Skip to contents

Aggregate hex-level indicator data to administrative shapes

aggregate_hex_to_shapes.Rd

Takes the tibble from [fetch_hex_data()] and aggregates each indicator column to every admin level present in `shp_dta`, using the per-variable weight × function strategy supplied by the deployer. Aggregation is always **hex → admin level directly** (never chained through intermediate levels).

Usage

aggregate_hex_to_shapes(hex_data, hex_layer, shp_dta, strategy)

Arguments

hex_data

Tibble returned by [fetch_hex_data()]: must contain a `hex_id` column, a `population` column (unless no indicator requests `weight = "pop"`), and one column per indicator.

hex_layer

The `admin9_Hexagon` sf layer (or plain tibble) from `shapes.rds`, as built by [make_hex_grid()] and enriched by [make_admin_lookup()]. Must contain `admin<N>Pcod` (own level) and parent Pcod columns for every admin level in `shp_dta`.

shp_dta

Named list of sf tibbles (or plain tibbles), one per admin level including the hex level. Names must follow the `admin<N>_<HumanName>` convention.

strategy

Named list specifying the aggregation weight and function for each indicator. **Must** contain a `.default` entry. Each entry is a named character vector with elements: - `weight`: `"pop"` (population-weighted), `"area"` (area-weighted), or `"none"` (unweighted). - `fun`: `"mean"`, `"sum"`, `"median"`, `"min"`, or `"max"`. Ignored when `weight` is `"pop"` or `"area"` (always uses weighted mean). Temporal variable columns (e.g. `nightlights_2020`) are matched by stripping the `_<year>` suffix to get the stem, then looking up the stem in `strategy` before falling back to `.default`.

Value

Named list of tibbles, one per slot in `shp_dta`. Each tibble contains `admin<N>Pcod`, `admin<N>Name`, parent Pcod columns, `population` (aggregated), and one column per indicator.

Details

Population always aggregates as an unweighted sum regardless of any `population` entry in `strategy` (a warning is emitted and the entry is ignored). The population column is preserved in every output tibble for downstream use by [build_hex_metadata()].

See also

Other data-input: build_hex_metadata(), fct_template_reader(), fetch_hex_data(), get_available_years(), get_shape(), list_hex_vars(), make_admin_lookup(), make_hex_grid(), pti_patch_admin_sheet(), use_hex_vars()

Examples

if (FALSE) { # \dontrun{
my_shp    <- readRDS("app-data/shapes.rds")
hex_layer <- my_shp$admin9_Hexagon
hex_ids   <- hex_layer$admin9Pcod
vars      <- use_hex_vars("flood_exposure_15cm_1in100")
hex_data  <- fetch_hex_data(hex_ids, vars)
aggregated <- aggregate_hex_to_shapes(
  hex_data  = hex_data,
  hex_layer = hex_layer,
  shp_dta   = my_shp,
  strategy  = list(
    .default                   = c(weight = "pop",  fun = "mean"),
    flood_exposure_15cm_1in100 = c(weight = "none", fun = "sum")
  )
)
names(aggregated)
} # }