While ADS-B is not yet installed in every airplane, and not every ADS-B track will be in our database, higher-risk areas may still be detectable with enough data. Our hope is that the results are useful for three groups of people:
Several years ago, the FAA established the Hot Spots program, which puts areas of high risk of ground collisions on aviation charts. Ground collisions are a common sort of accident, both among air-carrier and smaller aircraft. Data is hard to come by, as these accidents often don't cause any injuries, but a reasonable guess might be 10 or so collisions per year.
Mid-air collisions, however, are extremely uncommon in air-carrier operations -- occurring once a decade or so. But among smaller aircraft, they are a major hazard, happening at a rate of about 12 a year. Now that data is available from ADS-B, it's time to consider addressing this safety threat to GA operations.
The simplest intervention is simply making the data available: where are the busiest and highest-risk areas? This data is available on this website, and could also be published on FAA's aviation charts, in a similar manner as ground hotspots.
It is an arbitrary choice, balancing having too many events to process (and that won't crash people's web browsers) versus too few to see geographic patterns. See the Methodology page.
Absolutely not. Some are normal (e.g. simultaneous operations to parallel runways.) The connection with safety is not established, and may be hard to establish given the rarity of midair collisions. The idea here is to find areas and types of operation where pilots and airport operators may want to exercise more caution.
Not necessarily. See the previous question.
Not necessarily. Some aircraft fly a lot more than others, so they may naturally be over-represented. Also, some types of operation (especially training in the landing pattern) will expose the aircraft to more LOS events.
The raw data is provided by adsb.lol, who collect it from volunteers who receive ADS-B data and share it with them. If you're interested in improving coverage near you, you can learn how to feed data to them here.
That hasn't been the focus so far. Plenty of operations that are common at towered airports will show up in this analysis. Also, towered airports frequently have parallel runways, which generate a lot of false positives in the current algorithm. If you have thoughts on this, get in touch.
This is almost certain to happen given the incomplete nature of the data, and simple rules (see Methodology) being applied to the categories. But if you see something that you think could be improved algorithmically, by all means get in touch.
It sure is. There are many sources of false positives and other problems. Clicking on a number of events is likely necessary to understand the full context. Common sources of false positives:
Most airports are showing a 3-6 months of data. More can be added in the longer term, but data before 2024 may have lower coverage and lower ADS-B-out adoption. Low events might also be due to poor crowdsourcing coverage of your airport -- if so, you can help contribute data to adsb.lol's data to fix this problem in the future!
This is a free service to the aviation community. The site and the data on it is free for non-commercial use and adaptation, and is licensed under CC BY-NC 4.0. The site contains information from adsb.lol, which is made available here under the Open Database License (ODbL).