4) Hazard Assessment

In this entry, I will discuss how we can predict earthquakes of certain magnitude, type and also predict how hazardous the earthquake is by region. I am sure this section will be of interest to most people here as it covers how often earthquakes can and will strike in Ecuador and how these earthquakes may be more severe in certain regions.

There are two ways of predicting hazard for an earthquake; time-independent hazard and time-dependant hazard. This will split this entry into these two sections.

Time Independent Factors

This essentially means that this prediction gives the hazard of an earthquake striking at any given time and does not base the prediction on if previous events might set up the next ‘quake (these processes will be explained later). There are four major parts to these predictions:

• 1.       Sources
• 2.       Recurrence
• 3.       Attenuation 
• 4.       Probability of Exceedance ( this is more simple than it sounds don't worry)

I will start by explaining sources. Sources are areas that are known or could potentially produce earthquakes, for example at the subduction zone off the coast or the boundary between the North Andean plate and the Nazca plate. The subduction of the oceanic plate is the cause of the most frequent and biggest earthquakes.

Secondly recurrence. Simply put this is how often an earthquake of a certain magnitude will strike per given time (e.g. per day/ year / decade). So, for example, a small earthquake of magnitude 4.5 will strike in Ecuador probably twice a year. So if earthquakes of 4.5 magnitude are measured for let’s say 50 years with two striking every year on average, we could say it had a recurrence probability (likelihood) of two 4.5M earthquakes per year. From this data we can predict the recurrence of larger earthquakes through a relationship called the Gutenburg-Richter relationship.  Although the maths for this may be a little complex, this relationship can give us a value for how often an earthquake of any magnitude will strike in an area. For anyone interested and relatively good at maths, this website will give further information. http://www.eq.ccu.edu.tw/lab/lab105/public_html/members/master/93/GR.pdf.

Thirdly the attenuation. What this means is how much of the energy of the earthquake will reduce with distance. If we think back to our dropping a pebble into water analogy, the height of the wave after you drop the pebble will decrease as it moves from the impact of the pebble (the wave source). The same idea applies for earthquakes, the further from its source you go the weaker the wave will be.  This is the reason why body waves do not pose much threat. Body waves travel along the surface but also travel down into the earth in all directions.Therefore their energy is used more quickly than surface waves which just travel along the surface. Now with attenuation explained, we can see how this can be applied to hazard prediction. In simple terms, if we can predict how often a large earthquake can strike then we can also predict the area this earthquake will reach to and impact when it does strike.

Fourthly and finally there is the probability of exceedance. This is simply all the previous factors combined into one. It is how likely the ground is going to shake or move at any certain place. This probability is one of the most important factors into planning for earthquakes as it allows you to clearly see where the areas are that are most at risk from all of these factors. This information can be used for damage prevention in the areas most under risk, such as deciding where to put buildings or earthquake drills in schools. This type of hazard assessment is also good for using on seismic hazard maps, for example this is one for Ecuador.

(Bell, University of Edinburgh 2016)

The high hazard in the coastal region is obviously to do with the subduction on the Nazca plate. However the hazard in the mountains around Quito and Riobamba is due to the faulting of the minor plate, as mentioned in the first post.

One factor that can increase the hazard in the area is the rock on which you stand on (known as the bedrock). If the bedrock is strong and thick -lavas and other volcanic rocks are good examples of these- then the energy of the wave will be used up quicker as it tries to bend the rock. However, a weaker bedrock (for example river deposits or sandstone) can bend much easier due to their structure. The images of the recent earthquake in early 2016 show how the material below the surface can badly affect the impact of surface waves. For example in the photo below, look at the material below the road; it is mostly soft sands and mud which can easily be warped by surface waves.

(Sourced from http://www.independent.co.uk/news/world/americas/ecuador-earthquake-today-61-magnitude-quake-coast-days-deadly-tremor-a6992366.html)

Time Dependent Factors 

Unlike time independent factors, these factors are dependent of previous activity in the region. What this means is that the activities of other earthquakes (such as large earthquakes). The processes of working out these are a bit more complicated so I won't cover them in this blog but I'll explain them in general. For those of you who are still interested and quite good at maths, I recommend you read http://www.nat-hazards-earth-syst-sci.net/16/2177/2016/nhess-16-2177-2016.pdf as it gives a much more detailed description of these factors.

These factors are dependent on three parts:

• Time and slip predictable models
• Earthquake triggering
• Coulomb stress transfer

Forget the complicated names again as they are just jargon. 

Firstly the time and slip predictable models. What this in essence means is that with increase in time, the chance of the crust buckling and faulting increases.This is similar to coiling a spring very slowly but at a certain speed, after a predictable amount of time the stress in the spring will give way and it shoots back into it's original position. However this factor only works in areas where we know the average time between this stress build up and release in the earth, and unfortunately not every earthquake producing area has this predictability. Where this method does work, we can predict the recurrence of earthquakes quite successfully.

Secondly earthquake triggering. This is quite easy to understand. Essentially, when a large earthquake hits an area it weakens the crust all around it. This allows for smaller earthquakes to occur as the rock is weaker, it is more likely to slip and fault. It is this process that creates aftershocks, the mini-earthquakes after a major earthquake. It can work the other way round, where a smaller earthquake can weaken the rock just enough to be able to create a much larger, more dangerous earthquake.

Thirdly and finally we have Coulomb stress transfer. This is very similar to earthquake triggering. What this means is as the plate that creates an earthquake suddenly moves (think of the spring being released), the surrounding rocks and earth move with it and become part of the earthquake process. However there are parts in the crust that are adjacent to this movement but do not move, this part of the crust as a result becomes weakened. The part of crust that does not move has had the 'stress' transferred to it, and is more likely to fault and create another earthquake.

These processes are much more useful when they can be applied. They can be used along with the time independent factors to give an even more accurate prediction of the earthquake probability. Although the chances of an earthquake striking at any second are still very very low (around less the 1% chance), the time dependent factors can give an idea of how high an alert the area should be on, leading to stocking up supplies and temporary evacuations of vulnerable buildings (eg old churches, schools etc).

#Quaketips 3

Always be prepared if you live in an earthquake prone area. Check how strong your foundations are. Just a couple changes or fixes to the strength of it could save your house in a minor earthquake.