Rewetting peat next to a solar build: three years in

The ecology is ahead of plan and the maintenance cost is double what we budgeted. Both were surprises.

In 2022 we committed to rewetting 96 hectares of drained lowland peat adjacent to a 60 MW solar build. The commitment was made for two reasons that were both true and were not the same: it was the condition on which the consent was granted, and it was, on our own numbers, the single largest carbon item available to us on the site — larger than the generation itself over a thirty-year horizon.

What we measured

Three growing seasons in, with quarterly water table monitoring at fourteen dipwells and annual vegetation survey on sixty quadrats:

• Mean water table sits 8 cm below surface against a target of 20 cm. We are wetter than we planned to be, sooner.
• Sphagnum is established on 61% of quadrats against a year-three target of 30%.
• Two target species arrived without being introduced.
• Modelled emissions reduction is running at about 1.4 times the case we submitted.

That is the good surprise, and it is worth being precise about why it happened. Our hydrological model treated the drainage network as more effective than it was. Half of the field drains had already collapsed. Blocking a partly-failed network is a much cheaper way to raise a water table than blocking a functioning one, and we were doing the former while pricing the latter.

The bad surprise

Maintenance is running at €118 per hectare per year against a budget of €55. The overrun is almost entirely in three lines.

• Dam repair. Peat dams settle and are breached by the same high-flow events that make them necessary. We budgeted for inspection and occasional repair. The reality has been an annual campaign.
• Scrub control. Birch colonised the wetter margins faster than the survey predicted and has to be removed by hand, because the ground will not carry a machine — which is the point of the project.
• Access. We designed access for a dry site and then made the site wet. Two of the four routes are now seasonal, which roughly doubles the labour cost of anything done between October and March.

We made the site wetter than planned, faster than planned, and every one of our cost assumptions was written for a site that was neither.

What that does to the case

Over thirty years the maintenance overrun is about €180k in present value. The carbon outperformance, valued at the price we use internally, is worth substantially more than that. The project is better than we underwrote it, and it costs more than we underwrote it, and both statements are true at once.

The reason we are writing it up is that in the version of this that ends up in a sustainability report, only the first statement appears. That version is not useful to anyone deciding whether to do the same thing.

What we would do differently

• Survey the drains before modelling the hydrology. Two days of walking the network would have told us the drains had failed, and we would have designed a cheaper intervention and forecast a wetter outcome.
• Budget maintenance against the intended end state, not the current one. Every cost we got wrong, we got wrong because we priced the site as it was on the day we surveyed it.
• Put the access design after the hydrological design. Ours ran in parallel, which is how you end up with a track that works only in the condition you are trying to eliminate.
• Agree the monitoring period with the regulator up front. Ours is open-ended in practice, which makes the liability hard to close and hard to price for a buyer.

Would we do it again

Yes, and we are — 140 hectares adjacent to a project consented last year, with the drain survey in the first work package and a maintenance budget of €120 per hectare. The ecology case for this work is strong and the carbon case is stronger than we expected. What we no longer claim is that it is cheap.