2006 MDEQ-FEMA Hinds County Lidar Survey

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.

MGI requested the collection of lidar data over Hinds County, MS. In response EarthData International, Inc. acquired the data on April 11 and 12, 2006 using its aircraft with tail number N62912. Lidar data was captured using an ALS50 lidar system, including an inertial measuring unit (IMU) and a dual frequency GPS receiver. An additional GPS receiver was in constant operation over a temporary control point set by EarthData International, Inc. at Hawkins Airport which was later tied into a local network by Waggoner Engineering, Inc. During the data acquisition, the receivers collected phase data at an epoch rate of 1 Hz. The solution from Hinds County, MS was found to be of high integrity and met the accuracy requirements for the project. 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 A, section 8, Airborne Light Detection and Ranging (LIDAR) Surveys. Airspeed - 160 knots Laser Pulse Rate - 32900 kHz Field of View - 45 degrees Scan Rate - 18 Hz | Source Geospatial Form: model | Type of Source Media: firewire

Waggoner Engineering, Inc., under contract to EarthData International, Inc. successfully established ground control for Hinds County, MS. A total of 16 ground control points in Hinds County, MS were acquired. GPS was used to establish the control network. The horizontal datum was the North American Datum of 1983 (NAD83). The vertical datum was the North American Vertical Datum of 1988 (NAVD88). | Source Geospatial Form: diagram | Type of Source Media: electronic mail system

EarthData has developed a unique method for

processing lidar data to identify and remove elevation

points falling on vegetation, buildings, and other

aboveground structures. The algorithms for filtering data

were utilized within EarthData's proprietary software and

commercial software written by TerraSolid. This software

suite of tools provides efficient processing for small to

large-scale, projects and has been incorporated into ISO

9001 compliant production work flows. The following is a

1. The technician performs calibrations on the data set.

2. The technician performed a visual inspection of the

data to verify that the flight lines overlap correctly. The

technician also verified that there were no voids, and that

the data covered the project limits. The technician then

selected a series of areas from the data set and

inspected them where adjacent flight lines overlapped.

These overlapping areas were merged and a process

which utilizes 3-D Analyst and EarthData's proprietary

software was run to detect and color code the differences

in elevation values and profiles. The technician reviewed

these plots and located the areas that contained

systematic errors or distortions that were introduced by the

lidar sensor.

3. Systematic distortions highlighted in step 2 were

removed and the data was re-inspected. Corrections and

adjustments can involve the application of angular

deflection or compensation for curvature of the ground

surface that can be introduced by crossing from one type

of land cover to another.

4. The lidar data for each flight line was trimmed in batch

for the removal of the overlap areas between flight lines.

The data was checked against a control network to

ensure that vertical requirements were maintained.

Conversion to the client-specified datum and projections

were then completed. The lidar flight line data sets were

then segmented into adjoining tiles for batch processing

and data management.

5. The initial batch-processing run removed 95% of points

falling on vegetation. The algorithm also removed the

points that fell on the edge of hard features such as

structures, elevated roadways and bridges.

6. The operator interactively processed the data using

lidar editing tools. During this final phase the operator

generated a TIN based on a desired thematic layers to

evaluate the automated classification performed in step 5.

This allowed the operator to quickly re-classify points from

one layer to another and recreate the TIN surface to see

the effects of edits. Geo-referenced images were toggled

on or off to aid the operator in identifying problem areas.

The data was also examined with an automated profiling

tool to aid the operator in the reclassification.

7. The final bare earth was written to an LAS 1.0 format and

also converted to ASCII.

8. The point cloud data were delivered in LAS 1.0 format.

EarthData utilizes a combination of proprietary and COTS

processes to generate intensity images from the lidar

data. Intensity images are generated from the full points

cloud (minus noise points) and the pixel width is typically

matched to the post spacing of the lidar data to achieve

the best resolution. The following steps are used to

1. Lidar point cloud is tiled to the deliverable tile layout.

2. All noise points, spikes, and wells are deleted out of the

tiles.

3. An EarthData proprietary piece of software, EEBN2TIF

is then used to process out the intensity values of the lidar.

At this point, the pixel size is selected based on best fit or

to match the client specification if noted in the SOW.

4. The software then generates TIF and TFW files for each

tile.

5. ArcView is used to review and QC the tiles before

delivery.

6. The lidar intensity data were delivered in TIF format.

It should be noted that the breaklines developed for use in

the H&H modeling should not be confused with traditional

stereo-graphic or field survey derived breaklines. The

elevation component of the 3D streamlines (breaklines) is

derived from the lowest adjacent bare earth lidar point

and adjusted to ensure that the streams flow downstream.

The best elevation that can be derived for the 3D

streamlines will be the water surface elevation on the date

that the lidar data was acquired. The elevations in the 3D

streamlines will not represent the underwater elevations

for streams due to the fact that lidar data cannot collect

bathymetry information.

Watershed Concepts and EarthData have done

considerable research generating breaklines from lidar

data. Current H&H modeling practices rely heavily on

mass points and breaklines to create a realistic TIN

surface for hydrologic and hydraulic modeling. Lidar data

consists only of points, which are not suited to defining

sharp breaks on terrain. The problem is most pronounced

across stream channels, where lidar is not able to define

the stream banks clearly. Furthermore lidar does not

reflect off water; therefore, no reliable elevation points will

exist within the stream channel itself. The TIN surface

generated from lidar data alone is unsuitable for H&H

modeling.

Watershed Concepts engineers have studied the

sensitivity of the 100-year flood boundary to the definition

of stream channel geometry. The surface created with

both lidar points and breaklines improves channel

definitions for hydraulic cross section takeoffs and better

defines the stream invert. It is not necessary to create

breaklines on the top and bottom of stream banks; minor

modifications to the cross sections and stream inverts can

be made based on field survey data as necessary. In the

100-year flood, most of the flooded cross sectional area

occurs in the overbank; therefore, creating a more refined

channel definition from the lidar data is not cost effective.

The lidar TIN is used simply as the basis for the overbank

definition.

Our research indicates that breaklines are required at the

stream centerline for smaller streams with widths less than

50 feet. For larger streams (widths greater than 50 feet,

breaklines are needed on the left and right water edge

lines. Collection of photography and stereo compilation of

the breaklines is not cost-effective for this purpose.

Watershed Concepts and EarthData have developed

techniques to synthesize 3D breaklines using digital

orthophotos and lidar data. These breaklines can be

digitized in 2D from orthophotos, approximating the stream

bank in areas of significant tree overhang. A bounding

polygon, created from the edge of bank lines, is used to

remove all points within the channel. Automatic processes

assign elevations to the vertices of the centerline based

on surrounding lidar points. The lines are then smoothed

to ensure a continuous downhill flow. Edge-of-bank

vertices are adjusted vertically to match the stream

centerline vertices. A new TIN can then be created from

the remaining lidar points and newly created breaklines.

The new TIN clearly defines the stream channel.

For this project, breaklines were generated in the matter

described above for all streams draining greater than

approximately one square mile. 2D lines defining the

centerline and banks of those streams were manually

digitized into ESRI shape file format from 2005 imagery.

The streamlines were then processed against the bare

earth lidar as described above. The new 3D lines were

then viewed in profile to correct any anomalous vertices or

remove errant points from the lidar DTM, which cause

unrealistic "spikes" or "dips" in the breakline. The 3D

breaklines were delivered in ESRI shapefile format.

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

elevation and intensity measurements. The data were in Mississippi State Plane West (2301, feet) coordinates and NAVD88

(Geoid03) vertical datum (feet). OCM performed the following processing for data storage and

Digital Coast provisioning purposes:

1. The data were converted from State Plane (2301) coordinates to geographic coordinates.

2. The data were converted from NAVD88 (orthometric) heights to GRS80 (ellipsoid) heights using Geoid03.

3. 8 laz tiles had coordinates falling outside of the header boundary. These tiles were re-tiled to remove any data points falling outside of the header boundary.

4. All laz tiles were received with all points classed as Class 1 (unclassified); the laz tiles were put through lasground.exe (lastools) which uses an algorithm to define which points fall as class 2 (Ground).

5. The data were sorted by time and zipped to laz format.