Extreme monsoon fluctuations have frequented Pakistan in recent decades. Although the most stamped in memory are the floods of 2022 and 2010, the region also experienced a prolonged multi-year drought at the dawn of the 21st century.
Today, it is easy to attribute natural calamities like floods and droughts to climate change, especially excessive rains in regions where only traces of the monsoon are received on average. What's mostly ignored, however, is that floods and droughts in this region are much older than modern climate change.
So, is climate change driving all these uncharacteristic monsoon episodes?
If we compare monsoon variability, magnitude, and extremes in the first few ground-observation-rich decades of the 21st century with the ground-observation-scarce decades of the 20th century, monsoons over Pakistan undoubtedly look out of (little-known) character, apparently a textbook example of climate change mayhem. However, given the relatively short span of the elapsed 21st century period, establishing a robust link to climate change is challenging.
Western Margins of South Asian Monsoon
More importantly, the nature of the South Asian monsoon on its western margins, where most of Pakistan is situated, is poorly understood. Without first understanding the monsoon character along this peripheral belt, the out-of-character monsoon rains over these areas cannot be explained.
Pakistan's western half sits at the periphery of the South Asian summer monsoon, often out of the reach of vigorous monsoon low-pressure systems. Geographical orientation, atmospheric dynamics, and moisture transport pathways limit rain-producing weather systems' frequent penetration over those areas, making them some of the driest in South Asia during summer.
Instances of the western margins of the South Asian monsoon receiving excessive monsoon rainfall are those when anomalies in atmospheric dynamics and moisture transport allow monsoon lows to take unusual paths and sustain their vigor well beyond their typical limits. Both 2010 and 2022 were affected by such anomalous atmospheric conditions, but what makes 2022 distinct is that it suffered these conditions for much longer, leading to an even more catastrophic outcome.
Natural Atmospheric and Oceanic Modes of Variability
Interestingly, atmospheric anomalies facilitating the trajectories of monsoon lows along western margins beyond their climatological pathways are the two most common modes of variability of atmospheric dynamics over South Asia, one closely resembling a circumglobal wave pattern  and the other representing variability in (or meandering of) the subtropical westerly jet stream . Combined, these modes reposition the monsoon trough and allow anomalously westward movement of monsoon lows – a condition sustained for much of July and August 2022.
Animation of specific humidity (0.0123 KgKg-1) in the atmosphere colored by zonal winds. The strongest colors represent magnitudes greater than ± 10 m/sec. Warm (cold) colors represent eastward (westward) flow. The animation runs from July 1 to Aug. 31, 2022. The convergence zone of the winds represents the anomalously southward position of the monsoon trough. The illustration is based on Ashfaq et al. 2023 (https://doi.org/10.1038/s41612-023-00462-8).
But being the most common modes of atmospheric dynamics over South Asia, why does this natural variability not lead to such anomalies more often?
It is because while these meteorological patterns facilitate the uncharacteristic movement of monsoon weather systems, they may have little role in their genesis or excessive moisture content. Therefore, favorable atmospheric dynamics anomaly is necessary but may not be a sufficient condition for extreme monsoons over Pakistan.
Therefore, it is unsurprising that ENSO – one of the most well-known natural forces – was active in its negative phase, La Niña, in 2010 and 2022. La Niña represents colder than average sea surface conditions in the equatorial Pacific, often leading to wetter than normal monsoons over South Asia. Additionally, new evidence  suggests that sea surface temperature anomalies in the western tropical Pacific, independent of La Niña forcing, have an even more significant role in shaping atmospheric conditions and moisture transport necessary for extreme monsoons over Pakistan in recent decades. Likewise, abnormally warm waters in the adjacent Arabian Sea can maintain excess moisture transport.
When these naturally occurring atmospheric (jet stream meandering, circumglobal wave train) and oceanic modes (sea surface temperature variability in the equatorial and western tropical Pacific Ocean and Arabian Sea) of variability are combined as explanatory variables in a multi-linear regression model, they explain ~75% of the monthly precipitation variability over Pakistan in the 21st century. Their combined forcing reproduces abnormal atmospheric circulation patterns of extreme monsoon episodes over Pakistan and the spatial and area-average distribution of excessive rainfall in West South Asia.
The composites of wet precipitation anomalies in South Asia and associated atmospheric (850 and 200 hPa) anomalies. The left column represents actual precipitation (circulation) anomalies in observations (ERA5 reanalysis), and the right column represents corresponding anomalies in the regression model using natural atmospheric and oceanic forcings as explanatory variables. The vectors represent atmospheric winds, and the colors represent geopotential height. The wet and dry anomalies are defined using a 0.75 standard deviation threshold of the precipitation index over West South Asia. The illustration is based on Ashfaq et al. 2023 (https://doi.org/10.1038/s41612-023-00462-8).
The observed and regression model-based monthly (July-August-September) anomalies from 2001 through 2022 over West South Asia. The illustration is based on Ashfaq et al. 2023 (https://doi.org/10.1038/s41612-023-00462-8).
More Variable and Co-occurring Natural Modes
The question of why monsoons have become more extreme in Pakistan and why excessive rains have often occurred over western margins in recent decades can be answered by investigating the changes in the characteristics of natural atmospheric and oceanic variability affecting monsoons over Pakistan. Two changes have been noted in their characteristics. First, the variability of some of these modes has increased, with record strengths witnessed during episodes of extreme monsoon over Pakistan. Second, they have been co-occurring more often than before. The compounding of multiple forcings favoring excessive rains over Pakistan during the monsoon explains why La Niñas of the 20th century were less lethal as often as those in the 21st century.
A strong La Niña may be less impactful when occurring alone than a moderate La Nina co-occurring with other forcings.
The sum of all forcings after their standardized monthly indexes are transformed into +1/−1 or 0 using ±1 standard deviation as a threshold. The resulting +ve (−ve) number represents the net forcings at the monthly scale that positively (negatively) influence WSA precipitation. The illustration is based on Ashfaq et al. 2023 (https://doi.org/10.1038/s41612-023-00462-8).
The Role of Climate Change
Climate change may be responsible for changing characteristics of natural forces affecting summer monsoon variability over Pakistan, such as increased variability and co-occurrence. Extreme monsoon episodes may also be affected by climate change indirectly through the increase of water vapor in the atmosphere. However, establishing these plausible connections requires rigorous investigation.
So, coming back to the question of what causes extreme monsoons over Pakistan, one can note that it's the combined effect of multiple forces related to natural variability in the atmosphere and oceans. Why have recent decades more often experienced the combined effect of these forcings than before? It may be climate change, multi-decadal variability, or both. However, it may take substantially longer than two decades into the 21st century to establish either as the cause.
It is important to note, however, that the current generation of global climate models used in preparing the sixth Assessment Report of the United Nations Intergovernmental Panel on Climate Change fails to reproduce extremes even marginally comparable to what happened in 2022 over Pakistan, even at the highest level of radiative forcing of the late 21st century . Nevertheless, the lack of extremes in late 21st-century model simulations comparable to present-day observations is less useful as evidence for drawing meaningful conclusions, knowing that climate models exhibit limited skill in simulating South Asian monsoon rains along its western margins .
Can the occurrences of such extreme monsoon episodes be predicted well in advance?
The monsoon precipitation variations over Pakistan are significantly influenced by internal atmospheric variability, usually considered less predictable or unpredictable. Consequently, the question of the predictability of extreme South Asian monsoons along its western margins depends on the capability of the current seasonal forecasting systems to represent sub-seasonal variations in such relatively "unpredictable" atmospheric conditions.
These atmospheric modes of variability are more variable over time (~weekly) than the oceanic modes (months to seasons) associated with sea surface temperature anomalies. Without a scientific breakthrough in forecasting internal atmospheric variability, sub-seasonal predictability of extreme monsoons over Pakistan will remain a distant goal.
Furthermore, the role of lesser-known oceanic variability in the western tropical Pacific as the proximate forcing of extreme monsoons over Pakistan highlights the need to look beyond conventionally defined modes of variability to improve the predictability of extremes.
 Ding, Q. H. & Wang, B. Circumglobal teleconnection in the Northern Hemisphere summer. J. Clim.18, 3483–3505 (2005).
 Ashfaq, M., Johnson, N., Kucharski, F. et al. The influence of natural variability on extreme monsoons in Pakistan. npj Clim Atmos Sci 6, 148 (2023). https://doi.org/10.1038/s41612-023-00462-8
 Otto, F. E. L. Climate change increased extreme monsoon rainfall, flooding highly vulnerable communities in Pakistan. Environ. Res. Clim. 2, (2023). https://doi.org/10.1088/2752-5295/acbfd5
 Ashfaq, M., Rastogi, D., Mei, R. et al. Sources of errors in the simulation of South Asian summer monsoon in the CMIP5 GCMs. Clim Dyn 49, 193–223 (2017). https://doi.org/10.1007/s00382-016-3337-7