Quakes

This analysis will be on the Quakes dataset in base R. This dataset includes the locations, depths, stations, and magnitudes of 1000 different earthquakes off the coast of Fiji.

Variable Specifics

-(lat, long) Locations are given as latitude longitude pairs. -(depth) Depths are in kilometers. -(mag) Magnitude is given on the Richter scale. -(stations) The number of stations that reported the quake.

Source

As stated in the dataset documentation: This is one of the Harvard PRIM-H project data sets. They in turn obtained it from Dr. John Woodhouse, Dept. of Geophysics, Harvard University.

# Load the quakes dataframe into the R environment:
data("quakes")

# Create a copy of the data to work on:
quakes_df = quakes

# Finding the max depth and min depth for later graphing:
max_depth = max(quakes_df$depth)
min_depth = min(quakes_df$depth)

# Finding counts of each magnitude for pie chart:
count_by_mag = quakes_df %>%
  group_by(floor(mag)) %>%
  summarise(count = n(), .groups = "drop") %>%
  rename_with(.cols = 1, ~"mag")

Shape Analysis From the shape of the histogram we can see that most earthquakes tend to happen at the ends of the depth spectrum. In other words, most quakes are either very deep or very shallow, with a large dip in the number of quakes towards the middle depths.

While it is true that there are large ammounts of quakes at both the deep and shallow depths, we can also see that the majority of the quakes are still at the shallow depths, with a large number of them ocurring at around 50 km.

Conclusion Based on the fact that the majority of quakes happened around 50 km depth, there may be something at that depth off the coast of Fiji that is causing a large number of quakes.

summary(quakes_df$depth)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##    40.0    99.0   247.0   311.4   543.0   680.0

sd(quakes_df$depth)

## [1] 215.5355

Conclusion: This summary and standard deviation seem to support that quakes happen mostly at either shallow or deep depths. We can see this because the 1st and 3rd quartiles are very close to the min and max, and the standard deviation is quite large, suggesting the values are spread far from the center.

Slice Analysis We can see that the vast majority of quakes recorded were a magnitude of 4.0-4.9, since the magnitude 4 slice of the chart is far larger than the others. The next largest slice is the magnitude 5 slice, being much smaller than the magnitude 4 slice, but still noticible on the chart. The magnitude 6 slice is extremely small compared to the other 2.

Conclusion This, combined with the observations made on the histogram, could suggest that the magnitude 4 quakes are caused by whatever is at a depth of 50 km off the coast of Fiji since the depth of 50 and magnitude 4 quakes both have a large majority over the rest. It is also possible that the deeper depth quakes (which were a smaller, but still noticible bump on the histogram) could be caused by a different source that makes higher magnitude quakes (since the magnitude 5 slice on the pie graph is also a smaller, but still noticible slice).

Shape Analysis The points do seem to be clustered around certain areas on the graph. There appears to be a large line of points in the top left corner, as well as a large cluster in the top right with a line extending down from it.

Size Analysis There does not appear to be any pattern related to the size of the points on the graph. Points that are very close to one another can have very different sizes, and there don’t seem to be any areas in particular that have larger points than others.

Conclusion Based on the shapes made by the points, there do appear to be specific areas that are the sources of many earth quakes. The top left and right side regions of latitudes and longitudes seems to have lines of some sort that are large sources of quakes, which could mean that these areas are fault lines.

Depth Analysis We can see that the 2 clusters of points from the previous graph look quite different when depth is in the picture. The larger of the 2 clusters at the higher longitudes seems to be a large connected cluster of quakes that spans the entire depth range, while the smaller cluster at the lower longitudes seems to be broken into 2 separate clusters of shallow depth and just a few at the deep depths.

Conclusion It appears that whatever is the source of the large cluster of quakes must be a large source that spans the full depth range of all of the quakes. This could align with our previous conclusion that it is a fault line. The small cluster, on the other hand, seems to be mostly at the shallow depths, not reaching very far down. This could mean that these quakes are caused by something other than a fault line that is near these latitudes and longitudes.