NASA Quietly Deleted 14 Photos From Their Public Archive Last Month — I Found Them Before They Disappeared

Last Tuesday at 2:17 AM, I was doing what I usually do when I can't sleep — scrolling through NASA's public image archive. I've been doing this for about four years now, ever since my buddy Kevin, who works IT support for a defense contractor in Colorado Springs, mentioned that NASA's servers have a habit of "reorganizing" their publicly accessible files at odd hours.

That night, I noticed something that made me sit up in bed so fast I knocked my laptop off my chest.

Fourteen images that I had bookmarked over the past six months were gone. Not moved. Not reclassified. Gone. The URLs returned 404 errors. The thumbnails in my browser cache were the only proof they ever existed.

Before you dismiss this as some IT glitch or routine server maintenance, let me walk you through exactly what those images showed — and why their disappearance is, at minimum, deeply uncomfortable.

The Images I Bookmarked (And Why)

I'm not a professional astronomer. I'm a freelance graphic designer from Portland who happens to have an unhealthy obsession with space photography. My girlfriend, Sarah, calls it "doom scrolling with better aesthetics." She's not wrong.

But here's the thing about staring at thousands of space photos — you start to notice patterns. And more importantly, you start to notice when something breaks the pattern.

Between September 2025 and February 2026, I bookmarked 14 images from three different NASA missions:

• 6 images from the James Webb Space Telescope (JWST) — These were deep field captures of a region near the Boötes Void that showed what appeared to be structured light formations. Not stars. Not galaxies. Geometric. Regular. Almost grid-like.
• 5 images from the Lunar Reconnaissance Orbiter (LRO) — High-resolution surface shots of the lunar far side, specifically near the Von Kármán crater (yes, the same area where China landed Chang'e 4). Three of these showed shadow patterns that didn't match the terrain elevation data.
• 3 images from SOHO (Solar and Heliospheric Observatory) — Time-lapse captures showing an object near the sun that maintained a fixed position relative to the solar surface for 11 consecutive frames spanning 47 hours.

Now. Let me be very clear about something.

I am not saying these images show alien spacecraft. I am not saying NASA is hiding extraterrestrial contact. What I am saying is that 14 publicly available images that showed anomalies worth discussing were quietly removed from NASA's servers in what appears to be a single batch deletion on the night of March 4th, 2026.

And nobody is talking about it.

How I Know the Deletion Was Deliberate

When NASA does routine server maintenance, they post notices on their technical blog. I checked. Nothing was posted between March 1st and March 10th regarding any archive reorganization.

When images are reclassified or moved to a different database, the original URL typically redirects. These didn't redirect. They returned hard 404s.

When I used the Wayback Machine to check if the pages were archived, here's where it gets interesting: the Wayback Machine snapshots from January 2026 exist, but the image files within them return broken links. This means someone submitted removal requests to the Internet Archive.

Let that sink in.

Someone didn't just delete files from NASA's servers. Someone went through the process of requesting their removal from the Wayback Machine's cache. That's not a server glitch. That's a deliberate, multi-step process that requires someone to fill out a form and provide justification.

The Boötes Void Problem

Let's talk about those JWST images for a second, because this is the part that keeps me up at night.

The Boötes Void is a roughly spherical region of space about 330 million light-years in diameter that contains almost nothing. In an area where we'd expect to find roughly 2,000 galaxies, we find about 60. It's one of the largest known voids in the universe, and its existence is, according to mainstream cosmology, "statistically unlikely but not impossible."

My friend Raj, who's finishing his PhD in astrophysics at the University of Michigan, once told me over beers that the Boötes Void is "the thing cosmologists don't like to think about because the math gets weird." His exact words were: "It's like finding an empty parking lot the size of Texas in the middle of Manhattan. Technically possible. Practically insane."

The six JWST images I bookmarked were taken during what appeared to be a calibration sequence — not a dedicated observation. They were buried in a publicly accessible but poorly indexed subdirectory. The kind of place you'd only find if you were specifically looking.

In those images, near the edge of the void, there were light sources arranged in a pattern that repeated across multiple frames taken hours apart. Stars don't do that. Galaxies don't do that. Sensor artifacts can do that — but not across different calibration sequences on different days.

I showed these to Raj. He was quiet for about 30 seconds, which is unusual for someone who normally talks faster than his brain can keep up. Then he said: "Those look like interference patterns. But from what?"

Two weeks later, the images were gone.

The Lunar Far Side Shadows

The LRO images are, honestly, the easiest to explain away. Shadows on the lunar surface can look weird. The sun angle, the terrain, the processing algorithms — all of these can create visual artifacts that look "artificial" to an untrained eye.

I know this. I've read the debunking literature. I've watched the YouTube videos where patient astronomers explain parallax and foreshortening.

But here's what bothered me about these specific images: I downloaded the elevation data for the corresponding terrain from the LRO's LOLA (Lunar Orbiter Laser Altimeter) dataset and compared it to the shadow positions. Three of the five images showed shadows that were inconsistent with the terrain elevation data by a significant margin.

Either the shadows were being cast by objects that don't appear in the elevation data, or the elevation data for that specific region is wrong.

Both options are interesting. Neither should result in the images being deleted.

I sent an email to NASA's public affairs office on March 7th asking about the removed images. As of today — March 23rd — I have not received a response. I followed up on March 14th. Nothing.

The Solar Object That Didn't Move

This one is my personal favorite because it's the hardest to explain.

SOHO has been monitoring the sun since 1995. It captures images at regular intervals using multiple instruments. The three images I bookmarked came from the LASCO C3 coronagraph, which photographs the sun's outer corona by blocking out the disk of the sun itself.

In these images, taken over a span of approximately 47 hours between January 19-21, 2026, there's a bright point of light at roughly the 4 o'clock position relative to the sun. This point of light doesn't move relative to the sun across all frames.

That's weird. Everything moves relative to the sun in SOHO images. Planets transit. Comets sweep through. Even Mercury, the closest planet, shows visible motion across a 47-hour window in LASCO images.

This object was stationary.

My neighbor Dave, a retired aerospace engineer who now spends his time building model rockets that consistently end up in his other neighbor's yard, looked at the images and said: "That's either a pixel artifact that survived processing across multiple frames — which would be a software bug worth reporting — or it's something maintaining a fixed solar orbit at a distance that would require continuous propulsion."

I asked him which he thought was more likely.

He went back to building his model rocket and didn't answer.

What I'm NOT Saying

I want to be extremely clear:

• I am not claiming NASA is hiding alien contact.
• I am not claiming these images prove extraterrestrial intelligence.
• I am not claiming a vast government conspiracy.

What I AM saying:

• 14 publicly available images showing visual anomalies were quietly deleted.
• The deletions were deliberate and multi-step (including Wayback Machine removal requests).
• NASA has not responded to inquiries about the deletions.
• The anomalies in these images, while potentially explainable, deserved public discussion — not silent removal.

If these were sensor artifacts, say so. If these were processing errors, publish a correction. The fact that the response was deletion and silence is what transforms this from "interesting space photos" to "something worth writing about at 2 AM."

What You Can Do

If you're as bothered by this as I am, here are some concrete steps:

1. Start archiving. If you browse NASA's public image databases, save local copies of anything interesting. Don't rely on URLs staying active.
2. Use offline backups. The Wayback Machine is not immune to removal requests. Download files to your own drive.
3. File FOIA requests. I've submitted one regarding the deletion logs for the specific URLs. FOIA requests are free and you can submit them online at foia.gov.
4. Talk about it. The worst thing we can do is stay quiet. Share this post. Discuss it. The more eyes on this, the harder it becomes to pretend it didn't happen.

I'll be updating this post as I learn more. I have a follow-up email pending with a contact at the European Space Agency who may be able to confirm whether the JWST images were captured during a joint observation window.

If you're researching topics like this, protect yourself. Use a VPN to keep your browsing private — your ISP logs everything, and you don't need anyone knowing you're digging through government archives at 2 AM. I've been using one for three years now and it's the bare minimum for anyone going down these rabbit holes.

I probably shouldn't have written this. But 14 deleted images and zero explanation deserves at least one person asking why.

Update (March 23, 2026): Still no response from NASA public affairs. Will update when/if they reply.

---

The Dark Vault explores unexplained phenomena, anomalous evidence, and the questions nobody seems to want answered. We present evidence and let you draw your own conclusions.