Independent environmental-intelligence research software

From open environmental data to reproducible scientific evidence.

HeatSafe integrates deterministic analysis, geospatial workflows, compound-hazard research, forecasting benchmarks, uncertainty quantification and optional AI explanation layers.

✓ No paid AI required
✓ Reproducible baselines
✓ Optional local and BYOK AI
Open DataProvenance ContractScientific CoreHeatAQ NexusForecasts + uncertaintyCompound RiskSensitivity + evidenceOpen ToolsResearch Artifacts
FOR EVERYDAY PUBLIC USE

Looking for practical, non-commercial information?

Open a simple guide to free household tools, plain-language learning pages and important safety limits—no account, purchase or paid AI required.

Open the Everyday Public Guide
SCIENTIFIC PROGRAM

Research tracks

Every track is designed around explicit data identity, baselines, uncertainty, validation and reproducibility.

HeatAQ Nexus

Chronological and spatial forecasting benchmarks for temperature, PM2.5 and compound events.

Open benchmark →

Compound hazards

Transparent multi-hazard formulations with exposed assumptions and sensitivity analysis.

Research questions →

Climate extremes

Trends, anomalies, heatwaves, hot nights and cooling-degree metrics.

Climate analysis →

Air quality and smoke

Concentration-first summaries, explicit AQI standards and cautious wildfire context.

PM2.5 workflows →

Reproducibility

Experiment manifests, checksums, model cards, data cards and versioned artifacts.

Methods →

Reproducible experiments

Run one immutable specification to produce complete metrics, uncertainty figures, paper-ready reports, checksums and a self-contained candidate release.

Open experiment orchestrator →
ACCESSIBLE BY DESIGN

Three capability tiers

Tier 0

Standard

Deterministic core, local files, synthetic data, CPU baselines and browser tools. No paid AI API.

Tier 1

Local AI

Optional local language models explain existing outputs without changing decisions or measurements.

Tier 2

BYOK

Optional provider-neutral cloud models and credentialed data connectors using user-managed secrets.

OPEN DEMONSTRATIONS

Browser tools

Public tools translate the scientific core into inspectable decisions and local-first workflows.

🪟

Ventilation Decision

Balance heat removal with PM2.5, smoke, humidity, wind and airflow.

Open tool →
🏠

Home Heat Profile

Identify likely heat pathways without producing a certified score.

Open tool →

Cooling Cost

Expose energy, price and duty-cycle assumptions.

Open tool →
🌡️

Indoor–Outdoor Comparator

Inspect thermal lag and observational response without causal overclaiming.

Open tool →

Built to be cited, inspected and extended.

The repository includes typed APIs, CLI workflows, Docker, CI, CodeQL, tests, provenance contracts, citation metadata and a research roadmap.

PUBLISHED DATASET

First DOI-backed official-source PM2.5 benchmark

Explore the San Diego County station-level forecasting dataset, its verified research archive, citation, temporal coverage and scientific limitations.

Faramarz Kowsari
Author & Developer

Faramarz Kowsari

Author, Software Engineer and AI researcher based in Istanbul, developing open educational and research tools at the intersection of Artificial Intelligence, environmental intelligence and public learning.

About the author & developer →