merging.options: Options Controlling the merging parameters.

View source: R/cdtMerging_Options_functions.R

merging.optionsR Documentation

Options Controlling the merging parameters.

Description

Functions to handle settings used by the merging.

Usage

merging.options(...)

Arguments

...

using one or more arguments of the form name = value. Existing values can be retrieved by supplying the names (as character strings) of the components as unnamed arguments.

Details

Available options are

  • mrgMinNumberSTN: integer, minimum number of stations to perform the merging. If the number of stations is less than mrgMinNumberSTN, then no merging will be performed and the input netCDF will be taken

  • rkMinNumberSTN: integer, minimum number of stations to perform a regression kriging. If the number of stations is less than rkMinNumberSTN, then the merging method will be replaced by a simple bias adjustment

  • vgmMinNumberSTN: integer, minimum number of station to compute the empirical variogram. If the number of stations is less than vgmMinNumberSTN, then the variogram will not be computed and the interpolation method will be replaced by IDW

  • useLocalInterpolation: logical, use local interpolation, if FALSE a global interpolation will be use and the arguments nmin, nmax and maxdist from interp.method will not be considered

  • powerWeightIDW: numeric, inverse distance weighting power

  • powerWeightShepard: numeric, modified Shepard weighting power

  • powerWeightBarnes: numeric, Barnes interpolation weighting power

  • addCoarseGrid: logical, add a coarse grid to interpolate the residuals. The coarse grid will be created by resampling the grid of the input data. As CDT uses an additive model, the residual values at the coarse grid will be set to zero

  • saveGridBuffer: logical, save the buffer of coarse grid used to interpolate the residuals. Default is FALSE

  • saveRnoR: logical, save the rain-no-rain mask. Default is FALSE

  • RnoRModel: character, model to use to compute the rain-no-rain mask. Options are: "logit", "additive". Default is "logit"

  • RnoRCutOff: integer, the method to be used to define the decision boundaries of the rain-no-rain mask. Options are: 1, 2, 3

    • option: 1

      mask = \left\{ \begin{array}{l l} 0 & \quad \mbox{if } rnr < 0.5 \\ 1 & \quad \mbox{if } rnr \geq 0.5 \end{array} \right.

    • option: 2

      mask = \left\{ \begin{array}{l l} 0 & \quad \mbox{if } rnr < 0.1 \\ rnr & \quad \mbox{if } rnr \geq 0.1 \end{array} \right.

    • option: 3

      mask = \left\{ \begin{array}{l l} 0 & \quad \mbox{if } rnr < 0.25 \\ rnr & \quad \mbox{if } 0.25 \leq rnr < 0.75 \\ 1 & \quad \mbox{if } rnr \geq 0.75 \end{array} \right.

    where rnr is the interpolated rain-no-rain values. Default is 3.

  • RnoRaddCoarse: logical, use the coarse grid to create the rain-no-rain mask. Default is FALSE

  • RnoRUseMerged: logical, if TRUE the merged data is used to compute the rain-no-rain mask, otherwise the input gridded data is used. Default is FALSE

  • RnoRSmoothingPixels: integer, the number of pixels from the target pixel to be used to smooth the rain-no-rain mask. Default is 2.

  • blockType: character, the method to use to create the block estimation when the argument use.block from interp.method is TRUE. Available options are "gaussian" and "userdefined"

    • "gaussian": using Gaussian quadrature method

    • "userdefined": user defined block

  • blockSize: vector, the size of the block.

    • "gaussian": vector of length 2 in the form c(width_x, width_y)

    • "userdefined": vector of length 4 in the form c(width_x, by_x, width_y, by_y)

  • netCDFDataDef: named list, netCDF variable definition


rijaf-iri/CDT documentation built on July 3, 2024, 2:54 a.m.