Notes from a weather observer

I have always been obsessed by the weather. One of my earliest memories is of a thunderstorm in London, and my earliest datable memory is when I woke to the Great Winter of 1962-1963 arriving in Southampton on Boxing Day. While other people might mark their autobiographical memories by family events, I mark them by severe weather events.

As soon as my life settled into some semblance of post-student stability in the mid-eighties, I fulfilled my long-held dream of being able to set up my own weather station. Anyone can do this, although with most hobbies you can spend a great deal of money if you wish. You at least need a maximum-minimum thermometer, a rain gauge, and some way of measuring the wind direction. Traditionally conditions are measured at 9am GMT. And that is the problem; what if you're not there at 9am? The answer is obvious: you don't go away.

A serious, systematic hobby

Weather observers take their hobby seriously. There are several internet groups and forums devoted to the weather. I have studied the characteristics of the users, and it is perhaps not surprising that 95 per cent of them are male. Many keep their own weather records, and most observers are at least as dedicated as I am. They produce monthly and yearly summaries which are distributed among fellow observers, and many submit them to central databases. I have one colleague who has the benefit of having collected records from the same site for 51 years (a window of 30 years is thought to be sufficient to construct the current climate). We take great pride in the completeness, thoroughness, and accuracy of our records.

This kind of male obsession fits well with Simon Baron-Cohen's suggestion that men need to systemise their environment, and some more than others. The weather is, in survival terms, important: farmers and hunters need to predict the weather for crop planting, growing, and harvesting, and predicting the movement of animals and occurrence of nuts and berries. The construction of elaborate spreadsheets is similar to the other examples of extreme systemisation that Baron-Cohen describes. It is notable that weather observers also have other kinds of systemising hobbies, such as bird watching (or more accurately bird listing).

The good news is that with modern technology I no longer need to be there, or have a reliable weather sitter. Measuring instruments are now wireless, and a logger stores the data remotely and downloads it to a computer. Distressing accidents nevertheless still happen, such as cutting the wire from the anemometer (which measures wind direction and strength) when trimming the hedgerow, or the rain gauge filling with autumn leaves (invariably during an interestingly heavy rainfall). It is, however, difficult to shake off an obsession; extreme weather events are still best experienced in person, not through a graph.

A British preoccupation

It isn't just weather observers that are affected by the weather; of course we all are to differing extents. It is often claimed that the British are peculiarly obsessed by the weather; it is believed we talk about the weather more than the inhabitants of other countries, and use the topic of the weather as an ice-breaker and safe topic of conversation.

There are good reasons why we might be particularly fascinated by the weather. Britain's location at the edge of the continent, influenced by the Gulf Stream, and at the boundaries of large-scale convection cells, means that our weather is highly variable (and difficult to predict) and often underneath battling large-scale air masses. If you live in Russia you can predict what your winter is going to be like: very cold or extremely cold. As I write this sentence in November 2018 we have nothing but a glimmer of an idea of what our winter this year will be like (changeable December, slightly colder than average January and February). In spite of this putative obsession there is no clear data that we are indeed obsessed more than others; indeed, it seems that everyone everywhere is fascinated by the weather.

We should distinguish climate and weather. Climate is the underlying weather pattern, conventionally defined over 30 years. Britain has a temperate oceanic climate. Weather is what is happening day-to-day; today's weather in Dundee is best characterised as 'horrible' (windy, wet, and dull). The season is the typical weather for the time of year. Climate, weather, and season all affect our personality and how we behave, but, it turns out, not always in ways that people expect.

Confounds

There are many confounds that we need to consider when investigating the relation between climate and weather and behaviour. For example, the long dark nights of northern winters are invariably accompanied by cooler if not cold weather. The nearer the equator you go, day lengths are more equal across the year, and the sunnier it tends to be but also the hotter. Poverty tends to be higher in equatorial regions too; but perhaps the poverty is in some underlying way related to the climate? It's easy to go round in circles, struggling to tease out the causal effect of the variables of interest.

As an example of how climate affects our psychology, Wei et al. (2017), in a large-scale study in China and the USA, found that among weather variables temperature has most effect on the openness measure of the 'Big Five' personality constructs, particularly on socialisation and social growth. The causal reasoning sometimes strikes individual differences researchers as simplistic, but that does not mean that it is wrong. Warmer weather means that children are more able to wander around outside, exploring, meeting people, and taking more risks than they would do if stuck indoors. A a result a child growing up in southern California is more likely to be friendly, outgoing, and inquisitive than one growing up in Minnesota.

The season in which a person is born matters too. People born in summer tend to be healthier and more outgoing, with again temperature apparently being the most important variable. The results are complex however, and may be moderated by gender – a study in Japan found seasonality effects but only on females (Kamata et al., 2009). And season of birth affects both normal and pathological behaviour. People born in spring and early summer have an elevated risk of suffering from major depression later in life and completing suicide, particularly by violent means. People born later in the summer and in autumn are more at risk from suffering from obsessive-compulsive disorders. (I was born in August.)

In line with the complex ways in which season and climate at bitter affects later behaviour, the ways in which weather affects mood are also complex. Surely good weather leads to good mood, and bad weather bad mood? No, it's much complicated than that. The most famous modern study is that by Jaap Denissen and colleagues (2008), who found few effects of weather on mood. This much cited study important because it used an unusually large sample (over 1200 participants) and a sophisticated statistical analysis. Among the few reliable effects the researchers found were that the amount of sunshine, temperature, and humidity affects mood and energy levels – a little. But the complexity of the results is demonstrated by their finding that rising temperature also makes people less anxious and less sceptical.

Why are these results, at first sight, counter-intuitive? I think a particularly important study in the field was published in 2011 study by Theo Klimstra and colleagues. They found large individual differences in the type of weather people say they like, and this interacts with how the weather affects their behaviour. People can broadly be divided into 'summer lovers' and 'summer haters'. Rising temperatures affect these types differently. Consider also that it's very pleasant in spring when the temperature reaches an optimum 22°C, but very different in summer when it reaches 32°C. On a related note, how people say they will behave and how they actually behave may be very different: we might like the prospect of a hot summer's day, but the reality of struggling to work when it is blisteringly hot and humid (the interaction of temperature and humidity being another issue) may be very different. Weather observers themselves are motivated by recording extremes: in my study of their preferences, a weather observer loves severe weather, whether it is very hot, very cold, or very wet, while apart from at Christmas, most dread a long snowy cold winter.

What remains to be determined is where these particular differences come from. There is some intergenerational concordance: summer loving and rain hating mothers are more likely to have children who are summer loving and rain hating. But obviously this concordance could come about in many ways.

SAD and suicide

People with Seasonal Affective Disorder  (SAD) are particularly affected by the season, becoming depressed when the days shorten (which in the northern hemisphere is in winter, particularly January and February, although some people can suffer for 40 per cent of the year). Not all psychiatrists accept the existence of SAD, some suggesting it is another example of the pathologisation of normal behaviour. But most accept that some people are particularly affected by the lack of light, with DSM-5 labelling the disorder 'Major Depressive Disorder with Seasonal Pattern'. It is estimated that 5 per cent of the population may suffer from it. I look favourably on the existence of SAD partly because I experience it (it can be particularly grim living in northern Scotland; how I miss Hampshire sometimes) and also because there is a plausible biological mechanism for how day duration might affect mood in terms of light levels, the retinohypothalamic tract, and levels of melatonin and serotonin.

In the northern hemisphere days are at their shortest in late December, around the winter solstice, and as might be expected there is a significant seasonal variation in the suicide rate. However the suicide rate peaks not in the depths of winter but in late spring and early summer. We can currently only hazard guess as to why this might be. Victor Hugo notched that 'the miserable do not rebel' – revolutions tend to happen when things start to get a little better, and perhaps suicide is similar. Perhaps the rate is linked to seasonal variations in hormonal changes, particularly on the levels of melatonin. Here clearly the legendary 'more research is needed' applies.

As well as varying with time of year, one would expect suicide rates to vary with latitude, increasing the further north one goes. Very broadly there is such an increase, although as I have noted changing latitude is confounded with many other variables. In addition there is a broad east-west decline, such that suicides increase northeastwards across Europe. The suicide rate in Russia is the highest in Europe, with 26.5 suicides per 100,000 people, with the vast majority of these being men (WHO, 2018), but of course many other factors might be responsible for Russia's high rate. Within an individual country such as the USA rates generally are higher in northern states, but only in central ones, with northern coastal states generally being lower, with the exception of Alaska, which has a high rate. All we can conclude is that lack of light in winter might increase the likelihood of suicide the following spring, although it is just one of many variables. The weather also has a weak effect on the suicide rate. My colleague Fhionna Moore and I have looked at the effect of weather on suicide in Scotland, and the fairest summary is that it's very complicated. A summary is that suicide is slightly more likely at times of pleasant weather (reflecting the seasonality finding).

Hot heads

Violence can of course be directed to others as well as the self. In Spike Lee's film Do the right thing, a hot summer's day in Brooklyn leads to violence and temperature. There is a belief that hot weather leads to hot-headed behaviour, but as you will expect by now, the relationship is not a simple linear one.

In countries with a long north-south axis, with a reasonable difference in temperature along that axis (such as the USA and France), violent crime is indeed more common along in the hot southern areas. Civil conflicts were more common between 1950-2004 in tropical areas during El Niño events, which lead to more storms and hotter than usual weather in some areas. And violent crime anywhere generally increases when it gets hot. The relationship is not linear, however, but curvilinear: sometimes it's just too darned hot to do anything much (Bell, 1992). Perhaps paradoxically, higher rainfall increase violence as well as higher temperature (Hsiang et al., 2013).

Complex systems

People are first surprised and then bored when I try telling them that, yes, weather and climate affect us, but they do so in complicated ways that are modulated by many other factors, particularly individual differences. Perhaps we shouldn't be surprised by the small size of these effects. We are after all complex systems whose behaviour is determined by countless variables, and the weather comprises just a few of these. It is the very variability and unpredictability of weather and climate that makes its effects on us so complex – and fascinating.

- Trevor Harley is Emeritus Professor of Psychology at the University of Dundee. He is a full-time writer and science journalist, and has unbroken weather records since 1988. Inbetween staring at the sky and spreadsheets he has written several books: the latest is The Psychology of Weather. You can read a chapter here.

[email protected]

References 

Baron-Cohen, S. (2004). The essential difference: Men, women and the extreme male Brain. London: Penguin.

Bell, P.A. (1992). In defense of the negative affect escape model of heat and aggression. Psychological Bulletin, 111, 342–346.

Bridges, F.S., Yip, P.S.F., & Yang, K.C.T. (2005). Seasonal changes in suicide in the United States, 1971 to 2000. Perceptual and Motor Skills, 100, 920–924. doi:10.2466/pms.100.3c.920-924.

Catalano, R., Bruckner, T., & Smith, K. R. (2008). Ambient temperature predicts sex ratios and male longevity. Proceedings of the National Academy of Sciences of the United States of America, 105, 2244–7. http://doi.org/10.1073/pnas.0710711104.

Denissen, J. J. A., Butalid, L., Penke, L., & Aken, M. A. G. Van. (2008). The effects of weather on daily mood : A multilevel approach. Emotion, 8, 662–667. http://doi.org/10.1037/a0013497.

Hsiang, S. M., Burke, M., & Miguel, E. (2013). Quantifying the influence of climate on human conflict. Science, 341, 1212–1226. http://doi.org/10.1126/science.1235367

Kamata, M., Suzuki, A., Matsumoto, Y., Togashi, H., & Otani, K. (2009). Effect of month of birth on personality traits of healthy Japanese. European Psychiatry, 24, 86–90. http://doi.org/10.1016/j.eurpsy.2008.10.006.

Klimstra, T. A., Frijns, T., Keijsers, L., Denissen, J. J. . A., Raaijmakers, Q. A. W., van Aken, M. a G., … Meeus, W. H. J. (2011). Come rain or come shine: individual differences in how weather affects mood. Emotion, 11, 1495–1499. http://doi.org/10.1037/a0024649

Moore, F. R., Bell, M., Macleod, M., Smith, E., Beaumont, J., Graham, L., & Harley, T. A. (2018). Season, weather, and suicide - Further evidence for ecological complexity. Neurology, Psychiatry and Brain Research,  30, 110–116. http://doi.org/10.1016/j.npbr.2018.08.002

Wei, W., Lu, J. G., Galinsky, A. D., Wu, H., Gosling, S. D., Rentfrow, P. J.,  et al. (2017). Regional ambient temperature is associated with human personality. Nature Human Behaviour, 1, 890–895. http://doi.org/10.1038/s41562-017-0240-0.

World Health Organization (2016). Suicide rate estimates, age-standardized estimates by country.