The candidate rules for one species and stage, around link's own thresholds

scripts/0297 tuned CH rearing over a grid of three components; this builds that grid for any species and stage from link's default thresholds. A rule is a stream-branch test with three parts: a size test (a MAD floor in m³/s with no maximum, or a channel-width floor in m), a gradient maximum, and an exception admitting a segment that fails the size test when a river of at least exc_q m³/s lies within exc_d m downstream. default is the member with link's own width floor and gradient maximum and no exception; departures counts how many parts differ from it (a size test that differs in variable or value: 1; a gradient that differs: 1; an exception: 1). departures_flow counts the same from link's flow version of default (its MAD floor, its gradient maximum, no exception), the starting point #80 asks the pick to be read from as well; NA where default has no MAD floor.

Usage

knw_tune_grid(
  thresholds,
  stage = c("rear", "spawn"),
  grid_mad = c(0.1, 0.15, 0.2, 0.28, 0.35, 0.5, 0.75, 1),
  grid_width = c(1, 1.5, 2, 3, 4),
  grid_grad = c(0.03, 0.04, 0.0549, 0.07, 0.09, 0.11, 0.13, 0.15, 0.2, 0.25, 0.3),
  grid_q = c(1, 3, 5, 10, 20, 30, 50),
  grid_d = c(100, 250, 500, 1000, 2000, 5000),
  extra = NULL
)

Arguments

thresholds: one row of link's parameters_habitat_thresholds.csv for the species (a data.frame with species_code, <stage>_gradient_max, <stage>_channel_width_min, <stage>_mad_min).
stage: "rear" or "spawn".
grid_mad, grid_width, grid_grad, grid_q, grid_d: numeric candidate values; the species' own default width, MAD and gradient are added when absent.
extra: NULL, or one row of another bundle's thresholds for the same species, in the same columns.

Value

A tibble with rule, size_var, size_min, grad_max, exc_q, exc_d, departures, departures_flow, is_default, is_link_tuned; zero rows when the species has no stream rule for the stage.

Details

The size and gradient grids are the species' default values and the fixed steps around them, so every species is tuned against the same ladder; the gradient ladder runs to 30 % since #68 (#67 found CH rearing's best rule at the top of the 9 % ladder). A species whose default has no stream rule for the stage (KO and SK rear in lakes only) gets an empty grid, and the caller says so.

extra adds one more candidate: the rule of another bundle's thresholds row (link's default_tuned, whose MAD floors were set by a use rate on wet's flows, link#331), its MAD floor and gradient maximum joining the ladders and the rule flagged is_link_tuned, so the table shows how link's own flow numbers score beside the grid. A row with no MAD floor for the stage adds nothing.

Examples

thr <- data.frame(species_code = "CH", rear_gradient_max = 0.0549, rear_channel_width_min = 1.5,
                  rear_mad_min = 0.28)
g <- knw_tune_grid(thr, "rear")
nrow(g)
#> [1] 6149
g[g$is_default, ]
#> # A tibble: 1 × 10
#>   rule         size_var size_min grad_max exc_q exc_d departures departures_flow
#>   <chr>        <chr>       <dbl>    <dbl> <dbl> <dbl>      <int>           <int>
#> 1 width>=1.5|… width         1.5   0.0549    NA    NA          0               1
#> # ℹ 2 more variables: is_default <lgl>, is_link_tuned <lgl>
g[g$departures_flow == 0, ]
#> # A tibble: 1 × 10
#>   rule         size_var size_min grad_max exc_q exc_d departures departures_flow
#>   <chr>        <chr>       <dbl>    <dbl> <dbl> <dbl>      <int>           <int>
#> 1 mad>=0.28|g… mad          0.28   0.0549    NA    NA          1               0
#> # ℹ 2 more variables: is_default <lgl>, is_link_tuned <lgl>