LiDAR Elevation Data Collection - Putnam County, NY, 2008 (NYSDEC)

Create custom data files by choosing data area, product type, map projection, file format, datum, etc.

Simple download of data files.

This graphic shows the lidar coverage for the 2012 lidar project covering Brooks Camp of Katmai National Park in Alaska.

Date that the source FGDC record was last modified.

Converted from FGDC Content Standards for Digital Geospatial Metadata (version FGDC-STD-001-1998) using 'fgdc_to_inport_xml.pl' script. Contact Tyler Christensen (NOS) for details.

Partial upload of Positional Accuracy fields only.

Partial upload to move data access links to Distribution Info.

Data collection:

Using Sanborn's Optech ALTM 2050 (SN# 02D140) Light Detection And Ranging (LiDAR) system, 18 flight lines of standard density (1.4 meter ground sample distance) data were collected over areas in Putnam County, NY (approximately 111 square miles). Two returns were recorded for each laser pulse along with an intensity value for each return. The data acquisition occurred in 5 missions between April 02, 2008 , and April 08, 2008. During the LiDAR campaign, the Sanborn field crew conducted a GPS field survey to establish final coordinates of the ground control stations for final processing of the base-remote GPS solutions.

Airborne GPS processing:

Airborne GPS data was differentially processed and integrated with the post processed IMU data to derive a smoothed best estimate of trajectory (SBET). The SBET was used to reduce the LiDAR slant range measurements to a raw georeferenced point.

Airborne GPS is differentially processed using the GrafNAV v7.50 software by Waypoint Consulting of Calgary, Alberta, Canada.

The georeferenced point cloud was generated in REALM v3.5.4 software by Optech Inc., of Vaughan, Ontario, Canada.

IMU data processing:

IMU data provides information concerning roll, pitch and yaw of collection platform during collection event. IMU information allows the pulse vector to be properly placed in 3D space allowing the distance from the aircraft reference point to be properly positioned on the elevation model surface. IMU data is processed using the POSPac v4.2 software by Applanix Corporation of Richmond Hill, Ontario, Canada.

LiDAR point classification

First data in areas of overlapping scans is classified to separate overlap points into a separate class; this results in a more uniform point density in the point cloud from which the ground points will be extracted. This 'cut-overlap' process is performed with the aid of the aircraft trajectory (reduced to the sensor's trajectory) and the scan angle value within the LAS data. For this project, the value used for cutting the overlap was 25 degrees. This means that data from a single scan greater than 12.5 degrees off nadir gets classified to class 12, Overlap. Using the trajectory and embedded angle information, the software is able to properly reclassify the overlap points so that the remaining point cloud is edge-matched, i.e. there are no data voids within the point cloud with class 12 overlap points turned off.

Then comes the bare earth extraction from the rest of the point cloud. The extraction is the result of a morphological processing routine run in TerraScan: a set of user-defined distances and angles are used by the software to build an initial ground surface from established 'aerial low seed points'; iterative application of filters populates further the 'Bare Earth Point' class (Classification value 2) with those adjoining points judged not to be above-ground objects. The rest of the points stays in the Classification value 1, Unclassified Point class.

The classification algorithms used on the LiDAR point cloud involved several iterative steps including the removal of low points and other outlyers, the culling of overlap data, and finally the classification of the LiDAR Bare Earth Points. This process begins with automated routines and ends with a 100% manual edit and QC check of the data. Once the data set classification accuracy was deemed sufficient and no quality issues were found, a final vertical accuracy assessment was performed on the bare-ground class of the LiDAR.

The classification and quality control (QC) of LiDAR data is carried out using TerraScan software v. 8.003 by Terrasolid Limited of Helinski, Finland.

Output LAS files

The tiling and final LAS file creation was performed using LiDAR CuePac v5.0 from GeoCue Corporation of Madison, Alabama, USA. By client specification, the LiDAR point cloud data were cut to 750 x 750 meter tiles (0.56 sq km, 0.2 sq mi) tiles. The naming convention used for the tiles is based on the truncated grid coordinate at the point of origin of the tile, the Southwestern (lower left) corner, with an "u_" prefix.

Data Validation:

The final LiDAR LAS 1.1 was created in UTM Zone 18 North, referenced to NAD83 and NAVD 88, in meters. The final LiDAR DEM was verified against FEMA checkpoints in order to perform a redundancy check against the GPS solutions. These accuracy checks also verified that the data meets the guidelines outlined in FEMA's Guidelines and Specifications for Flood Hazard Mapping Partners and Appendix 4B, Airborne Light Detection and Ranging Systems.

LiDAR Network Overview:

NYS CSCIC contracted with Sanborn for the collection and production of LiDAR data over New York's Putnam County. In response, Sanborn acquired the data in the Spring of 2008.

Three dual-frequency GPS (Global Positioning System) units were used in this project; two on the ground operating over control stations and one on the aircraft. All three units operated concurrently during the LiDAR acquisition capturing raw positional information at an epoch rate of 1 second. The raw GPS data was differentially corrected using post-production kinematic GPS processing software to provide refined GPS solutions for proper georeferencing of the aircraft and LiDAR system. During the LiDAR acquisitions the GPS base stations were set up over National Geodetic Survey (NGS) control stations ARP (PID LX1523) located at the Dutchess County Airport and Fahnestock (PID AB3872) located southeast of Beacon, New York.

A ground control network was surveyed including 2 horizontal 0 order NGS stations: ARP (PID LX1523) and Fahnestock (PID AB3872), as well as 2 vertical First order NGS stations: 851 8924 D TIDAL (PID LX1848), and 851 8924 C TIDAL (PID LX1849). A minimally and fully constrained network adjustment was performed on these stations and the resulting final adjusted coordinates were used to post-process the airborne GPS and IMU data. The source information on the NGS control stations can be found by entering the PID information on the NGS website (http://www.ngs.noaa.gov/cgi-bin/ds_pid.prl).

A second network adjustment was performed after it was discovered the wrong control stations were constrained during the first adjustment. For the second network adjustment, station Fahnestock was constrained horizontally, station ARP was constrained both horizontally and vertically, and station 851 8924 D TIDAL was constrained vertically. The second adjustment yielded coordinates that differed from the first adjustment coordinates primarily in the vertical component. By the time the second network adjustment was computed, the LiDAR post-production process had advanced significantly and the LAS binary data was already tiled and classified in the final coordinate system. The solution to this problem was to derive a set of 3-D transformation parameters going from the first adjusted coordinates to the second adjusted coordinates. The transformation parameters were then applied to the tiled LAS data on a point by point basis. The process to derive the transformation parameters was as follows: The four first and second adjustment geographic coordinates were converted to UTM grid coordinates using CORPSCON v.6.0. The vertical input was referenced to GRS 80 and the vertical output was referenced to NAVD88 using Geoid03. Next, these two sets of four grid coordinates were read into TerraScan's 'Derive Transformation' utility (version 8.003) which used a least squares adjustment to create a set of 3-D rotate and translate transformation parameters. The resultant parameters were then applied to the entire LAS point cloud using TerraScan's transformation routine. The largest magnitude of change in the transformation was in the vertical component with a change of -0.072 meters. A text document named "3D_Translate_Rotate_Transformation" details the transformation paramters and is on file at Sanborn, at NYSDEC and at NY CSCIC.

Independent Accuracy Assesment and Quality Asurance

PUT HERE SUMMARY OF QA PROCESS

Dataset copied.

The NOAA Office for Coastal Management (OCM) received the files in laz format from USGS via an FTP online repository. The files contained lidar elevation and intensity measurements. The data were in State Plane Zone 2900, NAVD88 (orthometric) heights in feet. OCM performed the following processing for data storage and Digital Coast provisioning purposes:

1. The data were converted from UTM coordinates to geographic coordinates.

2. Erroneous elevations were removed.

3. Class 11 points were reclassed to class 15.

4. The data were converted from NAVD88 (orthometric) heights in feet to GRS80 (ellipsoid) heights in meters using Geoid 09.

5. The LAS data were sorted by latitude and the headers were updated.