fixes to NP summation

I realized that if the equation accepts any combination of subsets of subspecies, other equations later in the hierarchy might be completely skipped even though a more complete set of subspecies exist. To fix this, I had to basically give the function every possible combination of nutrient subspecies, in a larger hierarchy. This is now saved as a .csv in the package.

R/GenerateRefTables.R

@@ -715,6 +715,6 @@ TADA_UpdateMeasureQualifierCodeRef <- function() {
 #' @export
 
 TADA_GetNutrientSummationRef <- function() {
-  ref <- utils::read.csv(system.file("extdata", "Nsummation_key.csv", package = "TADA"))
+  ref <- utils::read.csv(system.file("extdata", "NPsummation_key.csv", package = "TADA"))
   return(ref)
 }

R/Transformations.R

@@ -325,7 +325,7 @@ TADA_HarmonizeSynonyms <- function(.data, ref, np_speciation = TRUE) {
 
 #' Calculate Total Nitrogen and Phosphorus
 #'
-#' This function uses the [Nutrient Aggregation logic](https://echo.epa.gov/trends/loading-tool/resources/nutrient-aggregation#nitrogen)
+#' This function applies the [Nutrient Aggregation logic](https://echo.epa.gov/trends/loading-tool/resources/nutrient-aggregation#nitrogen)
 #' from ECHO's Water Pollutant Loading Tool to add nitrogen subspecies together
 #' to approximate a total nitrogen value on a single day at a single site.
 #' Before summing subspecies, this function runs TADA_AggregateMeasurements to
@@ -334,9 +334,29 @@ TADA_HarmonizeSynonyms <- function(.data, ref, np_speciation = TRUE) {
 #' nitrogen subspecies expressed as nitrate, nitrite, ammonia, ammonium, etc. to
 #' as nitrogen based on the atomic weights of the different elements in the
 #' compound. The reference table is contained within the package but may be
-#' edited/customized by users. Future development may include total P summations
-#' as well.
-#'
+#' edited/customized by users. Nutrient equations are as follows:
+#' 
+#' NITROGEN:
+#' 1. TOTAL N (UNFILTERED)
+#' 2. TOTAL N (FILTERED) + TOTAL N (PARTICULE)
+#' 3. TOTAL KJELDAHL NITROGEN + NITRATE + NITRITE
+#' 4. ORGANIC N + AMMONIA + NITRATE + NITRITE
+#' 5. OTHER NITROGEN FORMS 
+#' 
+#' PHOSPHORUS:
+#' 1. TOTAL PHOSPHORUS
+#' 2. PHOSPHATE
+#' 3. OTHER PHOSPHORUS FORMS
+#' 
+#' Equations are applied in the order above. The function looks for groups of
+#' nutrients that exactly match each equation before looking for every
+#' combination within each equation (for example, a group of nitrogen subspecies
+#' including AMMONIA and NITRATE will be passed over in an intial sweep of
+#' groups of subspecies containing ORG N, AMMONIA, NITRATE, and NITRITE, but
+#' will be caught as the function moves down the hierarchy of equations to fewer
+#' and fewer subspecies). Eventually, even groups with only one subspecies will
+#' be used to represent a TOTAL N value for that site/day/depth.
+#' 
 #' @param .data TADA dataframe, ideally harmonized using TADA_HarmonizeSynonyms.
 #'   If user wants to consider grouping N or P subspecies across multiple
 #'   organizations, user should have run TADA_FindNearbySites and grouped all
@@ -420,43 +440,28 @@ TADA_CalculateTotalNP <- function(.data, sum_ref, daily_agg = c("max", "min", "m
       dplyr::group_by(dplyr::across(dplyr::all_of(thecols))) %>%
       dplyr::mutate(TADA.NutrientSummationGroup = dplyr::cur_group_id())
 
-    # Create list of summation equations in order of preference, can also accommodate phosphorus
-    eqns <- list(
-      "N" = list(
-        N0 = "TOTAL N (UNFILTERED)",
-        N1 = c("TOTAL N (FILTERED)", "TOTAL N (PARTICLE)"),
-        N2 = c("TKN", "NITRATE", "NITRITE"),
-        N2_a = c("TKN", "NITRATE + NITRITE"),
-        N3 = c("ORG N", "AMMON", "NITRATE", "NITRITE"),
-        N3_a = c("ORG N", "AMMON", "NITRATE + NITRITE"),
-        N4 = c("AMMON", "NITRATE", "NITRITE"),
-        N4_a = c("AMMON", "NITRATE + NITRITE"),
-        N5 = "OTHER N"
-      ),
-      "P" = list(
-        P1 = c("PHOSP"),
-        P2 = c("PO4"),
-        P3 = "OTHER P"
-      )
-    )
+    # bring in equations
+    eqns = utils::read.csv(system.file("extdata","NP_equations.csv", package = "TADA"))
+
 
     # dataframe to hold results
     summeddata <- data.frame()
     grps <- vector()
 
-    for (i in 1:length(eqns)) {
-      eq <- eqns[[i]]
-      for (j in 1:length(eq)) {
-        eq1 <- eq[[j]]
-        eqname <- names(eq[j])
-        nutrient <- ifelse(grepl("N", names(eq[j])), "Total Nitrogen as N", "Total Phosphorus as P")
+    for (i in 1:length(unique(eqns$Nutrient))) {
+      nut = unique(eqns$Nutrient)[i]
+      nutqns = subset(eqns, eqns$Nutrient==nut)
+      for (j in 1:length(unique(nutqns$EQN))) {
+        eqnum = unique(nutqns$EQN)[j]
+        eqn = subset(nutqns, nutqns$EQN==eqnum)$SummationName
+        nutrient <- ifelse(nut=="N", "Total Nitrogen as N", "Total Phosphorus as P")
         # for each equation, see if any groups contain all required subspecies, and for each pick the variant with the lowest rank.
         # combine group with other groups and remove group ID from consideration for the next equation
         out <- sum_dat %>%
           dplyr::filter(!TADA.NutrientSummationGroup %in% grps) %>%
           dplyr::group_by(TADA.NutrientSummationGroup) %>%
-          # dplyr::filter(all(eq1 %in% SummationName)) %>% # this line ensures that ALL subspecies are present within an equation group, not just one or more
-          dplyr::filter(SummationName %in% eq1) %>% 
+          dplyr::filter(all(eqn %in% SummationName)) %>% # this line ensures that ALL subspecies are present within an equation group, not just one or more
+          dplyr::filter(SummationName %in% eqn) %>% 
           dplyr::mutate(TADA.NutrientSummationEquation = paste0(unique(SummationName), collapse = " + "))
 
         out <- out %>%

inst/extdata/NP_equations.csv

@@ -0,0 +1,47 @@
+Nutrient,EQN,SummationName
+N,0,TOTAL N (UNFILTERED)
+N,1,TOTAL N (FILTERED)
+N,1,TOTAL N (PARTICLE)
+N,2,TKN
+N,2,NITRATE
+N,2,NITRITE
+N,3,TKN
+N,3,NITRATE + NITRITE
+N,4,ORG N
+N,4,AMMON
+N,4,NITRATE
+N,4,NITRITE
+N,5,ORG N
+N,5,AMMON
+N,5,NITRATE + NITRITE
+N,6,AMMON
+N,6,NITRATE
+N,6,NITRITE
+N,7,AMMON
+N,7,NITRATE + NITRITE
+N,8,TKN
+N,8,NITRATE
+N,9,TKN
+N,9,NITRITE
+N,10,ORG N
+N,10,AMMON
+N,11,ORG N
+N,11,NITRATE
+N,12,ORG N
+N,12,NITRITE
+N,13,ORG N
+N,13,NITRATE + NITRITE
+N,14,AMMON
+N,14,NITRATE
+N,15,AMMON
+N,15,NITRITE
+N,16,TKN
+N,17,NITRATE
+N,18,NITRITE
+N,19,NITRATE + NITRITE
+N,20,AMMON
+N,21,ORG N
+N,22,OTHER N
+P,1,PHOSP
+P,2,PO4
+P,3,OTHER P