FluxGapFill 1.0 · local-first scientific software

From processed fluxes to defensible, publication-ready results.

Import already-computed half-hourly/hourly flux or environmental data, verify the mapping, apply transparent QC, benchmark gap-filling methods, and continue only into the energy, water, carbon or footprint modules supported by your dataset.

EddyProCampbell TOA5AmeriFlux / FLUXNETCSV / TXT / TSVExcelManual mappingFiles stay local

?

Start hereUpload a processed flux/environmental file. A supplemental weather file is optional. Raw 10–20 Hz sonic/IRGA data are outside this tool.

1. Project data

Primary processed flux/environmental file plus optional supplemental weather data.

Primary processed data

EddyPro TXT, TOA5, CSV, TSV, TXT, XLSX/XLS/ODS

+

Supplemental weather / environmental data (optional)

CIMIS or another mapped half-hourly/hourly station file

+

No alternate-format files to test?Use the bundled synthetic examples:

Campbell TOA5 FLUXNET-style Generic CSV Excel Supplemental weather

2. Project policy

Time handling is explicit because logger conventions differ.

Project name

Site name

Timezone

Time basis in source file

Maximum numeric QC accepted

Scope

Privacy behavior. Selected files are copied only into the temporary Pyodide filesystem/memory for this browser session. Refreshing or closing the page removes the working copies; original files on your computer are unchanged.

Variable mapping

Verify before processing

FluxGapFill proposes mappings and units but does not silently assume unknown fields. Correct any suggestion here before loading the project.

?

Confirm what FluxGapFill detectedCheck timestamp convention, variable roles and units before processing. Correct uncertain mappings instead of accepting a guess.

Choose files and run inspection first.

Primary file

Not inspected

—

Primary timestamp convention

Role	Source column	Unit	Detection
Inspect a primary file first.

Supplemental file

Optional

—

Supplemental timestamp convention

Role	Source column	Unit	Detection
No supplemental file selected.

Input preview

First records only. Preview does not alter the file.

Inspect a file to preview it.

Data overview

Quality, completeness and gap structure

Standardized midpoint grid after mapping, unit normalization, QC screening and optional supplemental alignment.

?

Check whether the record is analysis-readyReview coverage, time resolution, missingness and gap structure. Calculations use the complete standardized record, not the downsampled chart.

Records—

Time step—

LE missing—

H missing—

LE longest gap—

H longest gap—

Period—

External source—

Measured fluxes

Downsampled only for visualization; calculations use every record.

Project details

Project—

Primary source—

Period—

Records—

Resolution—

Supplemental source—

Predictor availability

Variables visible to current/future analysis modules.

Variable	Complete	N

Quality control

Screen transparently before model training

Import-stage QC remains intact. This additional layer can reject implausible ranges and isolated robust spikes without altering the standardized source copy.

?

Screen conservativelyStart with the documented QC flags and physical limits. The extra robust spike screen is optional and should be enabled only when justified.

QC policy

Conservative defaults; every rejection receives an explicit reason flag.

Enable optional robust Hampel-style spike screening for LE, H and NEE

Rolling window

Spike threshold

LE plausible range (W m⁻²)

H plausible range (W m⁻²)

NEE plausible range (µmol m⁻² s⁻¹)

Principle

Reversible within the session. Re-running QC always starts from the standardized imported dataset, not from previously screened values.

QC summary

Import-stage rejection and secondary screening remain distinguishable in the audit trail.

Range rejects—

Spike rejects—

Variables—

QC audit

No rejected value is silently removed; the full provenance export retains pre-QC values and flags.

Variable	Before	Range reject	Spike reject	After	Missing after
Load a project and apply quality control.

Capability-aware workspace

Use what is ready. Add data where needed. Skip the rest.

Each analysis module checks required variables independently, so missing footprint or carbon information never blocks gap filling or ET work.

?

Run only what your data supportEach module declares its required inputs. Supply missing metadata/data when useful, or skip that module without blocking the rest of the workflow.

Optional site information

Needed by selected advanced modules; leave blank if not applicable.

Latitude

Longitude

Measurement height (m)

Canopy height (m)

How optional modules behave

Ready means the required data are already in the project. Needs data identifies exactly what is missing. Use “Add data” to return to the import workspace, or “Skip” to hide that module for this project. Skipping never deletes data.

High-frequency raw EC processingOut of scope

Processed flux / environmental dataSupported

Server-side storageNone

Process a dataset to evaluate module readiness.

Model lab

Compare established MDS with enhanced machine learning

Validation hides contiguous calendar outages and scores only observations that existed before masking. This is retrospective reconstruction validation—not forecasting.

?

Validate before fillingBalanced blocked validation is the recommended starting point. It hides contiguous calendar windows and evaluates reconstruction against observations that were actually measured.

Overall blocked-validation performance

RMSE, MAE, bias, R² and prediction coverage across the hidden calendar windows.

Flux	Gap	Model	N	RMSE	MAE	Bias	R²	Windows
Run validation to populate this table.

Day / night diagnostics

Day/night basis appears after validation.

Flux	Gap	Model	Segment	N	RMSE	Bias
Run validation first.

Selected model by duration

Site-specific adaptive rules from the validation results.

No site-specific validation has been run yet.

Seasonal diagnostics

DJF, MAM, JJA and SON are reported when a held-out window contains observations from that season.

Flux	Gap	Model	Season	N	RMSE	MAE	Bias	R²
Run validation first.

Results

Gap-filled fluxes with provenance

Measured observations remain unchanged; reconstructed values retain method labels, gap class, MDS donor diagnostics and empirical blocked-validation uncertainty when available.

?

Inspect provenance, not only the final lineMeasured values are preserved. Filled records retain method, gap duration, diagnostics and empirical interval information when validation supports it.

Final LE and H

Intervals: run blocked validation before final filling to calibrate empirical uncertainty.

Method provenance

Run gap filling to populate provenance.

Daily ET from final LE

Temperature-dependent latent heat of vaporization.

energy + water / ET

Energy closure and water-use analysis

Run only the modules supported by your data. Missing inputs are identified explicitly; optional methods can be skipped without affecting gap filling.

?

Optional energy and water analysisEnergy closure needs H, LE, Rn and G. ET can run from LE alone. Irrigation, ETo and SWC are optional context and can be added or skipped.

Energy-balance configuration

Raw/reconstructed H and LE are preserved. Corrections are separate sensitivity products.

Ground heat flux (G)

Positive raw G direction

Plate depth (cm)

Bulk density (kg m⁻³)

Dry-soil specific heat (J kg⁻¹ K⁻¹)

Radiation threshold (W m⁻²)

Soil heat storage. If you choose buried-plate correction, FluxGapFill requires soil temperature, SWC, plate depth and measured bulk density. It does not silently query or invent soil properties.

ET + advanced correction inputs

De Roo methods are enabled only when their scaling variables are available.

Daily ET completeness threshold (%)

ET temperature fallback (°C)

Effective measurement height (m)

PBL height sensitivity value (m)

Interpretation. Energy-balance corrections are sensitivity scenarios, not proof of the physical cause of non-closure. De Roo methods require u*, Monin–Obukhov length, measurement height and w* or the variables needed to derive it.

Optional irrigation / applied-water file

Use this when irrigation is stored separately from the primary or weather file. CSV/TXT/Excel logger-style files are accepted.

Optional

Water-input file

Timestamp convention

No separate irrigation file selected. Skip this step if irrigation/rainfall is already in the project or not needed. Download a daily-irrigation example.

Measured EBR—

OLS slope—

Closure R²—

Primary intervals—

Total accepted ET—

Mean ET/ETo—

Energy closure

Available energy (Rn − G) versus turbulent energy (H + LE).

Monthly energy-balance ratio

Correction-method diagnostics

Mauder 2013, Charuchittipan 2014, De Roo 2018 direct and De Roo 2018 EBR-rescaled.

Method	Status	Applied	Post-correction EBR	Notes / missing inputs
Run energy & water analysis.

Daily ET and ETo

Days below the selected completeness threshold are flagged rather than silently treated as complete.

Cumulative water

ET, ETo and rain + irrigation when available.

Water inputs and soil-water response

Precipitation and irrigation are shown when supplied; SWC is plotted as context, not as a complete soil-water-balance model.

Energy & water warnings and provenance

Run energy & water analysis to populate method availability, assumptions and warnings.

Energy & water exports

Generated locally in your browser.

advanced flux post-processing

u* threshold · carbon partitioning · footprint

Each module checks its own requirements. Missing inputs cause a clear “Needs data” result rather than a failed workflow.

?

Advanced optional modulesu* and carbon require NEE-related variables; footprint analysis additionally requires turbulence/site metadata. Missing inputs are reported explicitly.

u* threshold analysis

Seasonal MPT-style plateau detection plus independent CPD-style breakpoint detection with bootstrap uncertainty.

Selected method

Bootstrap iterations

Temperature classes

u* classes

Night radiation threshold (W m⁻²)

NEE sign convention

Carbon partitioning

Nighttime Lloyd–Taylor respiration modeling with time-varying Rref; GPP is reported positive for uptake.

Respiration window (days)

Window step (days)

Manual u* threshold override (m s⁻¹)

Requires NEE and temperature. u* and radiation are strongly recommended and used automatically when available.

Footprint climatology

Kljun et al. (2015) FFP parameterisation.

Measurement height (m)

Canopy height (m)

Displacement height (m)

PBL height fallback (m)

Domain half-width (m)

Grid points

Requires u*, wind direction, Monin–Obukhov length, σv, wind speed, measurement height, and PBL height.

AOI / land-cover overlay optional

GeoJSON polygons can represent fields, crops, roads, water, or exclusion zones.

GeoJSON AOI file

No AOI file loaded. Footprint climatology can run without AOIs.

Selected u*—

Night records—

Respiration E0—

Carbon model R²—

Valid footprints—

Median peak fetch—

u* response

Daily carbon components

Footprint climatology

AOI contribution

AOI	Exclude	Contribution
No AOI results.

Advanced module status

Run carbon & footprint analysis to populate module status and diagnostics.

Carbon & footprint exports

Export & figure studio

Export the audit trail and journal-ready figures.

All exports are created locally. Figure Studio produces print-oriented white-background figures independent of the on-screen theme.

?

Export data and journal figuresDownload the scientific audit trail, validation reports and publication figures. Journal PNG export uses a white print background and 600-DPI metadata; SVG is vector.

Gap-filled products

QC & validation diagnostics

Journal Figure Studio

Export any rendered chart with publication-oriented typography, margins and resolution.

600 DPI / SVG

FigureOnly figures already rendered in this browser session are listed.

Journal width

Format

Legend

Include in-figure title Keep light gridlines

Journal export style: white background, regular-weight sans-serif type, restrained line weights, automatic margins, no interface colors/backgrounds, and no forced bold formatting. PNG files are written at the selected physical width with 600-DPI metadata.

From processed fluxes to defensible, publication-ready results.

1. Project data

2. Project policy

Verify before processing

Primary file

Supplemental file

Input preview

Quality, completeness and gap structure

Measured fluxes

Project details

Predictor availability

Screen transparently before model training

QC policy

QC summary

QC audit

Use what is ready. Add data where needed. Skip the rest.

Optional site information

How optional modules behave

Compare established MDS with enhanced machine learning

Blocked validation design

Adaptive final fill

Overall blocked-validation performance

Day / night diagnostics

Selected model by duration

Seasonal diagnostics

Gap-filled fluxes with provenance

Final LE and H

Method provenance

Daily ET from final LE

Energy closure and water-use analysis

Energy-balance configuration

ET + advanced correction inputs

Optional irrigation / applied-water file

Energy closure

Monthly energy-balance ratio

Correction-method diagnostics

Daily ET and ETo

Cumulative water

Water inputs and soil-water response

Energy & water warnings and provenance

Energy & water exports

u* threshold · carbon partitioning · footprint

u* threshold analysis

Carbon partitioning

Footprint climatology

AOI / land-cover overlay optional

u* response

Daily carbon components

Footprint climatology

AOI contribution

Advanced module status

Carbon & footprint exports

Export the audit trail and journal-ready figures.

Gap-filled products

QC & validation diagnostics

Journal Figure Studio

Preparing scientific runtime