THE letter from Robin Larkham ('Nuclear waste' Review February 15) pretty much hit the nail on the head. Even if every plant functioned perfectly, Britain would be building up a stockpile of highly dangerous waste that can only accumulate and become more and more concentrated... and this is if we were only burying the waste from UK nuclear plants.

The fact that we are accepting waste from international plants exacerbates this problem tenfold.

Britain is a small island and once buried in some "deep geological site", nuclear waste must remain undisturbed for generations to come. The figure given to us by the nuclear industry is 1,000 years and it seems even this is a very conservative estimate. Figures obtained from neutral sources with no vested interest in 'massaging' figures estimate the time that high level waste remains radioactive to be closer to 160,000 years.

But, in reality, any figure over something like 1,000 years is purely academic. The point is that neither we nor several generations of our descendants will ever see a time when the threat of highly toxic nuclear waste leaking into the environment will not be hanging over us all.

How many times have you read news reports that this or that car manufacturer is recalling certain models because of a potentially dangerous design fault? Even then, the companies rarely respond simply because the fault has been reported to them. It usually takes an accident caused by the fault before they take action.

People who design and build cars are not idiots and they are not careless. They are simply human and humans make errors and overlook minor details that may result in major consequences.

How confident are you that the sealed canisters containing the nuclear waste in the Irish Sea will remain intact for 1,000 years? Their ability to survive this length of time can only be estimated because it is not possible to run a test for a period of 1,000 years before deciding it is suitable to store nuclear waste. The same is true for any geological site. The practical feasibility of this method of storing nuclear waste is only theoretical because it is not possible to realistically test the theory.

We are banking rather a lot on people making absolutely accurate calculations not only that the site is truly suitable at the outset but also that no natural geological events will compromise the integrity of the site for the next 1,000 years (or 160,000 years depending on which "nuclear experts" you believe). This is making "predictions" for the future based on probability.

When it comes to predicting future natural events based on probabilities, the most advanced specialists in this area are probably the Meteorological Office. They have at their disposal the most advanced technology for mapping weather patterns, computer 'models', mathematical processes, long term experience and finest brains available in the field of predicting the weather.

But, if you were going camping for the weekend, would you not bother to pack a coat just because the weather forecast says it will be fine? Despite their experience and ability and technology, it is not even uncommon for the Meteorological Office to get it spectacularly wrong! But, when they get it wrong, the consequences are being caught in the rain without an umbrella. When the nuclear industry get it wrong, people die, the environment becomes contaminated and, in worst-case scenarios, vast regions become uninhabitable effectively forever. How much of Britain can we spare?

Full scale reactor meltdowns are relatively rare... but not rare enough to be sure it wouldn't happen and devastating enough to make even the possibility unacceptable. Other accidents resulting in a leak of radioactive material into the environment are less rare. These can still contaminate land and can even remain undetected for a long time before it is discovered and addressed.

In the meantime, people die of cancer and there are continual battles in courts around the world to get compensation for cancers that are most likely attributable to nuclear power radiation leaks.

Operator errors are alarmingly common in all industries and the nuclear industry is no exception. Usually they result in an anomaly that can be quickly addressed.

Sometimes the consequences are considerably more serious... like the reactor meltdown on Three Mile Island in the US. That wasn't caused by freak weather conditions but operator error. The matter was quickly addressed and large scale contamination was averted.

As Mr Larkham pointed out, we are effectively guinea pigs for the plant proposed at Oldbury. Horizon assures us that the new plant will operate the most advanced technology of any nuclear plant anywhere in the world. That is only a proud boast as long as it works as well as the computer models imply it should. But another way of looking at Horizon's claim is that we shall be the first. They are effectively saying: "We've never tried this before but we're convinced it'll be brilliant".

Not very reassuring when you look at it that way, is it?

– Ron Tocknell, Lydney.