Data: U.S. Spring Temperature Trends

Please click here to download local temperature trend data for 240+ U.S. locations.

This dataset supports U.S., state-, and county-level charts downloadable from this page.
Embeddable interactive maps and custom data views are available here through Observable
Click here for a KML map download
For background on these trends, their connections to climate change, and their local impacts, click here for the most recent spring package from Climate Matters.

METHODOLOGY

Average temperatures and days above normal were calculated for each meteorological spring (March, April, May) from 1970 to 2025 using data obtained from the Applied Climate Information System (ACIS), which is developed, maintained, and operated by NOAA’s Regional Climate Centers. Spring days above normal are defined as the number of days with average temperatures above the 1991-2020 NOAA/NCEI climate normals.

The map of the contiguous U.S. shows the change in average spring temperatures by county since 1970 with data from NOAA/NCEI’s Climate at a Glance. Note that data for Connecticut is displayed by county because NCEI data are not yet available for Connecticut’s planning regions.

Climate Central's local analyses typically include data for 247 U.S. weather stations. For reported data summaries based on linear regression, however, only 241 stations are included in this brief due to data completeness measures that were not met by six stations: Bend, OR; Hazard, KY; Jefferson City, MO; Jonesboro, AR; Twin Falls, ID; and Wheeling, WV.

To estimate the spring warming (1970-2025) attributable to human-caused climate change in each city, we used Climate Central's Climate Shift Index system to compare two statistical models of spring temperature distributions (fit to ERA5 data): one calibrated to today's climate (with ~1.3°C of global warming above pre-industrial levels) and one representing a world without human-caused warming (0°C above pre-industrial levels). The difference between these two modeled temperature distributions is the expected spring warming attributable to human-caused climate change.

We then compare the expected spring warming due to human-caused climate change to the observed spring warming since 1970 from weather station data (ACIS) to calculate the percent of observed warming that is due to human-caused climate change.

Based on these data, the 241 cities analyzed fell into one of four categories based on whether and how two factors (human-caused climate change and secondary drivers) affected the observed spring warming in each city.

Human-caused climate change reflects the accumulation of heat-trapping greenhouse gases in the atmosphere due to human activities.
Secondary drivers of spring warming include natural climate variability, observational noise (from weather station data collection), and the urban heat island effect.

Across these four categories, the amount of warming due to human-caused climate change was:

Less than the observed warming (69 cities). In these locations, secondary drivers amplified the warming due to human-caused climate change.
Greater than the observed warming (163 cities). In these locations, at least 100% of the observed spring warming since 1970 can be explained by human-caused climate change; however, secondary drivers dampened the expected warming due to human-caused climate change. For simplicity, graphics for these 163 locations report that 100% of spring warming is due to human-caused climate change. For exact estimates, please download the full data file.
Equal to the observed warming (4 cities). In these locations, the magnitude of observed spring warming is equal to what we’d expect based on human-caused climate change.
Opposite in sign to observed warming (5 cities). All of the cities in this category have experienced spring cooling since 1970. In these locations, the cooling influence of secondary drivers has outpaced any warming that may have occurred due to human-caused climate change.