Skip to contents

NSF Water Quality Index (NSF WQI, prototype)

Source: R/nsfwqi.R
nsfwqi.Rd

Computes a prototype NSF WQI as a weighted arithmetic mean of parameter sub-scores (Qi) using simple piecewise rules. This is intended for quick demonstrations and is not a full replication of the original NSF curves.

Usage

nsfwqi(
  df,
  pesos = c(do = 0.17, fc = 0.16, ph = 0.11, bod = 0.11, temp_change = 0.1, po4 = 0.1,
    no3 = 0.1, turbidez = 0.08, sst = 0.07),
  na_rm = FALSE
)

Arguments

df

Data frame containing columns compatible with the mapping above.

pesos

Named numeric vector with parameter weights. Defaults follow a common NSF WQI variant: do=.17, fc=.16, ph=.11, bod=.11, temp_change=.10, po4=.10, no3=.10, turbidez=.08, sst=.07.

na_rm

Logical; allow NA per row and rescale weights to available parameters (TRUE) or error on missing inputs (FALSE).

Value

The input df with an added numeric column NSFWQI.

Details

The function accepts both NSF-style column names and common Brazilian aliases. The mapping tried (if present) is:

  • do <- od

  • fc <- coliformes

  • ph <- pH or ph

  • bod <- dbo

  • turbidez stays turbidez

  • sst <- solidos_suspensos

  • po4 <- po4 or p_ortofosfato

  • no3 <- no3 or n_nitrato

  • temp_change must be supplied as-is (delta T to reference)

If na_rm = TRUE, weights are rescaled per row to the parameters available in that row. If na_rm = FALSE (default), any missing required input leads to an error.

Examples

d <- wq_demo
# create minimal aliases so the prototype can run
d$do  <- d$od
d$fc  <- d$coliformes
d$ph  <- d$ph
d$bod <- d$dbo
# others are missing; use na_rm = TRUE to rescale weights by row
out <- nsfwqi(d, na_rm = TRUE)
#> Warning: 'drop' argument will be ignored
head(out$NSFWQI)
#> [1] 68.33333 87.46032 61.42857 79.60317 84.92063 63.88889