The Drainage-Layer Debate: Does a Gravel Bottom Actually Drain?

Nearly every terrarium guide, including others on this site, teaches the same bottom layer: gravel, LECA, or lava rock under the substrate, to give excess water somewhere to go in a container with no drainage hole. It's good, standard advice, repeated so consistently across care guides and forums that it rarely gets questioned. What gets skipped over far more often is whether that layer actually "drains" the way the word implies — and the honest answer is that it's genuinely debated, not settled, once you look at the mechanism underneath it.

The case for it

The argument in favour is simple and correct as far as it goes: a terrarium has no hole in the bottom, so if you overwater it, the excess water has nowhere to leave the container. A layer of coarse material below the substrate gives that excess somewhere to sit that isn't in direct contact with roots, buying a genuine margin of error against a watering mistake. That's real, useful protection, and it's why the layer remains standard advice rather than a discredited one.

The mechanism that complicates it

Here's where it gets more interesting. When a fine material — potting substrate — sits directly on top of a coarse one — gravel or LECA — water doesn't move freely across that boundary just because gravity is pulling down on it. Capillary forces hold water in the fine material against gravity, right at the point where the two materials meet, until the substrate immediately above the boundary is genuinely saturated. Soil scientists call this a perched water table, and it's a well-documented phenomenon in horticulture and civil engineering alike, not a fringe theory — the same physics shows up in raised garden beds, green roofs, and sports-field drainage design, wherever a fine material sits directly on a coarser one.

It's worth being precise about what the mechanism does and doesn't claim. It doesn't mean the drainage layer holds water forever with no limit, or that a terrarium built this way is doomed to stay soggy. It means the boundary between the two materials resists water movement until the fine layer right above it is saturated, at which point water does move into the coarse layer below — just later, and holding more water in the substrate above the boundary, than a simple "gravity pulls it straight through" mental model would predict.

In an outdoor pot with a real drainage hole, that perched layer eventually overflows the pot's threshold and the excess genuinely leaves the system through the hole. In a sealed terrarium with no hole anywhere, there's no threshold to overflow past — the water at that boundary just sits there. Practically, that means the bottom few centimetres of substrate directly above a terrarium's drainage layer can stay wetter for longer than most builders assume, even when the drainage layer itself looks dry.

A worked comparison: with a hole vs. without one

This site's own drainage layer calculator treats these two situations very differently, which is a useful way to see the debate reflected in real numbers. Take a 20cm-diameter jar floor (about 314cm²) with a 10cm substrate layer on top. If that container genuinely has a drainage hole — a pot, not a sealed terrarium — the calculator recommends a 1cm drainage layer, about 0.31 litres of material. Build the exact same footprint and substrate depth into a sealed vessel with no hole, and the recommendation roughly doubles: 2.5cm, about 0.79 litres. That gap isn't arbitrary — it's the calculator building in extra reservoir capacity specifically because a hole-less container can't flush a perched layer the way a drained pot can, so the margin of error has to come from a bigger buffer instead. It's a concrete, numeric version of exactly the mechanism described above: no hole means no overflow point, so the only lever left is building a bigger reservoir to begin with.

Depth compounds the effect. For that same hole-less jar, a shallow 5cm or 10cm substrate layer both hit the calculator's practical minimum of 2.5cm of drainage material, but stepping up to a 20cm substrate depth calls for 3.6cm of drainage, and a 30cm depth calls for 5.4cm — the reservoir has to grow faster than the substrate depth above it does, because a deeper, heavier substrate column holds a larger perched-water zone at its base once it's saturated.

What the substrate-layer ratios already build in

This same open-vs-closed distinction shows up one level up, in how much of a container's total height gets allocated to drainage in the first place. For a 25cm-tall vessel, this site's substrate-layer ratios give an open terrarium about 2.5cm of drainage against 19.3cm of substrate, while a closed terrarium of the same height gets about 3.8cm of drainage against a shallower 18cm of substrate. A closed vessel recirculates more of its own moisture and has less margin if that reservoir runs out of room, so it gets proportionally more drainage allowance — the same logic as the hole/no-hole comparison above, just expressed as a design ratio rather than a stark two-way split.

So is the layer worth building?

Yes — but it's worth being precise about what it's actually doing. It is not plumbing. It doesn't remove water from a sealed terrarium the way a hole removes water from a pot. What it is, honestly, is a reservoir: a place for excess moisture to collect below the root zone rather than against it, buying time and margin for a watering mistake to be corrected before it becomes root rot. That's a real and worthwhile function even though it isn't "drainage" in the everyday sense of the word.

Because it's a reservoir rather than an outlet, the layer depths above are a starting point, not a guarantee, and watching the terrarium's actual behaviour matters more than trusting the ratio alone. A closed jar with heavy, unclearing condensation and a substrate surface that stays visibly dark and wet needs venting and less water regardless of how generously its drainage layer was built — the reservoir can only hold so much, and a persistently overwatered terrarium will eventually saturate it no matter the depth.

Should you skip the layer, then?

No — and it's worth being clear about that, because "the mechanism is debated" is not the same claim as "the layer doesn't work." Even as an imperfect reservoir rather than true plumbing, a drainage layer still does something a terrarium with none at all can't do: it keeps standing water below the roots instead of directly against them, and it gives you a buffer to correct an overwatering mistake within, rather than having excess water sitting straight against root tissue from the moment it's added. Skipping the layer entirely removes that buffer altogether, which is a strictly worse position than building one, even an imperfect one. The debate is about precisely how the layer behaves and what claims are fair to make about it, not about whether to build it in the first place.

Where this actually changes practical advice is in how much weight to put on the layer depth alone. Treating a correctly-sized drainage layer as a guarantee against overwatering, and watering on a fixed schedule because "the drainage layer will handle it," is the mistake the perched-water-table mechanism argues against. Treating the same layer as a genuine but limited buffer, worth building carefully and then backing up with attentive watering and good airflow, is the position this mechanism actually supports.

An alternative worth knowing: wicking rather than reserving

Some more advanced builds sidestep the perched-water-table question differently, using a wick — a strip of absorbent cord or fabric run from a small water reservoir up into the substrate — to move moisture upward by capillary action instead of relying on a bottom layer to hold excess water passively. This is a genuinely different approach to the same underlying problem (a sealed container with no drainage hole), and it isn't something the calculators on this site model, since it's a specialised technique most builders won't need. It's worth knowing it exists mainly because it illustrates the same point from another angle: builders who've spent time thinking carefully about how water actually behaves in a sealed container keep arriving at solutions that treat the classic gravel layer as one reasonable option among several, not the only correct way to manage moisture in a terrarium.

A separate problem the drainage layer doesn't solve

It's worth clearing up one common point of confusion: a mesh or landscape-fabric barrier laid between the charcoal and substrate layers has nothing to do with the perched-water-table question. That barrier's job is to stop fine soil particles from washing down into the drainage material and clogging it over time — a completely separate, mechanical problem from how water behaves at the substrate/gravel boundary. Both are worth doing in a well-built terrarium; neither one solves the other's problem.

What actually keeps a terrarium balanced

Given that the drainage layer is a buffer rather than an outlet, the things that do the most real work in keeping a closed terrarium's moisture balanced are the ones covered elsewhere on this site in more depth: watering conservatively and re-checking rather than soaking heavily on a schedule, reading the condensation pattern on the glass as the main day-to-day signal, and airing the lid out periodically rather than relying on the substrate to self-correct. A springtail population helps too, grazing on the mould and decay that overly damp, stagnant conditions tend to produce, regardless of how the drainage layer underneath is built. None of that is a knock against building the layer in the first place — it's simply a reminder that the layer is one part of a system, not a substitute for the rest of it.

If you want to see how these layer depths and drainage volumes play out across a range of common vessel sizes rather than just the one worked example here, the vessel size reference runs the same substrate-layer and drainage-layer calculations across eight typical shapes and sizes side by side.