Different Spatial Models Are Used To Study The Distribution Of Pakistani Ruins

The ancient Silk Road has a long history and a large number of cultural heritages along the route.The archaeological sites that have been discovered so far are only a small part of the total cultural heritage.There are still a large number of archaeological sites that need further exploration and research.Using remote sensing to study the spatial distribution is of great significance for understanding the distribution law of ruins,discovering more important archaeological sites,and gaining a deeper understanding of the history and culture of the ancient Silk Road.Pakistan is an important country along the Belt and Road.The China-Pakistan Economic Corridor is an important part of the Belt and Road.It is also a key hub that runs through the North and South Silk Roads.Therefore,predicting the cultural heritage of Pakistan can provide science for the protection of the cultural heritage of the corridor Data and demonstration cases.Provide technical support for the study of other corridors along the Belt and Road.Pakistan has a long history and rich cultural heritage.Through research and analysis of the distribution of ruins in Pakistan,we have a deep understanding of the spatial distribution law of ruins in the region,and we have selected different spatial distribution models to predict the distribution of Pakistani ruins.To obtain the main environmental factors that affect the spatial distribution of the site and provide important guidance data for the protection of the cultural heritage of the China-Pakistan Corridor.In most cases,the Maxent model is used to analyze dynamic processes and spatial interactions,and is rarely used in archaeology.In view of the advantages of the Maxent model,which is easy to operate,less workload,high accuracy,and good stability,so this research attempts Use it to study the distribution of archaeological sites.In this study,based on 674 sites,environmental variables such as DEM,soil type,land use type,distance from water,ecological zoning,land cover type,slope,profile curvature,aspect,and ecological land unit were selected as independent variables.Construct a prediction model for the Maxent site.This paper further uses the relatively mature Logistic model of the research site distribution as a comparative study to verify the accuracy of the Maxent model,and uses the Kvamme gain value as an evaluation index to compare the accuracy of the two models.Finally,the local analysis method is used to select representative different landforms The type of research area validates the prediction results.Research indicates:(1)Comparison between Maxent model and Logistic model,and Kvamme gain value is used to evaluate the accuracy of the model.The results show that the Maxent model has high accuracy and stability in the study of site prediction in Pakistan.It is 0.869,and the value range of AUC is [0,1].The accuracy of the model is high from 0.7 to 0.9;the Logistic model directly obtains the accuracy value,and the accuracy of the Logistic model is 84.1%,which is also high.The Kvamme gain value is used to compare the accuracy of the two models.The gain value of the Maxent model is much greater than the Logistic model.The results of applying the same research method to the local area also show that the Maxent model has higher accuracy,while the gain value of the Logistic model is unstable in different sub-regions.In addition,comparing the probability values of the overall region and the local region,the results show that the maximum value of the local region and the overall region is not much different,ranging from-0.02 to 0.11.Therefore,the local area verification and the overall area have little difference in probability values,and the experimental results are credible.Therefore,for the study of the distribution of sites in Pakistan,the accuracy of Maxent model is higher and more stable.(2)The distribution law of the ruins in Pakistan is as follows: the distribution probability of sites in the Indus basin is high,and the distribution probability of sites in the northern highlands is low;the reason for this distribution is mainly due to the flat terrain,fertile soil and sufficient water sources in the Indus basin,suitable for ancient people to live,while the northern part The high terrain has high terrain,insufficient water resources,and a cold climate,making it unsuitable for ancient humans to sit on.(3)The selection of ancient human settlements is affected by various environmental factors.The Maxent model can predict the importance of different environmental factors on the distribution of sites.Under the overall research scale,based on the Maxent model,the degree of influence of different environmental variables on the distribution of Pakistani ruins is obtained.The order of importance is: soil type> land cover category> DEM>ecological land unit> ecological zoning> land use type> slope> profile curvature> water distance> slope direction.(4)The prediction results of the Logistic model for small data sets are unstable.Maxent can successfully build models with small data sets and the prediction results are credible.By analyzing the temporal and spatial characteristics of Pakistani ruins,a spatial analysis model is established to study and predict the distribution of Pakistani ruins,the prediction accuracy of different spatial models is analyzed,and a model more suitable for predicting the distribution of Pakistani ruins is obtained,and the distribution of Pakistani ruins is deeply analyzed According to the law,the importance of different environmental variables on the distribution of Pakistani ruins is obtained,which provides data support for the protection of Pakistani ruins;provides data support and decision-making suggestions for the protection and utilization of the cultural heritage of the China-Pakistan Corridor.At the same time,it can provide a reference for the research on the prediction of the distribution of other corridors along the Belt and Road.