Why Your Refugium Isn’t Working (And How to Fix It)

Why Your Refugium Isn’t Working (And How to Fix It)

I killed my first refugium setup over the course of about six months. Not dramatically — no sudden crash, no mass die-off. Just a slow, boring failure where the chaeto stopped growing, the nitrates crept up, and eventually I pulled the whole thing out in frustration and replaced it with a bucket of GFO and a prayer. I told myself refugiums were overhyped, that they were for people with too much time and not enough equipment.

That was three tanks ago. I’ve had a working refugium for the last four years now, and I can tell you exactly where I went wrong the first time. It wasn’t the concept. It was everything about how I set it up.

The Light That Fools You

Most people start with the wrong bulb. I did. I bought a clip-on desk lamp from a hardware store, screwed in a 6500K CFL, and assumed that was good enough because the package said “daylight.” It grew algae on my display tank’s glass just fine, but the chaeto in the refugium stayed pale and sad for weeks. I kept thinking it just needed time.

What I didn’t understand then — and what nobody told me — is that macroalgae needs a specific spectrum ratio, not just color temperature. The cheap CFLs put out a lot of green and yellow light, which is what land plants use. Marine macroalgae, especially chaetomorpha, wants more red and blue. The difference isn’t subtle once you see it. A proper refugium light — I use a Kessil H80 on the current tank, but even a decent PAR38 grow bulb from Amazon will do the job — turns the chaeto from washed-out green to almost neon within a week. Growth rate triples. Nitrate consumption follows.

I run mine on a reverse photoperiod: on at 10 PM, off at 10 AM. The pH stabilization during the night cycle is a bonus I didn’t plan for but won’t give up now. The tank’s pH swings about 0.15 points between day and night instead of 0.4.

The Flow Problem Nobody Warns About

Here’s the thing about refugiums that every article skips: the flow rate through the compartment matters more than the flow rate through the tank. I had a return pump pushing 600 GPH through the whole system, and the refugium section was just a pass-through. The water moved too fast. The chaeto tumbled, sure, but it didn’t have time to strip nutrients out of the water column before the water left.

I talked to a guy named Marcus at a reef club meetup in San Jose — he kept stony corals, SPS dominant, with a refugium the size of a shoebox. His nitrates were undetectable. I asked him about flow. He said he ran a separate pump just for the refugium, throttled way down, and had the water trickling through rather than rushing. “Think of it like a slow river through a marsh, not a fire hose through a drain pipe,” he said.

I went home, added a ball valve to the refugium feed line, and dialed it back until the water turnover in the refugium was maybe once every 45 minutes. The chaeto started darkening within days. “The macroalgae needs time to breathe the water, not just see it pass by.” I’ve repeated that line to probably a dozen people since.

What Grows in the Dark Corners

The refugium I have now is a 20-gallon long tank plumbed into the basement sump. It’s not pretty. It’s under the house, in a crawl space that smells like damp concrete and salt creep. The lighting is on a timer. There’s a small powerhead in one corner to keep dead spots from forming. I check it once a week, usually on Sunday mornings when the rest of the house is still asleep.

What I’ve noticed — and this took me about a year to figure out — is that a refugium develops its own microfauna population that matters way more than the macroalgae. Copepods, amphipods, little bristle worms, tiny spaghetti worms. The chaeto provides structure for them to live in, but the real nutrient export happens because these creatures are constantly processing detritus and waste. The chaeto is the visible part. The invisible part is the ecosystem living inside it.

I stopped cleaning the refugium so aggressively. I used to pull out all the chaeto every two weeks and rinse it in tank water, trying to keep things pristine. Now I only harvest about half the growth every three weeks, and I leave the bottom layer undisturbed. The pod population exploded after I made that change. The mandarin dragonet in the display tank started fattening up without any supplemental feeding.

The Sand Bed Question

I ran a bare-bottom refugium for two years. It worked fine for nutrient export, but I never got the biodiversity I wanted. Adding a two-inch layer of fine aragonite sand changed everything. The sand bed hosts worms, mini brittle stars, and a whole class of bacteria that the bare glass never supported. The trade-off is that you need to stir it occasionally to prevent hydrogen sulfide pockets from forming. I use a turkey baster — the same one I use for target-feeding corals — to gently poke through the sand once a month. A few bubbles come up, the water clears, and the sand stays aerobic.

The Harvesting Mistake Everyone Makes

You’re supposed to harvest macroalgae regularly. Everyone knows that. What nobody says out loud is how much and how often. The first year I ran a refugium, I harvested every two weeks, removing about half the chaeto each time. The nitrates stayed at 10-15 ppm. I couldn’t figure out why.

Then I read a thread on Reef2Reef from someone who ran a commercial frag operation. He said he harvested 80% of his chaeto every week. Eighty percent. I thought that was insane — surely you need enough biomass to maintain the biological filtration. But I tried it. Pulled out huge handfuls, left only a loose ball about the size of a softball in a 20-gallon refugium. Within three days, the nitrates dropped from 12 ppm to 2 ppm.

The logic is counterintuitive but makes sense: young, fast-growing chaeto consumes nutrients much more aggressively than old, dense mats. By harvesting aggressively, you’re constantly pushing the macroalgae into its exponential growth phase. The old stuff is just sitting there, consuming light and space without exporting much.

I now harvest every six days. Set a reminder on my phone. Pull out roughly three-quarters of what’s there. The chaeto grows back so fast I can see the difference between Saturday and Monday.

The Nutrient Balancing Act

There’s a trap that a lot of people fall into with refugiums, and it’s the opposite of the problem I had. Instead of not exporting enough, they export too much. The chaeto strips nitrates and phosphates so aggressively that the corals start to pale. You see it in SPS first — the tips start to lighten, then the whole colony looks washed out. The tank is too clean.

I hit this wall about a year ago. My nitrates were consistently below 1 ppm. Phosphates were undetectable on a Hanna checker. The acropora were alive but not coloring up. No pests, no disease, just… beige.

I had to start dosing nitrates. I use potassium nitrate — the same stuff sold for planted freshwater tanks, label and all. A quarter teaspoon dissolved in RO water, dripped into the display over an hour. I do it once a week now, aiming for 5 ppm nitrate. The corals colored up within three weeks. The chaeto kept growing, but I just harvested more frequently to compensate.

The balance is real. A refugium is a tool, not a set-it-and-forget-it solution. You have to adjust based on what the tank tells you.

The Noise You Learn to Ignore

The refugium in the basement makes a sound. It’s not loud — a constant trickle, like a small stream running over rocks. At first it annoyed me. I’d go down there and adjust the plumbing, try to silence it, never quite succeeding. After a while I stopped noticing. Now, when I’m up late working and the rest of the house is quiet, I can hear it faintly through the floor. It means the system is running, the chaeto is tumbling, the nutrients are being processed.

Last week I replaced the return pump with a slightly larger model. The flow changed. The sound changed. I spent an hour with a wrench and some PVC fittings getting the trickle back to exactly where it was before.

I’m planning some changes — a deeper sand bed, maybe chaeto and caulerpa together to see if the combination works better. If you’ve found a different approach that works for your tank, drop a comment below.