Research Protocol

Source markdown: brunei-indoor-air-quality-during-cooking.md

`r`n

Last updated:

Quick Answer

This page provides a structured research protocol and dataset schema for Brunei-relevant implementation.

Indoor Air Quality During Cooking in Brunei Homes

Executive Summary

This package defines a publishable research protocol for measuring cooking-related indoor air quality (IAQ) in Brunei homes. It is built for replication, transparent assumptions, and decision-useful outcomes.

Primary goal: quantify short-duration pollution spikes (PM2.5 and NO2) across kitchen configurations and ventilation behaviors, then convert findings into practical design guidance.

Claim Labeling Rules

  • `Measured`: Generated from instrument logs in this study.
  • `Cited`: Taken from external primary sources.
  • `Inference`: Derived from measured and cited evidence.

No claim in this document should be treated as measured local evidence until dataset rows are collected and validated.

Research Question

How much do PM2.5 and NO2 concentrations rise during typical cooking events in Brunei homes, and which ventilation controls produce the largest reduction in peak and one-hour exposure?

Decision Outputs

  • Minimum viable ventilation strategy for common kitchen layouts.
  • Expected reduction ranges for interventions (hood use, window strategy, door state).
  • Practical trigger thresholds for homeowner action.

Study Scope

  • Geography: Brunei-Muara, Belait, Tutong, and Temburong.
  • Home types: detached, semi-detached, terrace, apartment.
  • Kitchen types: wet kitchen, dry kitchen, hybrid/open-plan.
  • Fuels: gas, induction, electric coil (if present).
  • Pollutants: PM2.5 (ug/m3), NO2 (ppb).
  • Environmental covariates: temperature, relative humidity, occupancy, window/door state.

Methodology

Design

Prospective observational field study with repeated sessions per home.

Sample

  • Target: 30 homes minimum.
  • Sessions: 3 sessions per home (breakfast, lunch, dinner windows where possible).
  • Total events: 90+ cooking sessions.

Session Protocol

  1. Place monitors at breathing-zone height in kitchen and adjacent living area.
  2. Record 15-minute pre-cooking baseline.
  3. Record full cooking period with event markers:
  • Burner start
  • Hood ON/OFF
  • Window OPEN/CLOSED
  • Door OPEN/CLOSED
  1. Record 60-minute decay period after cooking stops.
  2. Export timestamped logs and event notes.

Primary Endpoints

  • `pm25_peak_ug_m3`: maximum PM2.5 during session.
  • `pm25_1h_mean_ug_m3`: one-hour mean from cook start.
  • `no2_peak_ppb`: maximum NO2 during session.
  • `time_to_pm25_baseline_min`: minutes to return within 10 percent of pre-cook baseline.

Secondary Endpoints

  • Adjacent-area transfer ratio.
  • Effect size by intervention state (hood/window/door).
  • Fuel-specific risk profile.

Data Sources

Primary measurement sources (to be generated)

  • Study logger exports and structured event logs.

Cited standards and guidance

  • WHO Air Quality Guidelines: https://www.who.int/publications/i/item/9789240034228
  • WHO Household Air Pollution: https://www.who.int/health-topics/air-pollution#tab=tab_3
  • U.S. EPA indoor air quality resources: https://www.epa.gov/indoor-air-quality-iaq
  • U.S. EPA PM basics: https://www.epa.gov/pm-pollution/particulate-matter-pm-basics
  • ASHRAE indoor environmental quality resources: https://www.ashrae.org/technical-resources/indoor-air-quality
  • CDC indoor environmental quality overview: https://www.cdc.gov/niosh/topics/indoorenv/
  • Brunei Darussalam Meteorological Department: http://www.met.gov.bn/
  • AMBD official publications index: https://www.ambd.gov.bn/publications/

Assumptions

  • Sensor placement is representative of occupant exposure in each zone.
  • Participant behavior in logged sessions reflects typical cooking behavior.
  • Weather and occupancy variability are partially controlled through repeated sessions.

Limitations

  • Observational design cannot isolate all confounders with causal certainty.
  • Low-cost sensors may drift; calibration checks are required.
  • Small sample subsets (for rare kitchen types) may widen uncertainty bands.

Independent Validation Status

Current status: protocol complete, field dataset pending.

Validation gate for publication-quality conclusions:

  1. At least 90 valid sessions.
  2. Calibration check logs attached.
  3. Outlier handling published before analysis.
  4. Reproducible codebook and data dictionary released.

Analysis Plan

  • Descriptive: median, IQR, 95th percentile for each endpoint by kitchen type.
  • Comparative: mixed-effects model with home-level random intercepts.
  • Intervention effects: within-home delta for hood/window/door state changes.
  • Robustness: sensitivity checks excluding sessions with instrument anomalies.

Quality Controls

  • Clock sync before each session.
  • Duplicate monitor spot checks on 10 percent of sessions.
  • Missing-data policy:
  • Keep rows with at least 80 percent valid minute-level readings.
  • Flag and document imputation where used.

What This Does Not Prove

  • It does not prove long-term disease risk from this dataset alone.
  • It does not prove national prevalence outside sampled households.
  • It does not prove one appliance brand is superior without controlled appliance trials.

Publication Outline

  1. Baseline exposure profile by kitchen type.
  2. Peak and one-hour concentration comparisons.
  3. Ventilation intervention effectiveness.
  4. Design guidance and homeowner action thresholds.
  5. Reproducibility appendix and dataset release notes.

Dataset Specification

Required fields are provided in:

`knowledge-base/research/data/brunei-iaq-cooking-study-template.csv`

JSON-LD Block

A production-ready JSON-LD graph is provided in:

`knowledge-base/research/data/brunei-iaq-cooking-study.jsonld`

Version

  • Version: 1.0.0
  • Last updated: 2026-03-04
  • Validation state: protocol-ready, no local measured rows published yet.

Changelog

  • 2026-03-04 (v1.0.0): Initial editorial-grade protocol package, dataset template, and JSON-LD graph.