Jonathan McDowell has been keeping a detailed running tally of what humans have put in orbit, and his catalog has become the de facto reference for journalists, regulators, and rival satellite operators trying to understand what is actually up there. The number that appears in nearly every recent news story about Starlink’s scale comes from this astronomer who ran the count from his home computer in Massachusetts for more than three decades, parsed U.S. Space Force tracking data by hand, and published the result on a personal website that became an essential resource.
That figure — including the Tech Times report on SpaceX’s 1,500th Starlink satellite deployment of 2026 — traces back to McDowell’s spreadsheet. So does the count cited when a Blue Origin upper-stage anomaly lost an AST SpaceMobile BlueBird 7 satellite in April.
The catalog started on a 1989 desktop
McDowell began compiling what he calls Jonathan’s Space Report while working as an astrophysicist at the Harvard-Smithsonian Center for Astrophysics. The original format was a weekly email newsletter sent to a handful of subscribers. The point was simple. Public satellite tracking existed — cataloging objects had been underway for decades — but the raw data was scattered, inconsistent, and rarely cross-referenced with launch records, mission descriptions, or fate-of-object information.
McDowell did the cross-referencing himself. Each object got a line. Each line got a source. When a satellite re-entered, he updated the entry. When a rocket body broke up, he logged the debris field. The work was tedious. It was also the only place a researcher could find a unified, plain-language record of what humanity had launched and what had happened to it since Sputnik.
That database is still running. The format has changed — the website now hosts machine-readable tables, launch logs by year, and a separate Starlink-specific page — but the curation remains a one-person job. McDowell parses orbital element sets released by the U.S. Space Force, checks them against launch manifests, and decides what counts as in orbit, decayed, or operational. Those decisions matter, because the categories are not always obvious.
Why governments and reporters cite a private database
Official catalogs exist. Space-Track.org, run by the U.S. Space Force, lists every tracked object above a certain size. The European Space Agency maintains its own database. Russia and China publish partial figures. But none of them, until recently, produced a clean public count of active Starlink satellites currently in service. The Space Force tracks objects. It does not distinguish a working V2 Mini from a decommissioned V1 drifting toward re-entry.
McDowell does. His Starlink page separates the constellation by version, shell, altitude, and operational status. When SpaceX adjusted satellite altitudes to comply with deorbit requirements, McDowell’s tables tracked the altitude shift week by week. Reporters covering the move had a single citable source. So did the FCC.
The result is that a private catalog, maintained by one astronomer without institutional funding for the work, sits at the center of how the orbital environment is publicly described. When SpaceNews reports on constellation growth, McDowell’s numbers appear. When congressional staffers brief members on satellite congestion, his figures are in the slide deck. When the count of active Starlinks appeared in the Tech Times story on the June 15 launch, it carried his name.
The 1,500-satellite year
The pace that made 2026 the year Starlink became impossible to ignore is also the pace that made an independent counter necessary. By June 15, 2026, the company had placed 1,500 satellites in orbit in a single calendar year — with half the year still to run. That averages to roughly one dedicated launch every three to four days, with Falcon 9 missions carrying batches of satellites into orbit. The launch cadence has reached unprecedented levels of reusability and frequency.
Space Daily has covered the cadence as it accelerated, including the Falcon 9 missions that pushed the active fleet to new milestones and the polar-orbit deployments expanding coverage at high latitudes.
How the work actually gets done
The mechanics are unglamorous. McDowell downloads two-line element sets — the standard format for describing an object’s orbit — from public Space Force feeds. He runs them through his own scripts to identify which catalog number corresponds to which payload from which launch. He cross-checks against launch broadcasts, FCC filings, and operator statements. He flags anomalies. He emails the operator when something looks wrong. He updates the database.
The process is what allowed him to quickly assess the New Glenn anomaly in April 2026 and determine that AST SpaceMobile’s BlueBird 7 satellite had been dropped into an unusable orbit. The Space Force tracking data showed the orbit. McDowell matched it to the payload, calculated the decay time, and posted the result. Blue Origin’s statement indicated the satellite was in a problematic orbit, but did not specify whether re-entry was imminent. McDowell’s reading of the tracking data did. AST confirmed the satellite would be deorbited later that day.
The same workflow applies to the Tesla Roadster episode. In January 2025, the Minor Planet Center cataloged what appeared to be a new near-Earth object. It turned out to be Elon Musk’s Tesla Roadster, launched on a Falcon Heavy test flight in February 2018. McDowell was among the astronomers who flagged the mismatch between the object’s trajectory and any known asteroid. The catalog entry was deleted within hours. The car remained on a heliocentric orbit, where it still is.
Why a one-person catalog became the standard
Three things made McDowell’s database harder to replicate than it looks. The first is continuity. A satellite catalog is only useful if it spans decades. McDowell’s runs from Sputnik to last week, in a single consistent format, maintained by the same person. Institutional databases change schemas, lose funding, get reorganized. His does not.
The second is editorial judgment. Deciding what counts as “active” requires reading operator statements, watching for maneuver patterns, and noticing when a satellite stops adjusting its orbit. Automated systems flag the object as still present. McDowell flags it as drifting. The distinction matters when the question is how much of low Earth orbit is actually being used versus how much is filling up with dead hardware.
The third is accessibility. The data is free. The website is plain HTML. The tables download as CSV. A graduate student in Bangalore, a regulator in Brussels, and a reporter in Washington can all pull the same numbers in under a minute. McDowell has spoken at Duke and elsewhere about the regulatory implications of constellation growth, and his framing — that orbital space is a finite resource being allocated faster than rules can adapt — has shaped how policy researchers describe the problem.
The constellation he is counting
What McDowell is tracking is no longer a satellite program in the traditional sense. It is the infrastructure layer underneath a consumer broadband business with millions of subscribers across more than a hundred countries, a direct-to-cell service that turns standard LTE phones into satellite handsets, and — as of June 12, 2026 — a newly public company. SpaceX’s public listing reached a valuation in the hundreds of billions. Starlink has generated billions in revenue with rapid year-over-year growth.
The technical reason the fleet has to be so large is geometry. A satellite at 550 kilometers crosses the sky in minutes. Continuous coverage requires enough satellites distributed across multiple orbital shells that as one sets, another rises. Direct-to-cell coverage, which routes phone signals through laser inter-satellite links to ground stations, requires hundreds of dedicated satellites to reach global text service. Voice and data require denser deployment still. Each Falcon 9 mission — including the recent ones Space Daily covered when SpaceX returned to flight days after a service outage — adds capacity to that layer.
The FCC has authorized thousands of satellites under the current Starlink license, with possible extensions to tens of thousands more. Amazon’s Kuiper and other rivals plan their own mega-constellations. The astronomy community has been raising alarms for years. Spectroscopic data is corrupted by satellite trails. Wide-field surveys, including the Rubin Observatory’s, have to develop new pipelines to subtract them. Radio astronomers face interference in bands they once had to themselves.
What the growing constellation looks like from the ground
The scale of the Starlink constellation is hard to grasp until you look up during twilight in June and watch the sky. Right after sunset, with the sun still illuminating objects hundreds of kilometers above you, Starlinks are everywhere. They appear as steady, moving points of light, often in lines for newly launched batches that have not yet spread into their operational shells. By the middle of the night, when Earth’s shadow extends past their altitude, they disappear.
That visibility was once a novelty. During the early COVID-19 lockdowns, watching a Starlink “train” race across the sky became a small ritual for stargazers stuck at home. It is now, for many naked-eye astronomers, something to avoid. The brightness has been reduced through design changes — sunshades, dielectric mirror films — but the count keeps climbing. SpaceX has stated its intention to deploy next-generation Starlink satellites using Starship V3, which completed its first test flight on May 22, 2026. Each Starship is designed to carry roughly ten times the Falcon 9’s payload to low Earth orbit.
The catalog as public infrastructure
The strange position McDowell occupies is that the public record of a privately owned constellation is being maintained by another private individual, working without commercial incentive, because no government agency produces the equivalent in a form people can use. Commercial space-situational-awareness companies have begun to fill some of the gap — platforms like Kayhan Space’s Satcat aggregate orbital data for operators — but they sell to industry, not to the public.
McDowell’s site has no paywall. There are no ads. When a launch like India’s Chandrayaan-3 lunar landing happens, or when an interstellar object appears, or when an AST satellite re-enters days after launch, the catalog absorbs the event, dates it, sources it, and moves on.
The Starlink count he maintains will be out of date by the time you finish reading this. Another Falcon 9 is on the pad. Another batch of satellites is on its way. The number will tick up, and the spreadsheet will record it. He will add the rows himself.
Sometime in the next year, if SpaceX hits the cadence its trajectory suggests, the count will continue to climb. The astronomy community’s projections from years ago are being realized. What was once a string of 60 lights moving in formation above the Atlantic has become a constellation that dominates the orbital environment — and that McDowell, watching the data come in from the Space Force, continues to log with the same rigor he brought to that first entry in 1989.
