CityWaterBalance (version 0.1.0)

getSolutions: Evaluate parameter uncertainty

Description

This function searches for acceptable model solutions within the uncertainty parameters and long-term storage balances using Latin hypercubes.

Usage

getSolutions(data, p, n, tol = 0.01, interc = c(0, 0.05), et_mult = c(1,
  1.1), flow_mult = c(1, 1.1), open_wat = c(0.01, 0.1), run_mult = c(1,
  5), run_css = c(0.1, 1), bf_mult = c(0.5, 1.5), nonrev = c(0.05, 0.2),
  ind_evap = c(0.01, 0.02), wast_gen = c(0.75, 0.9), pot_atm = c(0.1,
  0.15), npot_infilt = c(0.25, 0.75), slud_evap = c(0, 0),
  leak_css = c(0.05, 0.25), dgw = c(0.5, 0.5), dgw_rep = c(0, 1),
  global_bal = c(-500, 500), sw_bal = c(-500, 500), css_bal = c(-500,
  500), sgw_bal = c(-500, 500), dgw_bal = c(-500, 500))

Arguments

data

xts or zoo object. See CityWaterBalance function for details.

p

list of initial parameter values. See CityWaterBalance function or the inputs below for descriptions.

n

integer number of initial parameter sets to search

tol

tolerance acceptable difference mean flow solutions

interc

vector of min and max fraction of pet lost to interception

et_mult

vector of min and max multiplier for et

flow_mult

vector of min and max multiplier for outflow

open_wat

vector of min and max fraction of area that is open water

run_mult

vector of min and max multiplier for runoff

run_css

vector of min and max fraction of runoff diverted to sewers

bf_mult

vector of min and max multiplier for baseflow

nonrev

vector of min and max fraction of potable water supply lost to leaks

ind_evap

vector of min and max fraction of industrial use that evaporates

wast_gen

vector of min and max fraction of potable use that returns to sewers

pot_atm

vector of min and max fraction of potable use that evaporates

npot_infilt

vector of min and max fraction of nonpotable use that infiltrates

slud_evap

vector of min and max fraction of wastewater that evaporates from sludge

leak_css

vector of min and max fraction of wastewater effluent from gw infiltration

dgw

vector of min and max fraction of groundwater from deep, confined aquifers

dgw_rep

vector of min and max multiplier for deep groundwater pumping replacement

global_bal

vector of min and max acceptable global water balance values, cumulative over model run

sw_bal

vector of min and max acceptable surface water balance values, cumulative over model run

css_bal

vector of min and max acceptable sewer system water balance values, cumulative over model run

sgw_bal

vector of min and max acceptable shallow groundwater balance values, cumulative over model run

dgw_bal

vector of min and max acceptable deep groundwater balance values, cumulative over model run

Value

out numeric solutions

Details

The function creates n parameter sets using a Latin hypercube. It runs `CityWaterBalance()` with each set, accepting solutions that meet user-defined criteria for storage balances. It then computes the mean of flow solutions, and doubles n until the difference between the means of old and new solutions is less than tol for all flows. Defaults for parameter value ranges are set to reasonable values, but they should be reconsidered for each application. Defaults for storage balances are set high to allow for solution discovery, however, acceptable values must be determined on a case-by-case basis.

Examples

data <- cwb_data
data$cso <- 0
p <- list("interc" = 0,"et_mult" = 1,"flow_mult" = 1, "open_wat" = 0.02, 
          "run_mult" = 3.378, "run_css" = 0.35, "bf_mult" = 1, 
          "nonrev" = 0.08, "ind_evap" = 0.012, "wast_gen" = 0.85, 
          "pot_atm" = 0.13, "npot_infilt" = 0.5, "slud_evap" = 0,
          "leak_css" = 0.05, "dgw" = 0.5, "dgw_rep" = 0.5)
out <- getSolutions(data, p, 10, 0.1)