Frequently Asked Questions

AtmoFacts' FluxMapper is a software that turns standard eddy-covariance observations into spatially explicit flux maps with quantified uncertainty.

FluxMapper generates a fixed 301 × 301 pixel grid (≈ 90,000 pixels). Each pixel represents a spatially explicit flux time series derived from the station data.

FluxMapper turns a single flux station into a force multiplier, revealing spatial patterns and management-scale signals that are invisible in station-only time series. It enables: drawing areas of interest (fields, treatments, management units), extracting independent flux time series per area, comparing multiple areas simultaneously, and operates with no assumption of land-cover homogeneity.

FluxMapper uses the wind-directional information already embedded in high-frequency EC data to reconstruct where measured fluxes originate on the ground, turning one flux station into a spatially resolved flux map. Every 30-minute flux average contains tens of thousands of high-frequency observations sampled under changing wind conditions. Traditional EC processing collapses this information in time. FluxMapper instead exploits wind variability to attribute fluxes spatially, solving a highly over-determined inverse problem that converts temporal richness into spatial insight.

Pixel size scales with the effective measurement height (map extent is a 301 × 301 grid). For example, a ~3 m station produces ~3 m pixel resolution at 901 m × 903 m spatial extent. Taller stations cover exponentially larger areas with proportionally coarser pixels.

Daily outputs are delivered as GeoTIFF layers (48 half-hourly layers per day). Each pixel contains a flux time series with an associated uncertainty layer.

FluxMapper standard fluxes include H (sensible heat), LE (latent energy), and CO₂/NEE over ecosystem and terrestrial environments. Additional advanced configurations (e.g., urban, methane, aquatic landscapes, etc.) are available via AtmoFacts-direct projects and require additional consideration.

Yes, please reach out to AtmoFacts directly for CH₄ capabilities. CH₄ is not part of the initial standard package but is available through AtmoFacts-direct projects with additional considerations.

No additional hardware is necessary. All you need is access to raw high-frequency eddy-covariance data (≥10 Hz) for processing. Your data does not have to originate from LI-COR instruments, SmartFlux®, or EddyPro®—any ≥10 Hz eddy-covariance dataset can be ingested if it conforms to the required format. Historical ≥10 Hz data can also be processed if it meets formatting requirements.

Yes—historical ≥10 Hz data can be processed if it meets formatting requirements. Some historical workflows may require manual upload during early phases.

Yes. FluxMapper applies rigorous QA/QC consistent with established eddy-covariance standards, building on the open-source eddy4R ecosystem used by large national observatories like the US National Ecological Observatory Network. FluxMapper does not bypass EC QA/QC; it extends it into the spatial domain.

Yes. FluxMapper has been developed, tested, and applied over more than a decade across landscapes with strong spatial heterogeneity and contrasting flux patterns, as well as in controlled supercomputer simulations. Validation has focused on situations where traditional station-based analyses struggle most—heterogeneous surfaces, sharp transitions, and mixed source areas. References include peer-reviewed examples such as Table 1 in Xu et al. (2017) and Figure 2 in Xu et al. (2020), with additional publications available in the AtmoFacts resource library.

Yes—every pixel has an associated uncertainty (each FluxMap has a corresponding uncertainty map). Nominal uncertainty is approximately ~5% systematic and ~10% random per pixel-day. Random uncertainty decreases with sample size following the square-root law. In practice, random uncertainty can be reduced through temporal aggregation, spatial aggregation (e.g., geofenced areas of interest), or both.

Yes, this is a major breakthrough. The minimum resolvable plot is the pixel resolution = effective measurement height. For example, a 50m rice paddy is now feasible with proper tower height. This eliminates the traditional 200m radial distance requirement that previously restricted small-plot research. Each pixel has a quality layer reflecting the degree of bleed-over from adjacent areas.

AtmoFacts and LI-COR have partnered to deliver FluxMapper through the LI-COR Cloud platform as a foundational module. This integration allows seamless workflow with existing LI-COR instruments. Visit https://tinyurl.com/fm-cohort to complete the founding cohort interest form for early access.