Skip to contents

Filter National Forest Inventory Data

Source: R/filter_nfi.R
filter_nfi.Rd

The filter_nfi() function provides hierarchical and non-hierarchical filtering approaches for the complex structure of National Forest Inventory data based on user-provided condition expressions (expr_texts). This function enables effective filtering while maintaining the relationship between plot data (parent data) and other data (child data).

Usage

filter_nfi(data, expr_texts, hier = TRUE)

Arguments

data

: A list generated by read_nfi. Each dataframe should have a 'SUB_PLOT' column.

expr_texts

: @param expr_texts : A character vector; expressions specifying filtering conditions. Each expression should combine the dataframe name, dollar sign, and condition, with separate expressions for each data frame. (e.g., c("plot$OWN_CD == '5'", "tree$FAMILY == 'Pinaceae'"). Conditions must be valid R expressions.

hier

: A logical flag (default TRUE); indicates whether to apply hierarchical filtering (TRUE) or non-hierarchical filtering (FALSE). Hierarchical filtering ensures that connected dataframes are filtered based on the results of filters applied to the parent frame.

Value

A list of dataframes.

Details

This function parses expressions targeting specific columns in the dataframes within the provided list.

Expression requirements:

  • Each expression in expr_texts must start with a valid dataframe name in the list (e.g., "plot", "tree", "cwd") and combine the dataframe name, dollar sign, and condition (e.g. c("plot$OWN_CD == '5'").

  • Separate expressions must be provided for each dataframe being filtered (e.g. c("plot$OWN_CD == '5'", "tree$FAMILY == 'Pinaceae' | tree$WDY_PLNTS_TYP_CD == '1'").

Hierarchical filtering (hier = TRUE):

  • Filters applied to plot table affect all connected child data (tree, CWD, stump, etc.).

  • Filters applied to child data only operate within that dataframe and do not affect other dataframes.

  • Example: If only coniferous forest subplots are selected in the plot table, the child data will retain only the tree, CWD, stump, etc., associated with those subplots.

Non-hierarchical filtering (hier = FALSE):

  • Filters applied to the parent dataframe (plot table) do not affect the child data.

  • Filtering results from child data affect all other parent and child data.

  • Example: If only certain species are selected in the tree table, the plot table, CWD table, stump table, etc., will be filtered based on the remaining subplots from this selection.

Examples


data("nfi_donghae")

# Applying hierarchical filtering to select only privately owned forest subplots.
# Ensures all child tables' subplots match the filtered plot table's subplots.
# Expected results after filtering:
# all(nfi_donghae$tree$SUB_PLOT %in% nfi_donghae$plot$SUB_PLOT)  result: TRUE
nfi_donghae <- filter_nfi(nfi_donghae, c("plot$OWN_CD == '5'"), hier = TRUE)

# \donttest{ 
# Non-hierarchical filtering to select only privately owned forest subplots.
# Child tables remain unfiltered and may not correspond to the plot table subplots.
# Expected results after filtering:
# all(nfi_donghae$tree$SUB_PLOT %in% nfi_donghae$plot$SUB_PLOT)  result: FALSE
nfi_donghae <- filter_nfi(nfi_donghae, c("plot$OWN_CD == '5'"), hier = FALSE)

# Non-Hierarchical Filtering with only woody plants.
# Other tables remain filtered and correspond to the tree table.
# Expected results after filtering:
# all(nfi_donghae$plot$SUB_PLOT %in% nfi_donghae$tree$SUB_PLOT)  result: TRUE
nfi_donghae <- filter_nfi(nfi_donghae, c("tree$WDY_PLNTS_TYP_CD == '1'"), hier = FALSE)

# Combining multiple filters across different dataframes
nfi_donghae <- filter_nfi(nfi_donghae, 
                    c("plot$OWN_CD == '5'", 
                    "tree$FAMILY == 'Pinaceae' | tree$WDY_PLNTS_TYP_CD == '1'"))
# }