Portrait of Drew Polasky

Drew Polasky

Research Scientist I
Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado Boulder, embedded at the NOAA Physical Sciences Laboratory

I’m an atmospheric scientist at CIRES and the NOAA Physical Sciences Laboratory in Boulder, Colorado. My current work focuses on improvements to ensemble prediction of precipitation for streamflow modeling using statistical and machine learning methods. My research generally focuses on methods for downscaling and post-processing weather and climate models for specific use cases.

Research interests

Recent publications

  1. Discrepancies in precipitation trends between observational and reanalysis datasets in the Amazon Basin Andrew D. Polasky, Vikrant Sapkota, Chris E. Forest, and Jose D. Fuentes (2025) Scientific Reports, 15(1), 7268 DOI
  2. Statistical downscaling for precipitation projections in West Africa Andrew D. Polasky, Jenni L. Evans, and Jose D. Fuentes (2024) Theoretical and Applied Climatology, 155(1), 327-347 DOI
  3. CCdownscaling: A Python package for multivariable statistical climate model downscaling Andrew D. Polasky, Jenni L. Evans, and Jose D. Fuentes (2023) Environmental Modelling & Software, 165, 105712 DOICodeData

All publications →

Side projects

Birkebeiner Results

Every American Birkebeiner since 1973. How many finish, how fast, and how the weather and snow at home change race times.

  • skiing
  • data
  • python

US Transit Atlas

Transit maps for 14 US metros built from their published schedules, and dashboards comparing ridership, cost and density across US cities and against systems abroad.

  • transit
  • maps
  • data
  • python

Campaign ↗

A turn-based U.S. election campaign strategy game. Play friends over the network, or AI opponents trained with reinforcement learning.

  • game
  • python

All side projects →