Nepal: What can be done? What lessons can be learned?

Many people I care for live in Nepal.

Less than a month ago I was teaching in Nagarkot, in the hills about an hour out of Kathmandu. I recognize some of the buildings with cracks that come on the TV screen. I receive the tweets that say #prayforNepal.
I wonder, who does one pray to? Is there anyone at the end of that line? The earthquake has already happened. The buildings have collapsed. The people are dead. Are we to pray to an omnipotent one who can raise the dead or at least tell the response teams where under the rubble the few living are? Is this the same god who pushed the India Plate up against the Eurasian Plate, causing a subduction earthquake that caused the buildings to collapse and the avalanches to start?
So I will not pray. I will do what I can to apply my knowledge to the problem of reducing the risks of earthquake hazards in Nepal and elsewhere.


Earthquakes cannot be predicted in advance, despite all the hopes people invest in animal behavior and such. The science just does not exist. A few minutes may be possible, but that won’t make a difference in terms of evacuation. Sri Lanka is not located in an earthquake zone, though small tremors caused by reservoirs have been experienced in recent times.

The hazards that Sri Lankans are vulnerable to and are closest in effect to earthquakes are landslides. Here, the science is more advanced than decades ago and some degree of prediction is possible. And they are localized in effect, as against the broad-spectrum hazards such as cyclones, earthquakes and tsunamis.

Sri Lanka is at risk from tsunamis that care caused by earthquakes in distant locations, known as teletsunamis because they travel far. If all systems work, we can evacuate people to safety from teletsunamis, but that is because there is gap of around 90 minutes between an earthquake occurring in the Sunda Trench and a teletsunami generated by it reaching our shores.

So what can be done to reduce the risks posed by earthquakes per se? Buildings can be constructed to code. We have enough knowledge now of which buildings survive and which ones do not. Many modern condominiums did; some temples did not. In a poor country it is not possible to retrofit buildings, but at least we can ensure newly constructed buildings adhere to code. Sri Lanka can have more relaxed building codes than are required in earthquake zones. But it is important that we build appropriately in locations vulnerable to cyclones and tsunamis, for example, by ensuring availability of vertical evacuation options in hotels and public buildings on the East Coast.


Earthquakes happen. Even if most buildings survived, some would collapse. Even if people took the right actions, they would still get buried under the rubble. This is true for landslides as well as earthquakes.

Mobile phones are today ubiquitous. Given the instances of people being located under rubble because their phones kept communicating with the nearby towers, phone-signal detection appears to be a tool we need to add to the toolboxes of the first-response teams.

How quickly can the authorities locate the earth-moving equipment? How quickly can those with the necessary knowledge and experience be brought to the scene? This is where Sahana, the disaster response software suite developed in Sri Lanka after the 2004 Indian Ocean Tsunami, could play a role. If it was integrated into every regional government’s disaster response plan, the information about the resources needed for an effective first response would already be in the databases.

Even if the information is there, will it be possible to coordinate resources? Will the networks be functional? From the evidence, it appears that the telecom backhaul networks held up quite well. The Internet went down for some time for some of the smaller ISPs, but NTC continued to work without a break, according to Dyn Research. Sri Lanka did not suffer a loss of connectivity with the outside world in 2004.

How did the last-mile towers and antennae hold up? Did the backhaul network survive? It would be useful to do a full assessment of what parts of the network performed well and what parts did not.

One needs to look beyond simple congestion and look at where towers fell down and so on. This documentation was incomplete in Sri Lanka in the aftermath of the tsunami. Congestion is a natural result of a disaster which causes a spike in calling/communicating. Japanese researchers working in the aftermath of the Great Tohoku Earthquake and Tsunami have come up with novels solutions to dynamically reallocate network resources from unaffected regions to the disaster-hit region. This is the only available long-term solution to congestion.

Reports indicated that power was out in Nepal. It makes sense to shut down power in the immediate aftermath because you do not want people electrocuted. But it is also important that power be switched back on, as soon as possible. It may well be that difficulties in restoring electricity will a major barrier to effective response and recovery in Nepal. Because the tsunami affected only the coast line this was not an issue in Sri Lanka.

In the short term, it will be useful to give priority to solar power for base stations and also for charging stations for mobile communication devices. This could be useful niche contribution we in Sri Lanka can make, in addition to donating dry rations and medicines.

Action must be taken to harden the parts of utility networks that failed and to ensure that standards are adhered to throughout the networks. Every disaster is an opportunity to stress-test the systems.

Response should not focus only on bodies, but also on the relationships. It is critically important to help people locate their loved ones. This is a problem we have grappled with for years. The Facebook crowdsourcing solution that was rolled out in the immediate aftermath of the Nepal earthquake appears to have significant potential, especially in countries that have lots of Facebook users. When we last looked, Nepal had 4 million Facebook users, giving it 15 Facebook users for 100 people, the highest ratio in South Asia (Sri Lanka had over 12 per 100 people).


So the resources have been assembled and rescue efforts are underway. But now there are people with no roofs over their heads, whose homes and livelihoods no longer exist. They need to be housed and cared for. This requires resources and coordination. Again, Sahana and telecom networks can contribute.

Twelve hours after the main quake hit, it appeared congestion was still making calls difficult. In times of danger and uncertainty, it is natural for people to want to talk. What could the networks do to quickly supplement capacity? If buildings have collapsed and people have moved, it may be possible to shift resources around, based on analysis of network data to track their movements in the aftermath of the disaster.

I recall the thousands of bottles of water from what seemed to be the entire membership of the United Nations that piled up in relief-agency locations after the 2004 tsunami. It is generally accepted good practice that priority should be given to local-sourcing and jump-starting economic activity in the affected region; that this is superior to hauling in everything from outside, except in the immediate aftermath of the kind of devastation we are seeing in Nepal.

But giving money to people without ID documents in shelters without leakage and outright fraud is not easy. Keeping money, once received is also not that easy. Mobile payments with some cash-out facilities could provide a working solution.

Rohan Samarajiva heads LirneAsia, a regional think tank. He was also a former telecoms regulator in Sri Lanka. To read previous columns go to LBOs main navigation panel and click on the ‘Choices’ category.