Plot points with top and bottom 5%  mean kriging variance (use different point shapes for top and bottom points).

To plot points with top and bottom 5% mean kriging variance, you can follow these steps:

1. First, you need to calculate the kriging variance for each point. This can be done using a geostatistical software or programming language like R or Python.

2. Once you have the kriging variance for each point, sort the points based on their kriging variance.

3. Identify the top 5% and bottom 5% of points. This can be done by calculating the 95th and 5th percentile of the sorted kriging variances.

4. Now, you can plot the points. Use one shape (for example, a circle) for the top 5% of points and another shape (for example, a square) for the bottom 5% of points.

5. To do this in R, you can use the ggplot2 package. Here is a basic example:

# Assuming df is your data frame and it has columns 'longitude', 'latitude', and 'kriging_variance'
library(ggplot2)

# Calculate percentiles
top_5_percentile <- quantile(df$kriging_variance, 0.95)
bottom_5_percentile <- quantile(df$kriging_variance, 0.05)

# Create a new column in df that specifies whether each point is in the top 5%, bottom 5%, or neither
df$point_type <- with(df, ifelse(kriging_variance >= top_5_percentile, 'top 5%', 
                                 ifelse(kriging_variance <= bottom_5_percentile, 'bottom 5%', 'other')))

# Plot
ggplot(df, aes(x = longitude, y = latitude, shape = point_type)) +
  geom_point() +
  scale_shape_manual(values = c(16, 17, 15))  # Change these values to change point shapes

This will create a scatter plot with different shapes for the top 5%, bottom 5%, and other points.