April 2005 Lidar Point Data of Southern California Coastline: Long Beach to US/Mexican Border

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

Simple download of data files.

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.

air and ground GPS files

base station coordinates Easting, Northing, HAE in NAD83, Zone 11 (Latitude, Longitude):

Seal Beach (SEAL) = 399189.009, 3733584.462, -27.9778 (N 33 44 15.0510, W 118 5 17.8191)

Dana Point (DANA) = 434087.529, 3702982.315, 52.1756 (N 33 27 51.3542, W 117 42 33.5246)

Point Loma (LOMA) = 477398.387, 3614791.668, 90.1348 (N 32 40 14.01098, W 117 14 27.79485)

| Type of Source Media: digital file

raw lidar data from ALTM 1225 (all times UTC)

09405

Pass A (Oceanside to Dana Point) = 19:19-19:36

Pass B (Dana Point to Mexico) = 19:39-20:17

Pass C (Point Loma to Mexico) = 20:28-20:34

Pass D (Mexico to Dana Point) = 20:36-21:00 and 21:12-21:35

Pass E (Oceanside to La Jolla) = 21:45-21:59

Pass F (La Jolla to Oceanside) = 22:02 to 22:14

Calibration Passes = 21:02-21:11

09805

Pass G (Oceanside to Long Beach) = 21:37-22:20

Pass H (Long Beach to Dana Point) = 22:23-22:41

Pass I (Dana Point to Long Beach) = 22:45-23:06

Pass J (Long Beach to Dana Point) = 23:12-23:29

Calibration Passes = 23:34-23:47

| Type of Source Media: digital file

GPS and XYZ-Point Data Processing

The National Geodetic Survey's PAGES-NT software was used to compute double differenced, ionospherically corrected,

static GPS solutions for each GPS base station with precise ephemeredes from the International GPS Service (IGS).

As part of the solution tropospheric zenith delays were estimated and L1 and L2 phase biases were fixed as integers.

Aircraft trajectories were estimated with respect to all base stations using National Geodetic Survey's Kinematic and

Rapid-Static Software (KARS) software. Trajectories were double-differenced, ionospherically corrected, bias-fixed GPS

solutions computed with precise IGS ephemeredes. Coordinates for base stations and trajectories were in the International

Terrestrial Reference Frame of 2000 (ITRF00). The aircraft trajectory were transformed from the ITRF00 to North American

Datum of 1983 (NAD83) using the Horizontal Time Dependent Positioning (HDTP) software (Snay, 1999)

The 1Hz GPS trajectory and 50Hz aircraft inertial measurement unit (IMU) data were combined in Applanix's POSProc

version 2.1.4 to compute an aided inertial navigation solution (INS) and a 50Hz, smoothed best estimate of

trajectory (SBET) for day 09405. On the second day of data collection (09805), due to an equipment problem, the IMU data

was recorded with random data gaps onto the ALTM1225 hard drive. Because of these data gaps, the post-processed INS and

SBET for 09805 was judged not acceptable. The 1Hz aircraft trajectory computed with KARS and the real-time, aided

INS solution from POS-AV provided better results. The SBET (09405) and KARS trajectory (09805), laser range observations,

scanner position information, and GPS/internal clock files were processed in Realm 2.27 software suite to generate lidar

data points in the Universal Transverse Mercator (UTM) projection.

Lidar point data were compared to GPS ground survey data and 1998 ATM lidar data to estimate lidar instrument calibration

parameters: roll and pitch biases, scanner scale factor, and first/last return elevation biases. An iterative,

least-squares methodology was used to estimate calibration parameters so as to minimize differences between lidar and

ground GPS data. Samples of lidar data were used to create high-resolution digital elevation models (DEM); these DEM

were inspected for horizontal or vertical anomalies.

After system calibration and initial quality control step, the adjusted lidar x,y,z-point data were generated by REALM

software and output in UTM, zone 11 with elevations being heights above the GRS-80 reference ellipsoid (HAE).

The output format from REALM 2.27 was a 9-column ASCII file containing: the second in the GPS week, easting, northing

and HAE of the first lidar return, the easting, northing and HAE of the last lidar return, and the laser backscatter

intensity of the first and last returns. Each record contains 9 columns of data: time tag (seconds in the GPS week),

first return Easting, first return Northing, first return NAVD88, last return Easting, last return Northing,

last return NAVD88, first return intensity, and last return intensity. In some cases either the first or last return

values may be missing (5 columns).

Data Classification Processing

The classification of the lidar point data was accomplished with algorithms developed at the Center for Space Research

and implemented by C++ code running on PC computer using the LINUX operating system.

The ASCII lidar files were converted into binary and concatenated into a processing database. Data were separated into

ground and non-ground points using a lower envelope follower (LEF). A lower envelope detector is an electronic circuit

used to recover information in an Amplitude Modulated (AM) signal and the concept was adapted to the problem of

extracting the ground surface from the lidar signal by creating a computer analog: the lower envelope follower (LEF)

The LEF was used to detect ground points, or seeds, which include pixels located on open ground or on the ground surface

beneath vegetation penetrated by the laser, but excludes buildings and vegetation.

The LEF operation does not detect some ground surface areas with low gradients, so detected ground pixels are augmented

using an adaptive gradient flood fill procedure. The adaptive threshold value is determined as a function of surface

roughness and topographic relief.

The adaptive gradient flood fill procedure results in a ground mask which is used to parse individual lidar points

into ground or non-ground files. In some instances, hand editing is required to ensure accuracy of the ground mask.

This includes the addition of seed points along topographic ridges or removal of buildings not detected during previous

steps.

The 9-column binary dataset was pushed through the ground mask and each lidar point is classified as either ground

or non-ground depending on its elevation with respect to a threshold above or below the estimated ground surface.

Buildings are included as non-ground points. The final ground-only data points were parsed converted back into ASCII format.

Using the GEOID99 geoid model, heights above the GRS80 ellipsoid were converted to orthometric heights with respect to

the North American Vertical Datum of 1988 (NAVD88). The final step was parsing the data into quarter quadrangles.

Processing occurred 20050404-20050728.

Created initial metadata

The NOAA Office for Coastal Management (OCM) received files in ASCII format. The files contained LiDAR intensity

and elevation measurements. OCM performed the following

processing on the data to make it available within the LiDAR Data Retrieval Tool (LDART)

1. Data returned to ellipsoid heights from NAVD88, using GEOID99.

2. Data converted to LAS format.

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