Timeline
Recent updates
Plain-language summary of what's shipped recently. The full technical changelog with parameter-level detail lives in methodology section 8.
v1.8 — Three new LMIC climate archetypes (May 2026)
The archetype library doubles from two to four representative cities. Three new MERRA-2-derived archetypes ship alongside Abuja, all for calendar year 2022:
- Accra, Ghana — coastal-tropical Gulf of Guinea. Persistent South-Westerly marine inflow dominates year-round (36.6% SW), with the Harmattan more diluted at the coast than inland.
- Dhaka, Bangladesh — Ganges-Brahmaputra delta. Strongly bimodal Asian Monsoon (SW Jun–Sep / NE Nov–Feb) gives a softer S annual peak (18%); strongly neutral stability with a full diurnal range from clear-sky inland nights.
- Delhi, India — continental Indo-Gangetic Plain. NW post-monsoon / Western Disturbance flow dominates (19.8% NW), with the strongest diurnal stability range of the four archetypes (clear-sky days and nights inland).
The four cities span distinct dispersion regimes — users picking between them can screen sensitivity-to-met explicitly. The fetcher was hardened with an OS-level timeout wrapper after observing NASA OPeNDAP TCP hangs that no in-process timeout mechanism could interrupt; see methodology section 2.3 for full provenance + the pipeline source.
v1.7 — IQ dose-response simplified (May 2026)
The IQ map's ten-option dose-response dropdown has been retired in favour of a single hard-coded specification: Crump et al. (2013), concurrent BPb, log form. This is the published preferred form for the full observed BPb range, satisfies the ΔIQ = 0 boundary condition at BPb = 0, and is the only specification with a defensible case across the receptor BPb levels this tool screens. The other nine alternative specifications (peak BPb, lifetime-average BPb, linear forms, etc.) remain in the backend's coefficient table and unit tests for regression armour and internal sensitivity work, but are no longer exposed in the UI — the choice was methodologically complete but practically confusing for non-specialist users. Background-BLL slider and the 95% CI band toggle are unchanged. See methodology section 5.5.
v1.6 — IQ map refined for background BLL (May 2026)
The IQ pipeline now computes facility-attributable IQ loss as the marginal effect on Crump's dose-response on top of a user-supplied background BLL (default 3.0 µg/dL, the IHME GBD 2023 LMIC median). The previous implementation treated the facility's ΔBLL as if it were the child's total BPb — which overstated facility-attributable IQ loss by roughly 2× at typical backgrounds because Crump's log dose-response is concave. A background-BLL slider on the BLL Impacts tab now lets users tune to their region. CI band toggle relabelled "Lower / Upper 95% CI" — the band reflects statistical uncertainty in Crump's coefficient only, not the much wider total predictive uncertainty. Cursor readouts now show 1-decimal precision with an inline range. See methodology section 5.5.
v1.5 — IQ impact heat map (May 2026)
The BLL Impacts tab gains a new ΔIQ map layer showing predicted IQ-point loss per receptor for children chronically exposed at that location. The dose-response is from Crump et al. (2013) — a peer-reviewed reanalysis of the Lanphear pooled cohort data — with ten model variants selectable from a dropdown and a confidence-interval band sub-toggle for sensitivity assessment. Switching model is instant (no AERMOD re-run). See methodology section 5.5 for the equation, the rationale for ln(BPb + 1), and the four caveats about applying the dose-response to facility-attributable ΔBLL in LMIC contexts.
v1.4 — First real climate archetype: Abuja 2022 (May 2026)
The archetype menu now offers a real, MERRA-2-derived wind climatology for Abuja, Nigeria (calendar year 2022) — replacing the synthetic test fixture as the recommended option for West African ULAB scenarios. The full data ingest pipeline (NASA MERRA-2 OPeNDAP fetch → Pasquill stability classification → STAR-format wind rose) is documented in methodology section 2.3 and reproducible from the codebase. Accra, Dhaka, and New Delhi to follow.
v1.3 — Fugitive emissions (May 2026)
The tool now models fugitive emissions — uncaptured Pb releases from yard activities, charging, refining kettles, manual battery breaking, and slag handling — alongside the existing stack source. A user-editable fugitive fraction splits the total emission inventory between stack and fugitive components, with three scenario defaults (best-practice 10%, typical informal LMIC 35%, uncontrolled 60%). Fugitives are modelled as a ground-level area source centred on the stack with a coarser, lower-density particle size distribution reflecting mechanically generated dust. A master toggle on the Fugitive tab lets users keep the original stack-only behaviour for regulatory comparison runs. See methodology section 3.5 for the full description.
v1.2 — Blood lead level (BLL) impact heat maps (May 2026)
New BLL impact layer on the map showing modelled blood-lead-level increase from inhalation, soil/dust, and combined pathways. The pathway slope factors were rewritten around IEUBK-anchored piecewise interpolation; user-editable exposure duration and soil mixing depth let users explore sensitivity. See methodology section 5.
v1.1 — Default scenarios recalibrated (May 2026)
Quick-scenario presets and emission factors recalibrated against Manhart & Adjei (2026), reflecting the African-LMIC ULAB recycling spectrum: EU BAT, high-standard formal, substandard, smaller substandard, and informal artisanal. The uncontrolled emission factor was raised from 2 to 12 kg/t Pb based on lead-recovery data. See methodology section 8.
ULAB Plume Development Roadmap
This page describes Pure Earth's planned development trajectory for ULAB Plume — the milestones we're working toward and the research that will support tool validation.
Detailed timeline content is being prepared and will be published here shortly. Key planned milestones include:
- Field measurement campaigns at LMIC ULAB sites (West Africa, South Asia)
- Atmospheric measurement validation studies
- Population health surveillance linking ambient lead measurements to community blood lead surveys
- Methodology peer review through academic collaborations
- First case study field validation: target 2028
For collaboration on these research streams, contact chrisk@pureearth.org.