Climate Central

Climate MattersFebruary 11, 2026

Shorter Winters in 195 U.S. Cities

KEY FACTS

This Climate Matters analysis is based on open-access data from the National Oceanic and Atmospheric Administration (NOAA). See Methodology for details.

Download local data

CM: Chocolate Crunch: Bittersweet News For Chocolate Lovers Hilspire Video 2026

Click to download this ready-to-air video, Chocolate Crunch: Bittersweet News For Chocolate Lovers. This fully produced news story can be used by media outlets at no cost for broadcasts, streaming, and digital stories. Click here for other available packages.

Warming planet, shifting seasons

As the planet warms, all four seasons are warming and the seasonal clock itself is warping — with the hottest time of year lasting longer and the coldest time of year shrinking

There are several different ways to define a season and measure how they’re changing as our planet warms. 

Time-based seasons. Meteorological seasons — the default for climate science and Climate Central analyses — divide the year into four quarters of equal length. For example, meteorological winter is defined as the three-month period of December, January, and February. And all four meteorological seasons are warming across the U.S.

Temperature-based seasons. Seasons can also be defined based on temperature — for example defining winter as the coldest days of the year based on historical data — to understand how warming trends affect season length. 

In a new analysis, Climate Central used temperature-based seasons to assess how the length of the winter season has changed with climate warming since 1970. 

CM: Length of Winter 2026 (EN)
Click the downloadable graphic: Length of Winter

Shorter winters in 195 U.S. cities

Climate Central used weather station data from 245 major U.S. cities to define a temperature-based winter season in each location. 

This analysis defined winter as the coldest 90 consecutive days of the year during the past (1970-1997) and then compared the frequency of these winter-like temperatures during the most recent 28-year period (1998-2025). See Methodology for details. Results show that:

For example, in Erie, Pennsylvania — where winters are now 16 days shorter — winter-like daily temperatures (33.2°F or lower) now: start an average of 10 days later (starting Dec. 22 instead of Dec. 12) and end an average of six days earlier (March 5 instead of March 11) than during 1970-1997. 

Within the contiguous U.S., cities across the Southeast, Northeast, Upper Midwest, and South have experienced the largest average decrease in winter days. 

Juneau and Anchorage, Alaska have seen winters shrink the most (by 62 and 49 days, respectively), consistent with exceptionally rapid warming in Alaska and other high-latitude locations. 

Some 37 (15%) of the cities analyzed saw winters lengthen — particularly in coastal California (where the ocean’s influence limits large seasonal temperature swings) and in the Ohio Valley. But those cities were the exception to the national trend, which is consistent with prior studies.

Research shows that the trend toward longer, hotter summers and shorter, warmer winters across the Northern Hemisphere is mainly due to heat-trapping carbon pollution.

Warmer, shorter winters affect water supplies, crops, and ecosystems 

Warming winters can reduce mountain snowpack — a critical source of spring meltwater that refills reservoirs, irrigates crops, and helps meet peak water demand across the western U.S. during summer. Snow drought like the western U.S. has experienced this year can also boost fire risk during the following growing season. 

Warm winters disrupt sports, recreation, and related local economies. The multi-billion dollar winter recreation industry faces challenges from rising temperatures and reduced snow and ice cover. In some areas of Colorado, ski resort visits are down 20% this year amid severe snow drought. All of the Winter Olympics host cities since 1950 have warmed since then — threatening the reliability, safety, and fairness of outdoor winter sports competitions. 

Warmer, shorter winters can affect summer crop yields. Fruit and nut crops — a nearly $27 billion industry in the U.S. — depend on winter chill. Climate Central analysis found that this necessary winter chill has decreased since 1980 in top-producing counties for 10 most valuable non-citrus fruit and nut crops (especially in California, Washington, Florida, Oregon, South Carolina, and Georgia).

Cold winters can keep the populations of disease-carrying pests like mosquitoes in check. But warmer, shorter winters can worsen pest-related health risks. 

Warmer, shorter winters impact the growing season and allergies. And earlier spring thaw and later fall freeze gives plants more time to grow and release allergy-inducing pollen earlier in spring and later into the fall.

LOCAL STORY ANGLES

Is climate change influencing daily temperature extremes in your local area?

Climate Central’s Climate Shift Index (CSI) system provides tools, data, custom maps, and local alerts to answer this question in real-time. Here are three ways to use the CSI:

FIND EXPERTS

Submit a request to SciLine from the American Association for the Advancement of Science or to the Climate Data Concierge from Columbia University. These free services rapidly connect journalists to relevant scientific experts. 

Browse maps of climate experts and services at regional NOAA, USDA, and Department of the Interior offices.  

Explore databases such as 500 Women Scientists, BIPOC Climate and Energy Justice PhDs, and Diverse Sources to find and amplify diverse expert voices. 

Reach out to your State Climate Office or the nearest Land-Grant University to connect with scientists, educators, and extension staff in your local area. 

METHODOLOGY

Daily average temperature data (1970-2025) for 245 U.S. weather stations were obtained from the Applied Climate Information System, which is developed, maintained, and operated by NOAA’s Regional Climate Centers. 

This 56-year period was divided into two 28-year climatologies representing the average annual temperature cycle in each location during the past (1970-1997) and recent (1998-2025) periods. Both daily average temperature climatologies were smoothed to capture the seasonal cycle and reduce day-to-day variability (using R package “mgcv”).

The coldest 90-day window in the smoothed past (1970-1997) climatology was identified and used to define a baseline winter period. The warmest day during this 90-day window was used to define the temperature threshold below which days are considered to have historically winter-like temperatures.

Using the smoothed recent (1998-2025) climatology, we identified the start and end dates at which daily average temperatures remained at or below the historically winter-like temperature threshold (as defined above). The number of days that span the period between the start and end date (inclusive) represents the length of the recent winter period.

The difference in the number of days between the recent winter period and baseline winter period represents the change in winter at a given location. 

This analysis is based on an analysis by Brian Brettschneider, summarized in a 2020 blog post (Are Summer/Winter Longer/Shorter Than They Used To Be?) and covered in a 2025 article in The Washington Post (See how much longer summer is in your town). 

References to U.S. regions refer to the nine U.S. Climate Regions defined by NOAA (NCEI).

Climate Central's local analyses include 247 U.S. weather stations. For reported data summaries herein, however, only 245 stations are included due to data completeness measures that were not met by two stations: Hazard, Ky. and Wheeling, W.Va.