Data for article: New multisource data-driven Gap-filling approaches for meteorological data measured at eddy flux sites
Fuente:
Zenodo
Saved in:
| Main Author: | |
|---|---|
| Format: | Recurso digital |
| Language: | English |
| Published: |
Zenodo
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866901507654287360 |
|---|---|
| author | Anonymous |
| author_facet | Anonymous |
| contents | <p>These datasets accompany the article <em>“Multi-source Integrated Reconstruction of Missing Meteorological Data Enhances Flux Gap-Filling Accuracy across Sites and Timescales”</em>, which introduces the MIR (Multi-source Integrated Reconstruction) framework. MIR combines ERA5 reanalysis and in-situ correlated observations to improve the reconstruction of missing meteorological data, and has demonstrated high accuracy across 77 ChinaFLUX sites under various gap scenarios. The following three data archives provide the necessary inputs and outputs for the code package <code>01_MIR_Code.zip</code> at <a href="https://doi.org/10.5281/zenodo.16738283">https://doi.org/10.5281/zenodo.16738283</a>:</p> <ul> <li> <p><em><strong>02_Raw_Insitu.zip</strong></em><br>Contains raw half-hourly in-situ meteorological observations for each site-year. Each file is named in the format <code>YYYY_AAB_CXX.mat</code>, where <code>YYYY</code> is the year, <code>AAB</code> encodes the platform and general ecosystem type (e.g., A: cropland, F: forest), <code>C</code> denotes the footprint ecosystem, and <code>XX</code> is a two-digit site ID. Variables include air temperature, humidity, shortwave/longwave radiation, soil moisture, wind speed, and pressure, with missing values stored as NaNs.</p> </li> <li> <p><em><strong>03_ERA_Drivers.zip</strong></em><br>Includes hourly ERA5-Land reanalysis data extracted via Google Earth Engine, used as predictors in the MIR framework. Files are named <code>YYYY_AAB_CXX_NUPERA.csv</code>, corresponding to each site-year and containing variables such as 2m temperature, surface pressure, solar and thermal radiation, wind components, soil temperature, and total precipitation.</p> </li> <li> <p><em><strong>04_Gapfilled_Met.zip</strong></em><br>Provides MIR-reconstructed half-hourly meteorological datasets. Each file follows the naming format <code>filled_AAB_CXX_YYYY.mat</code>, and contains 17 fully gap-filled variables including temperature, relative humidity, net and component radiation, soil temperature and moisture, wind speed, pressure, and turbulent heat fluxes. The majority of original missing values have been filled using the best-performing model identified for each variable and gap duration.</p> </li> </ul> <p>All file and folder structures are aligned with the MIR codebase and support full pipeline reproducibility and future comparative studies in flux reconstruction.</p> <p><strong>Note: All author-identifying information has been removed from the dataset and code to comply with double-anonymous peer review; full metadata and authorship will be restored upon publication.</strong></p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_16738446 |
| institution | Zenodo |
| language | eng |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | Data for article: New multisource data-driven Gap-filling approaches for meteorological data measured at eddy flux sites Anonymous Eddy covariance Meteorological parameters Data-driven In-situ multi-layer observations ERA5 <p>These datasets accompany the article <em>“Multi-source Integrated Reconstruction of Missing Meteorological Data Enhances Flux Gap-Filling Accuracy across Sites and Timescales”</em>, which introduces the MIR (Multi-source Integrated Reconstruction) framework. MIR combines ERA5 reanalysis and in-situ correlated observations to improve the reconstruction of missing meteorological data, and has demonstrated high accuracy across 77 ChinaFLUX sites under various gap scenarios. The following three data archives provide the necessary inputs and outputs for the code package <code>01_MIR_Code.zip</code> at <a href="https://doi.org/10.5281/zenodo.16738283">https://doi.org/10.5281/zenodo.16738283</a>:</p> <ul> <li> <p><em><strong>02_Raw_Insitu.zip</strong></em><br>Contains raw half-hourly in-situ meteorological observations for each site-year. Each file is named in the format <code>YYYY_AAB_CXX.mat</code>, where <code>YYYY</code> is the year, <code>AAB</code> encodes the platform and general ecosystem type (e.g., A: cropland, F: forest), <code>C</code> denotes the footprint ecosystem, and <code>XX</code> is a two-digit site ID. Variables include air temperature, humidity, shortwave/longwave radiation, soil moisture, wind speed, and pressure, with missing values stored as NaNs.</p> </li> <li> <p><em><strong>03_ERA_Drivers.zip</strong></em><br>Includes hourly ERA5-Land reanalysis data extracted via Google Earth Engine, used as predictors in the MIR framework. Files are named <code>YYYY_AAB_CXX_NUPERA.csv</code>, corresponding to each site-year and containing variables such as 2m temperature, surface pressure, solar and thermal radiation, wind components, soil temperature, and total precipitation.</p> </li> <li> <p><em><strong>04_Gapfilled_Met.zip</strong></em><br>Provides MIR-reconstructed half-hourly meteorological datasets. Each file follows the naming format <code>filled_AAB_CXX_YYYY.mat</code>, and contains 17 fully gap-filled variables including temperature, relative humidity, net and component radiation, soil temperature and moisture, wind speed, pressure, and turbulent heat fluxes. The majority of original missing values have been filled using the best-performing model identified for each variable and gap duration.</p> </li> </ul> <p>All file and folder structures are aligned with the MIR codebase and support full pipeline reproducibility and future comparative studies in flux reconstruction.</p> <p><strong>Note: All author-identifying information has been removed from the dataset and code to comply with double-anonymous peer review; full metadata and authorship will be restored upon publication.</strong></p> |
| title | Data for article: New multisource data-driven Gap-filling approaches for meteorological data measured at eddy flux sites |
| topic | Eddy covariance Meteorological parameters Data-driven In-situ multi-layer observations ERA5 |
| url | https://doi.org/10.5281/zenodo.16738446 |