Linking Climate Change To Extreme Events Distracts From Forecasting

People are better served by early warnings rather than hearing about how climate change made their plight worse.

"People from Bangalore would have been better served by early warnings rather than hearing about how climate change made their plight worse."

Mention of climate change as an amplifier of extreme events has now become even more frequent than the extreme events themselves.

Do these reminders serve any real purpose? Or are they just a distraction from understanding whether the events were forecasted accurately or not; and if not, how can the forecasts can be improved?

Ability To Attribute Extreme Weather Events to Climate Change Is Limited

It is not clear what decisions can be initiated by these attribution exercises in addition to the general adaptation and mitigation efforts needed in any case.

Especially since the models lack reliable skill to inform us about the contribution of climate change to extreme weather events at local scales, where impacts are felt.

A constant focus on climate change impacts on extreme weather events may also be taking human and computational resources away from improving the forecasts of these extreme events.

The so-called attribution of climate events is an organized effort across many climate centers and they rely on global climate models to estimate if and how the probability of occurrence or the intensity, duration, frequency or the area covered by extreme events, may have been altered by climate change; climate change related to greenhouse gas emissions.

One has to also remember that these models are not perfect and their ability to attribute an event to climate change can be severely limited.

For example, temperatures tend to be similar over large regions and are more closely related to the greenhouse gas increases at large scales and thus can be better simulated by the models. Temperatures are depicted better than rainfall.

Climate Models Fail to Capture New Trends 

Rainfall tends to occur like kernels of corn popping randomly in a popcorn kettle – especially in the tropics. Even the historical changes in the frequency of heatwaves can be captured well by the models – but only at large scales.

Heatwaves are typically defined based on local temperatures and their exceedances over certain local thresholds. The difficulty of course is that such local temperature changes are often related to many local factors and not only to greenhouse gas effects.

Land use change such as urbanization and deforestation, and  pollution tend to determine whether a heatwave occurs at a given location. Irrigation for example is argued to affect  temperatures and rainfall by some studies while others find that they have negligible effects. Models do not capture such local processes in any case.

Attribution of climate events can only be done in the context of historical evidences of the likelihoods of such events in terms of their duration, intensity, and frequency.

For example, it was reported that the heatwave over India and Pakistan during March-April-May of 2022 were made 30 times more likely by climate change whereas the Pakistan floods have been mentioned to be made much worse by climate change.

A simple metric such as the trend in the maximum yearly temperatures across the globe shows that the US and India have seen many regions experience a decrease in annual maximum temperatures.

It is not entirely clear why but pollution and the solar dimming, land conversion to agriculture and irrigation are among the likely causes. Comparison of multiple datasets for the historic periods show some differences in these trends even in the data products. Models fail to capture the details of the trends as well.

Climate Scientists Should Focus On Improving Forecasts of Extreme Weather Events

Future projections of even large scale changes have uncertainties due to such model deficiencies for the next decade or two. But for the decades beyond 2050, projections are dominated by the uncertainties in the scenarios used for the future.

This again points to the fact that attribution of extreme events to climate change only based on greenhouse gas forcings are fraught with model deficiencies at this juncture.

Other human activities at local scales need to be represented better to improve the skills of attribution, if attributions are considered necessary at all.

For countries like Pakistan and India with their high climate vulnerabilities, drivers of extreme events being remote - such as the North Atlantic or Pacific warming, or the Arctic ice loss - means that their limited resources are better targeted towards adaptation to extreme events at local scales. And the management of these hydroclimatic hazards are better served by skillful early warning systems.

The concept of attributing responsibility to the emitters for the damages from extreme events such as the national scale flood over Pakistan made worse by climate change, do not seem to have any takers in the international negotiations. Simply attributing the likelihood of the event itself to climate change is of little value in this case.

Besides, climate scientists themselves can serve humanity better by focusing on improving forecasts of these extreme events than to spend human and computational resources on attribution exercises with limited skill and limited value. People from Bangalore would have been better served by early warnings rather than hearing about how climate change made their plight worse.

(Raghu Murtugudde is a Visiting Professor at IITB and Emeritus Professor at University of Maryland.)

(This is an opinion piece and the views expressed above are the author’s own. The Quint neither endorses nor is responsible for the same.)