If you’ve worked with lidar data, you know that point density (measured in points per square meter, for example) is a key quality metric. Until now, density has been evaluated in a very simplistic way: by taking the total number of points in a dataset or tile and dividing them by the area. This provides an average points-per-area and ensures it meets the minimum specifications (e.g., USGS Lidar Quality Level 2 requires ≥2 points/m² on average). However, an average can be misleading – you could meet the required density overall even if some spots have very few points and others have a lot. In other words, the uneven distribution of points is not flagged by the old method.
To address this, ASPRS (American Society for Photogrammetry and Remote Sensing) has approved a new lidar density validation method based on Voronoi polygons, referred to as the Voronoi Method. Without delving too deeply into computational geometry, a Voronoi-based approach essentially examines the spacing of points: it partitions the map into cells around each lidar point and calculates the area of those cells. If a lidar dataset has consistent coverage, each point’s Voronoi cell will be small and uniform. Conversely, if there are gaps, some cells will be large (indicating sparse areas). The new method uses this to create a thematic raster map of density and identify any locations that fall below the required density threshold. Instead of just one average number, you get a detailed view (pass/fail on a cell-by-cell basis) of whether density requirements are met across the entire project area.
Why does the Voronoi Method matter?
It makes quality assessment more rigorous and ensures there are no blind spots in lidar coverage. Data buyers (like USGS or state agencies) want to know that their minimum density is achieved everywhere, not just on average. ASPRS has tested and vetted this approach through public review, and it was formally approved as the new standard for evaluating point density. We anticipate that USGS will incorporate it into their lidar specifications by the end of 2025.
Sanborn’s Preparations
We didn’t wait for the mandate to arrive. Sanborn’s development team has been working on a custom software tool to perform Voronoi-based density analysis internally. Our goal is to run this check on all our lidar deliverables, even before it’s officially required, so that we can ensure and demonstrate superior data quality. In fact, our leadership has stated that as soon as our tool is ready, we will start including this validation in project workflows and even mention it in proposals to give clients added assurance. For clients, this means you can be confident that any lidar data from Sanborn will meet or exceed forthcoming standards for uniform density. Whether or not the USGS specification is updated to this standard, we’ll have you covered with more rigorous density validation.
Changes to Vegetated Vertical Accuracy (VVA) Standards
Another standard update comes in the realm of lidar accuracy reporting, specifically Vegetated Vertical Accuracy (VVA). VVA is a statistic that measures the vertical error of lidar points in vegetated areas (for example, under tree canopy). It’s defined as the 95th percentile error – in simpler terms, it tells you how bad the worst-case errors are (up to 95% of the points) in areas with vegetation. This complements the Non-Vegetated Vertical Accuracy (NVA), which measures error on open, hard surfaces, such as parking lots or bare ground. Together, NVA and VVA provide a more comprehensive picture of a lidar dataset’s vertical accuracy across different land cover types.
What’s changing is how VVA is treated in the standards and project requirements. Under the 2014 ASPRS accuracy standards, every lidar project needed to collect a certain number of checkpoints in vegetated areas and ensure the VVA met a maximum error threshold (it was essentially a pass/fail criterion, just like NVA). This was challenging – to get statistically significant results, it required a lot of extra field survey checkpoints (which means time and cost). Over the past decade, the community observed that VVA is influenced by many factors beyond just the lidar sensor (for example, GPS positioning errors of the checkpoints, the difficulty of measuring ground under dense foliage, etc.). In many cases, a lidar sensor could be performing well, yet the VVA could “fail” due to reasons outside the sensor’s control.
The new approach to VVA (2024 and beyond):
VVA is no longer a pass/fail requirement in the latest guidelines. Instead, it’s treated as a reported metric for informational purposes. Projects still need to collect checkpoints in vegetated areas and calculate the VVA, but there isn’t a hard threshold you must achieve to “pass”. What it means for data buyers: Data buyers (like government agencies) still get transparency. You will see the VVA statistic in reports and know how the data performs in vegetation, but it will be presented alongside context rather than as an absolute “must pass” number. It acknowledges that lidar sensors today are very capable and that a slightly high VVA in a highly dense forest, for example, doesn’t mean the dataset is bad – it might just reflect the reality of that environment.
Sanborn will continue to collect the required checkpoints and report VVA for all projects. We welcome discussions with clients about what level of VVA is acceptable for their needs. The key is that VVA is now informational; it supplements the Non-vegetated Vertical Accuracy but is not itself a contract performance metric.
The Upcoming LAS 1.5 Format – Fixes and Timeline
Anyone dealing with lidar data is familiar with the ASPRS LAS file format (las files), which is the standard way to store point clouds. The current version in wide use is LAS 1.4, which has been around for many years. Now, LAS 1.5 is on the horizon – and while file format updates don’t sound exciting, this one is important to enable the other changes we’ve been discussing.
Why introduce a new LAS 1.5?
It turns out LAS 1.4 has two critical technical flaws that will become problematic soon. First, the coordinate reference system information in LAS 1.4 uses an old standard (WKT v1) that does not support time-dependent coordinate systems. This is directly relevant to NATRF2022, which, as we noted, has a time component. LAS 1.4 can’t natively describe NATRF2022 coordinates, necessitating an update to handle the new datum properly. The second issue is with timestamps in lidar data: LAS encodes time (for each point, often as GPS time) in a field called “Adjusted Standard GPS Time.” With LAS 1.4, that time field is gradually losing precision as the numbers grow (we’re accumulating seconds since a reference). It’s projected that this will cause precision loss or even timestamp failures in the next few years. Estimates show it reaching a breaking point within <5 years if not addressed. Given that many lidar collections rely on precise timing (especially for systems like GPS/IMU integration), this is a ticking technical debt.
LAS 1.5 to the rescue
The new format will likely fix these issues by adopting an updated coordinate metadata standard (supporting WKT2 for modern datums) and modifying the way timestamps are stored with sufficient precision. The ASPRS has already drafted LAS 1.5 changes and put them out for public review and comment. This is the first step in ratifying the new version. Once the ASPRS formally approves LAS 1.5, we expect agencies like USGS to update their specifications to require it for new data deliveries (especially as NATRF2022 goes into use). How soon will this happen? From what we’ve learned, the timeline could be in the order of 6 to 9 months after approval for major players to begin adopting LAS 1.5. Since the process is already underway in 2025, it’s possible that by 2026, many projects will be delivering lidar data in LAS 1.5 instead of 1.4.
Sanborn’s production team has been tracking this closely (our internal experts are even involved in ASPRS committees). We’re ready to switch over to LAS 1.5 as soon as it becomes the official standard for deliverables. In practical terms, if you have a project kicking off late this year or next year, don’t be surprised if LAS 1.5 is specified – and rest assured, we’ll be prepared to deliver your data in that format without issue. Think of LAS 1.5 as a timely update that “future-proofs” lidar datasets for the new coordinate systems and the next decade of collections. The moment ASPRS finalizes the specifications and tools from LAStools, TerraScan, and Esri support the new format, we’ll integrate it into our workflow.
