气象观测站点数据下载与处理

news2024/11/22 23:45:06

一、下载途径

全国400多个气象站气候数据（1942-2022）
王晓磊：中国空气质量/气象历史数据 | 北京市空气质量历史数据
气象数据免费下载网站整理
中国气象站观测的气象数据怎么下载

二、R语言处理

2.1 提取站点文件

在这里插入图片描述

library(dplyr)
library(readxl)
library(openxlsx)
library(tidyverse)
require(devtools)
rm(list=ls()) 
setwd("E:\\program1\\03-项目进展-3模型模拟结果验证\\input\\20230628 中国气象观测数据")
Stations=read_excel("../city_pre.xlsx",sheet = 1, col_names = c("province","stations","city","lat","lon"), col_types = NULL, na = "", skip = 0)

####Pre 36cities#####

for (i in 1:dim(Stations)[1]) {
  station=Stations[i,2]
  province=Stations[i,1]
  city=Stations[i,3]
  station_data <- data.frame()
  
  for (year in 2000:2021) {
    infile <- paste0("china_isd_lite_", year, "/", station, "0-99999-", year)
    
    if (file.exists(infile)) {
      temp_data <- read.table(infile) %>% 
        as.tibble() %>% 
        set_names("Year", "Month", "Day", "Hour", "ATemp", "DPTemp", "SLP", "WDir", "WSpeed", "SkyCover", "LPD1", "LPD6") 
      station_data <- bind_rows(station_data,temp_data)
    } else {
      cat("No such file exists:", infile, "\n")
    }
  }
  if (nrow(station_data) > 0) {
    output_folder <- "E:/program1/03-项目进展-3模型模拟结果验证/output/PRE_files/"
    city_Pre <- paste(output_folder,"Pre_",province,"_",city,"_",station,"_data.csv",sep="")
    write.csv(station_data, city_Pre, row.names = T)
  }
}

####判断station是否存在####
#获取有文件的站点
row(Stations)#98 5
dim(Stations)[1]#行数
length(row(Stations))#总数行*列

station_data <- data.frame()
for (i in 1:dim(Stations)[1]) {
  city=Stations[i,2]
  infile <- paste0("china_isd_lite_2020", "/", city, "0-99999-", 2020)
  if (file.exists(infile)) {
    st <- as.data.frame(Stations[i, ]) 
    station_data <- bind_rows(station_data,st)
  }
  else {
    cat("No such file exists:", infile, "\n")
  }
}
output_folder <- "E:\\program1\\04-项目进展-4城市降水量整理\\output\\PRE\\"
city_Pre <- paste(output_folder, "Pre_stations_info.csv", sep = "")
write.csv(station_data, city_Pre, row.names = T)
Sys.setlocale(category = "LC_ALL",locale = "Chinese")
write.csv(station_data, city_Pre, row.names = FALSE,fileEncoding = "UTF-8")
####~~~~~~~~####
#查看结果文件数量
# 指定文件夹路径
folder_path <- "E:\\program1\\04-项目进展-4城市降水量整理\\output\\PRE_Process"
# 列出文件夹中的所有文件
files <- list.files(folder_path)
# 获取文件数量
num_files <- length(files)
# 打印文件数量
print(num_files)
```ruby
library(dplyr)
library(readxl)
library(openxlsx)
library(tidyverse)
require(devtools)
rm(list=ls()) 
setwd("E:\\program1\\04-项目进展-4城市降水量整理\\output\\")

Stations=read_excel("../input/city_pre.xlsx",sheet = 1, col_names = c("province","stations","city","lat","lon"), col_types = NULL, na = "", skip = 0)
head(file)

2.2降水数据处理

####Pre 36cities#####
folder_path <- "./PRE/"  # 替换为实际的文件夹路径
files <- list.files(folder_path, full.names = TRUE)
# 循环处理每个文件
for (file in files) {
  #查看字符串长度eg Pre_安徽_合肥_58321_data.csv
  a=str_length(file)
  #提取字符串从
  name <- str_sub(file, start = 7, end =a)
  Stations = read.csv(file)%>% 
    #将文件读取为一个tibble数据框，并修改表头
    as.tibble() 
  head(Stations)
  #将降水为-99999和-1的无效值赋值为0
  Stations$LPD6[Stations$LPD6 <0] <- 0
  #选取列，不要4-11列
  data=Stations[,-4:-11]
  
  #删除b,d列的处理方式,-which 可以用！代替
  #data[ , -which(colnames(data) %in% c("SLP","WDir"))]
  
 # head(data)
  # 将日期列合并为一个新的列
  data$Date <- as.Date(paste(data$Year, data$Month, data$Day, sep = "-"))
  # 按日期分组，并计算每天的 LPD6 总和
  sum_LPD6 <- aggregate(LPD6 ~ Date, data = data, FUN = sum)
  
 # head(sum_LPD6)
  # 将年、月、日的列与 LPD6 总和列合并以Date为标准
  result <- merge(data, sum_LPD6, by = "Date")
  #head(result)
  result=result[,-5]
  result$LPD6.y=result$LPD6.y/10
  data_unique <- unique(result)
  # 打印结果
 # head(data_unique)
  output_folder <- "./PRE_Process/"
  city_Pre <- paste0(output_folder,"Pro_",name, sep = "")
  write.csv( data_unique, city_Pre, row.names = FALSE)
}