4 Comments
User's avatar
Michael's avatar

No surprises. The multi-decade trend lines are clear. But I would not expect to see the apparent linear progression. Shouldn't the graphic show a curve in recent time?

Expand full comment
Rick Thoman's avatar

Unless I'm trying to make a specific point or there's a clear discontinuity in the data (e.g. October temperatures at Utqiaġvik), I look a variety of methods to assess change over time and then plot the "trend line" that best fits the data with the fewest assumptions (technically, the method that has the lowest value of the Akaike Information Criteria). For many variables in the Arctic, this a fit with little or no trend until the 1970s or 80s, and then increases, i.e a "hockey stick" type fit. For March temperatures though, a simple linear regression across the full 76 years period of record turns out to be best fit in this sense.

Expand full comment
FrankTrades's avatar

How do you summarize “climate” based on one month?

Expand full comment
Rick Thoman's avatar

"Climate summary" in this context means putting what happened in March 2025 into historical perspective, including differences from multi-decade averages or rankings.

Expand full comment