Drone topography you can check

Drone topography you can check

How land surveying and civil engineering firms use drones for topography, corridors, drainage and earthwork, and how Omniris takes a flight from surveyed ground control to checkpoint-tested surfaces and State Plane CAD files. Through all of it, the licensed surveyor decides what the data is fit for.

Summary

For surveyors and civil engineers, a drone is a fast way to measure the shape of the ground. One flight can capture what a field crew would otherwise collect shot by shot with a GNSS rover or total station. The deliverables are familiar: an orthophoto, a point cloud, a bare-earth terrain model, contours and a surface for the design software. So are the questions that come with them. Is the data tied to the project's control? Which datum and which foot is it in? How accurate is it, measured against points that were left out of the solution? And who signs for it?

In most of the United States, the answer to the last question is fixed by law. Producing topographic data from aerial imagery or LiDAR is the practice of land surveying in many states, and a licensed professional surveyor is responsible for it [8][9]. That responsibility stays with the surveyor. Omniris gives the surveyor and the engineer a complete, recorded path from photos and control to tested surfaces:

What the results are, and where the surveyor's own judgement and certification come in, is set out in What the results are, and what they aren't.

How surveyors and civil engineers use drones

Who flies, who buys, who signs

Drone topography serves the same people as any topographic survey: the engineers and designers who design on it [7] and, for an ALTA/NSPS survey, the insured owner, lender and title insurer the survey is certified to [6]. State transportation departments use drones for corridor mapping and for earthwork quantities during construction [12][27]. Two rules shape who may sell the result:

• Photogrammetric mapping is surveying in many states. NCEES, which maintains the Model Law on Surveying, revised its definition of surveying in 1995 to include photogrammetry [10]. By 2015, 22 states regulated photogrammetrically derived services or products [9]. North Carolina states the position plainly. Its surveying statute is "platform neutral", so it covers imagery and LiDAR from drones. Orthomosaics, point clouds, LAS files, terrain models, topographic and planimetric mapping and reality meshes are "derived survey products", and providing them requires a licensed North Carolina Professional Land Surveyor [8]. Rules differ by state, so the surveyor's own board decides.
• The certification is the surveyor's. An ALTA/NSPS Land Title Survey is certified to the insured, the lender and the title insurer over the surveyor's signature, seal and registration or license number. Topography is an optional item in it (see below) [6].

Drone operations themselves fall under the FAA's small-UAS rule, Part 107. Its rules include the 400 ft (122 m) altitude limit above ground level, except within 400 ft of a structure [25].

The missions

Drone LiDAR is used for ALTA topography, including for mapping under forest canopies or dense vegetation [7]. One survey firm writes that drone-based LiDAR can cover extensive terrain, "typically two acres or more", and that LiDAR and ground topography may be combined [7]. A supplemental field survey covers ground that aerial collection cannot see, such as under trees or water [11].

The standards that govern the work

Ground control and checkpoints in practice

There is no fixed count of ground control points (GCPs). Texas DOT's design-grade drone mapping specification says each project is unique and needs its own control layout that meets the required accuracy, with the spacing set by the consultant's ASPRS Certified Photogrammetrist. Control panels go in flat, open areas, are set before the flight and must be visible in the imagery, and the layout goes to the project manager before data is collected. Accuracy is then verified at checkpoints, in a report that follows the ASPRS standard [27]. Checkpoints should be spread evenly across the project area as far as possible [2], and they test the data against an independent source of higher accuracy [5]. The ASPRS standard asks for control twice as accurate as the product [2].

RTK, PPK and the base station

An RTK drone corrects its satellite position in flight from a base station or a correction network. One manufacturer specifies RTK-fix positioning of 0.4 in (1 cm) + 1 ppm horizontally and 0.6 in (1.5 cm) + 1 ppm vertically (vendor-reported) [13]. PPK achieves the same after the flight instead. Raw satellite logs recorded on the drone and on a base, without a live correction link, are processed together, and the photos are geotagged with the result [16]. Either way, the base's own position decides where everything lands. One way to fix it is to send a RINEX log of its observations to the National Geodetic Survey's OPUS service, which returns a position tied to the National Spatial Reference System [15]. Drone LiDAR systems quote their own accuracy figures. One specifies 2.0 in (5 cm) horizontal and 1.6 in (4 cm) vertical at 492 ft (150 m) with RTK fixed (vendor-reported) [14].

Datums, epochs and feet

The current US national datums are NAD83 for positions, which the State Plane Coordinate System of 1983 is built on, and NAVD88 for heights [17][26]. A drone's satellite positions are in a global frame (WGS84 / ITRF), with heights above the ellipsoid. Getting from one to the other is a geodetic transformation, not a constant offset:

• Heights. Orthometric heights come from NAD83(2011) ellipsoid heights through NGS's hybrid geoid model. GEOID18 covers the conterminous United States (NAVD88) and Puerto Rico and the U.S. Virgin Islands (their own vertical datums). Users in Alaska, Hawaii, Guam and the Northern Mariana Islands continue to use GEOID12B [20].
• The foot. The U.S. survey foot (1200/3937 m) was deprecated after December 31, 2022, in favour of the international foot (exactly 0.3048 m) [21]. The two differ by 2 parts per million, about 0.01 ft per mile, but on State Plane coordinates of millions of feet, mixing them up shifts positions by feet. NGS keeps the U.S. survey foot for State Plane 1983 coordinates in the states that specified it [26].
• Zones and conversions. NGS's Coordinate Conversion and Transformation Tool (NCAT) converts between latitude and longitude, State Plane, UTM and other systems, and transforms between datums [22].
• The datums are being replaced. NGS will replace NAD83 with four plate-fixed frames (NATRF2022 for North America) and NAVD88 with a gravity-based datum. NGS says NAD83 is misaligned to the earth's centre by about 7.2 ft (2.2 m), and NAVD88 is biased by about 1.6 ft (0.5 m) and tilted by about 3.3 ft (1 m) coast to coast [17]. NATRF2022, NAPGD2022 and SPCS2022 are marked finalized, but NGS says the preliminary beta products "are not yet final or official" and are open for public testing until at least March 2027 [18]. The beta tools were released on September 15, 2026 [19].

What drones replace, and the pain points

A drone takes over much of the shot-by-shot field topography, but not all of it. The Federal Highway Administration says drones map large areas relatively quickly compared with traditional survey and mapping, replacing boots on the ground and speeding up data collection [12]. What remains is just as clear: control and checkpoints are still surveyed on the ground [27], a field survey still covers ground hidden under trees or water [11], and the surveyor still certifies the result over a signature and seal [6]. The practical pain points are well known:

• Datum mistakes. A surface delivered in the wrong frame, the wrong height reference or the wrong foot looks right and is wrong by feet. NIST and NOAA describe errors from mixing the two feet as both common and costly [26].
• Accuracy that was never tested. Residuals at the control points show how well the map fits the points it was fitted to, not how accurate it is elsewhere. Only checkpoints test that [5].
• Hand-offs. Photos, control files, processing reports and CAD files move between field crews, processing software, CAD technicians and clients, with no single record of what was done to them.

Use case by use case: what Omniris does

A topographic survey, start to finish

Earthwork and as-builts

A saved volume boundary is measured again on every new flight of the site that covers it, giving a volume-over-time chart. Volumes are computed cell by cell over the surface model. You choose the base: a triangulated base through the boundary, a best-fit plane, the lowest or highest point, a custom elevation, or a design surface. Cut, fill and net each come with a plus-or-minus figure, and missing data is reported, not guessed. Against a design, the survey can be compared with an imported design model (IFC, DWG, DXF or DGN with its stated coordinate system), converted into the design's coordinates and heights cell by cell. It can also be compared with a LandXML, DXF or GeoTIFF surface, converted the same way: on a survey controlled in NAD83(2011), a State Plane LandXML in US survey feet with NAVD88 heights is placed on the survey's UTM map and its heights go through GEOID18. The datum rules in Coordinate systems and datums apply to volumes against a design as they do to grade checks. What the file doesn't declare (a DXF usually declares only its drawing unit) is asked in the volume tool, with the CAD export's coordinate-system picker and the design import's height-reference choices, and recorded with the measurement as stated by you. On survey-grade flights, Omniris can measure how much of a slab, pad or pavement footprint is at design height within a tolerance (±0.1 ft (3 cm) by default). The Design and Progress paper covers this in full.

Corridors

The mission planner's corridor pattern flies at least two parallel lines along the long axis of the area you draw, at 85% forward and 70% side overlap by default. Give it a width and it adds as many lines as that width needs. Long, narrow corridor flights are split into shorter sections for processing and merged, unless the photos report their positions to better than 1.6 ft (0.5 m), as RTK photos do. When a corridor crosses a State Plane zone boundary, the CAD export says so and names the zones, so the drawing is not silently projected into one zone across the line. The Mission Planning paper and the Stitching paper give the details.

Control, checkpoints and accuracy

Importing control

Omniris reads surveyed coordinates from thirteen file layouts, and suggests the layout and coordinate system from the file's contents:

Coordinates can be in latitude and longitude, UTM, a US State Plane zone (NAD83), any EPSG code or PROJ definition, or a project-local frame defined from reference points or an origin, rotation and scale. The original coordinate system is kept with the site, and each point keeps its stated horizontal and vertical accuracy. Each point also records its height reference: NAVD88, ellipsoidal, EGM96 or not stated. A mix that the stitch cannot reconcile produces a warning on the job, not a silent offset.

Control points

With ground control, the reconstruction is fitted to the surveyed points and the drone's own GPS is no longer used in the solve. A stitch controlled by fewer than three points is flagged on the stitch dialog and the job card, with the suggestion to mark more or use the points as checkpoints. After the stitch, the fit is shown point by point: the error at each control point (coloured by size), the average marking error in pixels, and every photo each point was marked in. Re-mark a poor photo and re-stitch.

The result also carries an accuracy class, set by the stated horizontal accuracy of the least accurate control point used:

These classes describe the control, not the map. They are not ASPRS accuracy classes.

Checkpoints: the independent test

A checkpoint is a surveyed point left out of the fit and used only to measure it. This is the test that the NSSDA and ASPRS standards describe [2][5]. In Omniris, checkpoints are imported and marked exactly like control points. A file can be imported as checkpoints, labels that look like checkpoints are offered as checkpoints, and any point can be switched between the two. Every switch is recorded.

• Measured on what was delivered. Each checkpoint is triangulated from its marks in the reconstruction the deliverables were made from. The height of the delivered surface model under it is read too, as a second vertical check.
• What is reported. The error east, north and up at every checkpoint. The horizontal, vertical and 3D RMSE. The mean error in each direction, so a systematic shift shows up as a bias rather than hiding in the RMSE. Which checkpoints lie outside the area the control encloses. A checkpoint whose surveyed height is not in the same height reference as the reconstruction counts horizontally only, and says why.
• Measured again without re-stitching. Checkpoints marked after delivery, or points switched later, are measured on the finished map at no charge. Every evaluation is kept, with who asked for it.
• The past does not change. Each stitch records exactly which points controlled it. Switching a point later never re-grades an earlier map, and a point that controlled a map is never reported as one of its checkpoints.
• RTK and GPS-only flights too. On a flight with no control points, the map is carried into the survey's coordinate system before it is measured. When no exact transformation exists, for example an RTK flight corrected in ITRF2014 measured against a NAD83(2011) survey, the raw offsets are shown and the results say why.

What a checkpoint summary contains. Illustrative example values.

When nothing was surveyed

• Ground reference points are points picked on a recognisable feature in satellite imagery. Unless you state otherwise, they are taken to be good to about 10 ft (3 m) horizontally and 33 ft (10 m) vertically, and a stitch that uses them is labelled basemap-grade.
• Align to a previous flight. A re-flight can be aligned to an earlier stitch of the same site, so the two line up for comparison without marking anything. The alignment page reports how closely the new surface matched the earlier one, and the result inherits the earlier flight's accuracy.

The stitch report

Every finished stitch files a report, as a web page and a PDF. It covers the flight (photos submitted and used, capture window, camera, resolution, area in acres) and the quality grades. It lists the control errors point by point and, when there are checkpoints, their measured accuracy first. It states the limits in plain sentences. Without ground control, it says that absolute accuracy cannot be stated from the flight alone. When a geometry check finds a problem with a solve, it warns that elevations, contours and volumes from that flight may be off. The Stitching paper describes the report in full.

Coordinate systems and datums

Most mistakes in drone survey data are datum mistakes, so Omniris records the frame and height reference of every flight and carries them through every export. It refuses rather than guesses.

The horizontal frame of every flight

Each stitch is stamped with its horizontal frame when it is submitted. It is one of four frames, or "unknown" with the reason: an RTK flight (its frame comes from the site's RTK settings, below), a project-local grid, NAD27, or another datum Omniris does not support.

Control surveyed in an older NAD83 realisation (HARN, NSRS2007, CORS96 and others) is recorded as such. A flight controlled in a project-local grid is exported only in its own coordinate system, untouched.

RTK flights

A photo is counted as RTK when the drone reports an RTK fix (DJI) or horizontal precision of 4 in (0.10 m) or better on both axes (Skydio X10), and a flight is treated as RTK when at least 90% of its photos are. RTK photos do not record which frame the corrections were in, so each site states it once: WGS84/ITRF, NAD83(2011), NAD83(PA11) or ETRF2000. The site also states whether the base sat on a known point or a network, or was set up on an unsurveyed point (autonomous). An autonomous base counts as uncontrolled. A single flight can override the site setting. California sites also state which NAD83(2011) epoch the corrections use, 2010.00 or 2017.50. There is no default for any of these: until they are set, RTK flights on the site are not graded against a design. Every change is recorded in the custody history and triggers re-measurement of the affected flights.

Heights

A stitch fits all its control heights in one reference, so Omniris puts them on one before the stitch:

• NAVD88 control heights are converted to NAD83(2011) ellipsoidal heights through GEOID18 in the lower 48 states and through GEOID12B in Alaska. This uses the named NGS conversion with its grid. It is allowed only for control surveyed in NAD83(2011).
• EGM96 heights are converted to WGS84 ellipsoidal heights. This is allowed only for control in WGS84 / ITRF.
• Ellipsoidal heights are used as they are.
• Heights whose reference is not stated are used as they are. The surface then follows the control's numbers. Terrain products and CAD exports made from it are labelled "heights relative — vertical datum unknown", and a CAD export needs an acknowledgement first.

If a conversion grid is missing, or an orthometric height is paired with the wrong frame, the height is not converted with a made-up value. It is refused when it is entered, or flagged on the job if the problem only shows up later. The geoid grids ship with Omniris, so conversion does not depend on an outside service.

From the drone's frame to the project's

The CAD export builds the conversion step by step rather than trusting a generic one. For a flight in WGS84/ITRF exported to State Plane or NAD83 UTM, it applies the published time-dependent transformation from ITRF2014 to NAD83(2011) (or to NAD83(PA11)) at the date the photos were taken. A generic conversion would treat the two frames as identical and drop the difference of roughly 4.9 ft (1.5 m) between them in the continental United States. A flight already in NAD83(2011) is never shifted a second time. The export dialog shows the transformation used, its stated accuracy and the epoch before you run it.

State Plane, zones and the legal foot

Omniris looks up the site's State Plane zone from NOAA's NCAT service, checking the centre of the survey and its four corners. When the corners fall in different zones, it says the site crosses a zone boundary and names the zones. The export defaults to that zone in the zone's legal unit, meaning the unit the state's law specifies. That is international feet in Arizona, Michigan, Montana, North Dakota, Oregon and South Carolina, metres in states and territories that legislated no foot, and U.S. survey feet elsewhere. You can choose any zone and any unit EPSG defines for it. U.S. survey feet and international feet are separate choices throughout Omniris, including when a design's coordinate system is stated. A design can also be placed in a project ground system using the scale factor printed on the plans.

CAD and surface deliverables

Every stitch delivers its maps as Cloud Optimized GeoTIFFs (orthomosaic, DSM, DTM) on the UTM grid for the area, with the point cloud as LAZ. Open the DTM or DSM and choose Export to CAD for the deliverables design software expects:

• Coordinate system. Choose State Plane (the default, in the zone's legal unit, or in U.S. survey feet, international feet or metres where EPSG defines the zone in that unit), UTM in metres in the flight's frame, or the surface's own coordinate system. The unit is used for X, Y and elevations alike.
• Labelled inside the file. Each file carries its coordinate system and EPSG code, unit, horizontal frame, the datum transformation and its epoch, and its height reference. DXF stores them in the drawing's custom properties, and LandXML in the file itself.
• The heights are named. Exported elevations are labelled as ellipsoidal heights on the named ellipsoid (for example "NAD83(2011) ellipsoid", including for a surface built from NAVD88 control), as EGM96 orthometric heights, or as "heights relative — vertical datum unknown". The last needs an acknowledgement before the export runs.
• Sized to the surface. A surface is never coarsened to fit a processing machine. A coarser grid is used only when you choose it. The DXF surface is limited to 1,000,000 faces and the LandXML surface to 4,000,000. The dialog tells you how fine a grid fits.
• Managed where created. Reopening the dialog lists the earlier exports of that folder with their settings. Each export is filed in its own folder beside the stitch with a custody record. CAD export is free.

Other ways out: download any original file (GeoTIFF, LAZ) for GIS, survey or CAD software, or the whole results folder as a STAC catalog that opens in QGIS. Publish map files to your ArcGIS Online or ArcGIS Enterprise organisation, or send files to a Procore project or an Autodesk Forma folder. Measurement profiles download as CSV. Contours, ponds and drainage lines from terrain analysis are GeoJSON.

Drainage and terrain analysis

Open a stitch's elevation model and choose terrain analysis to make, on demand, any of the following. The bare-earth terrain model is used by default, and the surface model can be chosen instead.

Slope and aspect are saved as map images and the rest as GeoJSON. All are filed beside the stitch with a custody record, become layers on the Omniris map, and state the height reference they were measured from, or say plainly that it is not known. A profile tool draws a cross-section of the surface along any line and downloads it as CSV. A comparison between two dates maps where the ground rose and fell, with the material added and removed and its uncertainty. It is graded for how well the two flights were aligned, because an offset between two flights looks like change.

These products are the terrain side of a drainage study. Watersheds to a design point, runoff and peak flows, and the sizing of pipes, culverts and ponds are computed in the engineer's hydrology and hydraulics software, fed by the exported terrain.

LiDAR and point clouds

Upload LAS or LAZ files from any scanner or provider into a folder and choose Process LiDAR. Before anything is charged, the dialog reads each file's point count and coordinate system. Files in different coordinate systems are refused up front. A file that does not say which system it uses is not guessed: you give its EPSG code. The job then:

• uses the provider's ground classification when the file has one, or finds the ground from the points;
• labels every point with a standard class (ground, low, medium and high vegetation, building, wire, tower, noise) and its height above ground;
• delivers a classified point cloud (COPC) with a terrain model (DTM), a surface model (DSM), a canopy height model and contours.

Heights are converted to metres but never moved to another reference. If the file says its heights are NAVD88, the results say NAVD88. If it says nothing, you can state the reference, and your answer is recorded as yours.

The point cloud opens in a browser viewer that streams only what is on screen, coloured by elevation, class, intensity, return number or natural colour, with a ruler. Where the automatic classification is wrong (a flat roof taken for ground, low brush left in the ground class), the classification editor lets you select points with a rectangle, lasso, polygon, box, brush or line, or in a side-on section view, and give them a class. Applying the edits writes a new version beside the original, which is never changed. Each version has its own custody history and can be regenerated into new terrain products. A point cloud from a photo survey can be classified the same way.

Two things matter for surveyors. The LiDAR terrain and contours are delivered as files; the CAD export and terrain analysis dialogs described above, and the volume and comparison tools, work on elevation models from photo surveys. And LiDAR results are delivered in a metre version of the file's own coordinate system: a State Plane file in feet comes back in the same zone and datum in metres, and a latitude-longitude or Web Mercator file in the UTM zone on the same datum. Omniris does not move them to another zone, datum or foot. The LiDAR and Point Cloud paper covers processing, the editor and pricing in full.

Records and chain of custody

A survey file is often asked about years later, in a boundary dispute, a construction claim or a board complaint. Omniris keeps the photos, the control, the processing and the deliverables in one evidence vault, with one record of what happened to each:

• Custody from upload. Every photo and every result has a tamper-evident custody history. Each result is recorded as an output of the job that made it, with its SHA-256 fingerprint. Downloads, exports and shares are recorded too. A signed custody certificate can be checked by anyone on a public verification page.
• The control trail. Every mark in every photo keeps a history of who placed, moved or removed it, and a change can be reverted. Every switch between control point and checkpoint is recorded. Every checkpoint evaluation is kept with who asked for it, and every change to a site's RTK settings is written to the custody record.
• What controlled what. Each stitch records which points actually controlled it, as the reconstruction loaded them. Each CAD export records its coordinate system, transformation, epoch and height reference, inside the file and in its record.
• Holds and retention. Results under a legal hold or retention policy are never replaced. A re-stitch that would overwrite them is refused before anything is charged.
• Encryption. Photos in an encrypted folder can be stitched, and the results are encrypted as they arrive. Encrypted files are not sent to Procore or Autodesk Forma unless someone allowed to open them releases them on purpose, and the release is recorded.

The Evidence Integrity paper describes the custody record, certificates, holds and encryption in full. A custody record shows how a file was made and handled. It does not make the data a certified survey.

Getting the flights flown

Firms that fly their own sites use the mission planner and pre-flight tools directly. Firms that need a pilot can post the job.

• Plan for mapping. Grid, crosshatch, corridor, perimeter and orbit patterns. The planner computes ground resolution, photo count, lines, flight time and batteries, and warns about motion blur. It warns above the 400 ft (122 m) Part 107 ceiling [25]. Terrain following keeps a steady height above the ground, with at least 33 ft (10 m) of clearance everywhere.
• Airspace and weather. The route is checked against FAA airspace, including the floors of Class B, C, D and E, and against no-fly areas drawn on the job. Omniris works out what LAANC authorisation is needed and records what was granted. A flyability forecast scores the days ahead.
• Before take-off. The pre-flight checklist includes the mapping checks: the take-off point matches the plan (the plan's heights are measured from it), the card has room for the planned photos, and the aircraft has a satellite lock.
• A precise-positioning requirement. In the marketplace, a job can be marked as needing precise location (RTK or PPK). It is offered only to pilots who have declared RTK or PPK equipment. Part 107 certification is required of US-based pilots.

The marketplace, payments and payouts exist in the Omniris cloud and data center editions. They are not part of on-premises installations, where firms fly their own sites. The Mission Planning paper and the Field App paper cover planning and field work in full.

Cloud, data center and on-premises

The survey tools in this paper (ground control and checkpoints, stitching, terrain analysis, CAD export, LiDAR processing and the classification editor) are part of Omniris wherever it runs:

The geoid grids (GEOID18, GEOID12B, EGM96) and the State Plane zone table are part of every installation. The automatic zone lookup asks NOAA's NCAT service. Where that service cannot be reached, the export says so, and you can still pick the zone by hand. The marketplace and payments are available only in the cloud and data center editions.

What the results are, and what they aren't

Survey data is relied on for design, quantities and title work, so Omniris states what its results are:

• Measured data, not a licensed survey. Omniris produces measured data and records how it was made. A licensed professional surveyor decides whether it meets a specification and certifies any survey product. Boundary work and plats are the surveyor's.
• Accuracy is reported as RMSE. Checkpoint results are per-point errors, bias and horizontal, vertical and 3D RMSE. Omniris does not assign an ASPRS Edition 2 accuracy class, add checkpoint survey error into the result, compute NSSDA 95% figures, split vertical testing into vegetated and non-vegetated areas, or enforce a minimum number of checkpoints.
• Heights are labelled as what they are. CAD exports carry ellipsoidal or EGM96 heights, named inside each file; NAVD88 elevations for the drawing are applied in CAD or survey software. Comparisons against a design convert into the design's NAVD88 heights.
• Frames are matched, or the comparison is refused. A flight positioned in WGS84/ITRF, or controlled in an older NAD83 realisation, is never graded or measured for volume against a NAD83(2011) design; that needs NAD83(2011) control. A design file is converted in the coordinate system, units and height reference it declares, and what it leaves out is stated by you in the volume tool and recorded. Designs in latitude and longitude, EGM96 heights on a NAD83 design, and California designs whose stated epoch doesn't match the control are refused rather than converted.
• Grids and surfaces. Orthomosaic and elevation-model GeoTIFFs are on the UTM grid, and LiDAR results in metres in the file's own zone and datum (UTM for latitude-longitude files); State Plane is offered for the CAD exports. Exported surfaces are regular-grid TINs sampled from the elevation model, without breaklines.
• A photo survey sees the top of the canopy. Bare earth under dense vegetation needs LiDAR or ground shots.

Sources

Industry facts in this paper come from the sources below. Figures that come only from a manufacturer or vendor are marked "vendor-reported" in the text.

1. American Society for Photogrammetry and Remote Sensing. ASPRS Positional Accuracy Standards for Digital Geospatial Data, Edition 2, Version 2 (2024), doi:10.14358/ASPRS.PAS.2024. doi.org/10.14358/ASPRS.PAS.2024
2. "The ASPRS Positional Accuracy Standards, Edition 2." LIDAR Magazine, October 6, 2023. lidarmag.com/2023/10/06/the-asprs-positional-accuracy-standards-edition-2/
3. "ASPRS approves Edition 2 of the ASPRS Positional Accuracy Standards for Digital Geospatial Data." GPS World. gpsworld.com/asprs-approves-edition-2-of-the-asprs-positional-accuracy-standards-for-digital-geospatial-data/
4. Penn State University, GEOG 892. "The New ASPRS Positional Accuracy Standards for Digital Geospatial Data." courses.ems.psu.edu/geog892/book/export/html/707
5. Federal Geographic Data Committee. Geospatial Positioning Accuracy Standards, Part 3: National Standard for Spatial Data Accuracy, FGDC-STD-007.3-1998. fgdc.gov/standards/projects/accuracy/part3/chapter3
6. American Land Title Association and National Society of Professional Surveyors. Minimum Standard Detail Requirements for ALTA/NSPS Land Title Surveys, effective February 23, 2026 (Section 7, Certification; Table A, item 5). alta.org/policies-and-standards/policy-forms/download.cfm?formID=721&type=pdf
7. Partner Engineering and Science. "ALTA Survey: Table A, Item #5 – Topography Surveys," November 17, 2025. partneresi.com/resources/articles/table-a-item-5-topography-surveys-purpose-technology-best-practices/
8. North Carolina Department of Transportation. "Rules & Regulations for Aerial Surveying in North Carolina with Unmanned Aerial Systems (UAS)," September 2020. connect.ncdot.gov/resources/Photogrammetry/Photogrammetry Documents/Aerial Surveying with Drones.pdf
9. "ASPRS Engages States." xyHt, February 26, 2015. xyht.com/aerialuas/asprs-engages-states/
10. ASPRS. "NCEES: ASPRS Membership, the Model Law, and More," January 19, 1999. old.asprs.org/a/news/archive/98-10-ncees.html
11. Colorado Department of Transportation. Drainage Design Manual, Chapter 6, "Data Collection" (2019), sections 6.3.2 and 6.4.1 (archived copy). web.archive.org/web/20230208160030/https://www.codot.gov/programs/environmental/water-quality/drainage-design-manual-documents-sept-2019/chapter-6-data-collection-6-29-2019-docx.pdf
12. Federal Highway Administration, Every Day Counts (EDC-5). "Unmanned Aerial Systems (UAS)" fact sheet. fhwa.dot.gov/innovation/everydaycounts/edc_5/docs/uas-factsheet.pdf
13. DJI Enterprise. Matrice 4 Series specifications (vendor). enterprise.dji.com/matrice-4-series/specs
14. DJI Enterprise. Zenmuse L2 specifications (vendor). enterprise.dji.com/zenmuse-l2/specs
15. NOAA National Geodetic Survey. OPUS: Online Positioning User Service. geodesy.noaa.gov/OPUS/
16. Emlid. Emlid Studio documentation: kinematic processing and drone data processing. docs.emlid.com/emlid-studio/
17. NOAA National Geodetic Survey. "New Datums." geodesy.noaa.gov/datums/newdatums/index.shtml
18. NOAA National Geodetic Survey. Modernized NSRS beta site. beta.ngs.noaa.gov
19. Iowa Department of Transportation. "NATFR2022 Beta Testing," Iowa Real-Time Network notice, October 8, 2026. iowadot.gov/announcement/2026-10-08/natfr2022-beta-testing
20. NOAA National Geodetic Survey. GEOID18 computation and coverage. geodesy.noaa.gov/GEOID/GEOID18/computation.html
21. National Institute of Standards and Technology. "U.S. Survey Foot." nist.gov/pml/us-surveyfoot
22. NOAA National Geodetic Survey. NCAT: Coordinate Conversion and Transformation Tool. geodesy.noaa.gov/NCAT/
23. U.S. Geological Survey, 3D Elevation Program. "Topographic Data Quality Levels (QLs)." usgs.gov/3d-elevation-program/topographic-data-quality-levels-qls
24. U.S. Geological Survey. Lidar Base Specification (2025 rev. A). usgs.gov/ngp-standards-and-specifications/lidar-base-specification-online
25. 14 CFR § 107.51, Operating limitations for small unmanned aircraft. law.cornell.edu/cfr/text/14/107.51
26. National Institute of Standards and Technology and National Oceanic and Atmospheric Administration. "Deprecation of the United States (U.S.) Survey Foot," Federal Register 85 FR 62698, October 5, 2020. govinfo.gov/content/pkg/FR-2020-10-05/html/2020-21902.htm
27. Texas Department of Transportation, Design Division. TxDOT UAS Aerial Mapping Specifications and Workflow for Design-Grade Mapping Applications, June 2025. txdot.gov/content/dam/docs/division/des/remote-sensing/uas-aerial-mapping-for-design.pdf

Glossary

ALTA/NSPS survey

A Land Title Survey made to the standards adopted jointly by the American Land Title Association and the National Society of Professional Surveyors, certified by a licensed surveyor. Topography is optional Table A item 5.

ArUco target

A printed square target carrying a machine-readable code, so software can find and identify it in photos automatically.

Checkpoint

A surveyed point left out of the fit and used only to measure the result's accuracy.

Control point (GCP)

A surveyed ground marker that the reconstruction is fitted to.

DSM / DTM

Digital surface model (the top of everything: ground, vegetation, structures) and digital terrain model (bare earth).

Ellipsoidal height

Height above the mathematical ellipsoid that satellite positioning uses. It is not the same as height above sea level.

Epoch

The date to which coordinates refer. It matters because the earth's crust moves, as in California's NAD83(2011) epochs 2010.00 and 2017.50.

GEOID18 / GEOID12B

NGS hybrid geoid models that convert NAD83(2011) ellipsoidal heights to orthometric heights: GEOID18 for the conterminous United States (NAVD88), Puerto Rico and the U.S. Virgin Islands; GEOID12B for Alaska, Hawaii, Guam and the Northern Mariana Islands.

GSD

Ground sample distance: how much ground one pixel covers.

ITRF / WGS84

The global reference frames that satellite positioning works in. Omniris treats ITRF2014 and the current WGS84 as one frame.

LandXML

An open XML format for civil and survey data, used here for TIN surfaces that civil design software reads.

NAD83(2011)

The current realisation of the North American Datum of 1983, fixed to the North American plate. State Plane coordinates are based on it.

NATRF2022 / NAPGD2022 / SPCS2022

The new US reference frame, height datum and State Plane system that NGS is introducing to replace NAD83, NAVD88 and SPCS83.

NAVD88

The North American Vertical Datum of 1988, the current US reference for elevations.

NSSDA

The FGDC National Standard for Spatial Data Accuracy, which reports accuracy at 95% confidence from checkpoint RMSE.

NVA / VVA

Non-vegetated and vegetated vertical accuracy, the two vertical tests of the ASPRS standard.

Orthometric height

Height above a geoid (roughly, mean sea level), such as NAVD88 or EGM96 heights.

PPK / RTK

Post-processed and real-time kinematic satellite positioning: corrections from a base applied after the flight, or during it.

RINEX

The standard file format for raw satellite observations, used for post-processing and for OPUS.

RMSE

Root mean square error: the square root of the average squared difference between measured and surveyed positions at the checkpoints.

State Plane

The State Plane Coordinate System: map grids by state and zone, in the unit each state legislated.

TIN

Triangulated irregular network: a surface made of triangles, as in a LandXML surface or DXF 3D faces.

U.S. survey foot

1200/3937 m, about 2 parts per million longer than the international foot (0.3048 m). Deprecated after 2022, but still the legal unit of many existing State Plane zones.