Algorithms and Software Tools for Extracting Coastal Morphological Information from Airborne LiDAR Data
With the ever increasing population and economic activities in coastal areas, coastal hazards have become a major concern for coastal management. The fundamental requirement of coastal planning and management is the scientific knowledge about coastal forms and processes. This research aims at developing algorithms for automatically extracting coastal morphological information from LiDAR data. The primary methods developed by this research include automated algorithms for beach profile feature extraction and change analysis, and an object-based approach for spatial pattern analysis of coastal morphologic and volumetric change. Automated algorithms are developed for cross-shore profile feature extraction and change analysis. Important features of the beach profile such as dune crest, dune toe, and beach berm crest are extracted automatically by using a scale-space approach and by incorporating contextual information. The attributes of important feature points and segments are derived to characterize the morphologic properties of each beach profile. Beach profiles from different time periods can be compared for morphologic and volumetric change analysis. An object-oriented approach for volumetric change analysis is developed to identify and delineate individual elevation change patches as discrete objects. A set of two-dimensional and three-dimensional attributes are derived to characterize the objects, which includes planimetric attributes, shape attributes, surface attributes, volumetric attributes, and summary attributes. Both algorithms are implemented as ArcGIS extension modules to perform the feature extraction and attribute derivation for coastal morphological change analysis. To demonstrate the utility and effectiveness of algorithms, the cross-shore profile change analysis method and software tool are applied to a case study area located at southern Monterey Bay, California, and the coastal morphology change analysis method and software tool are applied to a case study area located on Assateague Island, Maryland. The automated algorithms facilitate the efficient beach profile feature analysis over large geographical area and support the analysis of the spatial variations of beach profile changes along the shoreline. The explicit object representation of elevation change patches makes it easy to localize erosion hot spots, to classify the elevation changes caused by various mechanisms, and to analyze spatial pattern of morphologic and volumetric changes.
Gao, Yige (2009). Algorithms and Software Tools for Extracting Coastal Morphological Information from Airborne LiDAR Data. Master's thesis, Texas A&M University. Available electronically from