One reason may be the way water travels through your filtration system.

Every time CO₂-rich water falls through air, splashes, entrains bubbles or passes through a highly exposed water surface, it gets another opportunity to exchange gases with the atmosphere.

Different filtration designs can therefore require very different amounts of injected CO₂ to maintain a similar level in a planted aquarium.

The important point is that this is not simply about which filter is “best.”

It is about understanding where your water goes, how often it meets air, and what happens to the injected CO₂ along the way.

Why Does Injected CO₂ Leave the Aquarium?

A CO₂-injected planted aquarium deliberately raises dissolved CO₂ above the level it would normally reach in equilibrium with the surrounding air.

Once we do this, there is naturally a tendency for some of that additional CO₂ to move back out of the water.

This happens through gas exchange.

Where water and air meet, gases can move in either direction. Oxygen can enter the aquarium while excess dissolved CO₂ can leave it.

Increasing the interaction between air and water generally increases the opportunity for this exchange.

That is why splashing, falling water, bubbles and turbulent air-water mixing matter.

But there is an important distinction:

Water movement itself does not automatically mean high CO₂ loss.

Water can move quickly through a hose, filter or underwater current without being continually exposed to fresh air.

So instead of asking:

“How strong is my filter?”

A more useful question is:

“What happens to the water while it is moving?”

Circulation and Gas Exchange Are Not the Same Thing

Good circulation is important in a planted aquarium.

CO₂-rich water still needs to travel around hardscape, through plant groups and across leaves. Reducing circulation simply to retain more CO₂ can therefore work against what we are trying to achieve.

Consider two streams of water carrying the same flow.

One travels underwater.

The other falls several centimetres through air before hitting the aquarium surface.

Both are moving water.

But they create very different conditions for gas exchange.

The falling stream exposes water directly to air and, when it strikes the surface, may pull air beneath the water and produce bubbles.

This is known as air entrainment.

The greater the opportunity for air and water to interact, the greater the potential for gas transfer.

How Different Filters Affect Injected CO₂

There is no useful universal percentage such as “this filter loses 10% CO₂” or “that filter loses 50%.”

Tank dimensions, flow rate, water level, return position, overflow design, sump configuration and many other factors affect gas exchange.

But we can still understand the typical opportunities for CO₂ loss created by each filtration design.

The filter name gives us clues.

The behaviour of the water tells us much more.

1. Canister Filter

Generally favourable for CO₂ retention

A conventional canister filter provides a relatively simple water path.

Water leaves the aquarium through an intake, travels through tubing and the sealed filter body, then returns through another hose.

For most of that journey, the water is not directly exposed to atmospheric air.

This makes the filtration path itself relatively favourable for retaining dissolved CO₂.

However, the return to the aquarium still matters.

A submerged outlet can produce strong circulation throughout the aquarium while causing relatively controlled surface movement.

Point that same outlet aggressively upward and it can continuously disturb the surface.

The canister itself has not changed.

The gas-exchange conditions have.

What to do

Maintain good circulation throughout the aquarium.

Position the return so that the surface remains active without unnecessarily violent disturbance.

There is usually little benefit in sacrificing useful underwater circulation simply to conserve CO₂.

2. Hang-On Canister Filter

The return matters more than where the filter sits

A hang-on canister should not automatically be grouped together with a hang-on-back waterfall filter.

If water travels through a largely enclosed filter and returns through a submerged pipe or outlet, its CO₂ behaviour can be quite similar to a conventional canister.

If the return instead releases water above the aquarium surface and creates a cascade, the situation changes.

This gives us a useful general principle:

Judge the filtration system by the water path, not simply by the filter's name.

A filter hanging outside the aquarium does not automatically cause significant CO₂ loss.

The important question is whether the water remains enclosed before returning underwater—or repeatedly meets fresh air along the way.

3. Hang-On-Back Waterfall Filter

Water level can make a major difference

A typical hang-on-back filter creates another potential gas-exchange point when water returns over the outlet lip.

But even here, the effect is not always the same.

Consider a HOB filter on a nearly full aquarium.

The outlet may sit very close to the water surface. Water flows from the filter into the tank with relatively little drop.

Now allow evaporation to lower the aquarium water level by several centimetres.

The same return has become a small waterfall.

Water falls through air, hits the aquarium surface and may drag air beneath it.

You may even see a continuous cloud of small bubbles around the impact area.

The filter has not changed.

The water level has changed the way it interacts with air.

What to do

For a CO₂-injected aquarium using a waterfall-style HOB, maintaining an appropriately high water level can reduce unnecessary falling and splashing.

Do not assume the filter must immediately be replaced.

First look at what the returning water is actually doing.

Look Below the Surface

One of the easiest ways to understand your own system is simply to watch the water.

Gentle surface rippling is normal.

But if falling water is continuously dragging a large quantity of bubbles deep beneath the surface, the water is interacting much more aggressively with atmospheric air.

This is useful when the objective is aeration.

It is less efficient when we are simultaneously injecting additional CO₂ into the same water.

4. Overflow and Sump Systems

Not automatically bad for CO₂—but potentially inefficient

Sump systems deserve more explanation because simply saying “sumps lose CO₂” is misleading.

Consider the journey:

Display tank → overflow → drain → sump → return pump → display tank

There are potentially several places where water can interact with air.

Water crosses the overflow.

It travels down the drain.

It enters the sump.

It moves between sump chambers.

Then it is pumped back into the aquarium.

If these transitions involve falling water, splashing and large quantities of entrained air, the system can provide repeated opportunities for dissolved CO₂ to leave the water.

This principle is not unique to aquariums.

Systems designed specifically to remove dissolved gases from water deliberately increase contact between air and water.

But not every sump operates this way.

A Quiet Sump and a Splashing Sump Are Very Different

Imagine two sump systems.

In the first, the overflow continually sucks air. Water and air travel together through the drain before crashing into the sump and producing large quantities of bubbles.

In the second, most of the water travels through a flooded siphon with minimal air entrainment and enters the sump quietly below the water surface.

Both aquariums use sumps.

But their potential for gas exchange is very different.

That is why the statement:

“Sumps waste CO₂.”

is too simplistic.

A more useful statement is:

A sump provides additional opportunities for CO₂ loss, and how much is lost depends heavily on how the overflow, drain and sump are designed.

What About Covering the Sump?

Covering an open sump can reduce direct exchange between the sump water and constantly refreshed room air.

That can help reduce one source of unnecessary gas exchange.

But a cover cannot correct everything happening elsewhere.

If the display overflow is turbulent, the drain continually pulls in air or water crashes into the sump before reaching the covered section, gas exchange is already occurring.

A sump cover should therefore be considered one useful measure, not a complete solution.

Where Should CO₂ Be Injected in a Sump System?

There is no single injection position that is correct for every sump.

Different systems may use diffusers, inline atomisers, reactors or other methods of dissolving CO₂.

The more useful principle is:

Avoid creating CO₂-rich water immediately before an unnecessarily aggressive degassing stage.

If CO₂ is dissolved efficiently only for that water to immediately pass through a turbulent, air-filled transition, some of that effort is being wasted.

Depending on the design, CO₂ may instead be introduced into the return side or directly into the display so that CO₂-rich water can be delivered effectively to the planted area.

The objective remains:

Dissolve it → circulate it → deliver it to the plants → avoid unnecessarily stripping it out.

Does Higher Filter Flow Mean More CO₂ Loss?

Not necessarily.

This is an important distinction.

A high-flow closed circuit can move a large amount of water while providing relatively little opportunity for air-water interaction.

A much slower stream falling through air can create considerably more gas exchange.

So reducing filter turnover simply because CO₂ is being injected is not necessarily the correct response.

For planted aquariums, circulation remains valuable.

The objective is not to make the water still.

The objective is to move water efficiently without creating unnecessary gas exchange.

Don't Eliminate Surface Movement Just to Save CO₂

Once we understand how injected CO₂ escapes, it is easy to take the idea too far.

Reduce every ripple.

Stop surface movement.

Seal everything.

Retain as much CO₂ as possible.

That is not the goal.

Fish, shrimp, microorganisms and plants all depend on adequate oxygen availability.

Surface gas exchange is one of the ways oxygen enters the aquarium. Increasing gas exchange may make CO₂ injection less efficient, but eliminating too much gas exchange can reduce the aquarium's oxygen margin.

A well-run CO₂ planted aquarium therefore requires balance:

Good circulation.
Adequate gas exchange.
Sufficient oxygen.
Stable CO₂ delivery.

We are not trying to prevent CO₂ from ever leaving the aquarium.

We are trying to avoid wasting excessive amounts of it unnecessarily.

Before Increasing Your CO₂, Follow the Water

If your plants appear to need more CO₂, increasing the injection rate may eventually be necessary.

But first, spend a minute watching your aquarium.

Follow the water.

Is your HOB return several centimetres above the water?

Is your spray bar violently breaking the surface?

Is air constantly being sucked into an overflow?

Does water crash into your sump?

Are large quantities of bubbles being carried underwater?

Has evaporation quietly turned a smooth filter return into a waterfall?

These observations can tell you a great deal about how efficiently your aquarium is using the CO₂ you are already injecting.

Sometimes you genuinely need more CO₂.

Sometimes you need better circulation.

And sometimes you are already injecting plenty—

but your filtration system is giving that CO₂ too many opportunities to escape.

This article is part of Green Chapter’s Knowledge Hub, where we share practical guides on terrariums, aquascaping, and living ecosystems. If you’d like to go further, explore more guides or join one of our workshops to experience it hands-on.