I found eight first-order benchmarks, including the Japanese datum of leveling, on my walk from Tokyo to the National Diet Building!


The elevation in Japan is the height of Tokyo Bay from mean sea level.

In May 1894, a mark was placed at 24.500 meters above the mean sea level at the water level observation station on Reigan Island, and this mark was designated as the Japanese datum of leveling. The elevation of the whole of Japan is calculated from the difference in height between this point and the Japanese datum of leveling, and the value of 24.500m was chosen because it was in May of Meiji 24 (1891). However, due to earthquakes and other factors, the elevation of the level origin was revised to 24.3900m on October 21, 2011.

Today, I took a walk around the route from the water level observation station on Reigan Island, the key point of Japan’s level observation, to the Japanese datum of leveling.

Advance Preparation

Before going out for a walk, I checked how to measure elevation and the first level points around the target area.

How to measure elevation

The method of measuring a level point (first class) is as follows [1].

(1) Place a 3-meter leveling rod vertically at each place where you want to measure the difference in altitude.
(2) Place a level horizontally in the center of the leveling rod, and read the scales of both scales to the nearest 0.1 mm.
(3) The distance between the level and the leveling rod should be within 50 m.
(4) The distance between the level and the scale should be 2 km, and round trip observation should be carried out.
(5) For a distance of 2 km (=S), the round trip error should be within 3.5 mm (2.5 x √S).

Since the maximum distance between the leveling rod and the levelling rod is 100m, 40 measurements are required. And the error must be within 3.5mm. Surveying is a job that requires a lot of nerve.

First-order benchmarks around the target area

On the GSI page, there is a “Reference Point Data Browsing Service” that allows you to look up various reference points.

I found out that there are two first-order benchmarks, [15-001] and [Kou 7-7], in Ginza 3-chome and Tsukiji 4-chome, respectively, between the water level observation at Reiganjima and the Japanese datum of leveling, and decided to include them in my walking course.

Birthplace of the unskillful actor?

For the original purpose, I should have started my walk from the water level observation station at Reigishijima. However, since I had already walked from Reigishijima to Tokyo Station last time, I decided to start from Tokyo Station.


Grand Roof on the Pedestrian Deck with “Sails of Light” as its motif

I wanted to head towards Ginza, so I left the Yaesu exit and headed south along the station. Then I found a staircase to the top. I went up and found that it was a pedestrian deck (aerial walkway). It’s called the Grand Roof.

I wrote about my last walk, “Discovered the Reigishijima Water Level Observatory! in which I wrote about my last walk, I wrote that the exterior of Tokyo Station was “not designed in a Japanese way. This time, I found out that the grand roof was designed by Helmut Jahn, a German-American. It was, after all, designed by a foreigner. Intuition is rarely right.


Birthplace of Edo Kabuki

Cross the Kajibashi intersection and continue along Kajibashi Street to Chuo Street. Turn right, and after a while, you will see a monument to the “Birthplace of Edo Kabuki”.


Kyobashi Daikon Kashi Aomono Ichiba Ato

As I approached and read about the origin of the market, I saw a monument on the right hand side that said, “Kyobashi Daikon Kashi Aomono Ichiba Ato.” Nowadays, fashionable Kyobashi and Ginza used to be places where people mainly sold daikon (Japanese white radish).

By the way, there is a word, “daikon yakusha (unskillful actor).” There are various theories about the origin of the word, but since kabuki and daikon are used together in this area, this may be the birthplace of daikon yakusha.

I found the first-order benchmark [15-001] in front of Matsuya Ginza!


Installation state of the first-order benchmark [15-001]

The crowd, which was sparse around Kyobashi, became very crowded around Ginza 3-chome.
In the midst of all this, I found a first-order benchmark [15-001] on the corner of a planted area! It’s called 15-001 because it’s the first level point counting from Nihonbashi on Route 15.

I thought the level point was made of granite, so the metal marker did not come into view easily. And instead of a protective stone, it’s a plant like ophiopogon japonicus.

However, there’s a big 001 written on the surrounding stones.

This 15-001 is the only first-order benchmark that is within 2 km of both the water level observation station on Reigan Island and the Japanese datum of leveling, which is about 1.5 km and 1.8 km, respectively. Therefore, I presume that this level point is used to confirm the elevation of the current Japanese datum of leveling (personal estimation).

I found the first-order benchmark [Kou 7-7] near the intersection of the east side of the Uneme bridge!


Relief of a dog in Tsukiji Eto Building

I turned left along Matsuya, and on the corner of the next street after crossing the Metropolitan Expressway, there was a building with reliefs of the Chinese zodiac lining the walls. This is the Tsukiji Eto Building. Since the New Year is coming, I took a picture of a dog.

Turn left to the southwest along the Tsukiji To Building, and turn right when you hit the National Cancer Center, which leads you to the intersection at the east side of the Uneme Bridge. There is supposed to be a first-order benchmark [Kou 7-7] near here, but I can’t find it.


Installation state of the first-order benchmark [Kou 7-7]

It was actually buried underground. A map of Japan and a picture of a surveyor are drawn on the round iron cover in the foreground.

The name of the level point seems to have the character for “Kou (crossing)” because the level lines intersect. I’m not sure what 7-7 means.

The iron lid was marked “Basic”. Also, this benchmark is shown on the map of the Tokyo metropolitan area of the first-order benchmark route map in Reference [1]. Therefore, it seems to be one of the most important benchmarks.

Incidentally, there is an Inuyarai in the back of the photo. This is the second time I’ve seen an Inuyarai in Tokyo.

Found the first-order benchmark group!

Now it’s time to move on to the Japanese datum of leveling.


Ministry of Finance

As I passed the side of Hibiya Park, I saw the Ministry of Finance. It looks simpler than I expected.



From there, it was on to Shiomi-zaka. At this point, we gained more than 10 meters in altitude at once. The Diet Building is now slightly visible on the right.


Installation of the first-order benchmark “Tei”

I arrived at the front yard of the Diet and looked for the first-order benchmark ”Tei.”

It had the appearance of a level point I knew well, so I found it easily.

The Sakurada Moat of the Imperial Palace can be seen a little ahead. The road between the moat and the moat is Route 20. It is also the old Koshu Kaido. It’s around the willow well.

For some reason, there was a bamboo broom lying in front of it.

The other first-order benchmarks are located about 6 meters up the hill.


Installation of the first-order benchmark “Otsu”

First, I found ”Otsu” in the grass!


Installation of the first-order benchmark “Hei”

Then I found ”Hei” near the datum of leveling!


Installation of the first-order benchmark “Kou”

Furthermore, I found the “Kou” almost in front of the datum of leveling!


Installation of the first-order benchmark “Bo”

And I found “Bo” right near the Kensei Kinenkan!

Six reference benchmarks have been established in Tokyo. The standard benchmark is the next most important benchmark after the datum of leveling. The five points shown here, from “Kou” to “Bo”, are the standard benchmarks. The density is amazing.

The other one is located in Hachioji. It is located on the left side of the Keio Takao Line just before it enters a tunnel after leaving Takaosanguchi Station.

Actually, there are 13,722 first-order benchmarks in Japan, almost as many as the number of Lawson stores. It is not too unusual to find one. Finding 5 out of 6 reference benchmarks in Tokyo is much more rare. However, I used first-order benchmark as the title of the article because not many people search for the reference benchmark.

The second-order benchmark, by the way, has 3,111 points nationwide, which is less than a quarter of the first-order benchmark, so the second-order benchmark is rarer.

This relationship is different from that of the triangulation point. 975 points for the first order triangulation point and 5,028 points for the second order triangulation point, making the second order more than five times as much as the first order. The first order triangulation point is 975 and the second order triangulation point is 5,028.

The first-order benchmark and the second-order benchmark have different numerical resolutions: the first-order benchmark shows survey results in 0.1 mm units, while the second-order benchmark shows survey results in 1 mm units.

Hypothesis about reference benchmark

After finding the reference point, we walked counterclockwise along the Parliament building and returned from Kokkaigijidoumae station.

On the way, I saw the Prime Minister’s residence, so I took a picture of it. However, the photo was not in focus because of the backlighting and the reflection of the sun, so I did not post it.

Instead, I made some hypotheses about the reference benchmarks and discussed them. However, there may be some calculation errors.

Isn’t it possible that the iron lid was built at different times?

I have put together enlarged photos of the first level point I found at the top of the article. While I was compiling these photos, I came up with the following hypothesis.


(1) The iron covers of “Kou” and “Bo” were made together.
(2) The iron lids of “Otsu” and “Hei” were made at different times, and also different from (1).


The font of the iron lid of “Kou” and “Bo” is Gothic. But the font of the iron lid of “Otsu” and “Hei” is Mincho. If they were to be built together, they would not have bothered to change the font.

As for the Gothic font of “Kou” and “Bo”, they are almost the same and I could not tell them apart. Therefore, it can be assumed that they were manufactured together.

Then the lid of ”Kou” was damaged, so when the lid of “Bo” was made, it must have been renewed together.

The Mincho font of “Otsu” and “Hei” are quite similar, but “Otsu” has thinner horizontal lines. This means that they were made at different times.

Isn’t the “Japanese datum of leveling” and “Kou” along the magnetic north line?

Looking at the map of the Reference Point Data Browsing Service, “Kou” is located slightly westward and northward of the “Japanese datum of leveling”. Therefore, I made a hypothesis that the straight line connecting the “Japanese datum of leveling” and “Kou” shows the magnetic north line.

I calculated the angle of this deviation from latitude and longitude, and it is about 4 degrees. The magnetic north line has a deviation of about 7 degrees, so unfortunately, it does not seem to be related to magnetic north.

Isn’t “Kou”, “Otsu”, and “Hei” at the apex of an equilateral triangle?

The map of the Reference Point Data Browsing Service shows that “Kou”, “Otsu”, and “Hei” are at the top of an equilateral triangle.

I also calculated the distance between them based on latitude and longitude. The results are (“Otsu”-“Kou”: “Hei”-“Osu”: “Kou”-“Hei”) = (19.5 m, 22.4 m, 20.2 m). Although it is quite close to an equilateral triangle, it does not seem to be an equilateral triangle.

Can the elevations of the Japanese datum of leveling and the standard benchmark be directly compared?

In the photograph showing the state of “Kou”, there is a cylindrical stone in the middle of the datum of leveling. It is almost equidistant from the datum of leveling, “Kou”, “Otsu”, and “Hei”. If the level is installed here, it will be possible to directly compare them with each other.

The difference in elevation between “Tei” and the datum of leveling is 7.77m, so it cannot be directly compared with the 3m scale.

For “Bo”, the distance to the datum of leveling is calculated to be 91.7m, so it can be directly compared by placing the level at an appropriate position. The elevation difference is 139.4 mm.

Walking data

Course: JR Tokaido Line Tokyo Station -> Japanese datum of leveling → Tokyo Metro Marunouchi Line Kokkai-gijidomae Station
Distance: 6.7km
Time: 1h39m


[1] 国土交通省国土地理院:一等水準点検測成果集録、第59巻、平成28年3月