2022 City of Philadelphia Lidar: Philadelphia, PA

Create custom data files by choosing data area, product type, map projection, file format, datum, etc. A new metadata will be produced to reflect your request using this record as a base. Change to an orthometric vertical datum is one of the many options.

Bulk download of data files in LAZ format, geographic coordinates, orthometric heights. Note that the vertical datum (hence elevations) of the files here are different than described in this document. They will be in an orthometric datum.

Link to view the point cloud (using the Entwine Point Tile (EPT) format) in the 3D Potree viewer.

Link to custom download, from the Data Access Viewer (DAV), the raster Digital Elevation Model (DEM) data that were created from this lidar data set.

The Data Access Viewer (DAV) allows a user to search for and download elevation, imagery, and land cover data for the coastal U.S. and its territories. The data, hosted by the NOAA Office for Coastal Management, can be customized and requested for free download through a checkout interface. An email provides a link to the customized data, while the original data set is available through a link within the viewer.

This graphic displays the footprint for this lidar data set.

Entwine Point Tile (EPT) is a simple and flexible octree-based storage format for point cloud data. The data is organized in such a way that the data can be reasonably streamed over the internet, pulling only the points you need. EPT files can be queried to return a subset of the points that give you a representation of the area. As you zoom further in, you are requesting higher and higher densities. A dataset in EPT will contain a lot of files, however, the ept.json file describes all the rest. The EPT file can be used in Potree and QGIS to view the point cloud.

Link to the Merrick lidar report.

Link to the hydro breaklines.

Lidar acquisition was collected using fixed wing aircraft and two Optech Galaxy T2000 lidar sensors staging from a variety of airports around the project area. Up to eight return values are recorded for each pulse which ensures the greatest chance of ground returns in a heavily forested area. Lidar data collection was accomplished on March 29, 2022 (dates listed are in local time NOT UTC). Each mission represents a lift of the aircraft and system from the ground, collects data, and lands again. Multiple lifts within a day are represented by Mission A, B, C, and D.

GNSS/IMU Data

A five-minute IMU initialization is conducted on the ground, with the aircraft engines running, prior to flight, to establish fine alignment of the IMU. In air IMU calibration maneuvers were performed at the beginning and ending of all mission collections to ensure the best forward and reverse trajectory processing using the highest quality IMU calibration. During the data collection, the operator recorded information on log sheets which includes weather conditions, lidar operation parameters, and flight line statistics. Data is sent back to the main office for preliminary processing to check overall quality of GNSS / IMU data and to ensure sufficient overlap between flight lines. Any problematic data may be reflown immediately as required. The airborne GNSS data was post-processed using Applanix POSPac Mobile Mapping Suite version 8.x. A fixed bias carrier phase solution was computed in both the forward and reverse chronological directions. Whenever practical, lidar acquisition was limited to periods when the PDOP was less than 4.0. PDOP indicates satellite geometry relating to position. Generally, PDOPs of 3.0 or less result in a good quality solution, however PDOPs between 3.0 and 5.0 can still yield good results most of the time. PDOPs over 6.0 are of questionable results and PDOPs of over 7.0 usually result in a poor solution. Usually as the number of satellites increase the PDOP decreases. Other quality control checks used for the GNSS include analyzing the combined separation of the forward and reverse GNSS processing from one CORS station and the results of the combined separation when processed from two different CORS stations. An analysis of the number of satellites, present during the flight and data collection times, is also performed. The GNSS trajectory was combined with the raw IMU data and post-processed using POSPac Mobile Mapping Suite version 8.x. The SBET and refined attitude data are then utilized in the Optech LMS lidar processing software to compute the laser point-positions. The trajectory is combined with the laser range measurements to produce lidar point cloud data.

GNSS Controls

Virtual Ground GNSS Base Station(s) were used to control the lidar airborne flight lines. Post processed Trimble CenterPoint RTX correction service is a high-accuracy, satellite-delivered global positioning service. This technology provides high accuracy GNSS positioning without the use of traditional reference station based differential RTK infrastructure and delivers very high cm level accuracy. In addition, CORS are at times used to further QC or enhance the airborne GNSS solution.

Lidar Calibration

Merrick takes great care to ensure all lidar acquisition missions are carried out in a manner conducive to postprocessing an accurate dataset. Proper Airborne GNSS surveying techniques are always followed including pre- and post-mission static initializations. In-air IMU alignments (figure-eights) are performed both before and after on-site collection to ensure proper calibration of the IMU accelerometers and gyros. A minimum of one cross-flight is planned throughout the project area across all flightlines and over roadways where possible. The cross-flight provides a common control surface used to remove any vertical discrepancies in the lidar data between flightlines. The cross-flight is critical to ensure flightline ties across the project area. The areas of overlap between flightlines are used to boresight (calibrate) the lidar point cloud to achieve proper flightline to flightline alignment in all three axes. Each lidar mission flown is accompanied by a hands-on boresight in the office. After boresighting is complete a detailed statistical report is generated to check relative and absolute accuracies before filtering of lidar begins.

The lidar filtering process encompasses a series of automated and manual steps to classify the boresighted point cloud data set. Each project represents unique characteristics in terms of cultural features (urbanized vs. rural areas), terrain type and vegetation coverage. These characteristics are thoroughly evaluated at the onset of the project to ensure that the appropriate automated filters are applied and that subsequent manual filtering yields correctly classified data. Data is most often classified by ground and unclassified, but specific project applications can include a wide variety of classifications including but not limited to buildings, vegetation, power lines, etc. A variety of software packages are used for the auto-filtering, manual filtering and QC of the classified data.

Merrick used the ASPRS LAS Specification Version 1.4 R15 (ASPRS, 2011, published 09 July 2019), Point Data Record Format 6 for this project and classified the lidar point cloud in accordance with the following classification classes and bitflags.

Merrick has developed several customized automated filters that are applied to the lidar data set based on project specifications, terrain, and vegetation characteristics. A filtering macro, which may contain one or more filtering algorithms, is executed to derive LAS files separated into the different classification groups as defined in the ASPRS classification table. The macros are tested in several portions of the project area to verify the appropriateness of the filters. Often, there is a combination of several filter macros that optimize the filtering based on the unique characteristics of the project. Automatic filtering generally yields a ground surface that is 85 - 90% valid, so additional editing (hand-filtering) is required to produce an accurate ground surface. Lidar data is next taken into a graphic environment using MARS to manually re-classify (or hand-filter) noise and other features that may remain in the ground classification after auto filter. A cross-section of the post auto-filtered surface is viewed to assist in the reclassification of non-ground data artifacts. The following is an example of re-classification of the non-ground points (elevated features) that need to be excluded from the true ground surface. Certain features such as berms, hilltops, cliffs and other features may have been aggressively auto-filtered and points will need to be re-classified into the ground classification. Data in the profile view displays non-ground (Unclassified, class 1) in grey and ground in brown/tan (Class 2). At this point, individual lidar points from the original point cloud have now been parsed into separate classifications.

After hand-filtering has been completed and quality checked, a Checkpoint Report is generated to validate that the accuracy of the ground surface is within the defined accuracy specifications. Each surveyed ground check point is compared to the lidar surface by interpolating an elevation from a Triangulated Irregular Network (TIN) of the surface. The MARS derived report provides an in-depth statistical report, including an RMSE of the vertical errors; a primary component in most accuracy standards and a statistically valid assessment of the overall accuracy of the ground surface.

Hydro - flattening breaklines are captured per the USGS-NGP Lidar Base Specification 2021, Revision A. Final hydro-flattened breaklines features are appropriately turned into polygons (flat elevations) and polylines (decreasing by elevation) and are used to reclassify ground points in water to water (Class 9). The lidar points around the breaklines are reclassified to ignored ground (Class 20) based on the planned collected point density. The next step in the process is the hydro-flattening breakline collection required for the development of the hydro-flattened DEMs. Merrick will capture hydro-flattening breaklines for waterbodies greater than or equal to approximately eight-tenths (~0.8) hectare (e.g., ~100-meter diameter); double-sided streams and rivers that are greater than or equal to 30 meters in (nominal) width, and; any visible islands greater than or equal to approximately four-tenths (~0.4) hectare. Criteria for Non-Tidal Boundary Waters and Tidal Waters are assumed not applicable. No single-line streams or drainages will be collected, nor will any planimetric features that could be utilized as traditional breaklines. All downstream hydro-flattening breaklines require monotonicity (e.g., streams and rivers). Closed polygonal boundaries of water will maintain a fixed (i.e., flat) elevation.

The NOAA Office for Coastal Management (OCM) downloaded 752 las point data files from this PASDA site:

https://www.pasda.psu.edu/download/phillyLiDAR/2022/LAS/Classified_LAS/

The data were in Pennsylvania State Plane South (NAD83 2011), US survey feet coordinates and NAVD88 (Geoid18) elevations in feet. The data were classified as: 1 - Unclassified, 2 - Ground, 7 - Low Noise, 8 - Model Key Point, 9 - Water, 17 - Bridge Decks, 18 - High Noise, 20 - Ignored Ground. OCM processed all classifications of points to the Digital Coast Data Access Viewer (DAV). Classes available on the DAV are: 1, 2, 7, 8, 9, 17, 18, 20.

OCM performed the following processing on the data for Digital Coast storage and provisioning purposes:

1. The las files were converted to laz format using laszip

2. Internal OCM scripts were run to check the number of points by classification and by flight ID and the gps, elevation, and intensity ranges.

3. Internal OCM scripts were run on the las files to:

a. Convert from orthometric (NAVD88) elevations to NAD83 (2011) ellipsoid elevations using the Geoid18 model

b. Convert the laz files from Pennsylvania State Plane South (NAD83 2011), US survey feet coordinates to geographic coordinates

c. Convert the laz files from elevations in feet to meters

d. Assign the geokeys, sort the data by gps time and zip the data to database.