| Processing Steps
- GPS based surveys were utilized to support both processing and testing of LiDAR data
within FEMA designated Areas of Interest (AOIs). Geographically distinct ground points
were surveyed using GPS technology throughout the AOIs to provide support for three
distinct tasks. Task 1 was to provide Vertical Ground Control to support the aerial
acquisition and subsequent bare earth model processing. To accomplish this, survey-grade
Trimble R-8 GPS receivers were used to collect a series of control points located
on open areas, free of excessive or significant slope, and at least 5 meters away
from any significant terrain break. Most if not all control points were collected
at street/road intersections on bare level pavement. Task 2 was to collect Fundamental
Vertical Accuracy (FVA) checkpoints to evaluate the initial quality of the collected
point cloud and to ensure that the collected data was satisfactory for further processing
to meet FEMA specifications. The FVA points were collected in identical fashion to
the Vertical Ground Control Points, but segregated from the point pool to ensure independent
quality testing without prior knowledge of FVA locations by the aerial vendor. Task
3 was to collect Consolidated Vertical Accuracy CVA) checkpoints to allow vertical
testing of the bare-earth processed LiDAR data in different classes of land cover,
including: Open (pavement, open dirt, short grass), High Grass and Crops, Brush and
Low Trees, Forest, Urban. CVA points were collected in similar fashion as Control
and FVA points with emphasis on establishing point locations within the predominant
land cover classes within each AOI or Functional AOI Group. In order to successfully
collect the Forest land cover class, it was necessary to establish a Backsight and
Initial Point with the R8 receiver, and then employ a Nikon Total Station to observe
a retroreflective prism stationed under tree canopy. This was necessary due to the
reduced GPS performance and degradation of signal under tree canopy. The R-8 receivers
were equipped with cellular modems to receive real-time correction signals from the
Keystone Precision Virtual Reference Station (VRS) network encompassing the Region
1 AOIs. Use of the VRS network allowed rapid collection times (~3 minutes/point) at
2.54 cm (1 inch) initial accuracy. All points collected were below the 8cm specification
for testing 24cm, Highest category LiDAR data. To ensure valid in-field collections,
an NGS monument with suitable vertical reporting was measured using the same equipment
and procedures used for Control, FVA and CVA points on a daily basis. The measurement
was compared to the NGS published values to ensure that the GPS collection schema
was producing valid data and as a physical proof point of quality of collection. Those
monument measurements are summarized in the Accuracy report included in the data delivered
to FEMA. In order to meet FEMA budgetary requirements, 20 FVA points are necessary
to allow testing to CE95 ? 1 point out of 20 may fail vertical testing and still allow
the entire dataset to meet 95% accuracy requirements. In similar fashion, 20 CVA points
are necessary to test to CE95 as discussed above. 15 CVA points were collected with
the intention at the outset that 5 of the collected FVAs would perform double ?duty
as Open-class CVA points, to total 20 CVAs. The following software packages and utilities
were used to control the GPS receiver in the field during data collection, and then
ingest and export the collected GPS data for all points: Trimble Survey Controller,
Trimble Pathfinder Office. The following software utilities were used to translate
the collected Latitude/Longitude Decimal Degree HAE GPS data for all points into Latitude/Longitude
Degrees/Minutes/Seconds for checking the collected monument data against the published
NGS Datasheet Lat/Long DMS values and into UTM NAD83 Northings/Eastings: U.S. Army
Corps of Engineers CorpsCon, National Geodetic Survey Geoid09NAVD88. MSL values were
determined using the most recent NGS-approved geoid model to generate geoid separation
values for each Lat/Long coordinate pair. In this fashion, Orthometric heights were
determined for each Control, FVA and CVA point by subtracting the generated Geoid
Separation value from the Ellipsoidal Height (HAE) for publication and use as MSL
- Using an Optech Gemini LiDAR system, a total 111 flightlines of highest density (Nominal
pulse Spacing of 1.0m) were collected over the Concord area. A total of 405 square
miles was collected. A total of 12 missions were flown between December 2 and December
12, 2010. Two airborne global positioning system (GPS) base stations were used to
support the LiDAR data acquisition: BED A-AI5558,and ORH A-AI5600. Coordinates are
available in the Post-Flight Aerial Acquisition Report.
- Applanix software was used in the post processing of the airborne GPS and inertial
data that is critical to the positioning and orientation of the sensor during all
flights. POSPac MMS provides the smoothed best estimate of trajectory (SBET) that
is necessary for Optech's post processor to develop the point cloud from the LiDAR
missions. The point cloud is the mathematical three dimensional collection of all
returns from all laser pulses as determined from the aerial mission. Optech?s DASHMap
software and Leica?s ALS Post Processor software were used to create the Raw LIDAR
Flight Line strips. At this point this data is ready for analysis, classification,
and filtering to generate a bare earth surface model in which the above ground features
are removed from the data set. The GeoCue and TerraScan software packages are then
used for the automated data classification. Project specific macros are created to
classify the ground and to remove the side overlap between parallel flight lines.
LAS Class 2 (Ground) is used to check the surveyed control points against the Triangulated
LIDAR surface. Any bias is then removed using macro functionality within TerraScan.
Unclassified Point Cloud tiles are then created using TerraScan macro functionality.
These tiles are populated within GeoCue to ensure correct LAS versioning and LAS Header
information. LAS Class 2 is used to check the independent QC points against the Triangulated
LiDAR surface. If RMSE is not within guidelines TerraScan software is utilitzed to
remove any bias, and the check is performed again.
- Point Cloud data is manually reviewed and any remaining artifacts are removed using
functionality provided within the TerraScan and TerraModeler software packages. Additional
project specific macros are created and run within GeoCue/TerraScan to ensure correct
LAS classification prior to project delivery. Final Classified LAS tiles are created
within GeoCue to confirm correct LAS versioning and header information. In-house software
is then used to check LAS header information and final LAS classification prior to
delivery. LAS Class 2 is used to check the independent QC points against the Triangulated
- Created Multipoint feature class from Class 2 LAS files using ArcGIS 3D Analyst converion
tool LAS to Multipoint and stored results in APRS_Class_2_Bare_Earth feature dataset
within the Concord_HUC8_Bare_Earth_Data.gdb file geodatabase.
- Convert points to raster using the count option...assign one value to all data cells
with a conditional statement...fill small nodata areas with expand...reduce the extent
with shrink... vectorize with raster to polygon...clean up and create polygon for
LiDAR collection area.
- Run point file information tool on classified LAS files...measure tiles and create
polyline vector grid that covers collection area...feature to points for the label
points of the LAS Information polygon shapefile... feature to polygon using vector
grid as polyline input and LAS Info points as label points...clean up file.
- Created ESRI Terrain using class 2 multipoint feature class as mass point input and
Concord Collection Area as soft clip. This data is stored in the APRS_Class_2_Bare_Earth
feature dataset within the Concord_HUC8_Bare_Earth_Data.gdb file geodatabase.
- Created 5ft floating point raster DEM from ESRI Terrain by converting Terrain to Raster
using ArcGIS 3D Analyst. Save results as a ESRI GRID dataset.
- Created 10ft floating point raster DEM from ESRI Terrain by converting Terrain to
Raster using ArcGIS 3D Analyst. Save results as a ESRI GRID dataset.
- Create contours by Extracting by mask from the 5ft DEM using a HUC12 area. Save this
raster as HUC12 Name 5ft. Focal Statistics using Extracted 5ft DEM as input, Intermediate
Focal Raster as Output, Neighborhood should be set to weighted kernel, and the statistic
should be sum. Create contours using focal stats raster as input, output polyline
should be based on HUC12 name, Contour Interval of 2ft, Set base contour to 0.001.
Check results and store in file geodatabase under the HUC12 name feature dataset.
- Create HDEM from the Floating Point 10ft DEM derived from the Concord LiDAR data was
constructed to include all drainage areas for the Concord watershed. Used a U.S. Geological
Survey (USGS) National Hydrography Dataset (NHD) polyline featureclass containing
the stream channel locations was reviewed and compared with orthophotos to confirm
the stream channels were in the correct location. Using ArcHydro, the DEM and the
polyline shapefile were used as inputs to the DEM Reconditioning (AGREE) tool of Arc
Hydro. Arc Hydro burned the polylines into the DEM, and thus produced a hydrologically-correct
DEM (HDEM). Sinks are filled, a flow direction and flow accumulation analyses performed,
and a final stream GRID developed. The final output of the model is a stream network
shapefile. The stream network shapefile is reviewed and compared to orthophotos and
USGS NHD flow lines to confirm the stream channels are in the correct location.
- The NOAA Coastal Services Center (CSC) received the topographic files in LAS V1.2
format. The files contained lidar elevation measurements, classifications, intensity
data, return information, GPS time and scan angle. The data were received in NAD83
UTM Zone 19N (meters) and were vertically referenced to NAVD88 using the Geoid09 model.
The vertical units of the data were meters. CSC performed the following processing
for data storage and Digital Coast provisioning purposes: 1. All points in Class 11
were changed to Class 12 (Overlap) 2. The topographic las files' global encoding bit
was set to '1' to reflect use of the Adjusted Standard GPS Time format. 3. The topographic
las files were converted from a Projected Coordinate System (NAD83, UTM Zone 19N)
to Geographic coordinates (NAD83). 4. The topographic las files' horizontal units
were converted from meters to decimal degrees. 5. The topographic las files were converted
from orthometric (NAVD88) heights to ellipsoidal heights using Geoid09. 6. The data
were converted to LAZ format.
- The NOAA National Geophysical Data Center (NGDC) received lidar data files via ftp
transfer from the NOAA Coastal Services Center. The data are currently
being served via NOAA CSC Digital Coast at http://www.csc.noaa.gov/digitalcoast/.
The data can be used to re-populate the system. The data are archived in LAS or LAZ
The LAS format is an industry standard for LiDAR data developed by the American Society
of Photogrammetry and Remote Sensing (ASPRS); LAZ is a loseless compressed version
LAS developed by Martin Isenburg (http://www.laszip.org/). The data are exclusively
in geographic coordinates (either NAD83 or ITRF94). The data are referenced vertically
the ellipsoid (either GRS80 or ITRF94), allowing for the ability to apply the most
up to date geoid model when transforming to orthometric heights.