网格单元内的采样点数 [英] Count of sampling points within a grid cell

问题描述

计算空间网格的每个网格单元内的采样点总数.

Calculate the total number of sampling points within each grid cell of a spatial grid.

我想制作一个网格并计算每个网格单元内采样点的总数.我创建了一个随机生成的数据和网格，并尝试使用 sf 和 raster 包计算网格单元格内的记录数，使用以前类似的 SO 问题，但没有成功.我还研究了提取功能.我对空间分析相当陌生.

I would like to make a grid and calculate the total count of sampling points within each grid cell. I created a randomly generated data and grid, and tried to calculate the number of records within a grid cells using both the sf and raster packages, using previous similar SO questions, but wthout success. I have also looked into the extract function. Im fairly new to spatial analysis.

 library(sf)
library(raster)
library(tidyverse)
library(mapview)
library(mapedit)

#Trial with sf package 
# load some spatial data. Administrative Boundary
#https://stackoverflow.com/questions/41787313/how-to-create-a-grid-of-       spatial-points
aut <- getData('GADM', country = 'aut', level = 0)
aut <- st_as_sf(aut)
#Try with polygons
grid <- aut %>% 
 st_make_grid(cellsize = 0.5, what = "polygons") %>% 
  st_intersection(aut)                               

#fake data
 lat<-runif(1000, 46.5, 48.5)
 lon<-runif(1000, 13,16)
pos<-data.frame(lat,lon)

 ggplot() + 
  geom_sf(data = aut) + 
  geom_sf(data = grid)+
geom_point(data=pos, aes(lon, lat)) 
#how to count number of records within each cell?  
 ########################################
#Trial with raster package
#https://stackoverflow.com/questions/32889531/r-how-can-i-count-how-   many-points-are-in-each-cell-of-my-grid
 r<-raster(xmn=13, ymn=46.5, xmx=16, ymx=48.5, res=0.5)
r[] <- 0
#How do I use the pos data here
xy <- spsample(as(extent(r), 'SpatialPolygons'), 100, 'random')
tab <- table(cellFromXY(r, xy))
r[as.numeric(names(tab))] <- tab
plot(r)
points(xy, pch=20)
d <- data.frame(coordinates(r), count=r[])

我想获得一个包含采样点数的表格.

I would like to obtain a table with number of sampling points.

推荐答案

示例数据

library(raster)   
aut <- getData('GADM', country = 'aut', level = 0)
r <- raster(aut, res=0.5)
lat <- runif(1000, 46.5, 48.5)
lon <- runif(1000, 13,16)
# note that you should use (lon, lat), in that order!
pos <- data.frame(lon, lat)

解决方案

r <- rasterize(pos, r, fun="count")
plot(r)

要一张桌子，你可以这样做

To get a table, you can do

x <- rasterToPoints(r)
z <- cbind(cell=cellFromXY(r, x[,1:2]), value=x[,3])
head(z)
#     cell value
#[1,]   22    4
#[2,]   23   45
#[3,]   24   36
#[4,]   25   52
#[5,]   26   35
#[6,]   27   38

或者，na.omit(cbind(1:ncell(r), values(r)))

这篇关于网格单元内的采样点数的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！