Open Bounties

We are looking for someone who can combine technical rigor with practical field judgment. If you have done instrumentation checks in messy real-world conditions, this is a strong fit. Scope: - Co-locate low-cost sensors with reference stations for two 24-hour windows. - Capture installation orientation, airflow obstructions, and maintenance state. - Report raw and corrected PM2.5 readings plus anomaly flags. Deliverables: - Time-indexed CSV with paired readings and calibration parameters. - Site condition report with photos and mounting notes. - Drift summary that calls out sensor-specific confidence and replacement recommendations. Acceptance criteria: - Paired time-series has no missing timestamps over active windows. - Calibration method and coefficients are explicitly documented. - Anomaly flags include rationale suitable for external audit. Friendly note: we care about transparent documentation, including "what did not go perfectly." Honest logs are a plus, not a penalty.

Context: Community PM2.5 networks are highly valuable, but drift and installation variance can quietly degrade trust. This bounty funds a grounded co-location audit against reference stations.

If this sounds like your kind of fieldwork, we would love your help. The work is structured, but we want observations from people who notice details and document uncertainty honestly. Scope: - Run synchronized 90-minute observation windows at 3 habitat sites over 4 nights. - Log lux and spectral class for each observation block. - Record species-level observations where possible; otherwise use agreed genus-level taxonomy. Deliverables: - Observation table including site, spectral condition, timestamps, taxa labels, and confidence tags. - Short narrative on weather, disturbances, and observer uncertainty. - Calibration check notes for all handheld light meters. Acceptance criteria: - Every observation row includes spectral condition and confidence tag. - Site-level coverage complete across all required windows. - Data package can be merged directly into model training set without schema edits. Friendly note: precision matters more than speed here. If you are methodical, you are exactly who we want.

Context: Our current model predicts nocturnal pollinator response to artificial light spectra; we now need consistent observational data to validate those predictions in real habitats.

Thanks for looking at this project. We are looking for a field operator who can run a careful, low-drama sampling day and leave us with audit-ready evidence. Scope: - Collect paired surface-water and sediment samples at 6 pre-specified transect points. - Capture timestamped site photos, weather context, and recent discharge notes. - Run standardized field blanks and duplicate one site for QA replication. Deliverables: - Structured CSV with sample identifiers, coordinates, collection timestamps, and handling notes. - Photo bundle tied to sample IDs. - 1-2 page methods memo documenting deviations, contamination controls, and confidence limits. Acceptance criteria: - Complete metadata for every sample. - No broken sample-ID links between photos and rows. - QA fields present and interpretable for independent re-analysis. Friendly note: if you can only cover part of the corridor, still apply and specify your travel radius in your application.

Context: We have strong satellite-assisted estimates, but we need physical transect samples to verify concentration bands before publishing the next model update.