Of 354 included NPS visitor-statistics reporting codes, 205 (57.9%) recorded a higher April, May, September, and October share between the 2012 to 2015 baseline and 2022 to 2025. 154 (43.5%) rose by more than one percentage point. Share is not the same as visit count. 135 of the 205 also recorded more average annual visits during those four months. Yellowstone's four-month share rose from 30.9% to 38.5%. Its June through August share moved from 66.1% to 57.4%.
Shoulder and peak shares at 15 selected reporting units
These examples were selected to show different results, places, and access-policy contexts. They are not a representative sample or a ranking. The downloadable summary CSV contains all 354 reporting codes. "Shoulder" is this study's shorthand for April, May, September, and October. "Peak" means June, July, and August.
| Reporting unit | Shoulder 2012 to 2015 | Shoulder 2022 to 2025 | Δ shoulder (pp) | Δ peak (pp) | Access policy in reviewed set? | Avg annual visits, 2022 to 2025 |
|---|---|---|---|---|---|---|
| Glacier Bay National Park & PreserveAK · Cruise schedules strongly influence monthly visitation | 27.9% | 39.5% | +11.6 | −11.6 | No | 681,436 |
| Yellowstone National ParkID / MT / WY · National Park | 30.9% | 38.5% | +7.6 | −8.7 | No | 4,324,741 |
| Badlands National ParkSD · National Park | 25.7% | 32.5% | +6.8 | −6.5 | No | 1,071,704 |
| Glacier National ParkMT · National Park | 24.7% | 31.3% | +6.5 | −6.9 | Yes (2021) | 3,046,846 |
| Grand Teton National ParkWY · National Park | 28.5% | 34.3% | +5.8 | −6.4 | No | 3,413,050 |
| Acadia National ParkME · National Park | 35.8% | 41.4% | +5.6 | −5.4 | Yes (2020) | 3,972,782 |
| Mesa Verde National ParkCO · National Park | 34.7% | 40.0% | +5.4 | −6.6 | No | 487,045 |
| Olympic National ParkWA · National Park | 31.1% | 36.3% | +5.3 | −10.6 | No | 3,170,482 |
| Bryce Canyon National ParkUT · National Park | 42.1% | 47.4% | +5.3 | −7.6 | No | 2,320,343 |
| Rocky Mountain National ParkCO · National Park | 31.1% | 35.9% | +4.8 | −7.5 | Yes (2020) | 4,185,510 |
| Mammoth Cave National ParkKY · National Park | 33.7% | 36.3% | +2.7 | −10.4 | No | 681,343 |
| Mount Rainier National ParkWA · National Park | 27.7% | 29.4% | +1.7 | +3.8 | Yes (2024) | 1,638,009 |
| Great Smoky Mountains National ParkTN / NC · TRV includes substantial through-traffic | 38.3% | 39.7% | +1.4 | −5.6 | No | 12,488,763 |
| Arches National ParkUT · National Park | 43.7% | 43.5% | −0.3 | −7.9 | Yes (2022) | 1,480,241 |
| Shenandoah National ParkVA · National Park | 47.9% | 47.5% | −0.4 | −5.2 | No | 1,606,918 |
Shoulder share at 15 selected reporting units
The full distribution includes gains and declines
The result changes with the window and month definition
How this was computed
Source dataset. We used NPS Visitor Use Statistics Data Package, 2025, specifically Main_Data.csv. The file covers monthly data through 2025. This analysis uses the NPS Total Recreation Visits (TRV) statistic only. Data.gov marks the source package as CC0 1.0.
Windows compared. The baseline is 2012 through 2015. The comparison window is 2022 through 2025. We pool the complete years available within each window, then calculate each period's monthly shares from the pooled TRV totals.
Month definitions. This study uses "shoulder season" as shorthand for April, May, September, and October. That fixed definition does not fit every climate or travel market. Peak months are June, July, and August. Each share divides visits in those months by all annual TRV in the same window.
Inclusion and weighting. A reporting code needs at least one complete 12-month year in each window. Its average annual TRV across complete baseline years must be at least 10,000. 354 codes qualify, and each has equal weight. Three have fewer than four complete baseline years. Sacramento National Historic Trail and Rosie the Riveter/World War II Home Front National Historical Park use 2013 to 2015. Waco Mammoth National Monument uses 2015.
Identity metadata. Visitor-statistics reporting codes do not always match public Parks API codes. Names, types, and states use the official NPS unit registry, IRMA Stats reporting labels, and the package's state file. Reporting areas without an official unit designation are labeled as reporting areas or reporting sites rather than assigned a park type.
Share versus count. A higher share does not always mean more visits. Of the 205 codes with a higher four-month share, 135 also recorded more average annual visits in those four months. The full CSV includes both measures.
2025 shutdown limitation. NPS reports a 43-day partial federal government shutdown in 2025. October is one of the four months in this analysis. The 2022 to 2024 sensitivity result is higher than the primary 2022 to 2025 result, but the data cannot tell us how much of that difference came from the shutdown.
Access-policy limitation. The table flags a limited reviewed set of timed-entry or vehicle-reservation programs adopted after the baseline: Acadia, Rocky Mountain, Glacier, Arches, and Mount Rainier. It is not a complete policy inventory. Yellowstone had no timed-entry program in either window.
Other limits. TRV counts visits, not unique people. Counting methods differ among reporting codes and sometimes change. Glacier Bay depends heavily on cruise schedules. Great Smoky Mountains includes substantial through-traffic. The figures describe recorded shares, not what caused them.
What this finding does, and doesn’t, tell us
The majority rests on the zero-change threshold. 205 codes recorded any increase in four-month share, while 154 rose by more than one point. The median change was +0.56 points.
Yellowstone is a large measured change. Its four-month share was 7.6 points higher, while its June-through-August share was 8.7 points lower. Yellowstone had no timed-entry program in either period. The visitation data does not identify the cause.
The selected examples are mixed. Badlands, Glacier, Grand Teton, Acadia, Mesa Verde, Olympic, Bryce Canyon, and Rocky Mountain each rose by at least 4.8 points. Mount Rainier's four-month share rose 1.7 points, but its summer share also rose 3.8 points. Arches and Shenandoah recorded small declines.
Yellowstone's June through August share was 8.7 points lower
Quotable sentences
These sentences are available to quote with attribution to Nicholas Major, National Sites Guide. The methodology and downloadable rows show how the figures were calculated from the official NPS data package. A link back to nationalsitesguide.com is appreciated when you publish online.
“The latest NPS package shows 205 of 354 included reporting codes with a higher April, May, September, and October share. Of those, 135 also recorded more average annual visits in those months.”Nicholas Major, National Sites Guide
“Yellowstone recorded a 38.5 percent April, May, September, and October share in 2022 through 2025. Its June through August share was 57.4 percent.”Nicholas Major, National Sites Guide
“When 2025 or October is removed, the result changes. The data does not show that the shutdown caused the difference.”Nicholas Major, National Sites Guide
How to cite this analysis
If you reference these numbers in print, broadcast, or online, the following one-line citation is preferred:
Data and chart files
National Sites Guide's original charts and analysis are released under CC BY 4.0; please credit National Sites Guide when republishing. The underlying NPS source files retain their source terms.
Who did this work
Reach the analyst
Reach the analyst through the contact form. Include your outlet and deadline and we'll get back to you promptly.
Independent site. Not affiliated with or endorsed by the National Park Service. The underlying data comes from the official NPS Visitor Use Statistics Data Package, 2025 release; editorial analysis and methodology are ours. The NPS Arrowhead and other NPS marks are not used.