A drained marsh rarely looks like a marsh. More often it is a flat field of rye grass, straight ditches and a sluice gate rusted open. Rewetting sounds straightforward: close the drains, let the water back in. In practice, the water goes somewhere else, the plants that arrive are not the ones you hoped for, and the paperwork turns out to be the easy part. These five mistakes come up again and again, from half-acre field corners to whole floodplain compartments. None is fatal on its own. Two or three together can set a site back a decade.
A marsh is defined by how long water sits in it, not by whether water arrives. Block an outlet and you create a pond in the ditch and very little else; the rest of the field stays dry because the soil beneath is cracked, compacted and served by a network you cannot see.
Before designing anything, work out the water balance. Where does the water come from, how high does it stand in winter, and how quickly does it leave? Look at the catchment feeding the site, any tidal influence at the outfall, and whether groundwater is pushing up or draining away. On peat, remember that dried peat has shrunk and oxidised. The surface may now sit well below the original water table, and gravity will not help you.
Aim for control rather than a permanent flood. An adjustable structure at the outfall, with boards you can add or remove, lets you hold water high from autumn to spring and draw it down in summer. That wet-dry rhythm suits reed, sedge and rush communities far better than year-round stagnation, and it gives you a way to correct mistakes without calling in heavy machinery.
Old marshes were drained with more than ditches. Clay tiles, mole plough channels, rubble-filled grips and field drains can run for hundreds of metres and keep working long after the ditches are blocked. Water finds them. So does your newly rewetted site, which is why one corner stays stubbornly dry while another floods a metre deep.
Do the detective work in a dry spell. Walk the field after heavy rain and mark every strip that drains faster than its surroundings. Dig out old maps, estate records and aerial photographs. A soil auger will tell you more about the top half metre than any amount of guessing from the gate.
Where you find drains, decide whether to break them, fill them, or leave them and design around them. Removing nutrient-rich topsoil can help a fen community establish, but it is disruptive, releases carbon and can destroy buried seed banks. It is not a decision to take lightly, and not one to make without an ecologist on site.
Wet ground is a gift to invasive plants. New Zealand pygmyweed, parrot's feather, floating pennywort and Himalayan balsam all spread readily on damp machinery, boots and the muddy tracks of vehicles. Restoration sites are often the best remaining habitat in the area and the first place these species get a foothold. Assume every item of kit arriving on site is contaminated, then act accordingly.
Check the site at least twice in the first two growing seasons. Finding a small patch early is a morning's work; a mature infestation is a decade.
Rewetting a drained, fertilised marsh does not reset it to a species-rich fen. It often produces nettles, willow scrub, duckweed and mats of algae, because the nutrients are still in the soil and rewetting can release phosphate that was locked up in the dry profile.
Think about the whole catchment, not just the field. If water arrives from a road, a yard or improved pasture, it brings nutrients with it. A rough grass buffer, a silt trap or a shallow settling pool at the inlet can strip part of that load before it reaches the marsh. Within the site, avoid long periods of stagnant, warm, shallow water; a gentle through-flow or a summer drawdown limits the worst algal growth.
Set realistic targets. On a nutrient-rich site, a rushy wet grassland with good invertebrate interest may be a genuine achievement, and it may be what the land can support for many years yet.
The most common regret on restoration projects is not knowing what the site looked like beforehand. Without a baseline you cannot tell whether the rewetting worked, whether water levels are behaving, or whether that new plant is an arrival or a survivor.
Start six to twelve months before the works. Water levels first: a handful of dipwells made from slotted pipe, read monthly, will tell you more about your hydrology than any model. Then vegetation. Fixed quadrats, recorded at the same time each year, show change over time, as do fixed-point photographs taken from a marked spot with the date and direction noted.
Keep a works diary as well. Where the digger tracked, which drains were broken, the date the boards went in, the day the site first flooded. In three years that notebook will explain results that otherwise look baffling.
If you are at the beginning, keep the first year small and reversible.
Get the water right and the rest becomes possible. Get it wrong and no amount of planting will compensate. Check whether the site is designated, as a Site of Special Scientific Interest, a local nature reserve or land with protected species, and speak to the relevant statutory conservation body before works begin. If your plans affect flood risk to neighbours or alter a watercourse, consent is likely to be needed too. A few days spent on advice at the start is time well spent.
Photo: MabelAmber / Pixabay