Post-stratified estimator for custom variables in FIA Data

Produces estimates of population totals, ratios, and associated variances for custom variables using FIA's post-stratified inventories. Accepts tree- and condition-level summaries from rFIA estimator functions as input, with potential modifications to associated variables (e.g., custom allometrics can be applied to estimate tree biomass/carbon). See our website for example use cases.

Usage

customPSE(db, x, xVars, xGrpBy = NULL, xTransform = NULL, y = NULL,
          yVars = NULL, yGrpBy = NULL, yTransform = NULL, method = "TI",
          lambda = 0.5, totals = TRUE, variance = TRUE)

Arguments

db: FIA.Database or Remote.FIA.Database object produced from readFIA() or getFIA(). If a Remote.FIA.Database, data will be read in and processed state-by-state to conserve RAM (see details for an example).
x: data.frame; tree- or condition-list containing numerator variable(s). See details for more info on producing acceptable tree- and condition-lists using rFIA estimator functions.
xVars: name of variable(s) in x to be treated as numerator variables, unquoted. Multiple variables should be combined with c(), e.g., xVars = c(TPA, BAA).
xGrpBy: names of variables in x to be treated as grouping variables for the numerator, unquoted. Multiple variables should be combined with c(), e.g., xVars = c(SPCD, OWNGRPCD).
xTransform: function to be applied to plot-level summaries of numerator variables, e.g., sqrt or function(x){log(x + 0.001)}.
y: data.frame; tree- or condition-list containing denominator variable. See details for more info on producing acceptable tree- and condition-lists using rFIA estimator functions.
yVars: name of variable in y to be treated as denominator variables, unquoted. Maximum one denominator variable allowed at this time.
yGrpBy: names of variables in y to be treated as grouping variables for the numerator, unquoted. Multiple variables should be combined with c(), e.g., xVars = c(SPCD, OWNGRPCD).
yTransform: function to be applied to plot-level summaries of denominator variable, e.g., sqrt or function(x){log(x + 0.001)}.
method: character; design-based estimator to use. One of: "TI" (temporally indifferent, default), "annual" (annual), "SMA" (simple moving average), "LMA" (linear moving average), or "EMA" (exponential moving average). See Stanke et al 2020 for a complete description of these estimators.
lambda: numeric (0,1); if method = 'EMA', the decay parameter used to define weighting scheme for annual panels. Low values place higher weight on more recent panels, and vice versa. Specify a vector of values to compute estimates using mulitple wieghting schemes, and use plotFIA with grp set to lambda to produce moving average ribbon plots. See Stanke et al 2020 for examples.
totals: logical; if TRUE, return total population estimates (e.g. total area) along with ratio estimates (e.g. mean trees per acre).
variance: logical; if TRUE, return estimated variance (VAR) and sample size (N). If FALSE, return 'sampling error' (SE) as returned by EVALIDator. Note: sampling error cannot be used to construct confidence intervals.

Details

Workflow and intended use cases

customPSE is intended to be used in combination with standard rFIA estimator functions, like tpa(), area(), and volume(), among others. Standard rFIA estimator functions generate tree- and or condition-lists for standard variables of interest (see treeList and condList arguments in estimator functions). Users may make modifications to these standard variables, for example a variable representing tree crown area may be added to a tree-list produced by tpa() (via some suite of allometrics). Users may then hand their modified tree-list to customPSE to estimate the total and proportion of forested land area in their domain of interest that is covered by tree crowns.

customPSE may be used to estimate population totals for multiple variables simultaneously (total number of trees in a region), and given a denominator variable, the associated population ratios (trees per forested acre, where forested land area is the denominator). Estimation follows the procedures documented in Bechtold and Patterson (2005) and Stanke et al 2020.

Three general forms of ratio estimates may be produced: tree-tree, tree-area, and area-area ratios. For example, if tree height is specified as the numerator variable (adjusted for sampling area by multiplying by TPA), and TPA is specified as the denominator variable, a tree-tree ratio will be produced that represents the height of the average tree within a region of interest. Similarly, if stand age is specified as the numerator (adjusted for sampling area by proportionate area of the forested condition on the plot, i.e., PROP_FOREST), and the proportion area of the plot that is forested is specified as the denominator, an area-area ratio will be produced that represents average stand age within the region of interest. Tree-area ratios are more familiar, such as trees per acre, tree biomass per acre, etc, where a tree variable is specified as the numerator, and proportion of plot area occupied by forestland is the denominator. See our website for detailed examples of each of these ratio estimates.

Input requirements

Estimation of tree variables require the following columns be present in x and/or y: PLT_CN, EVAL_TYP, SUBP, TREE, and TREE_BASIS. Similarly, estimation of area variables require the following columns be present in x and/or y: PLT_CN, EVAL_TYP, CONDID, and AREA_BASIS. Each of these required variables will be returned in tree- and condition-lists generated by standard rFIA estimator functions.

IMPORTANT: Only one of TREE_BASIS or AREA_BASIS may be present x or y, as the presence of these columns are used to determine if variables to be estimated are tree variables or area variables. Some standard rFIA estimator functions will produce tree-lists with both TREE_BASIS and AREA_BASIS listed in output, as the tree-list will contain tree variables (e.g., TPA, BAA) as well as area variables (e.g., PROP_FOREST, proportion of plot represented by the forested condition where each tree is growing)). To produce a tree-area ratio with such an output, AREA_BASIS must be removed from the data.frame specified in x, and TREE_BASIS must be removed from that specified in y.

Value

YEAR: reporting year associated with estimates
*_RATIO: population ratio estimate, where * will be replaced with the name of each numerator variable
*_TOTAL: population total estimate, where * will be replaced with the name of each numerator/ denominator variable
*_RATIO_VAR: estimated variance of the population ratio
*_TOTAL_VAR: estimated variance of the population total
nPlots_x: number of non-zero plots used to compute numerator estimates
nPlots_y: number of non-zero plots used to compute denominator estimates
N: total number of plots (including zeros) associated with each inventory

References

rFIA website: https://rfia.netlify.app/

FIA Database User Guide: https://research.fs.usda.gov/understory/forest-inventory-and-analysis-database-user-guide-nfi

Bechtold, W.A.; Patterson, P.L., eds. 2005. The Enhanced Forest Inventory and Analysis Program - National Sampling Design and Estimation Procedures. Gen. Tech. Rep. SRS - 80. Asheville, NC: U.S. Department of Agriculture, Forest Service, Southern Research Station. 85 p. https://www.srs.fs.usda.gov/pubs/gtr/gtr_srs080/gtr_srs080.pdf

Stanke, H., Finley, A. O., Weed, A. S., Walters, B. F., & Domke, G. M. (2020). rFIA: An R package for estimation of forest attributes with the US Forest Inventory and Analysis database. Environmental Modelling & Software, 127, 104664.

Author

Hunter Stanke and Andrew Finley

Note

All sampling error estimates (SE) are returned as the "percent coefficient of variation" (standard deviation / mean * 100) for consistency with EVALIDator. IMPORTANT: sampling error cannot be used to construct confidence intervals. Please use variance = TRUE for that (i.e., return variance and sample size instead of sampling error).

Examples

# See our website for a more thorough suite of examples
data(fiaRI)

# Get tree-list from tpa
tree.list <- tpa(fiaRI,
                 treeList = TRUE)

# Estimate trees per acre and basal area per acre
customPSE(db = fiaRI,
          # Numerator variables
          x = dplyr::select(tree.list, -c(AREA_BASIS)),
          xVars = c(TPA, BAA),
          # Denominator variables
          y = dplyr::select(tree.list, -c(TREE_BASIS)),
          yVars = PROP_FOREST)
#> # A tibble: 6 × 14
#>    YEAR TPA_RATIO BAA_RATIO  TPA_TOTAL BAA_TOTAL PROP_FOREST_TOTAL TPA_RATIO_VAR
#>   <dbl>     <dbl>     <dbl>      <dbl>     <dbl>             <dbl>         <dbl>
#> 1  2013      467.      119. 171481307. 43626598.           367372.          960.
#> 2  2014      466.      120. 171464450. 44107127.           367884.          983.
#> 3  2015      444.      121. 164362240. 44586819.           369844.          808.
#> 4  2016      450.      123. 164787783. 45021353.           366364.          844.
#> 5  2017      441.      123. 162503315. 45388036.           368373.          863.
#> 6  2018      427.      122. 156585656. 44810915.           366959.          801.
#> # ℹ 7 more variables: BAA_RATIO_VAR <dbl>, TPA_TOTAL_VAR <dbl>,
#> #   BAA_TOTAL_VAR <dbl>, PROP_FOREST_TOTAL_VAR <dbl>, nPlots_x <int>,
#> #   nPlots_y <int>, N <int>

# Same as above, but rename variables for a clean output
customPSE(db = fiaRI,
          x = dplyr::select(tree.list, -c(AREA_BASIS)),
          # Variables can be renamed using c()
          xVars = c(NUM = TPA,
                    BA = BAA),
          y = dplyr::select(tree.list, -c(TREE_BASIS)),
          # Variables can be renamed using c()
          yVars = c(FOREST_AREA = PROP_FOREST))
#> # A tibble: 6 × 14
#>    YEAR NUM_RATIO BA_RATIO  NUM_TOTAL  BA_TOTAL FOREST_AREA_TOTAL NUM_RATIO_VAR
#>   <dbl>     <dbl>    <dbl>      <dbl>     <dbl>             <dbl>         <dbl>
#> 1  2013      467.     119. 171481307. 43626598.           367372.          960.
#> 2  2014      466.     120. 171464450. 44107127.           367884.          983.
#> 3  2015      444.     121. 164362240. 44586819.           369844.          808.
#> 4  2016      450.     123. 164787783. 45021353.           366364.          844.
#> 5  2017      441.     123. 162503315. 45388036.           368373.          863.
#> 6  2018      427.     122. 156585656. 44810915.           366959.          801.
#> # ℹ 7 more variables: BA_RATIO_VAR <dbl>, NUM_TOTAL_VAR <dbl>,
#> #   BA_TOTAL_VAR <dbl>, FOREST_AREA_TOTAL_VAR <dbl>, nPlots_x <int>,
#> #   nPlots_y <int>, N <int>

# Ensure the above matches expected output
tpa(fiaRI,
    totals = TRUE,
    variance = TRUE)
#> # A tibble: 6 × 14
#>    YEAR   TPA   BAA TREE_TOTAL  BA_TOTAL AREA_TOTAL TPA_VAR BAA_VAR
#>   <dbl> <dbl> <dbl>      <dbl>     <dbl>      <dbl>   <dbl>   <dbl>
#> 1  2013  467.  119. 171481307. 43626598.    367372.    960.    13.5
#> 2  2014  466.  120. 171464450. 44107127.    367884.    983.    13.8
#> 3  2015  444.  121. 164362240. 44586819.    369844.    808.    13.6
#> 4  2016  450.  123. 164787783. 45021353.    366364.    844.    13.1
#> 5  2017  441.  123. 162503315. 45388036.    368373.    863.    13.8
#> 6  2018  427.  122. 156585656. 44810915.    366959.    801.    13.9
#> # ℹ 6 more variables: TREE_TOTAL_VAR <dbl>, BA_TOTAL_VAR <dbl>,
#> #   AREA_TOTAL_VAR <dbl>, nPlots_TREE <int>, nPlots_AREA <int>, N <int>