Complete the following exercises within the lab period and submit to
Canvas before leaving. In addition to the points detailed below, 5
points are assigned to the quality of the annotation and to the
‘cleanliness’ of the code and resulting pdf document.
Exercise 1 – 2 points
We will again be working with the BC Parks dataset, which contains
information on the locations of Provincial Parks in British Columbia.
The parks belong to 5 different regions. There is also information on
elevation (in m) contained within the dataset.
- Import the BC park locations dataset and convert the data to a
ppp
object (being sure to include information on regions as
marks). – 0.5 point(s)
- Plot the resulting
ppp
object. The marks need to be
visually distinct. – 0.5 point(s)
- Inspect the spatial distribution of parks. Do you expect the process
to be homogeneous? Justify why you came to this expectation. – 1
point(s)
Note: You will need to load the maptools
or
sp
packages and make use of the as.owin()
function.
Exercise 2 – 2 points
- Under an assumption of homogeneity, what is the intensity of
parks/km\(^2\) in BC? – 1 point(s)
- Is the estimated intensity trustworthy? Why/why not? – 1
point(s)
Hint: see ?rescale
Exercise 3 – 2 points
- Use a quadrat test to determine whether the assumption of
homogeneity is met. – 1 point(s) Note: Be sure to set the number of
quadrats appropriately, to avoid issues with the quadrat test.
- Visualise both the quadrats and estimated intensity, being sure to
include the points in each figure. – 1 point(s)
- Is the estimated intensity from exercise 2 trustworthy, and why? – 1
point(s)
Exercise 4 – 4 points
- Estimate the intensity using kernel estimation with likelihood cross
validation bandwidth selection. – 1 point(s)
- Perform hotspot analysis to identify locations of elevated
intensity. – 1 point(s)
- Visualise the output (be sure to include the window). – 1
point(s)
- Based on the estimated intensity and hotspot analysis, where would
choose to go if you were planning a vacation to tour different
provincial parks. – 1 point(s)
Exercise 5 – 3 points
- Estimate \(\rho\) for the locations
of parks as a function of elevation. – 1 point(s)
- Plot \(\rho\) vs. elevation. Be
sure that the x-axis for elevation does not go below 0. – 1
point(s)
- Do you think that there is a relationship between elevation and park
intensity? Use the results/data to support your statements. – 0.5
point(s)
- Would you be more or less likely to find a park at 1500m compared to
the average intensity of parks across B.C.? Why? – 0.5 point(s)
Note: Estimating rho can be slow (\(\sim\) 1-2 min). Be sure to leave enough
time for the document to knit.
Exercise 6 – 5 points
- Using Ripley’s \(K\)-function, test
for a significant (i.e., \(\alpha\) =
0.05) correlation between park locations. – 4 point(s)
- Is there any evidence of correlations in park locations? Why? – 1
point(s)
Notes: Use border corrections (i.e.,
correction="border"
) and be sure the estimators assumptions
are being met.
Exercise 7 – 3 points
- Using simulation envelopes, estimate both the homogeneous and
inhomogeneous pair correlation functions. – 1.5 point(s)
- Visualise the results. – 0.5 point(s)
- Are the estimates comparable? Which of these would you use to draw
conclusions about the clustering of provincial parks? – 0.5
point(s)
- Are parks in BC clustered? – 0.5 point(s)
Note: These steps can be slow (\(\sim\) 1-2 min). Be sure to leave enough
time for the document to knit.
Exercise 8 – 3 points
- Based on these descriptive statistics, what have you learned about
the spatial distribution of parks in BC?