MPL – Marine Prediction Lab
Yellow SeaNov 4, 2025
  • Numerical Modeling
  • Observation
Development of Data Assimilation Techniques

✅Overview of Results

*The accuracy of numerical models tends to decline as the simulation period lengthens, which reduces forecast skill and can undermine user confidence.

*To improve accuracy, we merged observation data into the model— this process is known as data assimilation.

*Observation data for assimilation are grouped into satellite data and in- situ data.

* Satellite data provide broad spatial coverage but generally lower accuracy than in-situ measurements.

* In-situ data offer high accuracy but have strong spatial limitations.

* By combining these complementary datasets, we make a merging observation dataset that improve model accuracy over wide areas. Applying this merged data within our assimilation framework stabilizes and improves the reliability of the model’s performance.

Contact: JiMyeong Seo (tjwlaud4@mpl.inha.ac.kr)

Project
Sea Grant Program (Phase 5)
Funded by
Ministry of Oceans and Fisheries
← All research