Are humans increasingly responsible for extreme weather events?

Weather extremes are often so overwhelming that it is hard for us to imagine that we may actually be increasingly responsible for at least some of them

By Dilip Namboodiri
Last Updated: Saturday 04 July 2015

Roads in lower Rudraprayag destroyed by the Uttarakhand floods of 2013 (Photo: Soumik Mukherjee)

In the last couple of years, floods in Uttarakhand and Srinagar have wreaked havoc on lives and livelihoods. In 2014, north India faced tremendous hardship as it bore the brunt of the longest heat wave on record. More recently, unseasonal rainfall caused widespread damage to the kharif crop. These events make us dig deeper into the science of climate change and determine whether extreme events occurring world over are part of “nature’s fury” or they can be specifically linked to global warming and, consequently, human activities. Although no weather extreme can be solely attributed to humans, a growing body of research shows that extreme weather events like floods, droughts and heat waves have increased in frequency.

A recent study, by E M Fischer and R Knutti of the Institute for Atmospheric and Climate Science, Switzerland, tries to estimate the contribution of global warming in the occurrence of extremes of temperature and precipitation and find a significant correlation at a global scale. The paper also warns us that with increments in warming, the frequency as well as intensity of heat and heavy precipitation events will rise further. The study was published in the scientific journal Nature Climate Change on April 27, 2015.

The paper estimates the fraction of “moderate extremes” that can be attributed to global warming. Fischer explains via e-mail that the somewhat paradoxical term “moderate extreme” is defined as a weather event that occurs once in 1,000 days (or once in three years). The study finds that 75 per cent of hot days occurring globally and 18 per cent of heavy rainfall days are attributable to global warming.

The paper forecasts that if we reach a situation where there is an increase in the global average temperature to 2°C over pre-industrial levels, then around 40 per cent of heavy rainfall days and 95 per cent of hot days will be due to human influence.

The reason for use of the term “moderate extremes” is that rarer extreme weather events take place once in 30 or once in 100 years. The study shows that the more extreme the event, the stronger the correlation with global warming. Thus, the more extreme the event, the deeper is the human imprint on it.


It is very interesting to note how they arrived at these numbers. They plotted on graphs the frequency of extreme weather events in pre-industrial temperature conditions, present day warming and higher levels of warming using simulated data from several climate models. They found that the likelihood of extreme events has increased with warming. This increased probability can be attributed to global warming and, thus, largely human activities. They also forecast a further increase in extreme events as warming increases.
J Srinivasan, chairperson of the Divecha Centre for Climate Change at the Indian Institute of Science, Bengaluru, finds the methodology of the study quite robust. He explains that climate change models are quite accurate when it comes to larger regions and understanding global trends.

High rainfall extremes in India

Fischer tells us that the models show an increase in heavy rainfall extremes in India over the long run. This is consistent with findings of the Intergovernmental Panel on Climate Change (IPCC) and studies conducted in India. A widely cited paper published in the journal Science in 2006, with B N Goswami (former director of the Indian Institute of Tropical Meteorology, Pune) as the lead author, showed that the frequency of occurrence as well as the intensity of rainfall and heavy rainfall events have shown a significantly increasing trend over central India.

Another paper titled “Impact of climate change and extreme rainfall events and flood risk in India” led by P Guhathakurta of the India Meteorological Department (IMD), Pune, extended the scope of the study to the rest of India and found a similar increase in heavy rainfall events for most parts of India. But Chhattisgarh, Jharkhand and some parts of northern India showed a decrease in intensity as well as frequency of extreme rainfall events. The paper also shows that there has been an increase in the flood risk over the decades.

J Srinivasan tells us that there has been a 50 per cent increase in extreme rainfall events over the Indian subcontinent. However, he warns us against making simplistic assumptions about attributing localised extreme events solely to global warming. He says many other factors like aerosol content in the atmosphere and land use patterns also contribute to extreme events at a local level. A study led by C M Kishtawal, published in the International Journal of Climatology in 2009, studied the relation between urbanisation and occurrences of extreme rainfall and found that urban areas tend to have significantly higher occurrences of intense precipitation as compared to non-urban areas. The El Nino Southern Oscillation (ENSO) is also a factor which has an impact on extreme weather events in the subcontinent. However, the correlation between global warming and ENSO is still uncertain.

Heat extremes in India

In case of extreme heat events in India, A K Jaiswal and his colleagues at the IMD have shown that there has been an overall increase in high temperature days (greater than 37°C) in the summer even though there is lot of spatial variation with north-central and eastern regions actually showing a decrease in number of high temperature days during observation of the long-term trend. But the study also suggests that there seems to be an abrupt increase in summer temperatures in many regions since the mid-1990s. Their study also shows a striking correlation between urbanisation and number of summer high temperature days with mega cities showing a sharp increase in high temperature days, especially in the last two decades. A paper by V Krishnamurthy titled “Extreme Events and Trends in the Summer Monsoon” shows that in central India, the number of events of maximum temperature over 43°C have increased from 156 in 1969 to 320 in 2005.

Thus, the long-term trends on a global scale as well as in the Indian subcontinent indicate the unequivocal influence of global warming in temperature and rainfall extremes.

Subscribe to Weekly Newsletter :

Comments are moderated and will be published only after the site moderator’s approval. Please use a genuine email ID and provide your name. Selected comments may also be used in the ‘Letters’ section of the Down To Earth print edition.

  • The title of the article is:

    The title of the article is: "Are humans increasingly responsible for extreme weather events?". The answer is big "NO".

    Extremes are part of natural variation in light of the local & regional general circulation patterns in different seasons. If we use a truncated data series of natural variability cycle, the frequency change according to which part of the cycle the data series represent. During UPA-II, a central minister informed to parliament that Indian monsoon rains are decreasing. He used the period of descending part of the 60-year cycle.

    Models have no meaning as they are not verified with ground realities. Unfortunately, our Indian groups form part of this group. Many of them are part of Al Gore & IPCC.

    From AR5 ÔÇ£It is extremely likely that more than half of the observed increase in global average surface temperature from 1951 to 2010 was caused by the anthropogenic increase in greenhouse gas concentrations and other anthropogenic forcings together. The best estimate of the human-induced contribution to warming is similar to the observed warming over this periodÔÇØ. That means, 50.1% is also more than half; but it not only includes anthropogenic greenhouse gas concentrations and also by other anthropogenic forcings. That means the anthropogenic greenhouse gas component is still less than 50%. They are all qualitative but we need an answer in quantitative terms to postulate the associated impact on glaciers retreat, ice sheet melt, ocean rise, etc. So also the case with extremes.

    Please look in to IMD Red Book that provide climate normals and see whether temperature extremes have crossed these numbers. The answer is big no.

    Dr. S. Jeevananda Reddy

    Posted by: Anonymous | 5 years ago | Reply
  • In continuation of the above

    In continuation of the above post:

    The natural variability: Regularly we see in media on El Nino versus monsoon. This is a part of Southern Oscillation in equatorial Pacific ÔÇô around Mexico --. If we look at all India Southwest Monsoon precipitation data of 1880 to 2006, the rainfall is normal in 84 years, 18 years come under El Nino and 23 years comes under La Nina of 126 years. Also, if we look at 135 years data, in 94 years comes under normal, 23 years come under deficit and 18 years comes under surplus. In all these statistics, the main character is normal. Take for example this yearÔÇÖs IMD long-range forecast. The normal [long term average] is 890 mm based on 50 years. This selection of 50 years defines the resultant statistics ÔÇô 93%. If we use the 60-year cycle then the average is 852 mm for two cycles ÔÇô 1871 to 1990, IITM publication -- accordingly the 93% changes to 97%.

    I collected the onset dates for all the met sub-divisions from daily, weekly & monthly weather reports. We mean onset of monsoon by the onset over Kerala Coast [here normal is 31st May/1st June]. This data series presented a 52 year cycle similar to Fortaleza rainfall in Brazil [around the same latitude but in Southern Hemisphere]. Also, if we look at the regions that receive precipitation in southwest monsoon and northeast monsoon seasons, they presented 56 year cycle but in opposite direction. The cyclones in Bay of Bengal presented the northeast monsoon 56-year cycle patter. The yearly precipitation data series presented 132 year cycle ÔÇô during the below the average rainfall cycle part, out of 66 years 24 years are deficit rainfall years and 12 years are excess rainfall years; and during the above average rainfall cycle part, out of 66 years 24 years are excess rainfall years and 12 years are deficit rainfall years. Similar patterns are evident all over the globe. Also, ocean temperature presents 50-70 years cycles [PDO, etc.].

    Dr. S. Jeevananda Reddy

    Posted by: Anonymous | 5 years ago | Reply
  • Rubbish

    Posted by: Taylor Threser | 4 years ago | Reply