如何判断鸟类观测点位是否在新西兰多边形范围内并筛选数据

Got it, let's walk through how to solve this—identifying those offshore bird observation points using the NZ polygon from maps::map("nz") is totally doable, and we'll cover both mapping and prepping the data for your analysis.

Step 1: Load Required Packages & Prep the NZ Boundary #

First, we'll grab the NZ map data and convert it into a spatial object that makes it easy to check if points fall inside the landmass. We'll use sf (simple features) because it's great for straightforward spatial operations in R:

library(maps)
library(sf)
library(dplyr)

# Get NZ polygon data and convert to an sf object
nz_land <- map("nz", fill = TRUE, plot = FALSE) %>%
  st_as_sf()

Step 2: Convert Your Observation Data to Spatial Points #

Next, turn your Count.df into spatial points—we just need to tell R which columns are longitude/latitude, and set the coordinate reference system (CRS) to WGS84 (EPSG:4326), which matches the maps package's default:

# Convert Count.df to sf spatial points
count_points <- st_as_sf(Count.df, coords = c("longitude", "latitude"), crs = 4326)

Step 3: Flag Offshore Points #

Use a spatial intersection check to mark which points are outside the NZ land polygon. st_within() returns TRUE for points inside the landmass, so we flip it with ! to flag offshore locations:

# Check which points are inside NZ land
is_onshore <- st_within(count_points, nz_land, sparse = FALSE)[, 1]

# Add a flag column to your data
count_points <- count_points %>%
  mutate(is_offshore = !is_onshore)

Step 4: Map Onshore vs Offshore Points #

Now you can plot the points with different colors to highlight offshore locations. Here's a clean ggplot2 version:

library(ggplot2)

ggplot() +
  # Plot NZ landmass
  geom_sf(data = nz_land, fill = "olivedrab", color = "darkgreen") +
  # Plot onshore points (blue)
  geom_sf(data = filter(count_points, !is_offshore), color = "steelblue", size = 1, alpha = 0.7) +
  # Plot offshore points (red, more prominent)
  geom_sf(data = filter(count_points, is_offshore), color = "firebrick", size = 1.5) +
  labs(title = "NZ Bird Observations: Onshore vs Offshore",
       subtitle = "Red points = offshore/marine locations") +
  theme_minimal()

If you prefer base R plotting instead:

# Base R map
map("nz", fill = TRUE, col = "olivedrab", border = "darkgreen")
# Add onshore points
points(Count.df$longitude[is_onshore], Count.df$latitude[is_onshore], col = "steelblue", pch = 16, cex = 0.8)
# Add offshore points
points(Count.df$longitude[!is_onshore], Count.df$latitude[!is_onshore], col = "firebrick", pch = 16, cex = 1.2)
legend("topright", legend = c("Onshore", "Offshore"), col = c("steelblue", "firebrick"), pch = 16)

Step 5: Extract Offshore Data for Further Analysis #

Finally, pull out the offshore points into a separate dataset if you need it for downstream work:

# Extract offshore observations (remove spatial geometry if needed)
offshore_data <- count_points %>%
  filter(is_offshore) %>%
  st_drop_geometry() # Use this line to get a regular data frame without spatial info

内容的提问来源于stack exchange，提问作者jvan257