doQmapQUANT: Apply a quantile mapping

In Burke-Lauenroth-Lab/rSFSW2: Simulation Framework for SOILWAT2

Description

Whereas the function fitQmapQUANT estimates values of the empirical cumulative distribution function of observed and modeled time series for regularly spaced quantiles. doQmapQUANT.default_drs uses these estimates to perform quantile mapping.

Usage

doQmapQUANT.default_drs(
  x,
  fobj,
  type = NULL,
  lin_extrapol = NULL,
  spline_method = NULL,
  monthly_extremes = NULL,
  fix_spline = NULL,
  ...
)

doQmapQUANT_drs(
  x,
  fobj,
  type_map = NULL,
  monthly_obs_base = NULL,
  monthly_extremes = NULL,
  fix_spline = NULL,
  ...
)

Arguments

x	A numeric vector. The values to map.
fobj	An object of class fitQmapQUANT.
type	A character string. Type of interpolation between the fitted transformed values. See details.
lin_extrapol	A character string. Type of extrapolation when interpolation is linear. See details.
spline_method	A character string. Type of spline, passed to splinefun as method argument. The type “monoH.FC” is the only appropriate method here because quantile mapping requires a monotone function if possible.
monthly_extremes	A numeric vector of length two. The first element suggests a monthly minimum value and the second element a monthly maximum value for the mapped output.
fix_spline	A character string. See details.
...	Additional arguments are ignored.
type_map	A character vector. The type of interpolation, extrapolation, and spline passed to doQmapQUANT.default_drs. Possible values include “linear_Boe”, “linear_Thermessl2012CC.QMv1b”, “linear_none”, “tricub_fmm”, “tricub_monoH.FC”, “tricub_natural”, and “normal_anomalies”. See details.
monthly_obs_base	A numeric vector. Base values used to calculate t-scores of x which are only used if type_map is “normal_anomalies”.

Value

A numeric vector of the length of x. Return values differ among repeated calls with identical input arguments if jitter correction (using random numbers) is applied, i.e., type is “spline”, fix_spline is “attempt” and there are values outside the range suggested by monthly_extremes.

Details

• type takes one of two possible values

  • “linear”: linear interpolation using approx

  • “tricub”: monotonic tricubic spline interpolation using splinefun. Splines may result in abnormally high output values which appears to be due to at least two reasons:

    1. extrapolation errors

    2. huge oscillations in the spline-function which arise from non-monotone splines (spline_method is “fmm” or “natural”) or which arise from numerical instabilities in the exact monotonicity if spline_method is “monoH.FC”.

• lin_extrapol is the extrapolation for values of x that are outside range(fobj[["par"]]$modq). This argument is only in effect if type is “linear” and takes one of three possible values

  • “none”: no linear extrapolation is performed, i.e., output of approx for type = 2 is return; that is 'values at the closest data extreme'.

  • “Boe”: constant extrapolation from Boe et al. 2007

  • “Thermessl2012CC.QMv1b”: same extrapolation as Boe et al. 2007, but not including three largest/smallest values, from Themessl et al. 2012.

• fix_spline takes one of three values

  • “none”: No correction to mapped values is applied.

  • “fail”: If mapped values fall outside the range suggested by monthly_extremes, then an error is generated.

  • “attempt”: The spline-based mapping is repeated up to ten times where the values of the quantile map fobj are jittered.

• type_map with “normal_anomalies” represents a 'linear interpolation with extrapolation following Boe et al. 2007 and a correction using standard anomalies (i.e. number of standard deviations from the mean) for values outside the observed quantile map that is based on Tohver et al. 2014 (Appendix A)

References

Boe, J., L. Terray, F. Habets, and E. Martin. 2007. Statistical and dynamical downscaling of the Seine basin climate for hydro-meteorological studies. International Journal of Climatology 27:1643-1655.

Themessl, M. J., A. Gobiet, and G. Heinrich. 2011. Empirical-statistical downscaling and error correction of regional climate models and its impact on the climate change signal. Climatic Change 112:449-468.

Gudmundsson, L., J. B. Bremnes, J. E. Haugen, and T. Engen-Skaugen. 2012. Technical Note: Downscaling RCM precipitation to the station scale using statistical transformations - a comparison of methods. Hydrology and Earth System Sciences 16:3383-3390.

Tohver, I. M., A. F. Hamlet, and S.-Y. Lee. 2014. Impacts of 21st-Century Climate Change on Hydrologic Extremes in the Pacific Northwest Region of North America. Journal of the American Water Resources Association 50:1461-1476.

See Also

Based on code from doQmapQUANT v1.0.4 (Gudmundsson et al. 2012), but with additional methods and more granular control. See details.

Burke-Lauenroth-Lab/rSFSW2 documentation built on Aug. 14, 2020, 5:20 p.m.