Understanding Earthquake Data
The USGS maintains a world-wide record of earthquakes. This allows us to see a real-time snap shot of world earthquake activity.
Access the USGS website that reports recent earthquake activity: http://earthquake.usgs.gov/earthquakes/map/ (Links to an external site.)
When you view the USGS earthquake website, it should show you a list of the most recent earthquakes on the left. Since this assignment utilizes ‘real-time’ data, your answers will be unique and it is important to know when you accessed the data and started the lab.
1. Record the date and time that you are accessing the USGS website.
The most recent earthquake, which you should use to answer the next question, will be listed first. The magnitude is on the left, the depth is on the right. Other than time, date, and location, there are two additional pieces of data that you will find reported when an earthquake has occurred: depth and magnitude. Depth refers to how far below the Earth’s surface the event occurred (measured in kilometers). This value is the depth of the focus of the earthquake, or the hypocenter. (The point on the Earth’s surface directly above this is the epicenter.) Magnitude is a measure of the energy released during an earthquake (an indicator of size of the event).You can click on the earthquake to get more information and locate on a map as needed. Answer the questions below on the Answer Work-Sheet.
2. When was the most recent earthquake recorded? List the magnitude, location, date and time in UTC.
Next, investigate recent large earthquakes by resetting the map. First click on the ‘-‘ sign in the upper right hand corner to get a world view of earthquakes (this will take a few clicks). Then click on the gear symbol in the upper right hand corner to change the view of the type of earthquake. Choose the option for ‘7 Days, Magnitude 4.5+ Worldwide’. The list of earthquakes on the left will change and will have the information needed to answer the next question
3. How many earthquakes having a magnitude of 4.5 or larger have occurred over the past 7 days? Calculate the average ‘large earthquake per day’.
Hint: (Divide the total number of 4.5+ earthquakes by the number of days) to calculate the average. Your answer should be the same the level of precision as the data. Since the magnitude numbers are provided in tenth decimal (example 5.5), your answer should also be the same significant figures (Links to an external site.).
Next, investigate the large earthquakes that have occurred over the last 30 days by resetting the map. Choose the option for ’30 Days, Magnitude 4.5+ Worldwide’.
4. How many earthquakes having a magnitude of 4.5 or larger have occurred over the past 30 days? Calculate the average large earthquake per day.
Next, investigate whether the past week has been a ‘busy’ week for large earthquakes by comparing the averages you calculated in questions 3 and 4.
5. Has the past week been a significantly ‘busy’ week for large earthquakes? What are the limits of your conclusions? What data would you need to answer this question with more confidence?
Earthquakes and Plate Tectonics
Most earthquakes (but not all (Links to an external site.)) are caused by the slow motions of Earth’s plates. Along their margins, the plates interact with one another by diverging (moving apart), converging (moving together), or sliding horizontally past each other. Plate motions bend and deform the rocks near plate boundaries. This stores elastic energy in the rocks similar to the energy stored in a stretched spring. When deformation of the rocks increases to a point where the finite strength of the rocks is exceeded, the rocks break and slip suddenly along a fault plane.
The movement along plate boundaries is often not smooth, but more of a “stick-slip” process, in which sudden slips may, over a period of time, produce a range of earthquake magnitudes. The location of these earthquakes can help define the boundary between two plates. The spatial distribution of earthquakes can provide information about the geometry of Earth’s plates. Figure 3 provides some evidence for this. Exploring the distribution of recent earthquakes can provide additional insight.
Return to the USGS Earthquake website and the option ’30 Days, Magnitude 4.5+ Worldwide’. But now make sure you can see the entire work in your view and set the parameters to the ‘world view’ . You many need to click on the ‘-‘ sign in the upper right hand corner to get a world view of earthquakes and you may want to minimize the list of earthquakes (click on the ‘bars’ icon next to the world and gear icons in the upper right). Plate boundaries should appear as red lines on the map. If they are missing, check the ‘Plate Boundaries’ box in the Map Layers section of the settings.
6. What percentage of the recent earthquakes occurred at or near plate boundaries?
Hint: Start by counting all those NOT on plate boundaries and subtract that from total number of earthquakes (your answer to question 4).
My advice on the best approach: Zoom out to see the whole world in one view, then count only those earthquakes that are not near any plate boundary line. The earthquake dots do not have to be directly on top of the plateboundary line to qualify as as ‘on the plate boundary’. This is because the line represents the point where the twoplates meet but, in the case of subduction zones, the plates extend quite a bit further on either side of the line.
The West Coast of South America
The first plate boundary you will examine in detail is along the west coast of South America where the oceanic Nazca Plate collides with the continental South American Plate. Geologists have found that the Nazca Plate has one of the fastest relative velocities (about 17 cm/yr) as it moves eastward away from the Pacific Plate. The Nazca Plate is subducted under the South American Plate. The Nazca Plate/South American Plate convergence exhibits many classic features of a convergent oceanic-continental plate boundary. These include an offshore, deep-sea trench (the Peru-Chile Trench), a mountain range (the Andes), and a band of seismicity that ranges over a large series of depths. Analysis of earthquakes that occur along this boundary can provide insight into the physical and structural relationships between these two plates. You will look specifically at a cross-section in Chile that runs west-east (Figure 4).
Figure 4 Along the west coast of South America, the oceanic Nazca Plate is subducted beneath the continental South American plate. This zone of convergence is marked by a deep off-shore trench and the Andes mountains. These features are prone to seismic activity.
One of the first intriguing findings by seismologists was that, although many foci/hypocenters are situated at shallow depths (down to a few tens of kms in some areas), they can be hundreds of kms deep. The West Coast of South America is such a region. Earthquakes with foci/hypocenters occurring from depths of just a few kms to 70 kms are arbitrarily called shallow. Earthquakes with foci from 70 kms to 300 kms are called are called intermediate, and those below 300 kms are called deep.
Depth of Earthquake Definition
Shallow 0 to 70 km deep
Intermediate 71 to 300 km deep
Deep 301 to 800 km deep
Below is a graph you will need to analyze to answer the next set of questions. You should print it out if possible. This is a graph of what earthquakes are occurring under South America. Imagine if we just looked at the cross section under the earth of Figure 4. By looking at earthquakes, it allows us to see how the plates are interacting with each other (getting stuck and releasing stress) under the surface of the earth.
To view the graph in a separate window (with options for .pdf and .png files) , click on this link: The Nazca Plate/South American Plate Convergence.
7. Estimate the percentage of earthquakes that occur at a shallow depth, intermediate depth and deep depth.
Hint: This question is asking for an estimate only. I have an acceptable range of +/- 15% for each depth. I realize the ‘+’ signs are clustered which is why you must estimate. Provide your best estimate number (not a range, but a single number) for each depth (shallow 0-70 km, intermediate 71-300 km and deep 300+ km). Use the Depth of Earthquake Definition table above.
8. Describe the relationship between the locations of the epicenters and the depths of the hypocenters along this section of the west coast of South America.
Hint: Look at Figure 4 to understand the relationship between the plates in a cross section (West to East) that the graph is showing data for.
To answer the next question, you will need to print out the graph above, draw a best fit line from the the shallowest earthquakes on the left-side of the graph through the intermediate depth earthquakes.
Need to review what a best fit line is? (Links to an external site.)
Want to see a sample line?
9. What is the angle (relative to the top of your plot) defined by your best-fit line drawn through the shallow and intermediate depth earthquakes?
Hint: First draw your best-fit line and then you can calculate the approximate angle from horizontal that this represents by using trigonometry.
Trigonometry (Links to an external site.) is a really excellent and useful branch of mathematics. If you know the length of two sides of a triangle and one of the angles (in our case 90 degrees) then you can determine the angle we are looking for.
You will need to make two measurements on your graph:
Using the earthquakes on your line, subtract the depth of your deepest earthquake on your best fit line (in km) minus the depth of your shallowest earthquake on your best fit line (in km). These values are on the Y axis and represent side A on our right triangle (see below)
Using the earthquakes on your line and the top x-axis, subtract the horizontal position of your western most earthquake (in km) from the horizontal position of your eastern most earthquake (in km). This represents side B on our right triangle (see below)
Now you have determined the lengths of sides A and B on our right triangle. Most right triangle are portrayed ‘right side up’ like this:
However, on our graph Side B is on the top. You need to determine angle ‘a’ to answer the question. Below is an example of a similar calculation for a different graph.
Using simple trigonometry, you can now determine the approximate angle (from horizontal) of the plate boundary (as defined by the earthquake cluster).
You can use a graphing calculator, if you have one to do this calculation. The equation is Angle a = the inverse tangent (tan -1) of (A/B). If you don’t have a calculator you can plug your values for Side A and Side B into this online calculator and it will do the math for you. The number for angle a is the answer to question 9.
Online Trig Calculator: http://www.carbidedepot.com/formulas-trigright.asp (Links to an external site.)
Now complete the same steps to determine the angle from horizontal from the deepest intermediate earthquakes to the deepest earthquakes. Draw a best fit line from the the deepest intermediate earthquake to the deepest earthquakes.
10. What is the angle (relative to the top of your plot) defined by your best-fit line drawn from the deepest of the intermediate earthquakes to the deepest earthquake?
Hint: Do similar calculations to get sides A and B of your right triangle:
Using the earthquakes on your line and the Y-axis, subtract the depth of your deepest earthquake (in km) minus the depth of your shallowest earthquake (in km).
Using the earthquakes on your line and the top x-axis, subtract the horizontal position of your western most earthquake (in km) from the horizontal position of your eastern most earthquake (in km).
You calculations for the difference in Y-axis is Side A and the difference on the X-axis is Side B. You can use the same online trig calculator to calculate this angle as well.
Assuming you calculated different answers for questions 9 and 10; that means something changed in the earthquake profile under South America.
11. Discuss one rational hypothesis about why the angle of the plate boundary (approximately described by your two best-fit lines) might change as depth increases.
Hint: This question is asking about why your answers to questions 9 and 10 are different. Need some explanation about what a hypothesis is?
For the last several questions, we have been looking at how the plate changes as it subducts under South America. However, this angle of subduction does not stay the same up and down the entire west coast of South America. The next graph allows us to look at how the subduction zone changes.
Figure 5 Topography and earthquake epicenters recorded from 1975-1995 along the west coast of South America. The hypocentral depths of the earthquakes are given by the color code shown to the right. (U.S. Geological Survey National Earthquake Information Center)
12. As one travels to the north along the western margin of the South American Plate, the plate boundary changes (Figure 5). Using the color-coded depth scale, describe what happens to the angle of dip of the plate boundary between -40°S and -20° latitude. Is it getting steeper? more shallow? staying the same?
Hint: Focus on the different colors and the distance they are from the edge of the continent. If you just focus on the Orange (shallow), Green (intermediate) and Blue (deep), you can figure out the answer.
The West Coast of California
The second plate boundary you will investigate is along the west coast of the United States where a portion of the Pacific Plate is slipping past the North American Plate in a north-westerly direction. These plates have a relative movement of about 5.5 cm/yr. The San Andreas fault system in California is a family of parallel strike-slip faults located on this boundary (Figure 6).
Figure 6 Along the west coast of North America, the Pacific Plate is slipping past the North American plate in a north-westerly direction. The San Andreas fault system runs along the on-shore portion of the plate boundary from north of San Francisco (SF), CA to just east of Los Angeles (LA), CA.
Below is a graph you will need to analyze to answer the next set of questions. To view the graph in a separate window, click on this link: The San Andreas Fault System
Depths of earthquakes that have occurred along the San Andreas fault greater than 5.0 from 1973-2012.
13. Analyze the above graph to estimate approximately how many earthquakes occur at a shallow depth. Make this approximation for the intermediate and deep depths, as well.
Hint: Follow the same procedure and use the same table as question 7.
14. Describe the relationship between the locations of the epicenters and the depths of the hypocenters along this section of the west coast of the United States.
15. Why is the pattern of earthquakes in this area so different from the pattern observed in South America?
Investigating Earthquake Hazards
Because we know that many earthquakes are associated with movement along plate boundaries, locations where earthquakes have previously occurred are likely places for them to occur in the future. Knowledge of where earthquakes have occurred in the past, therefore, provides important information for assessing potential earthquake hazards. Considering the data that you’ve accessed in this module, you can reasonably state that, “Future seismic activity along the western margins of the South American and North American Plates is quite certain.” While this information is helpful for assessing future earthquake hazards, is it possible to make a more “site-specific” forecast?
Earthquake size and frequency are factors that will determine the likelihood of strong ground shaking and potential hazards at a given location. These factors are evaluated primarily by analyzing the historical record of earthquake activity. Historically in California, most of the larger earthquakes have occurred along the San Andreas fault system. Earthquake hazard research seeks to gain sufficient understanding of earthquake dynamics to minimize the loss of life and property resulting from such events. This research is made more complex because when strain is released along one part of the fault system, it may increase stress on another part. Ultimately, scientists would like to identify, with high probability, the location and time of a specific earthquake event.
Below is a graph you will need to analyze to answer the next set of questions. To view the graph in a separate window, click on this link: Recent Earthquakes in California
The region shown on this graph stretches from San Francisco (37.7°N, -122.5°W) to LA (34.1°N, -118.3°W). The green line on the graph represents the two best mapped faults in California (the San Andreas and the Hayward). Each red + is an earthquake higher than magnitude 3.0 that has occurred from 1973-2013.
To get a sense of this area, here is a map of California within the Latitude and Longitude area shown above:
16. What is the relationship between the location of recent earthquakes and major faults in California? Do most/some/a few of the earthquakes occur on the mapped faults? Is there any pattern to those that do not occur on the mapped faults?
Can we learn about earthquake risk by looking at history of ‘large’ (meaning bigger than 5.9 magnitude) earthquakes? Lets look at the history of big earthquakes (bigger than 5.9) in the San Francisco area:
Year Month Magnitude
1808 6 6.0
1836 6 6.8
1838 6 7.0
1858 11 6.1
1864 2 5.9
1865 10 6.3
1868 10 6.8
1884 3 5.9
1887 5 6.0
1889 4 6.0
1890 6 6.2
1906 4 7.8
1911 7 6.6
1926 10 6.1
1926 10 6.1
1979 11 5.9
1984 4 6.1
1989 10 7.1
17. Identify any stretches of time (two decades or longer) that are marked by the absence of ”larger“ earthquakes.
18. Identify periods of time that show a clustering (3 or more earthquakes within 10 years) of larger earthquakes.
19. Given your answers to questions 17 and 18, is there evidence that large earthquakes tend to cluster together? What are the limits of your conclusions? What data would you need to answer this question with more confidence?
Is it possible to make earthquake forecasts from plate tectonic theory?
Any scientific theory is valid to the extent that it can make predictions beyond the evidence that was accumulated to form the theory. An interesting validity test of what we know about earthquakes and plate boundaries comes from studying “quiet segments” of a fault along an active plate boundary. The idea is that, over the long run, all parts of a fault must average about the same amount of movement per time. This can happen either through the cumulative effects of a large number of small earthquakes or through the effects of a much smaller number of large earthquakes. With this in mind, the historical patterns of distance and time intervals between major earthquakes along plate boundaries can provide an indication of places where earthquakes might occur. Areas along an active fault zone where there has been a below-average level of earthquake activity are referred to as “seismic gaps.” A geographically large and longstanding seismic gap is generally interpreted to mean that a significant earthquake should be expected.
The graphic below should help you identify any seismic gaps (present and past). To view the graph in another window: Identifying Seismic Gaps in California.
The region shown on this graph stretches from San Francisco (37.7°N, -122.5°W) to LA (34.1°N, -118.3°W), same as the previous graph. The difference is that you can now see the difference between 2 sets of earthquakes based on the time frame in which they occurred. The October 1989, magnitude-7.1 Loma Prieta earthquake and its subsequent aftershocks had a dramatic impact on the seismic gap once located to the southeast of San Francisco.
How to interpret this graph:
Blue dots indicate earthquake activity from July 1989 to today. Blue = relatively recent earthquakes (Post Loma Prieta Earthquake)
Red dots indicate earthquake activity between 1973 and July 1989. Red = relatively older earthquakes (Pre Loma Prieta Earthquake)
Note: Blue dots do not hide, or cover-over, red dots on this plot.
20. What areas appear to have “gaps” (a below-average level of earthquake activity) pre-July 1989?
Hint: To identify pre-July 1989 gaps, find areas along the major faults that do not show very many “red” earthquakes. (These pre-July 1989 gaps can include areas along the fault with predominantly “blue” earthquakes or areas with NO earthquakes at all.) Identify these areas by latitude and longitude, specifically list the bottom of the gap and then the top of the gap, such as: (34.2 N, -120.5W – 35.3 N, -122.7 W)
21. Of the gaps you identified in question 20, were there any pre July-1989 seismic gaps that now appear to have had a high level of earthquake activity? If so, where were these gaps ‘filled in’?
Hint: Where you see predominantly “blue” earthquakes along a major fault line, a zone that was seismically quiet between 1973 and July 1989 has since become active. Identify these areas by latitude and longitude from bottom to top as you did in question 20.
22. Of the gaps you identified in question 20, what gap(s), if any, appear(s) to remain?
Hint: To identify gaps that still exist today, find areas along the major faults that have NO earthquake activity between 1973 and today. Identify these areas by latitude and longitude from bottom to top as you did in question 20.
Testing Earthquake Prediction: The Parkfield Gap
To develop the concept of “slip-deficit” further, you will now examine a region called the Parkfield Gap. Using historical data, scientists predicted that the Parkfield Gap would experience a “larger” earthquake in the mid-1980’s. You will now determine how they developed this hypothesis and assess whether or not they were correct.
To assign an earthquake risk to a given stretch of a fault, geoscientists attempt to estimate the “slip deficit” or how often it has moved in the past. These simple, direct approaches, however, only give rough estimates of the likelihood of when the next earthquake could occur. The area along the San Andreas fault system near Parkfield, CA (approximately 200 miles to the southeast of San Francisco, denoted as point SF on the map below), has had a fairly regular history of slippage over the last century and a half. As a result, this area was widely expected to experience a “larger” earthquake sometime in the last ten or twenty years. In expectation of this slippage, this section of the San Andreas fault system became one of the most heavily instrumented and watched stretches of fault on Earth. But, no large earthquake occurred in the 1980’s, nor in the 1990’s. Researchers began to jokingly refer to the Parkfield Gap as the seismological version of the proverbial watched pot that never boils! In the last 30 years, smaller earthquakes HAVE occurred, but none have been the magnitude that was expected — until recently.
Download and print this graphic here: Parkfield Seismic Gap
The Parkfield Seismic Gap is located along the San Andreas fault (green line), within the gray circle.
Note: The Parkfield Gap is a transition zone between “creeping” and locked sections of the San Andreas fault.
This area may not look like a ‘gap’ at all. However, take a look at the earthquake history from a different perspective. Shown below are two cross-sections along the San Andreas Fault. The upper cross section shows earthquakes that occurred along the fault prior to Loma Prieta (October 17, 1989). Three seismic gaps are seen, where the density of earthquakes appears to be lower than along sections of the fault outside the gaps. To the southeast of San Francisco is the San Francisco Gap, followed by the Loma Prieta Gap, and the Parkfield Gap. Because of the low density of density of earthquakes in these gaps, the fault is often said to be locked along these areas, and thus strain must be building. A magnitude 7.1 earthquake occurred in the Loma Prieta gap on Oct. 17, 1989, followed by numerous aftershocks. Note how in the lower cross-section, this earthquake and its aftershocks have filled in the Loma Prieta Gap. This still leaves the San Francisco and Parkfield gaps as areas where we might predict a future large event.
Here’s the record of earthquakes, magnitude-5.6 and larger, in the region presently referred to as the Parkfield Gap. This record spans approximately the last 150 years.
Year Magnitude
1857 7.93
1881 5.60
1901 6.40
1922 6.50
1934 6.00
1966 6.06
23. Based on the historical record of earthquakes above, estimate a specific year when the next magnitude-5.6, or larger, earthquake SHOULD HAVE occurred in the area of the Parkfield Gap.
Hint: Pretend it is 1967 and you are making a prediction about when the next big earthquake would occur. Your answer should be a year (example: 1776), there are range of acceptable answers.
24. Based on the historical record of earthquakes above, estimate the magnitude of the next “big” earthquake.
Hint: Your answer should be a magnitude number (example: 1.2), there are a range of acceptable answers.
25. Given your earlier observations about earthquake depth (hypocentral depth) in this area, make a prediction about the depth of the next “big” earthquake.
Hint: Your answer should be a specific number (557 km), there are a range of acceptable answers.
26. Seismically, which of the following options do you think will happen in the Parkfield Seismic Gap area over the next ten to twenty years?
A. Earthquakes are unlikely to occur in this area.
B. This area will experience an occasional small earthquake.
C. This area will experience a large number of small earthquakes.
D. This area will experience one very large earthquake, with a number of aftershocks.
