Editor's pick: Focus on methane to reach climate target
Our climate seems to be changing faster than estimated (28 July, p 5). The goal of climate policy, set out in the , is to keep the global average temperature rise well below 2°C. We would be more likely to attain this if we gave more weight to methane as a greenhouse gas. In climate policy reporting, emissions of all greenhouse gases are converted using their “” (GWP) over 100 years. On this horizon, methane's effect is about 25 times that of the same mass of carbon dioxide. Using this GWP indicates that it contributes about 20 per cent of the CO2 effect. But methane typically stays in the atmosphere for only 12 years.
The Intergovernmental Panel on Climate Change a GWP value of 85 for methane over 20 years.
Using this figure illustrates its contribution to warming more realistically, as methane's present influence is more than half that of CO2.
If we shorten the GWP time horizon, we emphasise reducing methane emissions and can thus accelerate the achievement of the Paris goal.
Citizen science is needed to track biodiversity (1)
Graham Lawton says it is often difficult to discuss biodiversity because of insufficient data (28 July, p 28). The reasons for this are unclear, but I expect time and money considerations play a role.
Could those in need of such data turn to citizen science? This involves members of the public giving such time as they have available to various surveys. Some, like me, are retired, research-trained scientists who have no intention of dying while watching TV and offer our knowledge and experience gladly.
Citizen science is needed to track biodiversity (2)
You quote UK TV naturalist Chris Packham saying he saw no butterflies over two days in his Hampshire garden. I recently joined a staff member of the Yorkshire Wildlife Trust and another volunteer on a butterfly counting “transect” near my home. It may give Packham hope that we recorded more than 100 butterflies belonging to 12 species in a little over an hour, though we saw only one example each of two of those species.
First class post 鈥� 15 August 2018
So if I supervise my robot, she can harm another human? Kool.
Susan Trexler on a suggested law that a robot may not injure a human… unless it is being supervised (4 August, p 38)
Consider the bluebell when defining 'species'
Robert Milne discusses the definition of a species (Letters, 30 June). This was interesting but, counter-intuitively, a species is not a natural concept.
It is a human construct, an idea to help categorise and log, an aid to our understanding of the rich variety of life. Its origins are in religious texts and the idea was a boon to post-Enlightenment collectors and taxonomists.
As with all constructs and models designed to represent natural phenomena, there are limits to their effectiveness. They are tools to aid elucidation, not ends in their own right.
It should come as no surprise that there are areas in nature where species' boundaries are unclear.
For example, the common bluebell (Hyacinthoides non-scripta) will inevitably interbreed in Britain with the Spanish bluebell (H. hispanica) to spawn the vigorously invasive and only weakly scented hybrid bluebell Hyacinthoides × massartiana. Could this make the bluebell more able to cope with climate change? The try-and-see, trial-and-error tinkerer that is evolution will answer that question over time.
The more eyes the better for medical diagnosis
Your interview with Regina Barzilay on her work using artificial intelligence to detect cancer was both interesting and encouraging (21 July, p 42).
When a friend went for a mammogram, it was inspected by a surgeon, who thought it was clear, and by two other members of staff, one of whom was not quite certain.
A further check found a problem. It was at a very early stage and was cleared relatively easily, quickly and permanently.
It seems to me that we can't get AI into service soon enough as a back-up partner for the hard-pressed teams making life-changing decisions. We will need to be alert, though, to the peril of getting so used to it that we sit back and trust it too much. We do need AI, but we also need people.
The danger of basing research on boring risks
Clare Wilson reports that low levels of the neurotransmitter dopamine are associated with risk-taking (21 July, p 10). From what I read about risk-taking research, it seems that most is based on money or gambling decisions. This may not be a good choice for behaviour sampling.
I, for instance, have been a skydiver, a scuba diver, have enjoyed motorcycle rides and launched into training for a profession in which I had zero female role models. All these are risky by some measures.
But I am fiscally conservative, would never gamble money and find research models based on this boring. I would never enrol in such a study. There must be others like me.
Why are diagnoses of autism increasing?
You say that the number of people diagnosed with autism has been climbing, leading to claims of an environmental cause, but that the increase is probably down to an expansion in the definition of autism and greater awareness (14 July, p 32). I'm sure this is true, but I still wonder about another likely cause: that couples are often having children much later in life. An association has been found between autism and paternal age. More recently, correlations with both maternal age and a wide disparity in age between parents have been shown. Could the trend towards starting a family later in life be a key factor?
I am convinced there is a cat killer in London
I was very disappointed to read Stephen Harris's claims that the work of the “cat killer” in London could all be explained by foxes and road traffic (21 July, p 26). As someone who has seen the work of all three, I can assure him that it is not that simple.
I was called by a neighbour to see if the dead cat in her back garden was mine. It wasn't, but I observed that the body had both its hind legs and forelegs arranged together in the geometric centre of the lawn, where it could be seen from the bedroom window. There was fur scattered around the lawn, but little near the body.
There was a single, straight wound from behind the ear to the shoulder, deep enough to expose the spine but not to decapitate. There was no blood anywhere.
I am sure this was not the work of a fox chewing with its weak jaws. Harris does not claim to have seen any of the recently killed cats, but dismisses the suspicion of a killer as “media hype” based on his observation of killings attributed to a satanist in the 1990s.
Use empty glacial valleys for pumped storage
Fred Pearce observes that hydroelectric dams have fallen out of favour with many environmentalists in recent years (23 June, p 36). This is partly because they drown habitats, which increases greenhouse gas emissions. They also take up land that could be put to other use.
Global warming is causing glaciers to retreat. The valleys they vacate seem to be ideal places to build hydroelectric dams – particularly for pumped storage – with no mature habitat to destroy and no human population or agriculture to disturb.
A disadvantage of most such sites would be their distance from customers, but high-voltage direct current power lines are lowering barriers to long-distance transmission (9 June, p 26).
Checking online urban folklore since 1994
Mick O'Hare asks whether anyone is prepared to vouch for the veracity or otherwise of his three tall tales (Old 快猫短视频, 21 July). I would direct him to , where he will find that the jet-assisted car and exploding cookie dough stories are urban myths, whereas the Rich Bastard incident took place largely as reported.
And machine shall speak garbage unto machine…
You quote Luciano Floridi saying there is no such thing as an independent fake-news bot unleashed on the web like some sort of alien life taking root (14 July, p 22). Does this mean it is impossible, or merely that no one is known to have done it yet? Is it impossible for someone to design a social media bot that transmits itself as a “worm” to invade as many poorly protected devices as possible, and makes random changes to its code after a certain number of replies?
Sooner or later a mutant might respond a thousandfold to its daughter bots' tweets or other postings, thereby drastically improving their chances of reproducing. The result could be the ultimate in futility, with machine speaking garbage unto machine and locking up the whole web. Then we have to think about the energy consumed by servers and its effect on the climate.
Its origin and purpose are still a total mystery
Colin Barras seems somewhat surprised that primates, including our ancestors, “stumble into the Stone Age by chance” (7 July, p 10).
How else might a species realise the usefulness of stone tools other than by trial and error? Or is he thinking that this cultural leap normally requires the presence of a large, black monolith, perhaps discovered in a Space Odyssey?
For the record 鈥� 15 August 2018
• The axis of rotation of Uranus lies close to the plane of its orbit, and by the time it and Neptune formed solid cores it would be too late to gather up a thick atmosphere (28 July, p 40).
• A fairly close shave: the star S2 gets within 20 billion kilometres of the black hole at the centre of our galaxy (4 August, p 14).