Then she yawned. At least she had forgotten what had woken her up originally. She looked a little bewildered.

I said, ‘I know, darling. You’re right. Switch out the light and let’s get some sleep.’

But I couldn’t sleep. I lay awake thinking about our marriage, and wondering whether I was being unfair to Mary, or she to me. I asked myself whether things might have been different if we had had children. I thought about salmon spawning in the highlands of the Yemen. Round and round my head went these thoughts, chasing each other like salmon parr wriggling in the shimmering water of a stream.

I got out of bed and came next door. I thought that maybe writing up my diary would help me sleep.

It hasn’t.

Proposal submitted by Dr Alfred Jones of the National Centre for Fisheries Excellence (NCFE) to Fitzharris & Price, 28 June

Executive summary

NCFE has been invited to advise and comment to Fitzharris & Price on the feasibility of introducing migratory salmonids into the wadi systems of the Yemen. The longer-term objective is to develop opportunities for good-quality angling tourism in the country. The Arabian peninsula has a rich natural fishery offshore which is harvested by all of the Gulf States. Fishery exploitation and, increasingly, good fisheries management is well understood in the region.

However, to date, angling for sport has not been accessible to most of the population. This could in theory change, if migratory fish such as salmon could be introduced into the river system. The proposal in this instance is to introduce salmon into the Wadi Aleyn in the western Yemen, as a pilot project. The longer-term objective is to develop a managed salmon fishery in this wadi, and subsequently in other watercourses where the right conditions can be found, or created.

It is accepted that the Yemen is, in many respects, not the ideal environment into which to introduce migratory fish whose natural breeding habitat is the northern edge of the temperate zone and whose feeding grounds are in the North Atlantic. Some obvious problems include:

Watercourses go from dry to spate conditions for relatively short periods of time and then only in the wet summer months in those parts of the Yemen which experience monsoon weather.

Mean average air temperatures indicate that water temperatures are likely to be significantly higher than those tolerated by the species Salmo salar without developing stress.

The migratory journey of the salmon, assuming the upper watercourse could be seeded with juvenile fish in the wet season, would be somewhat more challenging than its normal journey to the North Atlantic, being several thousand miles longer and involving a journey around the Cape of Good Hope and up the west coast of Africa before entering waters where salmon are normally found. The previous southern limit of the Atlantic salmon is the Bay of Biscay, and the southern limit of the northern Pacific salmon is northern California.

Once the rains end in September conditions in the watercourses would become dry and hot and it is unlikely any salmon still resident in the system would survive.

There are a number of other issues more technical in nature relating to the local ecosystem, lack of invertebrate life in the wadis (although there is an abundance of arthropods such as scorpions), bacterial issues, and the unknown question of predation. We speculate that buzzards, vultures and other local predators would quickly adapt to eating salmon stranded in relatively shallow water.

We have considered various closed-system models and our current proposals, based on desktop research only, are as follows:

Salmon from the North Sea would be trapped as they tried to enter their ‘home’ river and introduced into a cooled transport pod containing saline water from the North Sea. A condensation and recycling system would be installed to minimise evaporation losses. Means of controlling temperature and oxygen levels within the tank would also have to be found. The pod would be shipped by airfreight to the Yemen. The holding tank would be set up to have an outlet into the wadi which could be opened at need.

As fresh rainwater came into the wadi system, the outlet would be opened allowing it to flow into the holding tank. Salmon are anadromous-they adapt to both salt- and freshwater environments. We speculate that the salmon, on smelling the freshwater, would leave the saline environment and seek to migrate upstream to find spawning grounds. Although the salmon would not recognise the ‘smell’ of the water (the mechanism by which salmon at sea identify the estuarine water of the river in which they were hatched is still poorly understood) we believe there is a reasonable chance they would enter the freshwater. ‘Strangers’ are often found in English and Scottish rivers-salmon which have entered another river different to the one where they were spawned.

The upstream migration would depend on some civil engineering of the watercourse, subject to survey:

a.) to ensure that gradients and natural obstacles did not obstruct the movement of the fish along at least 10 kilometres of riverbed, regarded as the minimum distance suitable for a meaningful pilot experiment

b.) to ensure if possible some background level of flow from the aquifer to achieve a minimum level of water in the watercourse to avoid fish being stranded between spates

We understand that in the Wadi Aleyn there is an existing falaj system of stone conduits to allow the irrigation of a number of date palm groves which could be adapted for the purposes of the above.

Salmon seek gravel beds covered in relatively thin layers of well-oxygenated water for spawning. We understand there is an abundance of gravel in the Yemen, and in the Wadi Aleyn in particular. It is at least theoretically possible that fish could be encouraged to spawn, as we would be introducing summer⁄autumn-run salmon into the watercourse and they would be seeking to spawn at the end of their upstream journey if the right habitat presented itself. This gives rise to the exciting possibility that the introduced salmon could spawn naturally, or at least be electrofished and harvested for their eggs, either of which would allow a hatchery to be set up adjacent to the Wadi Aleyn in which the next generation of juvenile salmon would have an excellent chance of survival. This would create a generation of salmon whose true home was the Wadi Aleyn. How their migratory instincts could subsequently be managed must be a matter for further research. We speculate that the creation of a second holding tank filled with saltwater could be used to trick salmon returning downriver into thinking they smelled seawater, and trap and hold them in a saltwater environment.

At this stage we have not attempted to cost this project until the client has had a chance to consider and comment on the outline concept we have presented. We estimate that, excluding NCFE time and project management charges, the capital costs of this project would be in the region of £5 million. We have not yet considered operating costs. We await the client’s further instructions.

Letter

Fitzharris & Price

Land Agents & Consultants

St James’s Street

London

Dr Alfred Jones

National Centre for Fisheries Excellence

Department for Environment, Food and Rural Affairs

Smith Square

London

6 July

Dear Dr Jones,