Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 28, 2026

Surface Renewal: An Advanced Micrometeorological Method for Measuring and Processing Field-Scale Energy Flux Density Data
09:55

Surface Renewal: An Advanced Micrometeorological Method for Measuring and Processing Field-Scale Energy Flux Density Data

Published on: December 12, 2013

Hybrid spatial-field attention network for meteorological data downscaling.

Sheng Gao1,2, Yiming Ren3, Lianlei Lin4,5

  • 1School of Electronic Information Engineering, Harbin Institute of Technology, Harbin, 150000, China.

Scientific Reports
|June 2, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Immunodominant B-cell epitopes for precision surveillance of <i>Plasmodium vivax</i> malaria.

Emerging microbes & infections·2026
Same author

Engineering Organelle-Gated Reporters for Imaging Subcellular Enzyme Activity in Living Cells.

Journal of the American Chemical Society·2026
Same author

The association of oral health with anxiety symptoms among older adults in China: a cross-sectional study.

BMC geriatrics·2026
Same author

Mycoplasma pneumoniae infection impairs asthma control in pediatric patients and exacerbates allergic airway inflammation.

Respiratory research·2026
Same author

Biomedical Technologies for Androgenic Alopecia Treatment: Advance, Mechanism and Future Direction.

Research (Washington, D.C.)·2026
Same author

Longitudinal associations between exercise duration and sleep quality and poor sleep outcomes among college students: a latent growth curve modeling study.

BMC psychology·2026
This summary is machine-generated.

A new deep learning framework, HSFANet, improves meteorological data resolution for high-precision applications. This advanced downscaling method enhances accuracy and reconstructs future weather fields, benefiting engineering and production.

Area of Science:

  • Meteorology
  • Artificial Intelligence
  • Geospatial Analysis

Background:

  • Near-surface weather variables are crucial for production, but current data resolution is insufficient for high-precision needs.
  • Existing meteorological products lack the spatial detail required for modern engineering and decision-making.

Purpose of the Study:

  • To develop a deep learning framework for meteorological downscaling to enhance spatial resolution.
  • To improve the applicability of meteorological data in high-precision engineering and production scenarios.

Main Methods:

  • Proposed Hybrid Spatial-Field Attention Network (HSFANet) utilizing a hybrid spatial-field attention module.
  • Incorporated dynamic layer information integration for adaptive multi-level feature aggregation.
  • Integrated ground-object cross-attention to model meteorological fields and surface characteristics coupling.
Keywords:
Attention mechanismDownscalingEngineering applicationMeteorological variable

Related Experiment Videos

Last Updated: Jun 28, 2026

Surface Renewal: An Advanced Micrometeorological Method for Measuring and Processing Field-Scale Energy Flux Density Data
09:55

Surface Renewal: An Advanced Micrometeorological Method for Measuring and Processing Field-Scale Energy Flux Density Data

Published on: December 12, 2013

Main Results:

  • HSFANet demonstrated superior accuracy compared to state-of-the-art methods across various scaling factors on CLDAS V2.0 data.
  • The framework effectively reconstructs high-resolution future meteorological fields from coarse-resolution forecasts.
  • Achieved higher accuracy than existing methods in meteorological downscaling tasks.

Conclusions:

  • HSFANet offers a significant advancement in meteorological data downscaling.
  • The framework shows strong potential for practical deployment in engineering and production applications requiring high-resolution weather data.
  • The proposed deep learning approach effectively addresses the limitations of coarse spatial resolution in meteorological products.