我正在尝试计算R中空间对象的气候变量的平均值 . 挑战在于我正在尝试为世界上每个二级行政区域(www.gadm.org)计算这些方法,我需要一个有效的计算统计数据的方法 . 我已经计算了这些统计数据,对于跨越较少气候区/瓦片的较小区域定义没有问题,但是当试图将这项任务扩展到整个世界时,后勤问题已成为障碍 .
我尝试过将gadm.org的2级边界shapefile用于整个世界,然后导入并合并worldclim.org的完整生物气候栅格(在最高可用空间分辨率下)和区域/磁贴,但它似乎要求太高了资源 . 具体而言,将整套栅格区域/图块合并为一个全局栅格对象的操作永远不会完成 . 它似乎是最有效的方法,也是最有可能最大限度地减少错误,为整个世界合并栅格区域 .
我不确定如何从这里解决问题,因为按国家/地区计算这些统计数据似乎非常乏味和低效 . 此外,行政边界层中有大量形状与各个Worldclim区域/区块重叠,如果相关的气候对象在不完全位于单个区域/区块内的形状的计算中丢失,则会导致错误 .
我想知道如何根据操作的大小提出有效的解决方案 .
下载2级全局管理边界数据后,我尝试了以下代码:
library(raster)
library(rgdal)
library(maptools)
library(foreign)
#SET WORKING DIRECTORY
setwd("C:/gadm28")
#IMPORT GLOBAL ADMINISTRATIVE BOUNDARIES (LEVEL 2) DATA FROM HARD DRIVE
gadm <- readOGR(dsn="C:/gadm28", layer="gadm28")
#IMPORT GLOBAL (ALL TILES) BIOCLIMACTIC DATA DIRECTLY FROM WORLDCLIM.ORG
climatezone00 <- getData('worldclim', var='bio', res=0.5, lon=-180, lat=90)
climatezone01 <- getData('worldclim', var='bio', res=0.5, lon=-150, lat=90)
climatezone02 <- getData('worldclim', var='bio', res=0.5, lon=-120, lat=90)
climatezone03 <- getData('worldclim', var='bio', res=0.5, lon=-90, lat=90)
climatezone04 <- getData('worldclim', var='bio', res=0.5, lon=-60, lat=90)
climatezone05 <- getData('worldclim', var='bio', res=0.5, lon=-30, lat=90)
climatezone06 <- getData('worldclim', var='bio', res=0.5, lon=0, lat=90)
climatezone07 <- getData('worldclim', var='bio', res=0.5, lon=30, lat=90)
climatezone08 <- getData('worldclim', var='bio', res=0.5, lon=60, lat=90)
climatezone09 <- getData('worldclim', var='bio', res=0.5, lon=90, lat=90)
climatezone010 <- getData('worldclim', var='bio', res=0.5, lon=120, lat=90)
climatezone011 <- getData('worldclim', var='bio', res=0.5, lon=150, lat=90)
climatezone10 <- getData('worldclim', var='bio', res=0.5, lon=-180, lat=60)
climatezone11 <- getData('worldclim', var='bio', res=0.5, lon=-150, lat=60)
climatezone12 <- getData('worldclim', var='bio', res=0.5, lon=-120, lat=60)
climatezone13 <- getData('worldclim', var='bio', res=0.5, lon=-90, lat=60)
climatezone14 <- getData('worldclim', var='bio', res=0.5, lon=-60, lat=60)
climatezone15 <- getData('worldclim', var='bio', res=0.5, lon=-30, lat=60)
climatezone16 <- getData('worldclim', var='bio', res=0.5, lon=0, lat=60)
climatezone17 <- getData('worldclim', var='bio', res=0.5, lon=30, lat=60)
climatezone18 <- getData('worldclim', var='bio', res=0.5, lon=60, lat=60)
climatezone19 <- getData('worldclim', var='bio', res=0.5, lon=90, lat=60)
climatezone110 <- getData('worldclim', var='bio', res=0.5, lon=120, lat=60)
climatezone111 <- getData('worldclim', var='bio', res=0.5, lon=150, lat=60)
climatezone20 <- getData('worldclim', var='bio', res=0.5, lon=-180, lat=30)
climatezone21 <- getData('worldclim', var='bio', res=0.5, lon=-150, lat=30)
climatezone22 <- getData('worldclim', var='bio', res=0.5, lon=-120, lat=30)
climatezone23 <- getData('worldclim', var='bio', res=0.5, lon=-90, lat=30)
climatezone24 <- getData('worldclim', var='bio', res=0.5, lon=-60, lat=30)
climatezone25 <- getData('worldclim', var='bio', res=0.5, lon=-30, lat=30)
climatezone26 <- getData('worldclim', var='bio', res=0.5, lon=0, lat=30)
climatezone27 <- getData('worldclim', var='bio', res=0.5, lon=30, lat=30)
climatezone28 <- getData('worldclim', var='bio', res=0.5, lon=60, lat=30)
climatezone29 <- getData('worldclim', var='bio', res=0.5, lon=90, lat=30)
climatezone210 <- getData('worldclim', var='bio', res=0.5, lon=120, lat=30)
climatezone211 <- getData('worldclim', var='bio', res=0.5, lon=150, lat=30)
climatezone30 <- getData('worldclim', var='bio', res=0.5, lon=-180, lat=0)
climatezone31 <- getData('worldclim', var='bio', res=0.5, lon=-150, lat=0)
climatezone32 <- getData('worldclim', var='bio', res=0.5, lon=-120, lat=0)
climatezone33 <- getData('worldclim', var='bio', res=0.5, lon=-90, lat=0)
climatezone34 <- getData('worldclim', var='bio', res=0.5, lon=-60, lat=0)
climatezone35 <- getData('worldclim', var='bio', res=0.5, lon=-30, lat=0)
climatezone36 <- getData('worldclim', var='bio', res=0.5, lon=0, lat=0)
climatezone37 <- getData('worldclim', var='bio', res=0.5, lon=30, lat=0)
climatezone38 <- getData('worldclim', var='bio', res=0.5, lon=60, lat=0)
climatezone39 <- getData('worldclim', var='bio', res=0.5, lon=90, lat=0)
climatezone310 <- getData('worldclim', var='bio', res=0.5, lon=120, lat=0)
climatezone311 <- getData('worldclim', var='bio', res=0.5, lon=150, lat=0)
climatezone40 <- getData('worldclim', var='bio', res=0.5, lon=-180, lat=-30)
climatezone41 <- getData('worldclim', var='bio', res=0.5, lon=-150, lat=-30)
climatezone42 <- getData('worldclim', var='bio', res=0.5, lon=-120, lat=-30)
climatezone43 <- getData('worldclim', var='bio', res=0.5, lon=-90, lat=-30)
climatezone44 <- getData('worldclim', var='bio', res=0.5, lon=-60, lat=-30)
climatezone45 <- getData('worldclim', var='bio', res=0.5, lon=-30, lat=-30)
climatezone46 <- getData('worldclim', var='bio', res=0.5, lon=0, lat=-30)
climatezone47 <- getData('worldclim', var='bio', res=0.5, lon=30, lat=-30)
climatezone48 <- getData('worldclim', var='bio', res=0.5, lon=60, lat=-30)
climatezone49 <- getData('worldclim', var='bio', res=0.5, lon=90, lat=-30)
climatezone410 <- getData('worldclim', var='bio', res=0.5, lon=120, lat=-30)
climatezone411 <- getData('worldclim', var='bio', res=0.5, lon=150, lat=-30)
#COMBINE ZONES TO CREATE ONE COMPLETE CLIMATE OBJECT
climatemosaic <- mosaic(climatezone01, climatezone02, climatezone03, climatezone04, climatezone05, climatezone06, climatezone07, climatezone08, climatezone09, climatezone010, climatezone011, climatezone10, climatezone11, climatezone12, climatezone13, climatezone14, climatezone15, climatezone16, climatezone17, climatezone18, climatezone19, climatezone110, climatezone111, climatezone20, climatezone21, climatezone22, climatezone23, climatezone24, climatezone25, climatezone26, climatezone27, climatezone28, climatezone29, climatezone210, climatezone211, climatezone30, climatezone31, climatezone32, climatezone33, climatezone34, climatezone35, climatezone36, climatezone37, climatezone38, climatezone39, climatezone310, climatezone311, climatezone40, climatezone41, climatezone42, climatezone43, climatezone44, climatezone45, climatezone46, climatezone47, climatezone48, climatezone49, climatezone410, climatezone411, fun=mean)
#EXTRACT MEAN VALUES FOR BOUNDARY POLYGONS & ATTACH TO SPDF (WEIGHT AND BUFFER OPTIONS NOT USED HERE)
gadmMEANS <- extract(climatemosaic, gadm, fun=mean, na.rm=TRUE, small=TRUE, layer=1, nl=19, sp=TRUE)
1 回答
以下是我下载和镶嵌数据的方法:
首先,我将使用一个自动下载数据的循环,并为每次下载导出.tif栅格 .
之后,我将构建一个包含所有导出的.tifs的文件列表,并使用
gdalbuildvrt()
函数创建虚拟马赛克 . 这将为您节省一些时间和硬盘空间 .最后,您可以使用
extract()
函数来提取值 . 请注意,提取函数的速度很慢,并且对于像您这样的较大数据集而言需要永久 .我个人会在外部软件或其他语言(如Python,ArcGIS或OTB ToolBox)中执行此操作 . 目前,我正在使用OTB工具箱中的
otbcli_LSMSVectorization
函数进行大量工作,该功能使您能够基于区域输入栅格和值栅格提取区域统计(均值/变量) .Last word of advice: 尝试将镶嵌和shp拆分为较小的tile / AOI(尽可能好),然后运行
extract()
函数,可能与foreach
循环和%dopar%
结合使用 . 这应该极大地减少处理时间 . 有关详细信息,请查看以下链接 .Gdal BuildVRT
otbcli_LSMSVectorization
foreach %dopar% with extract