Home > Supply Pipeline > 2,839 Onsen Areas, 13,425 Onsen Inns: Japan’s Hot Spring Stock FY2024

2,839 Onsen Areas, 13,425 Onsen Inns: Japan’s Hot Spring Stock FY2024

Posted: 2026.09.02

Supply Pipeline

The Ministry of the Environment’s “Hot Spring Utilization Survey” is the only complete-count statistic that tallies Japan’s hot spring stock prefecture by prefecture. According to the latest FY2024 edition (as of 31 March 2025), Japan had 2,839 hot spring areas, 27,899 hot spring sources in total, and 13,425 lodging facilities using hot spring water. The number of hot spring areas fell by 95 over five years, while lodging facilities rose by 501 and annual overnight guest-nights climbed from 76.59 million to 125.41 million — a 1.6x increase. This article transcribes the prefecture-level figures from the official PDFs, compares all 47 prefectures on two density metrics — lodging facilities per hot spring source, and annual guest-nights per lodging facility — and then measures the gap between the statistical count of onsen inns and the facility counts MetroEngines Research observes on OTAs and similar channels.

Metric Definitions Used in This Article

  • Number of hot spring areas: Under the definition used in the Ministry of the Environment’s “Hot Spring Utilization Survey,” this counts places that have lodging facilities. Locations with day-use bathing facilities only are excluded.
  • Total hot spring sources: The sum of sources in use (free-flowing and pumped) and sources not in use (free-flowing and pumped).
  • Lodging facilities using hot spring water: The number of lodging facilities that use hot spring water, including hotels, ryokan, minshuku and similar.
  • Annual overnight guest-nights: The cumulative number of guests who stayed at hot spring lodging facilities during the fiscal year. The Ministry of the Environment notes this as a “reference figure.”
  • Reference date: FY2024 = as of 31 March 2025 (guest-nights are the fiscal-year total for April 2024 – March 2025).
  • ADR (average daily rate): An estimated settled rate (tax-excluded equivalent) calculated by applying a segment-specific adjustment factor to the lowest published plan level each facility lists on OTAs and similar channels (double occupancy, per-room rate, tax included). Cross-checked against property-level results disclosed by listed hotel REITs (91 properties, most recent three months), the median error is approximately 7%. These are estimates and differ from each facility’s actual transacted prices or accounting figures. Area-level ADR is the median of the target facilities (the level of a typical facility in that area).
  • Data sources: Ministry of the Environment, “Hot Spring Utilization Survey” (FY2020–FY2024) / MetroEngines Research
Key Takeaways
  • — 2,839 hot spring areas, 27,899 sources and 13,425 lodging facilities using hot spring water (FY2024, as of 31 March 2025). Hot spring areas fell by 95 over five years, while lodging facilities rose by 501.
  • — Lodging facilities per hot spring area rose from 4.40 to 4.73. Hot spring areas as places are consolidating, while lodging capacity concentrates in the areas that remain.
  • — Roughly 40% of hot spring sources (11,032) are unused. Kagoshima at 0.11, Oita at 0.18 and Aomori at 0.20 lodging facilities per source sit far below the national average of 0.48, leaving ample room to build lodging capacity on existing source infrastructure.
  • — Statistical onsen inns exceed OTA-visible inventory in Shizuoka (+504) and Oita (+101). Existing facilities have headroom in online exposure.
  • — Estimated settled ADR in the ryokan category rises as much as +30.5% from September (Ishikawa) into autumn. Hokkaido, however, is −3.7%: the autumn peak is not uniform nationwide.

Japan’s Hot Spring Stock — 2,839 Areas, 27,899 Sources, 13,425 Lodging Facilities

The FY2024 national totals (as of 31 March 2025) are 2,839 hot spring areas, 27,899 hot spring sources, 13,425 lodging facilities using hot spring water and guest capacity of 1,324,805. Annual overnight guest-nights came to 125,414,300. Year on year, hot spring areas fell by 18 and sources by 21 — both small declines — while lodging facilities rose by 246 and guest-nights by 4,708,332, both increases.

Of the 27,899 sources, 16,868 are currently in use and 11,032 are not. In other words, roughly 40% of hot spring sources are currently unused. That unused share represents room to add lodging capacity where flow volume, water quality and location allow, and it is an important denominator when gauging the supply potential of a hot spring area. That said, whether an unused source can carry lodging capacity is not determined by flow volume alone. How the three rights structures — own source, piped supply and centralised management — change acquisition price and operational freedom is set out in Who Owns the Hot Spring?.

Table 1: Japan’s hot spring stock and utilisation, FY2020–FY2024
Fiscal yearReference dateHot spring areasTotal sourcesHot spring lodging
facilities
Guest capacityAnnual overnight
guest-nights
Facilities per
hot spring area
FY2020 31 Mar 2021 2,934 27,969 12,924 1,313,024 76,592,711 4.40
FY2021 31 Mar 2022 2,894 27,915 12,909 1,316,694 78,038,804 4.46
FY2022 31 Mar 2023 2,879 27,932 12,999 1,330,203 109,857,625 4.52
FY2023 31 Mar 2024 2,857 27,920 13,179 1,326,618 120,705,968 4.61
FY2024 31 Mar 2025 2,839 27,899 13,425 1,324,805 125,414,300 4.73

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey” (FY2020–FY2024). Each fiscal year uses the figures published in that year’s PDF. Note: FY2022 annual overnight guest-nights were revised to 108,872,804 in the following year’s edition.

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey”

Over the five years, hot spring areas declined consistently from 2,934 to 2,839 (−95, −3.2%), while lodging facilities using hot spring water rose from 12,924 to 13,425 (+501, +3.9%). As a result, lodging facilities per hot spring area thickened from 4.40 to 4.73. Hot spring areas as places are consolidating, while lodging capacity gathers in the areas that remain.

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey”

Annual overnight guest-nights rose from 76.59 million in FY2020 to 125.41 million in FY2024, up 63.7% in five years. That span includes the recovery from FY2020 and FY2021, when movement was restricted, but growth continued in the most recent year as well, at +3.9% from FY2023 to FY2024. Per lodging facility the figure rose from 5,926 to 9,342, meaning the lodging demand each facility absorbs grew 1.58x in five years.

Top 15 Prefectures — Reading the Character of Hot Spring Areas Through Two Density Metrics

The prefectures with the most hot spring areas are Hokkaido with 226, Nagano with 193, Niigata with 135, Fukushima with 122 and Aomori with 120. The count alone, however, says nothing about how well a prefecture’s hot springs function as a lodging business. Two density metrics help. The first is lodging facilities per hot spring source (how many inns one piece of source infrastructure supports); the second is annual overnight guest-nights per lodging facility (how much lodging demand a single facility absorbs).

Table 2: Top 15 prefectures by number of hot spring areas, with two density metrics (FY2024, as of 31 March 2025)
RankPrefectureHot spring areasTotal sourcesIn useNot in useLodging facilitiesAnnual overnight
guest-nights
Facilities
per source
Guest-nights
per facility
1 Hokkaido 226 2,249 1,284 965 740 13,579,988 0.33 18,351
2 Nagano 193 955 636 319 998 7,081,036 1.05 7,095
3 Niigata 135 536 365 171 490 3,539,773 0.91 7,224
4 Fukushima 122 799 418 381 474 3,622,686 0.59 7,643
5 Aomori 120 1,089 550 539 216 1,453,144 0.20 6,728
6 Shizuoka 118 2,185 1,163 1,022 2,397 10,576,818 1.10 4,413
7 Akita 104 616 376 240 203 1,441,551 0.33 7,101
8 Gunma 98 459 363 96 631 5,925,649 1.37 9,391
9 Chiba 95 144 141 3 188 3,718,331 1.31 19,778
10 Kagoshima 87 2,735 1,202 1,533 311 2,136,463 0.11 6,870
11 Iwate 81 393 226 167 227 1,839,235 0.58 8,102
12 Hyogo 78 443 262 181 322 4,742,639 0.73 14,729
13 Mie 74 214 162 52 363 2,788,134 1.70 7,681
14 Yamagata 67 412 263 149 252 1,950,757 0.61 7,741
15 Tochigi 66 633 445 188 414 4,296,081 0.65 10,377
— National total 2,839 27,899 16,868 11,032 13,425 125,414,300 0.48 9,342

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey” (FY2024). Note: the in-use / not-in-use breakdown is reproduced exactly as printed in the original tables, so at the national level the two components differ from total sources by one.

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey” (FY2024). Circle size = number of lodging facilities using hot spring water.

Plotted on a scatter chart, the top 15 prefectures fall into roughly three groups.

1. Prefectures with thick lodging capacity per source — Mie has 363 lodging facilities against 214 sources, or 1.70 facilities per source, the highest in Japan. Gunma follows at 1.37, Chiba at 1.31, Shizuoka at 1.10 and Nagano at 1.05. These are highly consolidated hot spring areas where several inns share a limited number of sources.

2. Prefectures with a large draw per facility — Chiba absorbs 19,778 guest-nights per lodging facility, Hokkaido 18,351 and Hyogo 14,729, all far above the national average of 9,342. Large onsen hotels take in the demand here, and guest capacity is correspondingly deep, at 53,809 in Chiba and 124,348 in Hokkaido.

3. Prefectures where source stock outstrips lodging capacity — Kagoshima has 311 lodging facilities against 2,735 sources (0.11 per source), Oita 914 against 5,094 (0.18) and Aomori 216 against 1,089 (0.20). All are less than half the national average of 0.48, making this the group with ample room to build lodging capacity on existing source infrastructure. Kagoshima stands out in particular: its 1,533 unused sources are among the most in Japan, and the headroom in its hot spring areas shows up directly in the figures.

Fewer Hot Spring Areas, More Inns — Five Years of Thickening Density Per Area

Layering these figures over time reveals a consistent direction in Japan’s hot spring stock. Hot spring areas — places that have lodging facilities — fell by 95 in five years. Over the same period, total sources slipped only from 27,969 to 27,899, a decline of 70 and essentially flat. Lodging facilities nonetheless rose by 501.

Table 3: Five-year change in hot spring stock density metrics (national totals)
MetricFY2020FY2022FY2024Five-year change
Lodging facilities per hot spring area 4.40 4.52 4.73 +0.33 (+7.4%)
Lodging facilities per hot spring source 0.462 0.465 0.481 +0.019 (+4.1%)
Guest capacity per lodging facility 101.6 102.3 98.7 −2.9 (−2.8%)
Annual guest-nights per lodging facility 5,926 8,451 9,342 +3,416 (+57.6%)

Source: Calculated by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey” (FY2020, FY2022 and FY2024)

Worth noting is that guest capacity per facility edged down from 101.6 to 98.7. Facility counts rising while average size shrinks suggests that the properties newly counted as hot spring facilities are relatively small. Guest-nights per facility, meanwhile, jumped 57.6%, pointing to a shift toward absorbing demand through occupancy and turnover rather than scale. For small inns that carry the added value of a hot spring, this widens the room to design rates. How much the “onsen” label itself moves rates can also be seen as an estimated ADR gap in 23.5% of urban hotels have bathing facilities.

Behind the rise in facility counts are existing properties beginning to use hot spring water (new source use or piped supply) and new openings or rebrands of properties with hot springs. Given that total sources stayed nearly flat while sources in use moved from 16,960 to 16,868, the five-year picture is closer to lodging capacity being added while the existing source infrastructure is reshuffled.

The Gap Between Statistical Onsen Inns and Inns Visible on OTAs — 504 in Shizuoka, 101 in Oita

From here we match public statistics against our own data. The Ministry of the Environment’s “lodging facilities using hot spring water” is a complete count of inns using hot springs only, whereas the number of facilities for which MetroEngines Research can observe prices on OTAs and similar channels covers every segment regardless of whether a hot spring is present. The latter is therefore normally the larger figure.Where the former nonetheless exceeds the latter, a single segment — onsen inns — is outstripping the inventory visible on OTAs and similar channels, which reads as a structurally deep stock of onsen inns that are not listed.

Table 4: Lodging facilities using hot spring water (Ministry of the Environment) vs facilities observed on OTAs and similar channels (July 2026) — top 15 prefectures by hot spring areas
PrefectureHot spring areasHot spring lodging
facilities
(Ministry of the Environment)
Facilities observed on
OTAs and similar channels
(all segments, July 2026)
Difference (Ministry − observed)Ratio
Shizuoka 118 2,397 1,893 +504 1.27
Aomori 120 216 262 -46 0.82
Gunma 98 631 790 -159 0.80
Akita 104 203 256 -53 0.79
Fukushima 122 474 620 -146 0.77
Niigata 135 490 722 -232 0.68
Yamagata 67 252 374 -122 0.67
Iwate 81 227 360 -133 0.63
Mie 74 363 613 -250 0.59
Tochigi 66 414 810 -396 0.51
Nagano 193 998 2,101 -1,103 0.47
Kagoshima 87 311 770 -459 0.40
Hokkaido 226 740 2,142 -1,402 0.35
Hyogo 78 322 1,211 -889 0.27
Chiba 95 188 1,134 -946 0.17

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey” (FY2024, as of 31 March 2025) and MetroEngines Research (facilities for which prices could be observed on OTAs and similar channels in July 2026, all segments)

Source: Compiled by the HotelBank Editorial Team from the Ministry of the Environment’s “Hot Spring Utilization Survey” and MetroEngines Research

Only two prefectures exceeded a ratio of 1.00: Shizuoka (2,397 vs 1,893, a gap of +504 and a ratio of 1.27) and Oita (914 vs 813, a gap of +101 and a ratio of 1.12). Shizuoka holds one of Japan’s largest hot spring stocks, with 118 hot spring areas and 2,185 sources, and small ryokan and minshuku are densely distributed across Atami, Ito and the Izu Peninsula. Oita is the largest hot spring prefecture of all, with 5,094 sources — 18% of the national total — and onsen inns cluster around Beppu and Yufuin. In both, the base of onsen inns is broader than the inventory visible on OTAs and similar channels.

Ratios of 0.7 or higher line up next: Aomori at 0.82, Gunma at 0.80, Akita at 0.79 and Fukushima at 0.77. In these prefectures a substantial share of the observable lodging facilities are onsen inns, indicating that the prefecture’s lodging supply as a whole leans heavily on hot springs. At the other end, Chiba at 0.17, Yamanashi at 0.18, Hyogo at 0.27 and Hokkaido at 0.35 have deep non-onsen lodging supply — urban hotels, resort hotels, condominiums and the like.

This gap is more than a note that some inns are unlisted. The drawing power of a hot spring area rests on the size of its statistical stock, while the inventory that can actually be compared and booked online is only a part of it. The more onsen inns a prefecture has without a distribution channel, the more headroom there is in bringing existing facilities online and in designing rates within a visible market. In prefectures where the gap is wide — Shizuoka and Oita above all — greater regional exposure can also be expected to lift search demand for the hot spring area itself.

Note: the two figures use different populations (the Ministry of the Environment counts only facilities using hot spring water, in full; our figure covers all segments regardless of hot springs, limited to facilities for which prices could be observed on OTAs and similar channels that month). The ratio does not indicate a one-to-one correspondence of facilities and is used as a gauge of differences in supply structure between prefectures.

Toward the Autumn Hot Spring Season — Ryokan Rates Rise Up to 30% from September to November

This stock analysis connects to the demand season now approaching. MetroEngines Research’s estimated settled ADR for the ryokan category shows a clear lift from September to November in many prefectures with hot spring areas. Comparing September and November on 2025 confirmed figures, Ishikawa is +30.5%, Aomori +29.3%, Fukushima +24.9%, Yamanashi +22.6% and Yamagata +22.3% — a line-up of prefectures above 20%. For actual autumn price levels at major hot spring areas tracked year on year, Autumn Onsen Ryokan ADR 2026 is also useful for pinning down where Hakone, Kusatsu and Yufuin stand.

Source: Compiled from MetroEngines Research by the HotelBank Editorial Team (Oita, ryokan category. The base for estimated settled ADR varies by month, at N=246–323 facilities per month; confirmed months are N=296–323, while for estimated months from August 2026 onward the listed snapshot base narrows to N=246–314)

Take Oita, the prefecture with the most hot spring sources. Estimated settled ADR for the ryokan category rose from about ¥17,100 in September 2025 to about ¥19,800 in November and about ¥21,500 in December — up 25.8% from September. The same shape appeared in 2024, with the annual peak arriving between autumn and early winter. Autumn foliage, seasonal food and the added appeal of a hot spring as temperatures fall all overlap, making it the point in the year when onsen inns have the most price elasticity to work with.

Source: Compiled from MetroEngines Research by the HotelBank Editorial Team (by prefecture, ryokan category, 2025 actuals. N=72–501 facilities per prefecture)

What is interesting is that Hokkaido alone moves the other way, at −3.7%. Ryokan rates in Hokkaido peak in winter (January–February) and summer (August), leaving autumn as a shoulder season. “Autumn equals the hot spring high season” does not hold uniformly nationwide; it follows each prefecture’s own demand calendar — a premise worth holding on the supply side when building a pricing strategy.

Laid over the stock figures, the supply-side implications sort into three points. First, in prefectures below the national average for lodging facilities per source — Kagoshima at 0.11, Oita at 0.18, Aomori at 0.20 and Akita at 0.33 — room remains to build lodging capacity on the source infrastructure. Second, among those, Aomori (+29.3% autumn rate lift), Fukushima (+24.9%), Yamagata (+22.3%) and Akita (+14.0%) show a clear rate lift in the autumn demand season, meaning there is a window in which added supply can be absorbed without giving up rate. Third, in prefectures such as Shizuoka and Oita where statistical onsen inns exceed the inventory visible on OTAs and similar channels, raising the online exposure of existing facilities may be the shortest path to capturing autumn demand without any new development.

For Operators: Plan-Name Cues That Work for This Theme

Frequent appeal elements in published plan names for ryokan × onsen (N=385) —

Open-air hot spring bath 90%Kaiseki course 28%Two meals included 25%Room only 23%Breakfast included 20%Buffet 11%Private bath 9%

Examples of actual names (anonymised):

  • [5% OFF + Double Points] [Major Sale] ★Online Payment Only★ [Homemade Breakfast Buffet Included / Natural Hot Spring Public Bath]
  • [Winter Fishing and Hot Springs: Limited Period] Signature “Awaji Island 3-Year Torafugu Kaiseki” — Dinner in Your Room, Breakfast Buffet
  • [5% OFF] [Major Sale] Annex “Biyu” — Suite with Open-Air Hot Spring Bath / Dinner: Standard Kaiseki / Breakfast: Japanese Set Menu <In-Room Dining>

Note: these are tendencies based on aggregated published plan names and do not indicate that naming causes sales. For a deeper look at market × naming, please refer to our naming benchmark and event articles.

⚠ Note on ADR for future dates: Months from August 2026 onward in this article are estimated from the selling prices listed on OTAs and similar channels at the time of the survey, and will shift as the check-in date approaches. Because the basis differs from confirmed figures (July 2026 and earlier), no direct year-on-year comparison between the two has been calculated.

Summary

The FY2024 hot spring stock (as of 31 March 2025) comprises 2,839 hot spring areas, 27,899 sources, 13,425 lodging facilities using hot spring water and 125.41 million annual overnight guest-nights. Hot spring areas fell by 95 over five years while lodging facilities rose by 501, thickening lodging facilities per area from 4.40 to 4.73. Annual guest-nights per lodging facility grew 1.58x, from 5,926 to 9,342.

On the density metrics, prefectures such as Mie (1.70 facilities per source) and Gunma (1.37) concentrate lodging capacity on a limited number of sources, while Kagoshima (0.11), Oita (0.18) and Aomori (0.20) have ample room to build lodging capacity on their source stock. Comparing statistical onsen inns with facilities observable on OTAs and similar channels, Shizuoka (+504) and Oita (+101) showed a base of onsen inns wider than the visible inventory.

In the autumn demand season, estimated settled ADR for the ryokan category rises as much as +30.5% from September (Ishikawa). Prefectures where spare source stock and the autumn rate lift overlap hold a window worth working on from the supply side. The figures all point in the direction of what can still be built.

Related Reading

References and Sources

■ Data sources

Hot spring area counts, total sources, sources in use / not in use, lodging facilities using hot spring water, guest capacity and annual overnight guest-nights are taken from the prefecture-level tables in the Ministry of the Environment’s “Hot Spring Utilization Survey” PDFs for FY2021–FY2024 (FY2020 totals are carried in the FY2021 edition). Estimated settled ADR for the ryokan category and the number of facilities for which prices could be observed on OTAs and similar channels come from MetroEngines Research monthly aggregates.

■ Calculation assumptions

Both density metrics are calculated from the actual figures in the original tables. Lodging facilities per source = lodging facilities using hot spring water ÷ total sources (in use plus not in use). Annual guest-nights per lodging facility = annual overnight guest-nights ÷ lodging facilities using hot spring water. The fiscal-year reference date is FY2024 = as of 31 March 2025 (guest-nights are the fiscal-year total for April 2024 – March 2025). The autumn rate change is the ratio of estimated settled ADR in September and November 2025 for each prefecture’s ryokan category, comparing confirmed months in both cases.

■ Limitations and caveats

The Ministry of the Environment statistics and our observed figures use different populations (the former a complete count of facilities using hot spring water only, the latter all segments regardless of hot springs, limited to facilities for which prices could be observed that month). The ratio between them is not a one-to-one correspondence of facilities but a gauge of differences in supply structure between prefectures. The Ministry of the Environment notes annual overnight guest-nights as a “reference figure,” and the FY2022 value was revised in the following year’s edition. The in-use / not-in-use breakdown of sources is reproduced exactly as printed in the original tables, so at the national level the two components differ from total sources by one. Estimated settled ADR is an estimate and differs from each facility’s actual transacted prices or accounting figures. Months from August 2026 onward are estimates based on listed levels at the time of the survey; because the basis differs from confirmed figures, no direct comparison is made.

Related Articles

  • JNTO Announces March 2026 Foreign Visitor Arrivals to Japan Reached 3,618,900, Up 3.5% Year-on-Year and a Record High for March

  • Golden Week 2026 Hokkaido Hotel Price Analysis: Niseko +29% and the Drivers Behind the Surge in Sell-Out Rates

  • Post-Golden Week Hotel Prices Drop Up to 44%: Why Mid-May Is the Best Time to Book

  • Golden Week 2026 Hotel Price YoY Analysis Across Six Major Cities: Unpacking the Drivers Behind Kyoto (+20%) and Tokyo (+17%)