Bailey R, Viles H. Modelling the risk of deterioration at earthen heritage sites in drylands. Earth Surface Processes and Landforms. 2020;45(11):2401-16. Canuti P, Casagli N, Catani F, Fanti R. Hydrogeological hazard and risk in archaeological sites: some case studies in Italy. Journal of Cultural Heritage. 2000;1(2):117-25. Liu S, Huang S, Xie Y, Wang H, Huang Q, Leng G, et al. Spatial-temporal changes in vegetation cover in a typical semi-humid and semi-arid region in China: Changing patterns, causes and implications. Ecological Indicators. 2019;98:462-75. Richards J, Bailey R, Mayaud J, Viles H, Guo Q, Wang XJSR. Deterioration risk of dryland earthen heritage sites facing future climatic uncertainty. Scientific Reports. 2020;10(1):16419. Wolfe SA, Nickling WGJPipg. The protective role of sparse vegetation in wind erosion. Progress in physical geography. 1993;17(1):50-68. Moreno M, Ortiz P, Ortiz R. Analysis of the impact of green urban areas in historic fortified cities using Landsat historical series and Normalized Difference Indices. Scientific Reports. 2023;13(1):8982. Zhang W, Qi J, Wan P, Wang H, Xie D, Wang X, et al. An easy-to-use airborne LiDAR data filtering method based on cloth simulation. Remote sensing. 2016;8(6):501. Cai S, Zhang S, Zhang W, Fan H, Shao J, Yan G, et al. A General and Effective Method for Wall and Protrusion Separation from Facade Point Clouds. Journal of Remote Sensing. 2023;3:0069. Blaschke T, Hay GJ, Kelly M, Lang S, Hofmann P, Addink E, et al. Geographic object-based image analysis–towards a new paradigm. ISPRS journal of photogrammetry and remote sensing. 2014;87:180-91. Magnini L, Bettineschi C. Theory and practice for an object-based approach in archaeological remote sensing. Journal of Archaeological Science. 2019;107:10-22. Chen G, He Y, De Santis A, Li G, Cobb R, Meentemeyer RK. Assessing the impact of emerging forest disease on wildfire using Landsat and KOMPSAT-2 data. Remote Sensing of Environment. 2017;195:218-29. Haralick RM, Shanmugam K, Dinstein IH. Textural features for image classification. IEEE Transactions on systems, man, and cybernetics. 1973(6):610-21. Gadkari D. Image quality analysis using GLCM. 2004. Breiman L. Random forests. Machine learning. 2001;45:5-32. De Reu J, Plets G, Verhoeven G, De Smedt P, Bats M, Cherretté B, et al. Towards a three-dimensional cost-effective registration of the archaeological heritage. Journal of archaeological science. 2013;40(2):1108-21. Landeschi G, Nilsson B, Dell'Unto N. Assessing the damage of an archaeological site: New contributions from the combination of image-based 3D modelling techniques and GIS. Journal of Archaeological Science: Reports. 2016;10:431-40. Du Y, Chen W, Cui K, Zhang J, Chen Z, Zhang Q. Damage assessment of earthen sites of the Ming Great Wall in Qinghai Province: a comparison between Support Vector Machine (SVM) and BP Neural Network. Journal on Computing and Cultural Heritage (JOCCH). 2020;13(2):1-18. Lercari N. Monitoring earthen archaeological heritage using multi-temporal terrestrial laser scanning and surface change detection. Journal of Cultural Heritage. 2019;39:152-65. Richards J, Viles H, Guo Q. The importance of wind as a driver of earthen heritage deterioration in dryland environments. Geomorphology. 2020;369:107363. Richards J, Guo Q, Viles H, Wang Y, Zhang B, Zhang H. Moisture content and material density affects severity of frost damage in earthen heritage. Science of The Total Environment. 2022;819:153047. Richards J, Zhao G, Zhang H, Viles H. A controlled field experiment to investigate the deterioration of earthen heritage by wind and rain. Heritage Science. 2019;7:1-13. Frodella W, Elashvili M, Spizzichino D, Gigli G, Adikashvili L, Vacheishvili N, et al. Combining infrared thermography and uav digital photogrammetry for the protection and conservation of rupestrian cultural heritage sites in Georgia: A methodological application. Remote Sensing. 2020;12(5):892. Su T-C. Environmental risk mapping of physical cultural heritage using an unmanned aerial remote sensing system: A case study of the Huang-Wei monument in Kinmen, Taiwan. Journal of Cultural Heritage. 2019;39:140-51. Guo Y, Wang H, Hu Q, Liu H, Liu L, Bennamoun M. Deep learning for 3d point clouds: A survey. IEEE transactions on pattern analysis and machine intelligence. 2020;43(12):4338-64. Galantucci RA, Fatiguso F. Advanced damage detection techniques in historical buildings using digital photogrammetry and 3D surface anlysis. Journal of Cultural Heritage. 2019;36:51-62. Li L, Shao M, Wang S, Li Z. Preservation of earthen heritage sites on the Silk Road, northwest China from the impact of the environment. Environmental Earth Sciences. 2011;64:1625-39. Correia M, Guerrero L. Conservation of earthen building materials. The Encyclopedia of Archaeological Sciences. 2018:1-6. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 18 Oct, 2024 Read the published version in npj Heritage Science → Version 1 posted Editorial decision: Revision requested 18 Jul, 2024 Reviews received at journal 15 Jul, 2024 Reviews received at journal 12 Jul, 2024 Reviewers agreed at journal 27 Jun, 2024 Reviewers agreed at journal 26 Jun, 2024 Reviews received at journal 26 Jun, 2024 Reviewers agreed at journal 24 Jun, 2024 Reviewers agreed at journal 23 Jun, 2024 Reviewers invited by journal 23 Jun, 2024 Editor assigned by journal 19 Jun, 2024 Submission checks completed at journal 19 Jun, 2024 First submitted to journal 12 Jun, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4568335","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":319964378,"identity":"e01e8ab3-5dc1-45e2-a9be-db3bc95f1b36","order_by":0,"name":"Sheng Gao","email":"","orcid":"","institution":"Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Sheng","middleName":"","lastName":"Gao","suffix":""},{"id":319964379,"identity":"9a990ac2-e81e-4e7e-aad4-1c3f2c9fc6cf","order_by":1,"name":"Liang Tao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYHACxgMJQJKfmbGBOPU8QAzWItlMkhYQw+AAsa6yZz984MCDmjt2m48zN374wWCTL+/A/OwBXlt40hIOJBx7lrztMGOzZA9DmuXGA2zmBvgdlmNwIIHtcLLZYcY2ZgaGwwaGDTxsEni18L8Bavl3ONm4mWgtEkBbEtsO2xkwQ7XIMxDScuNZwoHEvsMJEmC/GKQZGDCzmeHVwt6ffPDhj2+H7fn7jz/88KPCxkC+vfkZXi0wkNgApoBBZXCYGPVAYA9nyTcQqWUUjIJRMApGDAAAzPdFwNg1m+IAAAAASUVORK5CYII=","orcid":"","institution":"Xi'an Academy of Conservation and Archaeology","correspondingAuthor":true,"prefix":"","firstName":"Liang","middleName":"","lastName":"Tao","suffix":""},{"id":319964381,"identity":"43fe82b9-548a-49a3-bb02-811e3d4b4eef","order_by":2,"name":"Fulong Chen","email":"","orcid":"","institution":"International Research Center of Big Data for Sustainable Development Goals","correspondingAuthor":false,"prefix":"","firstName":"Fulong","middleName":"","lastName":"Chen","suffix":""},{"id":319964382,"identity":"93ca7618-a75a-40fd-8c56-7b320bae7f1a","order_by":3,"name":"Xiaochen Zhou","email":"","orcid":"","institution":"Xi'an Academy of Conservation and Archaeology","correspondingAuthor":false,"prefix":"","firstName":"Xiaochen","middleName":"","lastName":"Zhou","suffix":""},{"id":319964385,"identity":"3d14478f-8986-4dd4-a9c0-5bd0ac870069","order_by":4,"name":"Pilong Shi","email":"","orcid":"","institution":"International Research Center of Big Data for Sustainable Development Goals","correspondingAuthor":false,"prefix":"","firstName":"Pilong","middleName":"","lastName":"Shi","suffix":""},{"id":319964387,"identity":"833fb6e7-fa3f-4646-a6a6-8229c1202e00","order_by":5,"name":"Xun Yao","email":"","orcid":"","institution":"Xi'an Academy of Conservation and Archaeology","correspondingAuthor":false,"prefix":"","firstName":"Xun","middleName":"","lastName":"Yao","suffix":""},{"id":319964391,"identity":"82f0deac-0c25-4a48-b8e2-bc2e9ca0aeb7","order_by":6,"name":"Meng Zhu","email":"","orcid":"","institution":"International Research Center of Big Data for Sustainable Development Goals","correspondingAuthor":false,"prefix":"","firstName":"Meng","middleName":"","lastName":"Zhu","suffix":""},{"id":319964396,"identity":"674893a1-2296-4724-83ba-6804021431f6","order_by":7,"name":"Wenbo Li","email":"","orcid":"","institution":"Xi'an Academy of Conservation and Archaeology","correspondingAuthor":false,"prefix":"","firstName":"Wenbo","middleName":"","lastName":"Li","suffix":""},{"id":319964399,"identity":"781bee71-b2b5-4898-9b46-0f40059c9cb7","order_by":8,"name":"Wei Zhou","email":"","orcid":"","institution":"International Research Center of Big Data for Sustainable Development Goals","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Zhou","suffix":""},{"id":319964401,"identity":"4ff91248-52a6-4998-a4e5-b50fdba0df40","order_by":9,"name":"Yansong Luo","email":"","orcid":"","institution":"Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yansong","middleName":"","lastName":"Luo","suffix":""},{"id":319964402,"identity":"e51981ee-082c-4cf0-b8ba-922263b335ca","order_by":10,"name":"Hongqiang Li","email":"","orcid":"","institution":"Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Hongqiang","middleName":"","lastName":"Li","suffix":""},{"id":319964403,"identity":"2efabd34-e4ba-4b25-a248-70faec272660","order_by":11,"name":"Caiyan Chen","email":"","orcid":"","institution":"Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Caiyan","middleName":"","lastName":"Chen","suffix":""},{"id":319964404,"identity":"2bc45bb4-6d58-4279-82d7-f5afd0eccb34","order_by":12,"name":"Xinru Zhang","email":"","orcid":"","institution":"Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xinru","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2024-06-12 07:36:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4568335/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4568335/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s40494-024-01485-x","type":"published","date":"2024-10-18T15:57:35+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":59698894,"identity":"30951b48-4ff6-4fca-aa06-91a795fb8856","added_by":"auto","created_at":"2024-07-05 03:53:48","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1176272,"visible":true,"origin":"","legend":"\u003cp\u003eOverview of Weiyang Palace and the location of the south palace wall.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/51e1d95531f651749a766563.png"},{"id":59698892,"identity":"a309ca83-9c52-4592-b743-9a149a689aaf","added_by":"auto","created_at":"2024-07-05 03:53:48","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2797383,"visible":true,"origin":"","legend":"\u003cp\u003e(a): Orthophoto of the south palace wall; (b): LiDAR point cloud of the south palace wall.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/e0ec5abdb98688018ad2f991.png"},{"id":59698893,"identity":"272850d7-fcc5-46ad-931c-54da8e83c6fa","added_by":"auto","created_at":"2024-07-05 03:53:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":211025,"visible":true,"origin":"","legend":"\u003cp\u003eIntegrated approach for analysis of the impact of vegetation types on wall deterioration\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/4808bee08e65cea262e589c4.png"},{"id":59699449,"identity":"9b78a4be-0b60-4738-8302-ca0f3c85ba1d","added_by":"auto","created_at":"2024-07-05 04:09:48","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1928556,"visible":true,"origin":"","legend":"\u003cp\u003e(a): 3D model of the eastern section of the wall. (b): Photograph of the eastern section of the wall. (c): 3D model of the western section of the wall and cracks on it.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/1c81a76c43e567745dc89bb6.png"},{"id":59698897,"identity":"18861cf3-2b7a-4170-ba04-66685ddb10e1","added_by":"auto","created_at":"2024-07-05 03:53:48","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":1058237,"visible":true,"origin":"","legend":"\u003cp\u003eCollapse locations identified by contour line analysis.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/4a16fa6606d2b3086a870611.png"},{"id":59698899,"identity":"24546484-76c3-4f78-a7e8-616b1e3625e0","added_by":"auto","created_at":"2024-07-05 03:53:48","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":1846936,"visible":true,"origin":"","legend":"\u003cp\u003eCollapse locations identified by façade grid analysis.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/e2f8f6a779114eb08948b60b.png"},{"id":59699214,"identity":"a92d5297-bca8-4109-a89f-f3bd209a5c8d","added_by":"auto","created_at":"2024-07-05 04:01:48","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":2913565,"visible":true,"origin":"","legend":"\u003cp\u003ePhotographs of collapses recorded on site and their corresponding digital identification numbers.\u003c/p\u003e","description":"","filename":"floatimage7.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/ff33ccf7cda7e30d70ca7273.png"},{"id":59698895,"identity":"780afa1d-c3af-49e1-9ef7-ef1791fb7234","added_by":"auto","created_at":"2024-07-05 03:53:48","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":303296,"visible":true,"origin":"","legend":"\u003cp\u003eResults of vegetation classification.\u003c/p\u003e","description":"","filename":"floatimage8.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/e609e2f3b5d056214fdc08ec.png"},{"id":59699216,"identity":"db621f95-92a7-430e-857d-925491751080","added_by":"auto","created_at":"2024-07-05 04:01:48","extension":"png","order_by":9,"title":"Figure 9","display":"","copyAsset":false,"role":"figure","size":171031,"visible":true,"origin":"","legend":"\u003cp\u003eLocations, lengths, and widths of cracks under different overlying vegetation type.\u003c/p\u003e","description":"","filename":"floatimage9.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/1c63dd0bfcd82428e609ec07.png"},{"id":59699217,"identity":"cce7aca0-df2e-4187-875c-4b055e5341e3","added_by":"auto","created_at":"2024-07-05 04:01:48","extension":"png","order_by":10,"title":"Figure 10","display":"","copyAsset":false,"role":"figure","size":2980709,"visible":true,"origin":"","legend":"\u003cp\u003eCracks in the sections of the wall directly exposed to the external environment.\u003c/p\u003e","description":"","filename":"floatimage10.png","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/c048bff4a5f8bedcb9dc5097.png"},{"id":67149005,"identity":"c8966105-52b9-4c62-888a-888c6ea33700","added_by":"auto","created_at":"2024-10-21 16:10:50","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":20836833,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4568335/v1/3bee50b5-8c0d-4b81-be48-1d90ae70426a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Extraction of Deterioration and Analysis of Vegetation Impact Effects on the South Palace Wall of Weiyang Palace","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eCultural heritage represents a vital asset to humanity, bearing the weight of history and playing a crucial role in sustainable development by substantially contributing to the socio-economic enhancement of local communities [\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. However, due to climatic conditions and human activities, both cultural heritage and its surrounding environment are increasingly facing deterioration [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. To address this, the United Nations has proposed the 2030 Agenda for Sustainable Development Goals (SDGs), which specifically mentions the need to further protect and safeguard cultural heritage in SDG 11.4 [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Cultural heritage encompasses a variety of categories, including but not limited to ancient ruins, stone carvings, and cultural landscapes. Each type of cultural heritage has its unique characteristics, presenting technical challenges in the preservation of various kinds of heritage [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Consequently, there is a need for universally applicable technical methods tailored to specific types of heritage, which can serve as benchmarks for heritage management practices. [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eEarthen heritage constitutes a significant category in the World Heritage list, accounting for 10% of the total entries [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. During the Central Plains Dynasties in China, rammed earth technology was widely used in large constructions like the Great Wall [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Most earthen sites, having been exposed to natural environmental factors such as wind, rain, and temperature variations over long periods, are progressively deteriorating [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Numerous studies have attempted to reveal the impacts of wind erosion and rain wash on these sites and to identify effective protective measures [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. However, constrained by technological limitations, previous studies primarily concentrated on small-scale extraction of deteriorated areas in earthen heritage, primarily employing manual methods [\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Such an approach, being labor-intensive and seldom supplemented by follow-up surveys, proves detrimental to the sustainable development of heritage sites. In recent years, the advancement of spatial technology and remote sensing has made it possible to automatically and repeatedly monitor and protect cultural heritage [\u003cspan additionalcitationids=\"CR18 CR19 CR20\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. Among these technologies, LiDAR stands out as it can non-contact acquire information about earthen sites using lasers, and its penetrative ability minimizes environmental interference during operation [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. Freeland et al. [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e] combined LiDAR and automated feature extraction techniques to extract the earthworks in the Kingdom of Tonga; Wang et al. [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] extract ancient city wall by deep learning from LiDAR data. Despite these advancements, the primary application of LiDAR has been in recording and identifying earthen sites, with less emphasis on detecting and analyzing deterioration.\u003c/p\u003e \u003cp\u003eIn addition, previous research scenarios on earthen sites have also been inadequate. Most earthen sites are located in arid or semi-arid areas, where they are directly exposed to environmental factors, with research focusing on the impacts of wind and rain [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan additionalcitationids=\"CR27 CR28\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. However, some earthen heritage lies in semi-humid and semi-dry regions, such as Xi'an, the eastern starting point of the Silk Road [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. In these areas, there are earthen sites where vegetation can grow. While it is generally accepted that vegetation mitigates the effects of environmental factors, the direct impact of the vegetation itself has been insufficiently explored [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. Moreno et al. [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] suggest that the impact of vegetation on earthen sites varies with the type of vegetation, being either positive or negative.\u003c/p\u003e \u003cp\u003eBased on these research gaps, this study employs the south palace wall of Weiyang Palace as a case study, proposing a spatial analysis method that integrates the locations of wall deterioration with the types of overlying vegetation to assess the impact of vegetation on rammed earth walls. This method utilizes LiDAR data to evaluate wall collapses and on-site data collection to document wall cracks, alongside optical image data for vegetation classification. By correlating deterioration locations with vegetation distribution, this research elucidates the varying impacts of different vegetation types on the southern palace wall, informing future preservation strategies.\u003c/p\u003e"},{"header":"2 Materials and data","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 The south palace wall of Weiyang Palace\u003c/h2\u003e \u003cp\u003eWeiyang Palace, once the royal palace of the Western Han Dynasty, is situated in Xi'an, Shaanxi Province, China, with geographic coordinates at 34\u0026deg;18\u0026prime;16\u0026Prime;N and 108\u0026deg;51\u0026prime;26\u0026Prime;E, covering a total area of 6.1 square kilometers (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Recognized as the eastern starting point of the Silk Road, it was designated a part of the Silk Roads: the Routes Network of Chang'an-Tianshan Corridor on the World Heritage list in 2014. The layout of the Weiyang Palace is characterized by palace walls, city walls, and archaeological remains of buildings. To date, archaeologists have identified 10 palace wall sites, with the south palace wall, particularly its western section, being relatively well-preserved.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe south palace wall, constructed using the rammed earth technique, is exposed to various environmental factors such as humidity, temperature fluctuations, salinization, wind erosion, and rain wash, compounded by human activities. These elements contribute to widespread deterioration, including cracking, powdering, erosion, and collapses. Vegetation on the wall also plays a significant role, with the canopy offering some protection against wind and rain, while the roots can destabilize the structure. Consequently, extracting the wall deterioration and understanding the interaction between vegetation and the wall are critically important for the conservation of the south palace wall. After conducting field surveys, we discovered that the south palace wall of Weiyang Palace hosts various types of vegetation, including paper mulberry, ziziphus jujuba, bromus, paulownia tomentosa, ailanthus altissima, and robinia pseudoacacia. Paper mulberry and ziziphus jujuba are the predominant species, whereas the others occur in significantly smaller quantities. Therefore, this study primarily focuses on the impact of paper mulberry and ziziphus jujuba on the deterioration of the wall. Additionally, the types of deterioration observed are numerous; this study categorizes them into small-scale deterioration, such as cracks, and large-scale deterioration, such as collapses.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Data description\u003c/h2\u003e \u003cp\u003eWe utilized the DJI M300 drone to design flight paths and captured a series of overlapping optical images, achieving a lateral overlap rate of 70% and a longitudinal overlap rate of 50%. These optical images were registered and stitched together using DJI TERRA software, producing an orthophoto of the southern palace wall with a resolution of 0.17 m (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). Additionally, the drone was equipped with a Livox LiDAR sensor to conduct low-altitude flights over the wall. By controlling the drone to direct laser pulses from both the top and the sides onto the wall, we obtained detailed information about the wall, especially its vertical facades (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThe scale of the cracks in the wall is too minute for existing spatial technologies to effectively extract. Therefore, data regarding these cracks were obtained through field surveys: we measured the location, length, and width of the cracks on the vertical facades of the wall on-site, and recorded the growth of root systems within these cracks. Subsequently, we digitized these records into spatial data with location and attribute information. For large-scale deterioration, such as collapses, we utilized LiDAR point cloud to pinpoint the locations of the collapses and verified the accuracy using on-site recorded positions and photographs of the collapses.\u003c/p\u003e \u003c/div\u003e"},{"header":"3 Methods","content":"\u003cp\u003eThis section introduces a combined analysis method using optical image and LiDAR data to study the impact of vegetation on ancient sites (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). We initiated by digitizing the wall's three-dimensional (3D) structure, followed by delineating the cracks at their corresponding locations on the wall model. Additionally, we employed the innovative application of contour line analysis and facade grid analysis to identify the collapse locations of the wall. For the optical image data captured by the drone, we conducted multi-scale segmentation to create image objects, and then calculated the internal texture features of each object. These texture features, along with the original image bands, were fed into a random forest classifier for classification, resulting in a classified map of the vegetation types covering the south palace wall. Finally, by integrating the results from the extraction of deterioration locations and the classification of vegetation, we conducted an analysis of how different vegetation types on the wall influence its deterioration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Wall deterioration detection\u003c/h2\u003e \u003cp\u003eFirstly, we applied top-down ground filtering methods to the LiDAR point cloud data to remove the vegetation point cloud, retaining only the ground point data. In this process, we employed a cloth simulation filter to separate the vegetation point cloud from the point cloud of the wall itself. The cloth simulation filter works by simulating the physical process of cloth draping over an object to separate ground and non-ground points [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. This process starts by inverting the point cloud, and then determining the final shape of the cloth through analyzing the interaction between the nodes of the cloth and the corresponding LiDAR points, with the final shape of the cloth representing the ground points. Once the ground points were obtained, we created a multipatch feature in ArcGIS software based on this ground point data, resulting in a spatially referenced 3D model. We then mapped the field-recorded locations, lengths, and trends of wall cracks onto this model, completing the spatial digitization of the wall cracks.\u003c/p\u003e \u003cp\u003eTo digitally identify collapse locations on the wall, we generated contour lines on the 3D model. These contour lines were segmented at approximately 2-meter intervals horizontally to form a series of curved segments. We calculated the curvature of these segments and categorized them by different colors based on their curvature, which was determined by dividing the length of each contour segment by the straight-line distance between its endpoints. In the color-graded map of contour line curvature, areas of the wall collapse were identified as regions where high-curvature contour lines clustered and concaved towards the inside of the wall. Following this principle, we recorded the identified collapse locations on the wall.\u003c/p\u003e \u003cp\u003eIn the initial top-down ground point filtering process, we considered the ground point cloud data as representing the wall itself, and used contour line analysis to extract collapses observed in the vertical direction. However, collapses such as partial indentations of the wall facade were not observable from a vertical perspective. To address this issue, Cai et al. [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e] proposed a method involving the horizontal placement of the facade, followed by the use of existing ground filtering algorithms to separate the wall surface and protrusions. In this study, we manually separated the point cloud data containing the wall facade from original point cloud data, rotated it 90 degrees to horizontally place the facade, and then applied the cloth simulation filter to obtain the point cloud data of the wall facade. Based on this facade point cloud data, we located areas of local indentation, thus identifying collapse locations on the wall from a horizontal perspective.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Vegetation classification\u003c/h2\u003e \u003cp\u003eWe employed an object-oriented approach for vegetation classification, a rapidly emerging technique in Geographic Information System (GIS) and archaeology, which provides a more accurate representation of geographical objects compared to single pixels [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Initially, the drone's optical images were subjected to multi-scale segmentation in eCognition software. The results of this segmentation were influenced by the scale of segmentation, shape, and compactness parameters. Based on preliminary research and subsequent experiments, we set the segmentation scale to 50, the shape parameter to 0.5, and the compactness to 0.8 [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eAfter obtaining image objects, using only the original red, green, and blue bands of the drone's image as classification features was insufficient. Therefore, we calculated the internal texture features of these objects as supplementary features to improve the classification results. We constructed a gray-level co-occurrence matrix (GLCM) for the pixels inside the image objects. GLCM is a tabulation of the frequency of different combinations of pixel gray levels [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Statistical measures derived from the GLCM, such as mean, dissimilarity, second-order moment, contrast, correlation, variance, homogeneity, and entropy, were used to effectively enhance image classification [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eFinally, we constructed a random forest classification model using these features to obtain the vegetation classification results. The random forest model achieves robust classification by combining multiple weak classifiers. These weak classifiers were built using decision tree algorithms, with the Gini index serving as the criterion for constructing the decision tree nodes [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The Gini index is calculated as follows:\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$Gini=1-{\\sum }_{i=0}^{I}{{p}_{i}}^{2}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(Gini\\)\u003c/span\u003e\u003c/span\u003e is the Gini index, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(I\\)\u003c/span\u003e\u003c/span\u003e represents the number of classes, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(i\\)\u003c/span\u003e\u003c/span\u003e denotes the class index, and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({p}_{i}\\)\u003c/span\u003e\u003c/span\u003e is the probability of occurrence of class \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(i\\)\u003c/span\u003e\u003c/span\u003e. A smaller Gini index indicates a higher purity of the sample set. Therefore, in constructing each tree node within the random forest classifier, the feature that results in the largest decrease in the Gini index is chosen for splitting, continuing until a leaf node is reached.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Analysis of the impact effects of vegetation on wall deterioration\u003c/h2\u003e \u003cp\u003eWe analyze the overlapping digital data products of vegetation types, cracks, and collapses on the wall to establish their spatial interrelationships. This analysis involves conducting spatial statistical evaluations to digitally represent and quantitatively assess how vegetation growth and its types are spatially correlated with the deterioration patterns of the wall. This approach enables us to understand how different types of vegetation correlate with and potentially contribute to collapses of the wall. Additionally, based on field survey data, which includes measurements of the lengths and widths of cracks as well as the conditions of root system growth within them, we further identified the impact of vegetation on wall cracks through statistical analysis.\u003c/p\u003e \u003c/div\u003e"},{"header":"4 Result","content":"\u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e4.1 Extraction of Wall Deterioration\u003c/h2\u003e \u003cp\u003eThe south palace wall of Weiyang Palace is divided into eastern and western sections, separated by a road. According to the 3D model (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea), the model of the eastern section of the wall is uneven and almost lacks the wall's original and regular shape. Upon field inspection, it was observed that the eastern section suffered significant damage (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eb). Additionally, the presence of large trees in this area impeded the penetration of the LiDAR signal. In contrast, the 3D model of the western section more clearly depicts the wall\u0026rsquo;s overall structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). Consequently, our subsequent research and analysis focused on the western section. We established the starting point of the western section of the wall as the origin, with westward direction serving as the positive direction for positioning wall deterioration. Based on the cracks recorded during field surveys, we marked these cracks on the corresponding locations of the wall model.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThrough field surveys and analyses between orthophoto and LiDAR point cloud data, we found that the western section of the wall experienced significant collapses from 130 meters to 740 meters, where the collapsed earth formed slopes. The vegetation growth on these slopes was notably faster than on the rammed earth walls, leading to the wall's sides from 130 meters to 740 meters being obscured by large paper mulberry trees. These large trees not only made field sampling challenging but also significantly obstructed the LiDAR signal. For these reasons, the extraction of collapses in this study excluded the part from 130 meters to 740 meters. After excluding the area from 130 meters to 750 meters, we divided the remaining areas into three parts for ease of display. Part 1 is from 50 meters to 130meters; Part 2 is from 750 meters to 1050 meters; Part 3 is from 1200 meters to 1300 meters.\u003c/p\u003e \u003cp\u003eOn the facade of the western section of the wall, we generated contour lines at 0.2-meter elevation intervals and segmented these lines at a horizontal distance of 2 meters, subsequently calculating the curvature of these segments and applying color grading based on their curvature. In the color-graded curvature map of the contour lines, the areas of wall collapse are indicated by the clustering of high-curvature contour lines that concave inward towards the wall. Based on this criterion, we identified the areas of collapse on the wall. Additionally, the protruding paper mulberry trees on the side of the wall caused the contour lines to first bulge outward and then concave inward. This necessitated a comparison with the LiDAR point cloud data to manually exclude any misidentified collapse areas. Ultimately, using the contour line method, we identified a total of 29 collapse locations., as illustrated in Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBased on the 3D model, we manually extracted the point cloud data of the facade of the wall, and after rotating it by 90 degrees, we re-applied the cloth simulation filter. Utilizing the filtered point cloud data, we constructed a grid model and visually interpreted areas of local indentation, identifying them as horizontal collapse locations on the wall. In total, we additionally found 10 collapse locations (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eThrough the analysis of contour lines and the facade grid, we identified a total of 39 collapse locations on the wall. We physically photographed and recorded 20 instances of wall collapse on-site. Upon detailed comparison with the digitally identified collapse locations, 18 out of the 20 recorded instances matched those detected by digital methods (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The detection accuracy for the collapse of Weiyang Palace's south palace wall was determined to be 90%.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e4.2 Vegetation classification\u003c/h2\u003e \u003cp\u003eTo simplify the research, we limited our classification targets to ziziphus jujuba, paper mulberry, and 'others' categories. The 'others' category includes bare earth and less frequent vegetation types like bromus. After performing multi-scale segmentation, we obtained 179,202 image objects, delineating 1,075 patches as paper mulberry samples, 336 as ziziphus jujuba samples, and 186 as samples of the 'others' category. We divided the sample set into a training set and a test set in 7:3 ratio and evaluated the classification model on the test set. We used precision, recall and f1-score as metrics (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). The final results of the vegetation types covering the wall are as presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e. The final classification ratios for paper mulberry, ziziphus jujuba, and 'other' were 0.78, 0.07, and 0.14, respectively. Paper mulberry and ziziphus jujuba are in a competitive state on the wall, with paper mulberry occupying a dominant position.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAccuracy of vegetation classification.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCategory\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePrecision\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eRecall\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eF1-score\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePaper mulberry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e90.2%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e97.5%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e93.8%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZiziphus jujuba\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e82.1%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e67.6%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e74.2%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOthers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e84.4%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e71.7%\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e77.6%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e4.3 Analysis of the impact effects of vegetation on wall deterioration\u003c/h2\u003e \u003cp\u003eCombining the collapse locations extracted in Section \u003cspan refid=\"Sec10\" class=\"InternalRef\"\u003e4.1\u003c/span\u003e with the vegetation classification results from Section \u003cspan refid=\"Sec11\" class=\"InternalRef\"\u003e4.2\u003c/span\u003e for an overlay analysis, we found that among the identified collapses, 75.9% were located under paper mulberry, 10.3% under ziziphus jujuba, and 13.8% were in areas without vegetation coverage. Comparing this with the overall proportions of vegetation types on the wall facade (paper mulberry: ziziphus jujuba: others\u0026thinsp;=\u0026thinsp;0.78: 0.07: 0.14), it appears that overlying vegetation type does not directly influence the probability of collapse occurrence.\u003c/p\u003e \u003cp\u003eWe conducted an overlay analysis of the wall cracks' attributes in conjunction with the types of vegetation covering the wall. Figure\u0026nbsp;\u003cspan refid=\"Fig9\" class=\"InternalRef\"\u003e9\u003c/span\u003e displays the locations, lengths, and widths of the cracks on the wall and the types of vegetation above them. It is evident that the cracks without vegetation above them are significantly larger in both length and width than those with vegetation, and these cracks are primarily concentrated in the 750 to 800 meters area. Based on our field investigations, we discovered that this particular section had undergone artificial treatment, resulting in the absence of vegetation growth above it. Consequently, due to the erosion caused by rainwater, this area experienced a significant number of large cracks. In the other cases, cracks under paper mulberry averaged 0.95 m in length and 3.37 cm in width, while those under ziziphus jujuba averaged 1.17 m in length and 1.41 cm in width, indicating that cracks under paper mulberry tend to be wider, whereas those under ziziphus jujuba are longer. Additionally, we noticed that 32% of the cracks had ziziphus jujuba growing above them, a proportion much higher than its occurrence on the wall (7%), suggesting a higher likelihood of crack formation in areas with ziziphus jujuba.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eDuring field investigations, we discovered that 62% of the cracks had vegetation roots growing directly within them; of these, 37% were from paper mulberry and 63% from Ziziphus jujuba. This distribution suggests that Ziziphus jujuba plays a significant role in the formation of wall cracks. Further statistical analysis reveals that, compared to conditions without roots, the roots of paper Mulberry and ziziphus jujuba simultaneously increase the average length and width of cracks. Specifically, paper mulberry roots significantly enhance the width, while ziziphus jujuba roots notably increase the length (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAverage lengths and widths of cracks under different roots conditions\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eRoots conditions\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAverage width (cm)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eAverage length (m)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePresence of paper mulberry roots\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e3.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.01\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003ePresence of ziziphus jujuba roots\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.98\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e1.25\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eAbsence of Roots\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e1.33\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe analysis above indicates that the occurrence of collapses on the wall is not influenced by the type of vegetation, suggesting that vegetation type is not a significant factor in these large-scale phenomena. In contrast, the development of cracks in the wall is highly dependent on the type of vegetation present. Paper mulberry is associated with wider cracks, whereas ziziphus jujuba is linked to longer cracks, with areas hosting ziziphus jujuba being particularly susceptible to crack formation. Importantly, despite the adverse effects of Paper Mulberry on crack severity, it still offers better protection than exposing the wall directly to external environmental conditions.\u003c/p\u003e \u003c/div\u003e"},{"header":"5 Discussion","content":"\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e5.1 Advancement and limitation\u003c/h2\u003e \u003cp\u003eSpatial technology is increasingly being applied in the field of cultural heritage [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. However, its use is often limited to digitalizing and demonstrating archaeological sites [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e], with only a minority of studies employing intelligent algorithms to extract useful information for heritage monitoring and protection from spatial data [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e, \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. The deterioration of earthen sites results from a combination of multiple factors [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Therefore, alongside using spatial technology to monitor these sites, it is crucial to investigate the underlying forces driving their deterioration to inform effective protection strategies. Regrettably, research in these areas tends to be conducted in isolation: studies utilizing spatial technology typically concentrate on data acquisition and site monitoring, whereas those analyzing deterioration mechanisms focus primarily on these processes [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e, \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e], often using small samples in laboratory settings, which do not provide a holistic understanding of the sites or practical conservation solutions.\u003c/p\u003e \u003cp\u003eThis study integrated multi-source spatial data to monitor and analyze the deterioration and driving forces of the Han Dynasty Weiyang Palace's south palace wall. Using the cloth simulation filter, especially after the 90-degree rotation, we developed three-dimensional data of the wall itself and portrayed its deterioration through a series of spatial analysis methods. These methods are significantly relevant for data processing and information extraction for earthen sites like the wall. By combining intelligent algorithms with high-resolution drone image, we classified the overlying vegetation on the wall. Through overlay and statistical analysis, we identified the impact of the vegetation on the deterioration of the wall. The approach of integrating multi-source data for overlay analysis in this study can be extended to similar research on earthen sites. Beyond the impact of vegetation, sensors carried by drones can infer temperature, humidity and so on [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e], which can be overlaid with the extracted locations of wall deterioration for further analysis, identifying driving forces and formulating scientific conservation recommendations.\u003c/p\u003e \u003cp\u003eOur research has developed a method for analyzing the driving forces affecting earthen sites using spatial technology, but it has some limitations. With the advancement of deep learning algorithms [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e], it has become possible to automatically extract wall deterioration from LiDAR point cloud data or 3D models. However, at this stage, our study only includes the case of Weiyang Palace's south palace wall, which is insufficient to build an adequate sample set for deep learning. In reality, the Silk Road is home to a multitude of earthen sites. Incorporating these sites into comprehensive research could significantly enhance the potential for automated and accurate identification of site deterioration. Moreover, during our research process, we found that the scale of the cracks in the wall is too small for detection by the data from airborne platforms, so our data on cracks was manually recorded. In winter, when vegetation leaves wither, close-range photogrammetric techniques can be used to identify cracks [\u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. The identified cracks can be digitally mapped onto the surface of the wall, allowing for direct calculation of their lengths and widths. This is also a direction for our future research.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e5.2 Suggestions for the protection of the south palace wall\u003c/h2\u003e \u003cp\u003eThe deterioration of earthen sites in natural environments is complex, making it challenging to completely prevent deterioration [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. Our experimental data shows that the south palace wall of Weiyang Palace is currently undergoing gradual deterioration, with collapses occurring regardless of whether vegetation is present or not. Therefore, it is necessary to use appropriate materials to reinforce the structure of the wall to prevent further extensive collapses [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e]. As for the cracks, ziziphus jujuba significantly contributes to the formation of cracks in the wall, while paper mulberry, although slightly exacerbating the cracks, offers better protection than exposing the wall directly to external environmental conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig10\" class=\"InternalRef\"\u003e10\u003c/span\u003e). On the wall, paper mulberry and ziziphus jujuba are in competition, with paper mulberry in a dominant position. Considering these factors, we suggest trying chemical methods to eradicate ziziphus jujuba to prevent the formation of more cracks in the wall. Until an appropriate man-made shelter is created for the wall, the paper mulberry trees on it should be preserved to avoid direct erosion by wind and rain.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"6 Conclusion","content":"\u003cp\u003eThis study has offered comprehensive insights into the deterioration of the south palace wall of Weiyang Palace. By integrating multi-source spatial data, including LiDAR and high-resolution drone image, we have successfully identified the deterioration along the wall and analyzed the impact effects of vegetation on wall deterioration. Utilizing innovative methods such as contour line analysis and facade grid analysis, we effectively pinpointed the locations of deterioration along the wall. These techniques enabled a detailed spatial representation of the wall's condition, demonstrating the value of advanced spatial analysis in cultural heritage preservation. And our research employed an object-oriented approach, integrating texture features to classify the types of overlying vegetation on the wall. This classification played a crucial role in understanding the interaction between vegetation and the wall's deterioration. Finally, we discovered that paper mulberry and ziziphus jujuba compete for dominance on the wall and their impact on the wall's structure is distinct. Paper mulberry, despite slightly exacerbating the cracks, offers a level of protection against direct environmental exposure, whereas ziziphus jujuba contributes significantly to the formation of cracks.\u003c/p\u003e \u003cp\u003eThe study underscores the complexity of conserving earthen sites, especially those continuously exposed to natural elements and human interventions. It is clear that a one-size-fits-all approach is not feasible for the preservation of such earthen sites. Our research provides a case study on the detailed analysis of the driving forces behind wall deterioration using spatial technology and intelligent algorithms. This approach can guide more precise conservation strategies for earthen sites.\u003c/p\u003e \u003cp\u003eIn conclusion, our study not only advances the understanding of biotic factors affecting earthen heritage deterioration but also sets a precedent for the application of integrated spatial and botanical analysis in the conservation of cultural heritage worldwide.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgements\u003c/h2\u003e\n\u003cp\u003eWe would like to express our sincere gratitude to the Xi\u0026apos;an Academy of Conservation and Archaeology for their generous support and funding.\u003c/p\u003e\n\u003ch2\u003eAuthors\u0026rsquo; contributions\u003c/h2\u003e\n\u003cp\u003eConceptualization, L.T. and F.C.; methodology, S.G., F.C. and L.T.; software, S.G., P.S and Z.X; formal analysis, S.G. and F.C.; investigation, S.G., W.L., Y.L, H.L and C.C.; resources, X.Y. and W.L.; data curation, S.G. and X.Z; writing\u0026mdash;original draft preparation, S.G.; writing\u0026mdash;review and editing, S.G. L,T and F.C.; visualization, S.G., W.Z., and M.Z.; supervision, F.C. and L.T.; project administration, F.C. and L.T.; funding acquisition, F.C. and L.T. All authors have read and agreed to the published version of the manuscript.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis work was jointly support by the National Natural Science Foundation of China (NSFC) (grant no. 42271327) and \u0026apos;Silk Road: Digital Preservation of Cultural Heritage Sites in the Xi\u0026apos;an Section of the Chang\u0026apos;an-Tianshan Corridor Network\u0026apos; project of the Xi\u0026apos;an Academy of Conservation and Archaeology.\u003c/p\u003e\n\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e\n\u003cp\u003eData sharing is not applicable to this article as no datasets were generated during the current study.\u003c/p\u003e\n\u003ch2\u003eEthics approval and consent to participate\u003c/h2\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBowitz E, Ibenholt K. Economic impacts of cultural heritage \u0026ndash; Research and perspectives. Journal of Cultural Heritage. 2009;10(1):1-8.\u003c/li\u003e\n\u003cli\u003eGravagnuolo A, Micheletti S, Bosone M. A Participatory Approach for \u0026ldquo;Circular\u0026rdquo; Adaptive Reuse of Cultural Heritage. Building a Heritage Community in Salerno, Italy. Sustainability. 2021;13(9):4812.\u003c/li\u003e\n\u003cli\u003eWang C, Fulong C, Wei Z, Huafen Y, Di WJNRSB. Sequential PSInSAR approach for the deformation monitoring of the Nanjing Ming Dynasty City Wall. National Remote Sensing Bulletin. 2022;25(12):2381-95.\u003c/li\u003e\n\u003cli\u003eWang X, Li H, Wang Y, Zhao X. Assessing climate risk related to precipitation on cultural heritage at the provincial level in China. Science of The Total Environment. 2022;835:155489.\u003c/li\u003e\n\u003cli\u003eShao M, Li L, Wang S, Wang E, Li Z. Deterioration mechanisms of building materials of Jiaohe ruins in China. Journal of Cultural Heritage. 2013;14(1):38-44.\u003c/li\u003e\n\u003cli\u003ePetti L, Trillo C, Makore BN. Cultural Heritage and Sustainable Development Targets: A Possible Harmonisation? Insights from the European Perspective. Sustainability. 2020;12(3):926.\u003c/li\u003e\n\u003cli\u003eGuo H, Chen F, Tang Y, Ding Y, Chen M, Zhou W, et al. Progress toward the sustainable development of world cultural heritage sites facing land-cover changes. The Innovation. 2023;4(5).\u003c/li\u003e\n\u003cli\u003eRichards J, Wang Y, Orr S, Viles H. Finding Common Ground between United Kingdom Based and Chinese Approaches to Earthen Heritage Conservation. Sustainability. 2018;10(9):3086.\u003c/li\u003e\n\u003cli\u003eGuti\u0026eacute;rrez-Carrillo ML, Arizzi A. How to deal with the conservation of the archaeological remains of earthen defensive architecture: the case of Southeast Spain. Archaeological and Anthropological Sciences. 2021;13(8):131.\u003c/li\u003e\n\u003cli\u003eRichards J, Zhao G, Zhang H, Viles H. A controlled field experiment to investigate the deterioration of earthen heritage by wind and rain. Heritage Science. 2019;7(1):1-13.\u003c/li\u003e\n\u003cli\u003eXie L, Wang D, Zhao H, Gao J, Gallo T. Architectural energetics for rammed-earth compaction in the context of Neolithic to early Bronze Age urban sites in Middle Yellow River Valley, China. Journal of Archaeological Science. 2021;126:105303.\u003c/li\u003e\n\u003cli\u003eRainer L. Deterioration and pathology of earthen architecture. Terra Literature Review. 2008:45.\u003c/li\u003e\n\u003cli\u003eFodde E, Khan MS. Affordable Monsoon Rain Mitigation Measures in the World Heritage Site of Moenjodaro, Pakistan. International Journal of Architectural Heritage. 2013;11(2):161-73.\u003c/li\u003e\n\u003cli\u003eParisi F, Asprone D, Fenu L, Prota A. Experimental characterization of Italian composite adobe bricks reinforced with straw fibers. Composite Structures. 2015;122:300-7.\u003c/li\u003e\n\u003cli\u003eDu Y, Chen W, Cui K, Zhang K. Study on Damage Assessment of Earthen Sites of the Ming Great Wall in Qinghai Province Based on Fuzzy-AHP and AHP-TOPSIS. International Journal of Architectural Heritage. 2019;14(6):903-16.\u003c/li\u003e\n\u003cli\u003eZhang Y, Ye WM, Chen B, Chen YG, Ye B. Desiccation of NaCl-contaminated soil of earthen heritages in the Site of Yar City, northwest China. Applied Clay Science. 2016;124-125:1-10.\u003c/li\u003e\n\u003cli\u003eChen F, Lasaponara R, Masini N. An overview of satellite synthetic aperture radar remote sensing in archaeology: From site detection to monitoring. Journal of Cultural Heritage. 2017;23:5-11.\u003c/li\u003e\n\u003cli\u003eLuo L, Wang X, Guo H, Lasaponara R, Zong X, Masini N, et al. Airborne and spaceborne remote sensing for archaeological and cultural heritage applications: A review of the century (1907\u0026ndash;2017). Remote Sensing of Environment. 2019;232:111280.\u003c/li\u003e\n\u003cli\u003eRemondino F. Heritage Recording and 3D Modeling with Photogrammetry and 3D Scanning. Remote Sensing. 2011;3(6):1104-38.\u003c/li\u003e\n\u003cli\u003eCampiani A, Lingle A, Lercari N. Spatial analysis and heritage conservation: Leveraging 3-D data and GIS for monitoring earthen architecture. Journal of Cultural Heritage. 2019;39:166-76.\u003c/li\u003e\n\u003cli\u003eChen F, Guo H, Ma P, Tang Y, Wu F, Zhu M, et al. Sustainable development of World Cultural Heritage sites in China estimated from optical and SAR remotely sensed data. Remote Sensing of Environment. 2023;298:113838.\u003c/li\u003e\n\u003cli\u003eMallet C, Bretar F. Full-waveform topographic lidar: State-of-the-art. ISPRS Journal of Photogrammetry and Remote Sensing. 2009;64(1):1-16.\u003c/li\u003e\n\u003cli\u003eComer DC, Comer JA, Dumitru IA, Ayres WS, Levin MJ, Seikel KA, et al. Airborne LiDAR Reveals a Vast Archaeological Landscape at the Nan Madol World Heritage Site. Remote Sensing. 2019;11(18):2152.\u003c/li\u003e\n\u003cli\u003eFreeland T, Heung B, Burley DV, Clark G, Knudby A. Automated feature extraction for prospection and analysis of monumental earthworks from aerial LiDAR in the Kingdom of Tonga. Journal of Archaeological Science. 2016;69:64-74.\u003c/li\u003e\n\u003cli\u003eWang S, Hu Q, Wang S, Ai M, Zhao P. Archaeological site segmentation of ancient city walls based on deep learning and LiDAR remote sensing. Journal of Cultural Heritage. 2024;66:117-31.\u003c/li\u003e\n\u003cli\u003eGuo Q, Wang Y, Chen W, Pei Q, Sun M, Yang S, et al. Key Issues and Research Progress on the Deterioration Processes and Protection Technology of Earthen Sites under Multi-Field Coupling. Coatings. 2022;12(11):1677.\u003c/li\u003e\n\u003cli\u003eDu Y, Chen W, Cui K, Gong S, Pu T, Fu X. A Model Characterizing Deterioration at Earthen Sites of the Ming Great Wall in Qinghai Province, China. Soil Mechanics and Foundation Engineering. 2017;53(6):426-34.\u003c/li\u003e\n\u003cli\u003eRichards J, Mayaud J, Zhan H, Wu F, Bailey R, Viles H. Modelling the risk of deterioration at earthen heritage sites in drylands. Earth Surface Processes and Landforms. 2020;45(11):2401-16.\u003c/li\u003e\n\u003cli\u003eCanuti P, Casagli N, Catani F, Fanti R. Hydrogeological hazard and risk in archaeological sites: some case studies in Italy. Journal of Cultural Heritage. 2000;1(2):117-25.\u003c/li\u003e\n\u003cli\u003eLiu S, Huang S, Xie Y, Wang H, Huang Q, Leng G, et al. Spatial-temporal changes in vegetation cover in a typical semi-humid and semi-arid region in China: Changing patterns, causes and implications. Ecological Indicators. 2019;98:462-75.\u003c/li\u003e\n\u003cli\u003eRichards J, Bailey R, Mayaud J, Viles H, Guo Q, Wang XJSR. Deterioration risk of dryland earthen heritage sites facing future climatic uncertainty. Scientific Reports. 2020;10(1):16419.\u003c/li\u003e\n\u003cli\u003eWolfe SA, Nickling WGJPipg. The protective role of sparse vegetation in wind erosion. Progress in physical geography. 1993;17(1):50-68.\u003c/li\u003e\n\u003cli\u003eMoreno M, Ortiz P, Ortiz R. Analysis of the impact of green urban areas in historic fortified cities using Landsat historical series and Normalized Difference Indices. Scientific Reports. 2023;13(1):8982.\u003c/li\u003e\n\u003cli\u003eZhang W, Qi J, Wan P, Wang H, Xie D, Wang X, et al. An easy-to-use airborne LiDAR data filtering method based on cloth simulation. Remote sensing. 2016;8(6):501.\u003c/li\u003e\n\u003cli\u003eCai S, Zhang S, Zhang W, Fan H, Shao J, Yan G, et al. A General and Effective Method for Wall and Protrusion Separation from Facade Point Clouds. Journal of Remote Sensing. 2023;3:0069.\u003c/li\u003e\n\u003cli\u003eBlaschke T, Hay GJ, Kelly M, Lang S, Hofmann P, Addink E, et al. Geographic object-based image analysis\u0026ndash;towards a new paradigm. ISPRS journal of photogrammetry and remote sensing. 2014;87:180-91.\u003c/li\u003e\n\u003cli\u003eMagnini L, Bettineschi C. Theory and practice for an object-based approach in archaeological remote sensing. Journal of Archaeological Science. 2019;107:10-22.\u003c/li\u003e\n\u003cli\u003eChen G, He Y, De Santis A, Li G, Cobb R, Meentemeyer RK. Assessing the impact of emerging forest disease on wildfire using Landsat and KOMPSAT-2 data. Remote Sensing of Environment. 2017;195:218-29.\u003c/li\u003e\n\u003cli\u003eHaralick RM, Shanmugam K, Dinstein IH. Textural features for image classification. IEEE Transactions on systems, man, and cybernetics. 1973(6):610-21.\u003c/li\u003e\n\u003cli\u003eGadkari D. Image quality analysis using GLCM. 2004.\u003c/li\u003e\n\u003cli\u003eBreiman L. Random forests. Machine learning. 2001;45:5-32.\u003c/li\u003e\n\u003cli\u003eDe Reu J, Plets G, Verhoeven G, De Smedt P, Bats M, Cherrett\u0026eacute; B, et al. Towards a three-dimensional cost-effective registration of the archaeological heritage. Journal of archaeological science. 2013;40(2):1108-21.\u003c/li\u003e\n\u003cli\u003eLandeschi G, Nilsson B, Dell\u0026apos;Unto N. Assessing the damage of an archaeological site: New contributions from the combination of image-based 3D modelling techniques and GIS. Journal of Archaeological Science: Reports. 2016;10:431-40.\u003c/li\u003e\n\u003cli\u003eDu Y, Chen W, Cui K, Zhang J, Chen Z, Zhang Q. Damage assessment of earthen sites of the Ming Great Wall in Qinghai Province: a comparison between Support Vector Machine (SVM) and BP Neural Network. Journal on Computing and Cultural Heritage (JOCCH). 2020;13(2):1-18.\u003c/li\u003e\n\u003cli\u003eLercari N. Monitoring earthen archaeological heritage using multi-temporal terrestrial laser scanning and surface change detection. Journal of Cultural Heritage. 2019;39:152-65.\u003c/li\u003e\n\u003cli\u003eRichards J, Viles H, Guo Q. The importance of wind as a driver of earthen heritage deterioration in dryland environments. Geomorphology. 2020;369:107363.\u003c/li\u003e\n\u003cli\u003eRichards J, Guo Q, Viles H, Wang Y, Zhang B, Zhang H. Moisture content and material density affects severity of frost damage in earthen heritage. Science of The Total Environment. 2022;819:153047.\u003c/li\u003e\n\u003cli\u003eRichards J, Zhao G, Zhang H, Viles H. A controlled field experiment to investigate the deterioration of earthen heritage by wind and rain. Heritage Science. 2019;7:1-13.\u003c/li\u003e\n\u003cli\u003eFrodella W, Elashvili M, Spizzichino D, Gigli G, Adikashvili L, Vacheishvili N, et al. Combining infrared thermography and uav digital photogrammetry for the protection and conservation of rupestrian cultural heritage sites in Georgia: A methodological application. Remote Sensing. 2020;12(5):892.\u003c/li\u003e\n\u003cli\u003eSu T-C. Environmental risk mapping of physical cultural heritage using an unmanned aerial remote sensing system: A case study of the Huang-Wei monument in Kinmen, Taiwan. Journal of Cultural Heritage. 2019;39:140-51.\u003c/li\u003e\n\u003cli\u003eGuo Y, Wang H, Hu Q, Liu H, Liu L, Bennamoun M. Deep learning for 3d point clouds: A survey. IEEE transactions on pattern analysis and machine intelligence. 2020;43(12):4338-64.\u003c/li\u003e\n\u003cli\u003eGalantucci RA, Fatiguso F. Advanced damage detection techniques in historical buildings using digital photogrammetry and 3D surface anlysis. Journal of Cultural Heritage. 2019;36:51-62.\u003c/li\u003e\n\u003cli\u003eLi L, Shao M, Wang S, Li Z. Preservation of earthen heritage sites on the Silk Road, northwest China from the impact of the environment. Environmental Earth Sciences. 2011;64:1625-39.\u003c/li\u003e\n\u003cli\u003eCorreia M, Guerrero L. Conservation of earthen building materials. The Encyclopedia of Archaeological Sciences. 2018:1-6.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-heritage-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hsci","sideBox":"Learn more about [Heritage Science](http://heritagesciencejournal.springeropen.com)","snPcode":"40494","submissionUrl":"https://submission.nature.com/new-submission/40494/3","title":"npj Heritage Science","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Weiyang Palace, earthen heritage, LiDAR, vegetation, spatial technology","lastPublishedDoi":"10.21203/rs.3.rs-4568335/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4568335/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWeiyang Palace, as the royal palace of the Western Han Dynasty, is a part of the Silk Roads: the Routes Network of Chang'an-Tianshan Corridor on the World Heritage list. The south palace wall of Weiyang Palace is a well-preserved earthen site within the palace, but it is undergoing continuous deterioration due to the influence of vegetation and external environmental factors. This study pioneers the integration of high-resolution three-dimensional LiDAR scanning with multi-source data analysis, including unprecedented on-site botanical surveys, to elucidate the nuanced impacts of different vegetation types on the structural integrity of the south palace wall. Through contour line analysis and facade grid analysis, we extracted the deterioration locations of typical sections of the earthen sites. And we classified the overlying vegetation types on the wall using an object-oriented classification algorithm. Our findings reveal a complex interaction between vegetation and earthen structures: paper mulberry exhibits protective qualities against erosion, while ziziphus jujuba significantly exacerbates structural vulnerabilities by inducing cracks. Methods employed in this study for extracting earthen site deterioration and combining multi-source spatial data analysis can serve as a technical application model for monitoring and analyzing the driving forces of surface earthen sites along the entire Silk Road network, thereby better guiding the conservation of earthen sites.\u003c/p\u003e","manuscriptTitle":"Extraction of Deterioration and Analysis of Vegetation Impact Effects on the South Palace Wall of Weiyang Palace","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-05 03:53:43","doi":"10.21203/rs.3.rs-4568335/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-07-18T19:33:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-15T12:24:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-07-12T06:00:44+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"261043246669437280545246421900969859757","date":"2024-06-27T12:26:08+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"249278336708540655629148231561112232905","date":"2024-06-26T15:04:17+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-06-26T11:30:57+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"56741871677881821580381070859894740301","date":"2024-06-25T01:49:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"61557274935946747091543553058956974318","date":"2024-06-24T00:13:11+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-06-23T21:16:43+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-06-19T11:33:49+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-06-19T11:31:58+00:00","index":"","fulltext":""},{"type":"submitted","content":"Heritage Science","date":"2024-06-12T07:34:36+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-heritage-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"hsci","sideBox":"Learn more about [Heritage Science](http://heritagesciencejournal.springeropen.com)","snPcode":"40494","submissionUrl":"https://submission.nature.com/new-submission/40494/3","title":"npj Heritage Science","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8f3e968e-c3e1-467e-84bf-e17dcbe0b3b6","owner":[],"postedDate":"July 5th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2024-10-21T16:02:14+00:00","versionOfRecord":{"articleIdentity":"rs-4568335","link":"https://doi.org/10.1186/s40494-024-01485-x","journal":{"identity":"npj-heritage-science","isVorOnly":false,"title":"npj Heritage Science"},"publishedOn":"2024-10-18 15:57:35","publishedOnDateReadable":"October 18th, 2024"},"versionCreatedAt":"2024-07-05 03:53:43","video":"","vorDoi":"10.1186/s40494-024-01485-x","vorDoiUrl":"https://doi.org/10.1186/s40494-024-01485-x","workflowStages":[]},"version":"v1","identity":"rs-4568335","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4568335","identity":"rs-4568335","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}