摘要: |
The goal for this project is to provide an advanced QL2 LiDAR-based tidal wetland prediction method and automation tools based on ArcGIS for the North Carolina (NC) coastal region. Based on the North Carolina Department of Transportation's (NCDOT's) needs the project proposes a scope of work as follows: (1) conduct a literature review and investigate the status of existing methods and models of LiDAR-based tidal wetland prediction and use of the QL2 standard LiDAR data; (2) determine the optimal resolution of DEM and subsequent terrain derivatives, and any other variables needed to predict wetlands via orthogonal test design approach and SCAD method on QL2 LiDAR data and other related data; (3) develop appropriate methods to model tidal wetland boundary locations via first-hand experience of wetland scientists, regression method Logit (logistic regression), machine learning method RF (random forest), and/or statistics Bayes estimation method; (4) develop tools to automate the process of interpolation, variable creation, and model development and application where it is appropriate and feasible to do so; (5) validate the developing methods and models through field testing; and (6) prepare products including the proposed methods, models, algorithms, and tools. |