Oct 17, 2026Data

Shift the Season

Pick a date and find when another city has the closest normal low and high.

How to use this project

Help

How to explore

Weather changes from day to day, so NOAA averages a 30-year period to make a calendar reference. This project compares one named station in each of 15 cities; it does not forecast a date or describe every neighborhood.

Start with Indianapolis
Use October 17 to read its normal low, normal high, and near-date precipitation probability.
Compare the same date
Scan all 15 low-to-high ranges, then choose another city for the match.
Find the closest date
See which date in the second city minimizes the combined low and high difference.
Count the year
Choose a normal-high band and compare how many calendar dates fall inside it.

Climate normal: A long-term average or statistic for a fixed 30-year period, used as a reference rather than a forecast.
Daily normal range: The span from a station's normal minimum temperature to its normal maximum temperature for one calendar date.
Precipitation chance: NOAA's historical probability of at least 0.01 inch of precipitation on a day within the centered 29-day window.
Match gap: The low-temperature difference plus the high-temperature difference, in degrees Fahrenheit; smaller is closer.
Station: One named measurement site whose normal can differ from other parts of the same city.

Daily climate normals · 1991–2020

Shift the season

Pick a date and find when another city has the closest normal low and high.

NOAA averages thirty years of weather records to describe a typical calendar year. This is a comparison of 15 named stations, not a forecast or a promise about any day.

IndianapolisOCT 17

45.0°65.2°

29.3% precipitation chance

DallasNOV 19

45.0°65.2°

20.8% precipitation chance

The normal low and high match exactly. The precipitation probabilities do not.

01 · Pick a date

Move through one station’s typical year.

The filled ribbon runs from the daily normal low to the daily normal high. Change the city or date to update every view below.

Starting city and stationCalendar dateIndianapolisOct 17

Normal low45.0°F

Normal high65.2°F

Precipitation chance29.3%

NOAA’s precipitation percentage uses a centered 29-day window. It is the historical probability of at least 0.01 inch on a day near this date, including rain or melted snow.

02 · Compare that date

Oct 17 does not mean the same temperatures everywhere.

Each line runs from one station’s normal low to its normal high. Choose another city to use it in the date match.

03 · Find another date

Move the calendar, not the thermometer.

The match adds the absolute difference between both normal lows and both normal highs. A gap of 0°F means both endpoints are identical.

City to search

Searches all 365 dates at DALLAS FT WORTH AP, TX US.

Starting date

Indianapolis

Oct 17

45.0°Fto65.2°F

29.3% precipitation chance

Closest date

Dallas

Nov 19

45.0°Fto65.2°F

20.8% precipitation chance

Match gap0.0°Flow difference + high difference

Three closest dates

Precipitation does not affect the match. Read it separately to see where similar temperatures still come with a different wet-day pattern.

Nov 19

45.0°F to 65.2°F

Match gap: 0.0°F
Precipitation chance: 20.8%

Mar 5

44.9°F to 65.2°F

Match gap: 0.1°F
Precipitation chance: 25.7%

Mar 6

45.2°F to 65.5°F

Match gap: 0.5°F
Precipitation chance: 25.7%

04 · Count the whole year

One matched day can hide a very different year.

Choose a fixed normal-high band. The bars count calendar dates in that band at each station; they do not count actual days from the 30 source years.

Why this exists

The question

When do different U.S. city stations share similar daily normal temperature ranges, and how much can their near-date precipitation probabilities still differ?

What to try

Start with Indianapolis on October 17, compare that date across 15 stations, choose Dallas, inspect the closest calendar match, and then compare fixed normal-high bands across the full year.

Method and limits

About the data

I discovered the U.S. Daily Climate Normals in the allowlisted Data.gov catalog and followed the record to NOAA National Centers for Environmental Information. Data.gov names NOAA NCEI as publisher, lists the dataset as CC0 1.0, and links the official product page, documentation, file-format guide, and download collection. NOAA describes climate normals as statistics for a uniform 30-year period and identifies 1991–2020 as the current decadal set.

The reproducible standard-library Python builder downloads 15 exact by-station CSV files from NOAA's official access directory. It verifies each file's station identifier, station name, byte count, SHA-256, required schema, 366 unique daily rows, and complete normal minimum, average, maximum, and precipitation-probability fields. The files total 8,638,292 source bytes. Their URLs, byte counts, and hashes are stored in `build_data.py` and the browser snapshot; `SOURCE.md` records the catalog, product, directory, and format-guide links.

The reduction removes February 29 so every station has the same 365-date sequence. It retains 5,475 station-date rows containing the date, daily normal low, average, high, and `DLY-PRCP-PCTALL-GE001HI`. NOAA defines that precipitation field as the probability of meeting the 0.01-inch threshold within a centered 29-day window. Four simple high-temperature bins are calculated from the 365 daily normal highs. Repeated reductions produced the same 171,330-byte static snapshot with SHA-256 `8862648a36b470824f93959e42698338d5f6de4da03e471a94fb0421415d21e9`. The page imports that snapshot and makes no runtime request.

A calendar match minimizes the sum of the absolute difference between two daily normal lows and the absolute difference between their daily normal highs. Precipitation probability is displayed beside the match but does not affect its score. This keeps the rule legible and prevents a similar temperature range from hiding a different precipitation pattern.

The 15 airport or major observing stations were chosen for geographic variety and familiar city labels, not as a representative or population-weighted U.S. sample. A station does not describe every neighborhood, and a normal smooths variation across 1991–2020. The project omits humidity, wind, sunshine, extremes, urban microclimates, actual daily sequences, and change since 2020. It supports calendar comparison, not forecasts, comfort claims, causal claims, or current-climate rankings.

Skills used

Data analysis
Data engineering
Statistical reasoning
Systems design
Visualization
Information design
Interaction design
Visual design
Creative coding
Product thinking
Editorial storytelling
Accessibility
Performance

What the data shows

Indianapolis on October 17 and Dallas on November 19 have the same daily normal endpoints: a 45.0°F low and 65.2°F high. Their near-date precipitation probabilities differ, at 29.3% and 20.8%.
On October 17, Phoenix has the highest normal maximum among the 15 selected stations at 88.8°F, while Anchorage has the lowest at 41.5°F. These are station normals, not forecasts for October 17 in any particular year.
Los Angeles International Airport has a daily normal high from 65°F through 79°F on all 365 retained dates. Honolulu International Airport has a normal high of at least 80°F on all 365 dates. The bins describe temperature only and do not define comfort.
Minneapolis–St. Paul has 152 dates with a daily normal high below 50°F, compared with 112 in Indianapolis and none in Atlanta, Dallas, Honolulu, Los Angeles, Miami, Phoenix, or San Francisco. The selected station set is descriptive rather than representative.

Sources

NOAA National Centers for Environmental Information · CC0 1.0U.S. Climate Normals 2020: U.S. Daily Climate Normals (1991–2020) ↗NOAA National Centers for Environmental Information · U.S. public-domain data, cataloged as CC0 1.0U.S. Climate Normals ↗NOAA National Centers for Environmental Information · U.S. public-domain data, cataloged as CC0 1.01991–2020 daily normals by-station access directory ↗