The Simple Answer That Isn't
Earth has one permanent natural satellite. Everyone learns that in elementary school. The Moon is big, it's tidally locked, and it's been doing its thing for 4.5 billion years. But if you're actually asking How Many Moons Does Earth Have, the answer gets messy fast, and most people stop looking after the basic count. There are other objects. Temporary captures. Minimoons. Quasi-satellites. These aren't sci-fi concepts — they're tracked by real observatories and catalogued in databases I check regularly when orbital mechanics gets interesting. The problem is that the Solar System doesn't respect human categories, and Earth's gravitational sphere of influence is bigger and messier than most people assume.
How Many Moons Does Earth Have
Officially, one. The International Astronomical Union says one. If you're filling out a form or writing a textbook, that's the answer. But if you want to know what's actually orbiting us at any given time, you need to look at the data. Several dozen near-Earth asteroids have shown temporary captured orbits around Earth at some point. 2006 RH120 was captured around 2006-2007 and then escaped. 2020 CD3 was another mini-moon, only about 2-3 meters across, captured briefly in 2020. These objects typically stay in Earth's vicinity for a few months to a couple years before solar gravity kicks them back out into heliocentric orbit. Then there are quasi-satellites like 3753 Cruithne and 2016 HO3. These aren't technically moons — they orbit the Sun, not Earth. But their orbits are synchronized with Earth's in a way that makes them appear to dance around us from our rotating reference frame. Cruithne has been in this configuration for thousands of years and will stay that way for thousands more. 2016 HO3, also called Kamo'oalewa, is the most stable known Earth quasi-satellite, likely locked in this relationship for at least several thousand years based on simulations. I've spent more time than I'd like to admit chasing 2024 PT5 through orbital simulation software when it made a brief appearance as a temporary capture candidate. The object was tiny — roughly 3 meters — and only detectable because the Catalina Sky Survey caught it during its flyby. Here's the thing nobody tells you: these objects are almost never visible to amateur equipment. They're too small and too far away. The only way to track them reliably is through sky survey data from professional telescopes, and even then, you're working with limited observation arcs. When I ran the ephemeris projections for 2024 PT5, the uncertainty cone was so large that the object could have missed Earth entirely — it was a statistical best guess based on maybe six nights of observation data. That's the reality of tracking these things: you're working with thin evidence and a lot of orbital propagation algorithms eating up CPU cycles.
So yes, there is a formal definition of a moon, and it matters for classification. The IAU requires a body to clear its orbital neighborhood, which immediately disqualifies most quasi-satellites and temporary captures. Earth's actual moon checks every box: it's massive enough for hydrostatic equilibrium, it's orbited by nothing comparable in its zone, and its host planet isn't a star. Everything else is a visitor. Here's a counter-intuitive point that most beginner astronomy resources miss: the definition of "moon" is actually the less interesting problem. The harder question is where you draw the line between an object orbiting Earth versus orbiting the Sun while Earth happens to be nearby.Objects like 2020 CD3 spent time inside Earth's Hill sphere — about 1.5 million kilometers — but they were still fundamentally solar-orbiting bodies that briefly got pulled in. Some astronomers call these "temporarily captured orbiters." Others call them flybys with a long lunch break. The terminology depends on whether you prioritize the object's primary gravitational partner or its temporary proximity to us. There's also the L4 and L5 Lagrange point question. Theoretical models suggest Earth could host trojan asteroids at these gravitational stable points, but none have been confirmed yet. Jupiter has thousands of trojans. Mars has a couple. Earth, despite being the third rock, has basically nothing at its Lagrange points. This might just mean we haven't looked hard enough — these objects would be faint and difficult to distinguish from background noise — or it might mean Earth's position in the Solar System makes trojan accumulation unlikely over geological timescales.
Get the Full Details

The practical limitation here is observation bias. Most sky surveys are designed to find dangerous near-Earth objects, which means they prioritize bright, fast-moving rocks on collision courses. A faint quasi-satellite drifting quietly in a synchronized orbit isn't going to trigger any alerts. You have to know where to look, and the looking requires either dedicated computational resources or a very patient researcher willing to cross-reference months of survey data against ephemeris predictions. If you want to track these objects yourself, start with the Minor Planet Center's database and JPL's Small-Body Database. Both are free. Download the orbital element sets, run them through a propagator like SkyCalc or the free version of Orbiter, and you'll see the picture emerge. It takes time — roughly 30 to 45 minutes per object to get clean visualizations of the orbit relative to Earth — but it's the only way to really understand what's happening rather than reading a summary that inevitably oversimplifies things.