mytestpkg: Demonstration of concepts

#
# # Interpolation
#
# ##Test data: Precipitation.RData
#
# #View(test) # shows the table of data
# #library(sp)
#
# #coordinates(test)=~lon+lat    -> produces error: Error in coordinates(test) = ~lon + lat : could not find function "coordinates<-"
#
# test=data.frame(test) # makes a data frame because of error:Error in coordinates(test) = ~lon + lat :
# #could not find function "coordinates<-"
# coordinates(test)=~lon+lat
# plot(test)
#
# proj4string(test) <- CRS("+init=epsg:4326") # to apply coordinate references
# test2 <- spTransform(test,CRSobj = CRS("+init=epsg:32634"))  # transforms the coordinate reference system to our local projected one
#
#
# library(mapview) # shows the map
# mapview(test2)
#
# library(raster) # needed to use getData
# srtm <- getData('SRTM', lon=mean(coordinates(test)[,1]), lat=mean(coordinates(test)[,2])) #elevation model
#
# plot(srtm) # elevation map
#
# extent(test)  # shows the coordinates
# srtm2=crop(srtm, extent(test)+1) # extends area
# mapview(srtm2)  # shows chosen area
#
# srtm3=projectRaster(srtm2, crs=CRS("+init=epsg:32634")) # first crop, then project, as raster might take ages to be build
#
# #######################################################################
# # Kriging
#
#
# library(gstat)
# rain_idw=idw(mean_r~1, test2,        # idw = makes the kriging funcion possible / mean of r with behaviour 1
#              as(srtm3,"SpatialGridDataFrame")) # raster has to be transformed to Spatial grid data frame as gstat doesnt like raster
# #rain_idw=idw(mean_r~1, test2,        # idw = makes the kriging funcion possible / mean of r with behaviour 1
# #as(srtm3,"SpatialGridDataFrame"), #idp= 10 -> looks like thysenpolygon) -> default here is idp=2
# # mapview(rain_idw) # shows precipitation --> creates error
#
# plot(rain_idw)
#
# plot(variogram(mean_r~1, loc=test2, cloud=T))
#
# va <- variogram(mean_r~1, loc = test2,
#                 cutoff = 100000, width = 10000)  # cutoff: after what distance are we not interested anymore?
#                   # width = distance between points, sensitive parameter
# plot(va)
# plot(va, vgm(8e+04, "Exp", 4e+04, 0)) # playing around with data, adds the result of the empirical variogram
#
# vmf <- fit.variogram(va, vgm(8e+04, "Exp", 4e+04, 0), fit.method = 7) # 7 -> ordinary squares
# plot(va,vmf) # fits the curve to the points
#
#
# rain_krige=krige(mean_r~1, test2,
#                   as(srtm3,"SpatialGridDataFrame"),
#                   vmf) # uses ordinary kriging
# plot(rain_krige)
#
# rain_krige_brick=brick(rain_krige) # transforms it into a raster brick
#
# plot(rain_krige_brick)
#
# library(automap)
# srtm4=aggregate(srtm3, 10)
#
#
# autkrig=autoKrige(mean_r~1, test2,
#                    as(srtm4, "SpatialGridDataFrame"))
#
# plot(autkrig)
#
# ###### KED = Kriging with external Drift
#
# names(srtm4)="altitude"   # renames srtm4 (aggregated srtm!)
# names(srtm4)  # checks on the name
#
# test3=test2[!is.na(test2$altitude),]  # sorting values from NA´s
#
# autked=autoKrige(mean_r~altitude, test3,
#                  as(srtm4,"SpatialGridDataFrame"))
#
# plot(autked)
#
#
#
#
#
#