Potential for western US seasonal snowpack prediction

Area of Science:

• Climate science
• Hydrology
• Agricultural meteorology

Background:

• Western US snowpack is vital for regional water supply and agriculture, with 80% of snowmelt runoff used for farming.
• Current climate projections and weather forecasts lack seasonal snowpack prediction capabilities (months to 2 years), hindering agricultural decisions.

Purpose of the Study:

• To demonstrate the feasibility of seasonal snowpack predictions beyond 3 months.
• To quantify the limits of predictive skill for snowpack 8 months in advance.

Main Methods:

• Utilized observations, climate indices, and a suite of global climate models.
• Developed a physically based dynamic system for seasonal snowpack prediction.
• Compared model performance against observation-based statistical predictions.

Main Results:

• The dynamic system demonstrates feasible seasonal snowpack predictions up to 8 months ahead.
• The system outperforms statistical predictions made on July 1 for March snowpack, except in the southern Sierra Nevada.
• High hydroclimate variability in narrow maritime mountain ranges poses challenges for seasonal snowpack prediction.

Conclusions:

• Seasonal snowpack prediction is advancing, offering optimism for societal needs.
• Natural snowpack variability may limit predictability in certain regions at seasonal timescales.
• Further development is needed to improve predictions in challenging geographical areas.