Counts and Aggregation • riskworkflowr

Overview

Many spatial risk workflows ultimately rely on counts aggregated to analysis units.

Typical workflows involve:

point events
→ spatial assignment
→ counts by unit
→ join counts back to polygons
→ calculate risk metrics

riskworkflowr provides helper functions to support reproducible and consistent aggregation workflows.

The main aggregation functions include:

risk_count_units()
risk_join_counts()
risk_build_counts()

risk_count_units() summarises events by analysis unit.

data <- data.frame(
  unit_id = c("A", "A", "B", "C", "C")
)

risk_count_units(
  data = data,
  unit_id_col = "unit_id"
)

##   unit_id event_count
## 1       A           2
## 2       C           2
## 3       B           1

risk_join_counts() joins aggregated counts back to spatial analysis units.

The units input is expected to be an sf object. The example below is not evaluated in this vignette because it uses placeholder spatial data.

risk_join_counts(
  units = analysis_units,
  counts = unit_counts,
  unit_id_col = "unit_id"
)

into a single workflow.

risk_build_counts(
  points = event_points,
  units = analysis_units,
  unit_id_col = "unit_id"
)

Zero-event areas are important in spatial risk analysis.

riskworkflowr supports workflows where:

This is important for:

The aggregation workflow is designed to encourage:

A core package principle is:

"units are units"

The same workflow should operate consistently across:

Aggregation workflows assume:

Potential issues include:

Analysts should carefully review join logic and QA outputs before interpretation.

Alternative workflows may include:

The appropriate approach depends on the analytical context and decision requirements.

Future versions of riskworkflowr may include: