15 December 2022

The occurrence of landslides in Pakistan

Posted by Dave Petley

The occurrence of landslides in Pakistan

The combination of topography, rainfall, seismicity and people means that Pakistan is heavily affected by landslides.  However, until now I have not been aware of a systematic analysis of landslides at a national level in a way that has been performed for many other countries, a major gap in our understanding of this hazard.  Pleasingly, a paper (Shabbir et al. 2022) has just been published in the journal Environmental Science and Pollution Research that seeks to fill in this gap.  Even better, it is open access.

The researchers have compiled a dataset of landslides across Pakistan from 2003 to 2019 inclusive.  The dataset includes all reported landslides, not just those that have caused fatalities (as in the case of my own research).  In total, 1,089 landslides have been identified, of which 180 led to loss of life.  In total, 1,072 people were killed by landslides over this period.  Rainfall triggered 63% of the recorded events, although interestingly the seasonal effects that are seen in other parts of South Asia (for example Nepal) is less strong in Pakistan.

The map below shows the distribution of recorded landslides:-

The distribution of recorded landslides in Pakistan from 2003 to 2019 inclusive.

The distribution of recorded landslides in Pakistan from 2003 to 2019 inclusive. Map from Shabbir et al. (2022).


There is huge richness in this map.  The most obvious conclusion is the immense focus of landslides on the northeast of the country, in the high mountain areas.  Pakistan is particularly interesting as the Karakoram Highway means that many people live beyond the frontal range of the high mountains.  This is reflected in the landslide distribution, which extends a long way to the north.  But even here there are two distinct clusters, one located in the area of Muzaffarabad and associated districts, and the other in the far northeast, around Baltistan.  The first cluster is at least in part in the area affected by the dreadful 2005 Kashmir Earthquake, which triggered many landslides (most of which are poorly constrained, and so are not captured in this dataset), whilst the other is the area in which the 2010 Attabad landslide occurred.

This is a really valuable study, and I commend the authors for putting it together.  I would really like to see further analysis, in particular of the fatal landslides, which are likely to form a more consistent dataset.  Understanding these patterns in time and space is incredibly valuable as a step towards managing the risk.



Shabbir, W., Omer, T. & Pilz, J. 2022.  The impact of environmental change on landslides, fatal landslides, and their triggers in Pakistan (2003–2019). Environmental Science and Pollution Research. https://doi.org/10.1007/s11356-022-24291-z