Research progress and prospect of karst geomorphology in China based on digital elevation model

Digital elevation model (DEM) contains rich morphological information because the process of digital terrain modeling and extracting a series of terrain parameters based on DEM data with various spatial resolutions, known as Digital Terrain Analysis (DTA), has achieved a number of advances in topographic analysis, morphological modeling, distribution patterns and evolution patterns of karst landforms. This paper summarizes the new progress in spatial analysis of DEM based on the key technical methods of applying DEM to the previous Chinese studies on karst landform in recent years. The future research of DEM in karst geomorphology is also predicted in order to expand its application and provide technical ideas for the study of karst geomorphology from topology to the process and mechanism of topographic evolution.In this study, the applicability of the data model and the method of determining the optimal scale of analysis are discussed. Both DEM and DTA are obviously scale-dependent, and it is indisputable that the slope scales, watershed scales, and regional scales of karst topography require different resolution of DEM. Higher spatial resolution does not necessarily guarantee better automatic karst detection. Therefore, the main features of regional topography must be considered. When local terrain parameters are calculated and extracted, determining an optimal analysis window is necessary to reflect the completeness of the landform and guarantee the representativeness of calculation results.For the identification and classification of karst landforms, four effective methods for automatic extraction of karst landform units with DEM data and the advantages and disadvantages of these methods are analyzed. These methods are respectively based on cell statistics of composite factors, the extraction of terrain feature points, contour trees, and positive and negative topography. The first method is to divide the study area into a series of grids, calculate the topographic attributes of each cell grid, and construct an identification model including the establishment of peak forests, peak clusters and other karst geomorphic units. The second one is to construct a spatial relationship model of terrain feature points of different geomorphic units, and then automatically extract these units. The third one is to extract geomorphic contour lines by building a contour line recognition model for classification. The fourth one can reduce and simplify the complex and diverse terrains into positive and negative terrains to effectively highlight the morphological differences, and then divide them according to the characteristics of geomorphic units such as peak forests, peak clusters, etc.As for the morphology and pattern analysis of karst landform, the typical index parameters used to characterize the spatial pattern of karst landform are sorted out. In addition, the main applications of DEM in the ecological environment of karst area have been outlined. DEM can be used for topographic statistics, spatial pattern, spatial estimation and 3D visualization of meteorology, hydrology, soil and habitat, and then for the analysis of ecological environment changes in karst areas.However, the present classification system of karst landform is mainly focused on abstract expression and qualitative description, which can hardly quantify the detailed information of spatial structure of karst landform. This kind of system also cannot meet the demand of digitalization and intellectualization. In addition, with good recognition effect, the current methods for morphological and quantitative expression of karst landform have been widely applied to typical karst landforms such as peak forests, peaks and depressions. But for atypical karst landforms, DTA has rarely been tried.In conclusion, the prospective direction of DEM-based karst geomorphology research has been proposed in this study. First of all, a more effective quantitative classification system needs to be constructed for karst geomorphology and its topographic elements. A digital indicator system of karst geomorphology should be established to describe its spatial and geological properties in terms of spatial distribution relationships, change patterns, etc. Secondly, more comprehensive geomorphological information should be extracted from regional, watershed and slope scales to reveal the relationship between each indicator and karst development in the study on mechanism and evolution of karst development. Moreover, it is necessary to integrate remote sensing with other multi-source data to enrich the geological attributes of DEM and to delve into more information in soil, vegetation, geological structure and other fields.