<aside> ❗ Geotiff output format (data_tiff, score_tiff, forecast_tiff) will be deprecated between now and model release v9.

</aside>

<aside> ❗ Bulk Download via /download/{paths} will be deprecated with the model release v9, planned for Sept. 3, 2024. It will continue to work with v8.

• forecast_timeseries replaces download.  In salientsdk you can vectorize your call to grab years of forecasts with a single line of code.  See notebook validate. </aside>

Dec. 3, 2024

salientsdk v0.3.13 is published to pypi

• new function extrapolate_trend returns expected-value timeseries for any point in time
  • example: used in the new notebook trend.ipynb
• new function Location.cluster divides large location files into downscale-friendly smaller lists
  • example: used in the 2024-25 ski-cast
• hindcast_summary now supports expanded set of model values.
• Package dependencies are relaxed for greater compatibility inside other environments

November 18, 2024

• Tabular view of forecasts is available in addition to a map view
• salientsdk v0.3.9 is now live on pypi
  • improved validate notebook
  • crps function has backwards-incompatible changes to the return value:
    • Returned Dataset data_val is now named crps_<timescale> to match inputs syntax like vals_<timescale>
    • Returned Dataset contains new data_val crps_<timescale>_all to show the progression of skill over time
    • If your notebook called crps and then inspected the results, it may need to be updated

November 4, 2024

• Forecast Evolution (Weekly, Monthly, Quarterly)
  • View the forecast for a selected valid date range over multiple initialization dates
  • Auto-play (in addition to manual advancing)
• Improved response time for seamless data visualization
  • Forecast view
  • Population and Unweighted averages using Area
  • Forecast distribution chart
• Low latency GEFS and ECMWF ENS medium range as a new data feed that can be licensed. Forecasts are available within 30 minutes of public availability delivered to an AWS or GCS bucket. Contact your account rep for additional information.

October 7, 2024

New Forecast Views include:

• Model Comparison
  • Compare forecasts across multiple models in a single view
• Forecast Evolution (Weekly)
  • View the forecast for a selected valid date range over multiple initialization dates
  • The initial release supports weekly calendar forecasts with a weekly initialization date cadence
  • Future releases will support monthly calendar and quarterly calendar forecasts