SIC-Prediction

산지니 조

• DS 대학원 2025-1 딥러닝 프로젝트

---

주제: Spatio-Temporal Forcasting Monthly Arctic sea ice concentration

• Data Source:
  • ERA5 monthly averaged data on single levels from 1940 to present (https://cds.climate.copernicus.eu/datasets/reanalysis-carra-single-levels?tab=overview)
  • NSIDC Sea Ice Concentrations from Nimbus-7 SMMR and DMSP SSM/I-SSMIS Passive Microwave Data, Version 2 (https://nsidc.org/data/nsidc-0051/versions/2)
• Data Range: 1995/01/01 ~ 2024/12/31
• Using Variables
  • 10m u component of wind
  • 10m v component of wind
  • 2m temperature
  • Mean sea level pressure
  • Total precipitation
  • Sea surface temperature (Delete)
  • Surface sensible heat flux
  • Snow albedo
  • Snowmelt
  • Total cloud cover
  • Total column water vapour
• Our Model Architecture

실행 방법

1. 다음과 같이 가상환경 생성

conda create -n sic python==3.11.11

2. Repository를 Clone

git clone [해당 Repository 주소]

3. requirements.txt 설치

pip install -r requirements.txt

4. 기존의 기후 변수 전처리 파일 download / 직접 원본 데이터 다운로드 후 Climate_variable_Preprocessing.ipynb 실행

• 전처리 완료 파일 다운로드 링크: (https://www.dropbox.com/scl/fi/179413qvyenr4fxqr9yli/feature_real_final.npz?rlkey=b18ycpier7iifl5v8j24if7ge&st=4q31ru01&dl=0)

5. main.py를 실행

Results

All results images exist in folder 'results'

Input Climate Variable

Input only the past SIC variable

Input Climate + SIC variable