CovariateSetup: Preparing Covariate Data for Use in Kriging
View source: R/CovariateSetup.R
| CovariateSetup | R Documentation |
Preparing Covariate Data for Use in Kriging
Description
This function is used to setup products of covariate data ready for use in Kriging. This functionality can either be applied to user-supplied covariate data or ready-made data products such as the global dataset of soil hydraulic and thermal parameters for earth system modeling (available at http://globalchange.bnu.edu.cn/research/soil4.jsp) and the median statistic of the Global Multi-resolution Terrain Elevation Data (GMTED2010; available at https://topotools.cr.usgs.gov/gmted_viewer/gmted2010_global_grids.php). In case of the latter, the data is downloaded at 30 arc-sec latitude/longitude grid cells and subsequently resampled to match training and target resolutions specified by the user.
Usage
CovariateSetup(
Training,
Target,
FilePrefix = "",
Covariates = "GMTED2010",
Source = "Origin",
Extent,
Buffer = 0.5,
Dir = getwd(),
Keep_Global = FALSE,
FileExtension = ".nc",
Compression = 9
)
Arguments
Training |
A SpatRaster file containing the data which is to be downscaled. Covariate data will be resampled to match this. |
Target |
Either numeric or a SpatRaster. If numeric, a single number representing the target resolution for the kriging step (i.e. wich resolution to downscale to). If a SpatRaster, data that the covariates and kriged products should align with. In case of a numeric input, covariate data is aggregated as closely as possible to desired resolution. If a SpatRaster, covariate data is resampled to match desired output directly. |
FilePrefix |
Character. A file name prefix for the produced files. |
Covariates |
Either character or a SpatRaster. If character, obtain frequently used and provably useful covariate data (i.e., GMTED2010 and soil data) and prepare for use in Kriging. Supported character values are "GMTED2010", "tksat", "tkdry", "csol", "k_s", "lambda", "psi", and "theta_s". If a SpatRaster, a user-supplied set of covariate data to be prepared for use in Kriging. |
Source |
Character. Only comes into effect when Covariates argument is specified as a character. Whether to attempt download of covariate data from the official sources (Source = "Origin") or a static copy of the data set on a private drive (Source = "Drive"). Default is "Origin". |
Extent |
Optional, prepare covariate data according to desired spatial specification. If missing/unspecified, maximal area of supplied data and covariat sets is used. Can be specified either as a raster object, an sf object, a terra object, or a data.frame. If Extent is a raster or terra object, covariates will be prepared according to rectangular extent thereof. If Extent is an sf (MULTI-)POLYGON object, this will be treated as a shapefile and the output will be cropped and masked to this shapefile. If Extent is a data.frame of geo-referenced point records, it needs to contain Lat and Lon columns around which a buffered shapefile will be created using the Buffer argument. |
Buffer |
Optional, Numeric. Identifies how big a circular buffer to draw around points if Extent is a data.frame of points. Buffer is expressed as centessimal degrees. |
Dir |
Character/Directory Pointer. Directory specifying where to download data to. |
Keep_Global |
Logical. Only comes into effect when Covariates argument is specified as a character. Whether to retain raw downloaded covariate data or not. Default is FALSE. |
FileExtension |
Character. A file extension for the produced files. Supported values are ".nc" (default) and ".tif" (better support for metadata). |
Compression |
Integer between 1 to 9. Applied to final .nc file that the function writes to hard drive. Same as compression argument in terra::writeCDF(). Ignored if FileExtension = ".tif". |
Value
A list containing two SpatRaster objects (Training and Target) ready to be used as covariates for kriging, and two files called Covariates_Target and Covariates_Train in the specified directory.
The SpatRasters produced and stored when specifying the Covariates argument as a character string and setting the Keep_Global argument to TRUE contain metadata/attributes as a named vector that can be retrieved with terra::metags(...):
Citation - A string which to use for in-line citation of the data product.
See Also
Kriging, Plot.Covariates.
Examples
## Not run:
## Rectangular Covariate data according to input data
CDS_rast <- terra::rast(system.file("extdata", "CentralNorway.nc", package = "KrigR"))
Covariates_ls <- CovariateSetup(
Training = CDS_rast,
Target = 0.01,
Covariates = "GMTED2010",
Keep_Global = TRUE,
FileExtension = ".nc"
)
Plot.Covariates(Covariates_ls)
## Shapefile-limited covariate data
data("Jotunheimen_poly")
CDS_rast <- terra::rast(system.file("extdata", "CentralNorway.nc", package = "KrigR"))
Covariates_ls <- CovariateSetup(
Training = CDS_rast,
Target = 0.01,
Covariates = "GMTED2010",
Extent = Jotunheimen_poly,
Keep_Global = TRUE,
FileExtension = ".nc"
)
Plot.Covariates(Covariates_ls, SF = Jotunheimen_poly)
## buffered-point-limited covariate data
data("Mountains_df")
CDS_rast <- terra::rast(system.file("extdata", "CentralNorway.nc", package = "KrigR"))
Covariates_ls <- CovariateSetup(
Training = CDS_rast,
Target = 0.01,
Covariates = c("tksat", "tkdry", "csol", "k_s", "lambda", "psi", "theta_s"),
Source = "Drive",
Extent = Mountains_df,
Buffer = 0.2,
Keep_Global = TRUE,
FileExtension = ".nc"
)
Plot.Covariates(Covariates_ls)
## End(Not run)
