Matt Simon

Geog 593

October 4, 2006

A LiDAR Processing Toolkit:

An Automated and Centralized Approach to

LiDAR Processing and DEM Grid Creation

  1. INRODUCTION

Habitat fragmentation has led to an increasingly patchy landscape. This has forced land managers to adopt a metapopulation approach to land management. One strategy for effective habitat conservation is to focus on those lands which will protect or create connections between existing patches of habitat. The Strategic Environmental Research and Development Program (SERDP) is the Department of Defense’s (DoD) environmental science and technology program. Currently at UNC-Chapel Hill,in conjunction with N.C.State, Duke, and Virginia Tech, is working on a SERDP project titledMapping Habitat Connectivity for Multiple Rare, Threatened, and Endangered Species on and Around Military Installations. The purpose of this project is to develop a spatially explicit decision-support system for identifying lands that enhance the connectivity for multiple threatened and endangered species. The conservation value of the landscape is dependent upon how well a species may disperse through it as well as the landscapes overall contribution to habitat protection and connective value. One method to quantify the dispersal ability is with movement resistance surfaces. These models place a value on movement through a landscape based on various parameters including vegetation canopy structure. In an effort to model the vegetation canopy structure this project will derive landscape characteristics from LiDAR (Light Detection and Ranging) data.

LiDAR is highly accurate method to remotely characterize the landscape. By emitting between 5,000 – 50,000 laser pulses per second in a scanning array from an airborne sensor LiDAR data is able to accurately measure the distance to objects on the ground.

Figure 1: LiDAR over hillshade in the Ft.Brag Area

One difficulty of working with LiDAR data is the shear volume of data returned and how to interpret the results. The aim of this project is to create a set of tools which will allow the user to create a DEM from one to multiple asci or text files, clip to smaller areas of interest, create contours, slope, and model the potential wetland areas.

  1. RESEARCH PROBLEM

LiDAR data is often difficult to work with as it consists of values for multiple returns. Post-processed LiDAR data is in .xyzi,.asci, or .txt format. The data consists of a list of points with their x and y coordinatepairs, the z-value (vertical elevation) and an intensity of return value, i. Currently in ArcGIS 9.1 there is not a way to batch process ascii files into GRID fomat. This project aims to create such a process in a user-friendly interface. An interesting question to ask to attempt to answer is how we can better understand the vegetative structure of a plant community using LiDAR to measure the height of the canopy and understory. Currently it is cumbersome to work with LiDAR in ArcGIS and this project hopes to remedy this.

Another research problem is how to most accurately map salamander habitat. The LiDAR flown for Ft.Brag should be very useful for identifying small depressional areas which may be classified as wetlands. However, currently most terrain-modeling or hydrology network modeling software fill these depressions, or sinks, early on in order to create a smooth surface with continuous flow. This projects aim is to allow the user to identify these potential wetland areas for further field validation and export them into a coverage or shapefile.

  1. DATA AVAILABILITY

LiDAR data for this project is available from two sources. One set of data is available through the North Carolina Flood Mapping Program (NCFMP). The other set of data was collected for the SERDP project at Ft.Brag. There is no cost to obtaining and using either of these readily available sets of data.

North Carolina Floodplain Mapping Data

North Carolina is the first state to be designated as a Cooperating Technical State (CTS) with the Federal Emergency and Management Agency (FEMA). This means that North Carolina assumes primary responsibility for Flood Insurance Rate Maps. In an effort to both update these maps and enhance their accuracy, the NCFMP is creating digital elevation models from LiDAR for the entire state of North Carolina. Currently 84 counties have this data available. The third and final phase of this project will be completed by the end of the year. The N.C. Floodplain Mapping Information System ArcIMS viewer at access to this data. What isn’t available online is available directly from the agency by Charity Peterson at (919) 715-5711 x113.

Three datasets are available for download. Bare-earth data with an average spot-spacing of five meters was created in shapefile format. A triangulated irregular network (TIN) can be created from this data and represents the most accurate depiction of the earth’s surface. From the other two datasets, a 20 and 50 foot DEM can be created. This data is partitioned into a 10,000 ft tiles and stored in asci text format.

Ft.Brag SERDP Data

For the SERDP LiDAR data, the spot-spacing (average distance between readings) is 1 meter, with 95% of spot heights within 18.5 cm of true heights. For this dataset first and last returns for bare-earth and extracted features are available. This data could be resampled for even spacing or kriging can be implemented to create an even spaced DEM. Also available is a processed bare-earth 1 m DEM. This data was processed by the vendor, Airborne1.

  1. APPLICATION DESIGN and OUTPUT FORMAT

The application will be written in either Visual Basic for Application (VBA) or Python and executed in ESRI’s ArcGIS. A GUI (graphical user interface) will be created which will allow the user to input multiple asci or text files to create a DEM. The user will then be allowed to clip the resulting grid if desired and then perform a series of spatial analysis on the grid including but not limited to slope, hillshade, and contour creation.

Figure 2: Flow Chart showing the possible calculations the user will be

able to make starting from a single or multiple LiDAR text files.

  1. INERPRETATION OF RESULTS

This application will be designed for a two audiences, so the interpretation of the results will vary accordingly. Specifically related to the SERDP project, the results of the sink digitization will need to be verified in the field. A protocol for this procedure is set forth in the SERDP project proposal. Generally, this protocol entails visiting roughly 20 known amphibian breeding habitats and setting up ground-water monitoring wells as well as stream gauges and then combining these results with the bare-earth DEM, and soil characteristics to produce inundation models. Automating the process of creating the sinks will save us time in the research process.