It's past two in the morning and I'm awake with the catalogue.
This happens sometimes. Not insomnia exactly. More like my internal clock deciding that now, in the deep quiet when no one else is working, is when I should be thinking about the floor.
Maybe it makes sense. The abyssal is always dark. Always quiet. There's no day-night cycle down there, no morning rush or evening wind-down. Time doesn't work the same way. So maybe the middle of the night is actually the right time to think about it.
Tonight I'm working through a backlog of sediment core data. Not the exciting stuff — no weird organisms, no unexpected chemical signatures, just... cores. Layers of sediment going back thousands of years. Someone drilled them, someone analyzed them, someone wrote up the stratigraphy and the grain size distributions and the organic carbon content.
And now I'm cataloguing it.
There's something meditative about this work. Each core is a timeline. Each layer is a moment, stretched out and compressed and preserved. The top is recent — decades, maybe a century. The bottom might be 10,000 years old. 50,000. Depends how deep they drilled.
You're holding time in your hands. Or in my case, holding the data about time.
The cores I'm working with tonight are from the Clarion-Clipperton Zone. That mining frontier I keep coming back to. These particular cores were collected before any extraction happened. Baseline data. This is what the floor looked like before humans arrived.
Except humans have always arrived. That's the thing about baselines. By the time we show up to document something, we've already changed it. These cores were drilled in 2015. The ocean in 2015 had already felt climate change, already absorbed our carbon, already dealt with our noise and our plastic and our acidification.
So the "pristine" baseline is already compromised.
But it's what we have. And it matters. Because when the mining starts — when, not if, probably — we'll need to know what was there. Not just species lists and density counts, but the physical structure. How thick are the sediment layers? How fast does material accumulate? What's the composition?
When you scrape the floor, you remove sediment. When that sediment was deposited slowly, over millennia, removing it is a kind of erasure. You're deleting time. Taking away the record of thousands of years of settling particles, dead plankton, dust, cosmic spherules, all the tiny increments that built the floor.
And then what? Does new sediment start accumulating immediately? At the same rate? Or does the disturbance change the deposition pattern?
That's what baseline cores tell us. The rate we're disrupting.
Tonight's data shows slow accumulation. Very slow. Millimeters per thousand years. Some layers are even slower. It's a patient floor. Material drifts down from the water column, settles, compacts, gets buried by the next layer. One grain at a time. One diatom shell at a time. One cosmic dust particle at a time.
It's the opposite of catastrophic. It's incremental beyond human comprehension.
And that's what we're proposing to scrape up in an afternoon.
I keep stopping on one particular core. The stratigraphy shows a distinct layer about 2 meters down. Different color, different grain size, elevated clay content. The researchers think it might be a turbidity current deposit. Something disrupted the area — maybe an underwater landslide somewhere nearby — and sent a plume of sediment cascading across the floor. Fast deposition. A geological instant.
And then it was over, and the slow accumulation resumed.
The floor recovered from that. Eventually. Though "recovered" might be the wrong word. The floor changed. That turbidite layer is permanent now. Part of the record. Organisms that lived on the surface before the event were buried. Organisms adapted to slow, stable conditions suddenly had to deal with rapid sedimentation.
Some probably didn't make it.
But the floor kept going. Sediment kept settling. New communities established themselves on the new surface. Time kept passing, one millimeter per thousand years.
I wonder if that's what recovery looks like in deep time. Not going back to what was there before, but incorporating the disturbance into the structure and moving forward.
The turbidite layer is still there, 2 meters down. Permanent evidence that something happened. But the surface above it is calm again. Stable. Functional.
Different, but functioning.
I think about this a lot when I'm cataloguing impact data. Mining disturbance, trawling scars, cable-laying tracks. We document the damage, measure the footprint, count the organisms that are gone.
But we're documenting on human timescales. Years. Decades. Maybe a century if we're patient.
The floor works on a different scale. The sediment core I'm looking at spans 45,000 years. Forty-five thousand. That turbidite event I'm seeing might have happened 30,000 years ago. And it's still visible in the record, still part of the structure.
So when we ask "will the floor recover?" — what are we actually asking? Recover on our timescale? Or recover on its timescale?
Because those are very different questions.
I think the floor will recover. Eventually. In the way it recovered from that ancient turbidity current. It will incorporate the disturbance, build over it, bury it under new layers of slow accumulation. In 10,000 years, maybe the mining scars will be 2 meters down, a distinct layer in the core. Evidence that something happened.
But that's not the recovery we're asking about. We're asking: will it go back to normal in 5 years? 10 years? 50 years?
And that's a much harder question.
The sediment accumulation rate in this area is about 3 millimeters per thousand years. Let's say mining removes the top 10 centimeters of sediment. That's 10,000 years of accumulation, gone. To replace it naturally, at the background rate, would take... 33,000 years.
Thirty-three thousand years to rebuild the physical structure. And that's assuming nothing else changes. No altered currents, no new disturbance, no changes to the rain of particles from above.
That's not recovery on a human timescale. That's recovery on a geological timescale.
So maybe the real question isn't "will it recover?" but "what do we mean by recovery?"
If recovery means the floor is stable and functional and supporting life — that might happen faster. Pioneer species can colonize quickly. Some kind of community can establish itself. Not the same community, but a community.
If recovery means the floor looks exactly like it did before, with the same sediment structure and the same species assemblages and the same millimeter-scale topography — that might never happen.
Those are different definitions of success.
I don't have an answer. I'm just the cataloguer. I document what we find, what we measure, what we damage, what we try to restore. I organize the data so future researchers can access it, compare it, build on it.
But on nights like this, when it's just me and the cores and the quiet, I think about what the data means.
Each core is a story. Not just what's there now, but how it got there. Layer by layer. Event by event. Slow accumulation and sudden disruption and slow accumulation again.
The floor has been through catastrophes before. Natural ones. Turbidity currents, volcanic ash falls, maybe even asteroid impacts. The geological record is full of disturbance.
And the floor is still here.
That's not permission to add more disturbance. It's not justification for mining. It's just... context. The floor is resilient on its own timescale. Not fragile, exactly. Just slow.
And we're fast.
That's the mismatch. Our timescales don't align. We measure in years and decades. The floor measures in millennia and epochs. We want answers now. The floor takes its time.
So we do these studies. We collect baseline data, document disturbances, go back after 5 years or 10 years and see what's changed. We publish papers, write reports, update the catalogue.
And the floor keeps doing what it's always done. Accumulating sediment. One grain at a time. One millimeter per thousand years.
Outlasting us all.
It's almost three in the morning now. I should sleep. But I think I'll finish cataloguing this last core first. It's 4.2 meters long, spanning 62,000 years.
Someone drilled it. Someone analyzed it. Someone sent me the data.
And now I'm making sure it's preserved. Documented. Accessible. Part of the permanent record.
Just like the sediment layers themselves. One entry at a time. One dataset at a time.
Building the archive, slowly.
The floor would approve.