Drone mapping spans flight activity, photogrammetric accuracy, terrain modeling and land classification. The available evidence combines a 2024 FAA U.S. survey with peer-reviewed studies from Alpine terrain, no-ground-control DEM testing, RTK-GNSS topographic mapping and Indonesian post-mining sites.
Key Drone Mapping Statistics
The most broadly useful figures are:
- 38.4 million estimated UAS flights were made by recreational and Part 107 nonrecreational operators in the contiguous United States in 2024.
- 16.6 million estimated flights came from Core Part 107 operators in the United States in 2024.
- 21.7 million estimated flights came from recreational operators in the United States in 2024.
- 56% of 8,? surveyed Part 107 registrants were classified as Core Part 107 operators in the FAA’s 2024 survey.
- 36% of surveyed Part 107 registrants were Recreational Only in the 2024 U.S. FAA survey.
- 2,481 active FRIAs were recorded in the United States at the end of 2024.
- 131.3 annual flights per respondent were reported by Core Part 107 operators in the 2024 FAA survey.
- 95.1 annual nonrecreational flights per respondent were reported by Core Part 107 respondents in 2024.
- 15 cm positional RMSE was achieved at the smallest Alpine study site during 2018–2021 surveys.
- 135 cm positional RMSE was recorded at the largest Alpine study site during 2018–2021 surveys.
- 4 to 40 GCPs per km² were used across four Alpine study sites surveyed from 2018 to 2021.
- 20 cm spatial resolution was used for all Alpine UAV orthophotos and DSMs from 2018 to 2021.
- 70% to 80% lateral overlap was recommended for mountainous UAV survey planning in the Alpine study.
- 2.8 m average elevation accuracy was reported across no-GCP UAV DEMs and DSMs in the study dataset.
- 2.1 m overall mean horizontal shift was measured across all no-GCP measurements in that study.
- 0.04 m mean horizontal error was reported for additional models using GCPs in the no-GCP study.
- 0.6 cm average 2D coordinate RMSE was reported at independent checkpoints in the RTK-GNSS comparison.
- 25 cm 3D checkpoint RMSE was reported in the same RTK-GNSS topographic mapping study.
- 92.3% overall classification accuracy was reported for the drone-based map in the Indonesian post-mining case study.
- 78.1% overall classification accuracy was reported for the comparison SAR land-cover map in that same case study.
Contents
- U.S. drone mapping activity and operator adoption
- Drone mapping data trends across operator groups
- Accuracy, terrain size and ground control
- Resolution, coverage and image overlap
- Terrain models without ground control
- Surveying and land-classification outcomes
U.S. Drone Mapping Activity and Operator Adoption
The FAA’s CY 2024 Small Unmanned Aircraft Systems Survey Report provides context for the scale of U.S. drone operations, but its flight totals are not mapping-specific. It estimated 38.4 million UAS flights by recreational operators and Part 107 nonrecreational operators in the contiguous United States during 2024.
The same report estimated more than 16.6 million flights by Core Part 107 operators and more than 21.7 million recreational flights in the United States. These are broad operating categories, so they should not be read as counts of mapping missions.
| FAA measure | 2024 result | Population and scope |
|---|---|---|
| Estimated UAS flights | 38.4 million | Contiguous U.S. recreational and Part 107 nonrecreational operations |
| Core Part 107 flights | More than 16.6 million | U.S. operators reporting at least one nonrecreational flight |
| Recreational flights | More than 21.7 million | U.S. recreational operations |
| Active FRIAs | 2,481 | United States, end of 2024 |
Source: CY 2024 Small Unmanned Aircraft Systems Survey Report. The FAA figures describe U.S. UAS operations broadly, not drone-mapping missions.
Among surveyed Part 107 registrants, 56%, or 4,559 respondents, were classified as Core Part 107 operators. Another 36%, or 2,982 respondents, were Recreational Only, while 8%, or 659 respondents, reported no flights.
The survey also identified 382 Part 107 registrants operating UAS for emergency-response organizations. Those programs had an average age of 3.5 years, and the longest-running reported program was 15 years; emergency response is broader than mapping.
Drone Mapping Data Trends Across Operator Groups
The FAA survey shows how activity and equipment differed across recreational and Part 107 respondent groups. These figures describe self-reported flights and aircraft, not mapping output, sensor payloads or mapping businesses.
- Within-FRIA recreational respondents reported 103.7 annual flights on average, compared with 37.9 for Outside-FRIA respondents in 2024.
- Within-FRIA respondents reported owning 10 aircraft on average, versus 2.5 outside FRIAs.
- Within-FRIA respondents operated 5.9 aircraft on average, versus 1.7 outside FRIAs.
- Within-FRIA respondents operated 7.47 fixed-wing aircraft on average, compared with 0.47 outside FRIAs.
- Self-identified drone operators reported 30.1 annual flights on average, compared with 132.3 for model-aircraft operators and 116.5 for respondents identifying as both.
- Self-identified drone operators averaged 1.46 quadcopters operated, versus 0.17 for model-aircraft operators and 1.97 for respondents identifying as both.
Core Part 107 respondents reported 131.3 annual flights on average, compared with 26.0 for Recreational Only Part 107 respondents. Core Part 107 respondents also reported 95.1 annual nonrecreational flights and operated 2.3 quadcopters on average, compared with 1.2 among Recreational Only Part 107 respondents.
More than 500 Core Part 107 operators rated real estate and construction as somewhat active, active or very active, making it the most common listed commercial activity. The activity scale was subjective and did not quantify mapping jobs.
Source: CY 2024 Small Unmanned Aircraft Systems Survey Report. Survey percentages and respondent averages are subject to the report’s self-reporting, weighting and classification definitions.
Drone Mapping Accuracy Statistics by Terrain Size and Ground Control
The Alpine study examined 13 UAV surveys across complex glacial and periglacial terrain at four sites during 2018–2021. Its positional results varied substantially with terrain, survey size, ground-control density and image quality.
| Alpine survey context | Area or range | Reported result |
|---|---|---|
| Smallest site | 0.2 km² | 15 cm RMSE |
| Largest site | 2.5 km² | 135 cm positional RMSE |
| Four-site GCP density | 4–40 GCPs/km² | Site-dependent |
| Four-site absolute control | 8–22 GCPs/site | Site-dependent |
At the two larger Alpine sites, XY-RMSE ranged from 0.97 to 1.34 m, while Z-RMSE ranged from 0.30 to 1.75 m. At the two smaller sites, both horizontal and vertical RMSE were below 0.30 m.
The smallest 0.2 km² site used 40 GCPs per km² and adequate overlap and contrast, producing 15 cm RMSD. At the largest 2.5 km² site, 15 GCPs per km² and 50–70% lateral overlap were associated with roughly 100 cm RMSD and artefacts; the study did not isolate the effect of either factor.
A 620 m UAV-DSM cross-section at Kanderfirn had an elevation RMS deviation of 0.5 ± 0.5 m from dGNSS elevations. About 40% of comparison points had deviations below the general dGNSS accuracy, but this applies only to the reported cross-section.
A prior study cited by the Alpine paper reported 10–15 cm accuracy on a 0.7 km² survey using 120 GCPs per km². Because that result is a secondary report with unusually high control density, it should not be treated as a universal benchmark.
Source: Accuracy of UAV Photogrammetry in Glacial and Periglacial Alpine Terrain: A Comparison With Airborne and Terrestrial Datasets. These are study-level observations from different sites and conditions.
Drone Mapping Resolution, Coverage and Image-Overlap Data
All Alpine UAV orthophotos and DSMs were generated at 20 cm spatial resolution during the 2018–2021 surveys. Spatial resolution is pixel size in the mapping product, not the same measurement as positional accuracy.
Insufficient lateral image overlap was reported in the 50–70% range at problematic Alpine survey areas. The study recommended 70–80% lateral overlap as a planning target for mountainous surveys.
The four sites used 4–40 GCPs per km², and the study noted that more than 10 GCPs per km² should generally support decimetre-level orthophoto and DSM accuracy. That threshold was cited guidance, and actual performance still varied widely across the sites.
The largest 2.5 km² site combined 15 GCPs per km² with 50–70% lateral overlap and produced roughly 100 cm RMSD plus artefacts. The smallest 0.2 km² site combined 40 GCPs per km² with adequate overlap and contrast and produced 15 cm RMSD.
Source: Accuracy of UAV Photogrammetry in Glacial and Periglacial Alpine Terrain: A Comparison With Airborne and Terrestrial Datasets. The 70–80% overlap figure is a recommendation from one mountainous study context, not a universal standard.
Drone Mapping Terrain Models Without Ground Control Points
A separate study evaluated UAV-derived DEMs and DSMs without ground control points using a Mavic 2 Pro benchmark system. Across the study dataset, average elevation accuracy was approximately 2.8 m.
An error band from −1 m to +1 m covered an average of 41% of each no-GCP model. Errors above +1 m occupied an average of 52% of each model, while areas with errors of −1 m or less occupied an average of under 17%.
| No-GCP measure | Reported result | Scope |
|---|---|---|
| Average elevation accuracy | Approximately 2.8 m | All UAV DEMs/DSMs in study |
| Area within −1 m to +1 m | 41% average | Each no-GCP model |
| Area above +1 m error | 52% average | Each no-GCP model |
| Area at −1 m or less | Under 17% average | Each no-GCP model |
| Overall mean horizontal shift | 2.1 m | All no-GCP measurements |
No-GCP horizontal shift values ranged from 0.18 m to more than 5 m across test sites, while average horizontal shifts for individual sites ranged from 1.1 to 3.6 m. With GCPs, additional models in the same study had a mean horizontal error of 0.04 m and vertical MAE of 0.47 m.
Source: Accuracy of UAV-based DEMs without ground control points. The results apply to the tested Mavic 2 Pro, terrain and processing setup rather than every drone workflow.
Drone Mapping Outcomes in Surveying and Land Classification
RTK-GNSS integration produced highly precise results in one large-scale topographic mapping experiment. Independent checkpoints had average 2D coordinate RMSE of 0.6 cm, average 2D MAE of 0.4 cm and reported 2D coordinate accuracy of 1 cm.
The same experiment reported 25 cm RMSE, 22 cm MAE and 43 cm accuracy for 3D checkpoint coordinates. After integrating the drone with a mobile RTK base station, average 2D orthomosaic error was below 2 mm in the reported experiment.
RTK-GNSS and UAV feature comparisons across five mapped topographic features differed by −0.26% in perimeter and −0.23% in area on average. These results came from one controlled study and are not directly comparable with ordinary consumer flights or the Alpine RMSE values.
A post-mining land-cover case study in Indonesia reported 92.3% average overall classification accuracy for the drone-based map. The comparison SAR land-cover map reached 78.1% overall accuracy in the same study, while the SAR map subset had an average Kappa of 0.375.
These classification figures measure agreement in one landscape-analysis case study, not positional accuracy, market adoption or a general performance rate for drone mapping. The drone and SAR results came from the specific datasets and sites examined in that Indonesian study.