analyzeResult: Analyze performance of estimated time series.

Description Usage Arguments Details Value

View source: R/analyzeResult.R

Description

The function analyzeResult conducts a first level analysis of estimated time series.

Usage

1
2
3
analyzeResult(modelOutput, zero.val = 0.001, cens.level = 0.005,
  min.obs = 0.01, ErrLevs = seq(0.01, 1.5, 0.01), bins = seq(0, 1, 0.01),
  FlowStat = F)

Arguments

modelOutput

A list of model results, in the format returned by estDAR or estQPPQ.

zero.val

(optional) The censoring value used to indicate a censored streamflow value. The default is 0.001.

cens.level

(optional) The value below which a streamflow value is considered censored. (Rounded down rather than up.) The default is 0.005.

min.obs

(optional) The minimum non-censored value of streamflow. The default is 0.01.

ErrLevs

(optional) A vector of percent error values that should be used to construct the cumulative distribution of absolute percent errors. The default is seq(0.01,1.5,0.01).

bins

(optional) A vector of the bin limits that should be used to characterize the the percent errors along the flow duration curve. The default is seq(0,1,0.01).

FlowStat

(optional) A logical indicating if the streamflow statistics should be evaluated. This will increase run time. The default is FALSE.

Details

The single-value performance metric calcuated include: The Nash-Sutcliffe efficieny of daily streamflow, the Nash-Sutcliffe efficiency of the logarithms of the daily streamflow, the root-mean-square error statistics from rmse_like for natural and logarithm streamflows, the average percent errors of the natural streamflows, the average differences of the logarithms of streamflows, and the Pearson and Spearman correlations between observed and simulated streamflow.

Value

A list of several elements:

PerfMat

A data frame of single-value performance metrics for each target location. See details.

CumDistErr

A list of two elements:

  • LevelsThe input ErrLevs.

  • CumFreqA matrix summarizing the quantiles of the absolute percent errors at each site.

ErrAlongFDC

A list of several elements:

  • binsThe input bins.

  • centersThe plotting centers of each bin.

  • meanErrThe mean percent error for each bin for each target location.

  • medianErrThe median percent error for each bin for each target location.


wfarmer-usgs/PUBAD documentation built on May 4, 2019, 5:21 a.m.