The emburden ecosystem

Thirteen R packages, all open-source under MIT, all hosted at github.com/ericscheier.

JOSS-readiness overview

Package LOC Tests Vignettes JOSS score Status
emburden 4,340 218 3 100 READY
emburdenweather 7,330 242 3 90 READY
emburdengeo 1,077 24 1 80 READY
emburdenhealth 4,521 49 4 80 READY
emburdendata 24,122 269 2 75 READY
emburdensynth 19,148 137 2 75 READY
emburdenstats 1,943 9 0 50 needs vignette
emburdenutil 1,681 3 0 45 needs vignette
emburdenvis 3,732 3 0 45 needs vignette
emburdenpub 774 3 0 45 needs vignette
emburdentest 483 3 0 25 dev-tool
emburdenplus 57 — — 10 meta-imports
emburdener — — — 0 pre-DESCRIPTION

Install everything

# One at a time
remotes::install_github("ericscheier/emburden")
remotes::install_github("ericscheier/emburdendata")
remotes::install_github("ericscheier/emburdensynth")

# Or the meta-package (installs all 13):
remotes::install_github("ericscheier/emburdenplus")

The pipeline in 5 lines

library(emburdensynth)
run_global_pipeline("USA", years = 2023)   # → output/global/USA_synthetic.parquet
attach_physics_baseline("USA")             # → adds physics_required_kwh
calculate_energy_insecurity_gap("USA")     # → adds eig_kwh + energy_limiting_flag
country_unified_metrics("USA")             # → EJ unification via retail prices

Design principles

Open data first

Every downloader prefers open, direct-download APIs over commercial data. Fallback chains ensure graceful degradation when a source is temporarily unavailable.

Reproducible caching

All downloads write to ~/.cache/emburdendata/ via rappdirs. Pipelines are fully reproducible if internet is available, fully offline if the cache is populated.

Net Energy Return math

The emburden::neb_func function embeds the canonical aggregation math correcting a documented bias in naive weighted-mean- of-ratios.

Colorblind-safe visualization

Every palette is Color-Universal-Design compliant. See emburdenpub::get_burden_palette and emburdenvis::theme_emburden_emrgi.

Tidyverse + snake_case

All column names snake_case. All data-frame handling via dplyr/tidyr with the base-R pipe (|>).

Testing scaffolding

emburdentest provides synthetic-data generators + test utilities that let any downstream package be unit-tested against reproducible fixtures.

Contribute

  • Bug reports + feature requests: GitHub Issues on the relevant package repository
  • Pull requests: welcome; please open an Issue first to discuss
  • Ecosystem-wide questions: info@emburden.org