Python Scripts and Forecasted Data (2018-2023) for Daily Arctic Sea Ice Concentration Forecasting

1. Data Description

The data include both codes, forecasted and observed Arctic SIC. The forecasted data, contains the forecasted SIC from January, 1, 2018 to December, 31, 2023.

2. SIC data

• Forecasted Data:
  
  • Temporal Coverage: January 1, 2018, to December 31, 2023.
  •  Data Formats: GeoTIFF format.

• Projection System: Northern Polar Stereographic.
• Spatial Resolution: 6.25 km.

3. code

• Data Formats: py format.
• Function:

  • SIC_plot.py: Visualizes sea ice concentration using NetCDF data, applying a polar stereographic projection, custom colormap, and land/coastline features for better geographic context.  

  • SIM_plot.py: Compares forecasted and observed sea ice margin using binary segmentation, erosion, and contour extraction, overlaying results on a polar stereographic map with land masking.

  • SIC_forecast.py: Implements long-term sea ice concentration forecasting by combining Least Squares (LS) trend modeling with an AutoRegressive (AR) model for residual correction. LS fits a quadratic trend with periodic components, while AR captures temporal dependencies in residuals, improving forecast accuracy. The method extrapolates future SIC values by predicting both trend and residual components.