ON the west's mountain rivers, the first thaw of winter snow sends ice water bouncing down foaming white beyond the glow of holly berries. No fish of any consequence, you'd think, could want to pass Christmas up there among the boulders, the glittering gravel the bare, peaty banks.
Tucked up under the scarp of Nephin Beg, in north Mayo, the highest reaches of the Deel Rivet were, indeed, empty of fish in winter for some unguessable number of centuries. Salmon ascending from the Moy to spawn (by way of Lough Cullin and Lough Conn) found the way finally barred by a falls in the Deel that slanted up for five steep metres. Leaping again and again, to the insistence of ancient co-ordinates, the fish eventually tired and fell back in defeat.
Such, at least, was the picture until this autumn, when the long reach of a pneumatic rockbreaker, chipping away noisily for weeks, sculpted the old red sandstone into a series of pools like the tiered basins of a fountain. Standing there in early November, biologist Dr Martin O'Grady of the Central Fisheries Board watched the first salmon leaping up to enter the final five miles of stream. Cost: £5,000; benefit: about 3,000 extra salmon smolts a year.
South of Lough Cullin, in the Castlebar River, a long, flat sill of rock set another kind of barrier to migrating salmon and trout: only a rare fish, gasping and wriggling, reached the higher miles of river beyond. For another £5,000, the rockbreaker engraved a channel through the sheets of rock, and autumn found salmon and many big lake trout queueing up to make the passage to new (or perhaps very old) spawning grounds.
The North-Western Regional Fisheries Board, whose development all this is, could have opted to build concrete fish passes at both barriers, which would have looked horrible and cost up to 20 times as much. Not the least pleasing aspect of their £2 million programme of "enhancement" works, co-funded from Europe, is how unobtrusive - how ingeniously natural, indeed - they are proving to be.
The quick fix of so much local fishery development has been to re-stock with artificially-hatched fish. But there's nothing wrong with the birth-rate of the salmon and trout in the great network of rivers, lakes and streams that make you the Moy catchment. What is needed is to help more baby fish to survive to a decent size, and to create more good places for catching them.
An exceptionally thorough survey by the Central Fisheries Board divided the river and their scores of tributary streams into spawning and nursery zones, with a separate, costed plan (and its benefit) for each. A bonus was the discovery, by Dr Andrew Ferguson of Queen's University, that all the brown trout in the catchment have remarkably similar genes, so that extra fish reared in any one stream will benefit the catchment as a whole.
What is so wrong with the streams that they rear, in some places, only one-tenth of the fish they could, and that thousands of smolt are doomed to be nomads in summer, searching for shelter and food?
Pollution causes its own, local problems, most of them well-identified. But the big challenge of "enhancement" is to restore the streams to a more natural flow, and to furnish them with the shade, shelter, vegetation and insect life.
Travelling round with Martin O'Grady, I learned an important new word - "thalweg", from the German. This is the natural valley that any stream, left alone, carves into its bed. The thalweg swings from one bank to the other, steering the floods in winter and saving a flow of deep water for the fish in summer.
The contour is missing from dozens of streams that flow through the Mayo farmland. They were dug out in the drainage schemes of the 1960s and left as wide, shallow, flat-bottomed channels. In winter, the floods undermine the banks, spreading the bed still further. In summer, long stretches of the streams dry out. Cattle on the bare, unfenced, unvegetated banks, tread them down as they wander in to drink and graze the water-plants to their roots.
Repairing and managing the streams for recovery involves something called "fluvial morphology" - knowing what will happen for a given flow, volume, gradient, substrate, shape of obstruction. You use it to create thalwegs, meanders, riffle-glidepool sequences, places for hiding, places to wait for food to drift by. You learn to think like a fish.
The American wildlife engineers, says Dr O'Grady, have got very good at repairing nature in this way. His own experience in the US has guided the placing of thousands of boulders, great beds of cobbles and gravel, a small forest of logs of Japanese larch, long-lasting under water.
PINNED down into banks, the logs check erosion and yet create shadowy shelter for fish. Willow slips stuck in behind them quickly grow into bushes with a network of roots to grip the soil. A fence behind the willows gives the whole "riparian zone" along the bank a chance to grow the streamside plants that nourish the insects that feed the fish. Even in December, the vegetation of these newly-protected streams catches the eye with its lushness: a Ph. D. for somebody in the botany of recovery!
Still a long way off, alas, is recovery of water quality in the lakes at the hub of the Moy catchment and into which the Deel, the Castlebar and all these little streams flow.
The Cullin has been virtually unfishable for the last two seasons and anglers have left in disgust at the blankets of algae on the water. In Conn, the trout, waxing fat on the lake's enrichment, lurk at the bottom to feed on snails instead of coming up for flies. Even the Moy itself is growing green algae, all the way to Ballina.
Cullin will start to clear when Castlebar takes the phosphate out of its sewage. Lough Conn is beset not only by Crossmolina's waste but by farmers double-dosing their land with slurry and fertilizer, always just ahead of the rain. Improvement, if it ever comes, will be too late for Conn's pollution-sensitive Arctic charr which, after almost 10,000 years in the lake, must now be reckoned extinct.
