Four launches in seven days across three coasts — and a Falcon 9 booster about to fly for the 30th time

Four orbital launches are scheduled across three launch sites between today and Sunday — one from Taiyuan, China, and three from Florida and California — and the most operationally interesting thing about any of them is not the payload. It is the rocket. The Falcon 9 booster slated to fly Starlink Group 10-39 out of Cape Canaveral on Thursday, tail number B1078, will be making its 30th flight, putting it within three flights of SpaceX's current reuse record and putting the Falcon 9 fleet as a whole at a level of reusability no other orbital-class vehicle has reached. The week ahead is also where the cadence of August's launches stops looking like a streak and starts looking like the new normal.

Week ahead: what's launching, where, and on which rocket

The Vandenberg cadence is doing the work this week. The 38-minute SpaceX doubleheader on August 15 was the most operationally significant launch gap in company history — a record that will likely be tied or broken in the next several months — and SpaceX returns to the same pad four nights later with Starlink 17-50. The Friday Vandenberg window opens at 1:25 AM PDT for Starlink 15-20, which means three Falcon 9 launches from the West Coast in roughly four days if all three fly on schedule. The KTLA-cited "three launches in seven days from California" frame is the right one; the actual count is closer to three Vandenberg flights in roughly 96 hours once Thursday's Cape launch is included in the cadence picture. (For context on why Vandenberg specifically matters for the Starlink shell architecture, see Orbital Shells 101.)

The China launch is a separate question, and an unanswerable one in the open literature. The Long March 2C at Taiyuan's Launch Complex 9 is flagged on the Spaceflight Now launch schedule and on every tracker with "unknown payload," which is the standard notation for missions whose payloads are not pre-announced — typically a remote-sensing satellite or an experimental platform for a Chinese government or military customer. CASC does not publish a flight-by-3 manifest, which means the cadence of these flights (how many per year, on which vehicles, from which sites) is something the open-source community reconstructs by counting the launches and matching them to known optical or radio signatures afterward. The launches themselves are not classified. The payloads are.

The booster reuse story underneath this week's launches

What makes this week a useful data point is not the launch count. It is the count of flights per booster. B1078, flying Thursday from Cape Canaveral, will be making its 30th flight. B1097 on Tuesday will be making its 12th. B1100 on Saturday will be making its 9th. The median flight count of the three Vandenberg and Cape boosters flying this week is 12; the arithmetic mean, with B1078 as the outlier, is about 17 flights per booster.

To put that in perspective: the Space Shuttle fleet, across 30 years of operations, flew 135 missions across five orbiters. The most-flown individual orbiter, Discovery, flew 39 missions. B1078 alone will have flown 30 missions by Thursday, and is expected to fly several more before retirement. Each Falcon 9 flight is a smaller payload than a Shuttle mission on average, and the engineering challenge of carrying humans is absent on a Starlink flight — but the comparison is the closest one available for what 30 flights on a single first stage means operationally. The economic logic that the Shuttle program tried to engineer in the 1970s and never quite reached is one SpaceX has now reached at the booster level. (The launch cost curve just bent is the longer version of the same argument.)

The other thing to watch is what happens at the recovery zone. Two of the three Falcon 9 missions this week are landing on droneships — OCISLY in the Pacific for the two Vandenberg missions, and ASOG in the Atlantic for the Cape mission. None of the three are returning to land. That is the standard pattern for Starlink missions and it is worth noting because it is also the pattern that makes the booster-reuse math work. Landing on a droneship is a controlled splashdown followed by a tugboat trip back to port. Landing on land is a controlled ground touchdown and a short tow back to the hangar. The droneship missions are operationally cheaper than the land-return missions but logistically slower to turn around. The current Starlink cadence is calibrated to the droneship pace.

The bigger week behind this one

If the rhythm of the next seven days looks familiar, it should. August is on track to be the highest-cadence orbital month in history, and the work behind that number is not concentrated in one company or one country. The launches this week are a sample of the broader pipeline: one Chinese state launch on a small-to-medium vehicle; three SpaceX Starlink missions on a medium-to-heavy reuse-first vehicle; and on the horizon at the end of August, two missions of more than routine interest — NET August 27, an Arianespace Ariane 62 carrying the second Meteosat Third Generation Imager satellite (MTG-I2) to geostationary transfer orbit from Kourou, and NET August 30, a SpaceX Falcon Heavy with NASA's Nancy Grace Roman Space Telescope from Kennedy Space Center's Launch Complex 39A.

The Roman launch is the larger story. The telescope carries a 300-megapixel infrared Wide Field Instrument and a coronagraph, with a primary mirror the same size as Hubble's (2.4 meters) but a field of view roughly 100 times wider. The mission's planned lifetime is at least five years and the science program is built around three core surveys — a statistical census of dark energy through type Ia supernovae, a microlensing survey for exoplanets including the population of cold planets the Kepler and TESS missions could not see, and a wide-area infrared survey of the sky for time-domain discovery. The Falcon Heavy award for the mission was set at $255 million in 2022 for launch services, comparable to the Europa Clipper award at $178 million and well below the early-estimate range. The Roman launch is the second of August's two marquee astrophysics missions, with the Chang'e 7 lunar south pole mission having launched earlier this month.

The point is that the week ahead is a useful four-day cadence test, and the week behind is a useful reminder that the cadence is the data, not the news. Four launches in seven days from three coasts with three boosters of meaningfully different flight counts is what the orbital launch industry now produces as a default. The story this week is the same story as last week, and the week before, and the week before that — and the story worth reading underneath the cadence is that the cadence has now reached a level where the question is no longer "how do we make orbital access cheaper" but "what do we do with the access we have." The answer to that question is what changes between this decade and the next. (For the longer arc on the same problem, see The launch rate finally matched the rhetoric.)

Sky this week, briefly

The Perseids peaked in dark, near-New Moon skies on August 12 and 13. This week the Moon is waxing — first quarter is around August 20 — which means the post-midnight window will be too bright for the deep-sky season's faint targets, but the evening window before Moon rise is still useful for the bright planets and the Milky Way from a dark site. Venus remains a brilliant low-western evening object through the third week of August, slowly losing altitude as it moves toward the Sun in the evening sky. Jupiter and Saturn are well-placed for late-evening observation; Mars does not rise until the small hours. The waxing crescent Moon will pass through a few useful configurations this week — first quarter on or around August 20 will give a clean terminator for lunar imaging — but the headline observation window remains the brief period after sunset before the Moon rises, when the western sky belongs to Venus and the Milky Way core is still visible from a dark site in the southern half of the country. For full sky-week coverage, Friday's observation edition will have detailed pass times for the ISS and any conjunction events worth marking on the calendar.

---

Sources

• Spaceflight Now — Launch Schedule (week of August 17–23, 2026)
• Spaceflight Now — SpaceX doubleheader recap, August 15, 2026
• SpaceX — Starlink Group 17-50 mission page
• Space Launch Schedule — Vandenberg SFB
• Space Launch Schedule — Falcon 9 launches
• Next Spaceflight — Long March 2C, August 17, 2026 (Taiyuan SLC-9, unknown payload)
• KTLA — SpaceX plans rare stretch of 3 launches in 7 days from California
• NASA — Nancy Grace Roman Space Telescope mission page
• NASA — Roman launch services contract announcement (2022)
• EUMETSAT — Meteosat Third Generation program
• NASA Science — What's Up: August 2026 Skywatching Tips