Sure, give me a few hours. You’d speed up that process if I know in what country you live. That way I only have to loop through the cities in your country instead of the entire world. Or you could tell me your city.
I feel the issue you mentioned with the coordinates not representing the actual city centres heavily benefitted the US here. Looking at the image you showed for Tulsa, the downtown area isn't even shown. Milwaukee has Dineen Park in it which is 7km+ away from the city center. That's obviously not right as American suburbs are much more green than downtowns
I think the reason that the US dominated the list is because suburban communities are so common here (and an unfortunately large part of the culture); like you said, each house has its own yard and its own tree. It's funny how it's reflective of the individualist culture here -- how many people want their own picturesque home.
Yeah exactly. I live in Sydney, Australia and while I haven't been, I've had friends that have been to a number of those cities in the US and they've all said Sydney is way more leafy. It also has to do with density, the suburban spaces in the US are so huge that it skews the results, so the city centre can just be a concrete jungle if there's large enough suburbs. Meanwhile here in Sydney you can be in the CBD and have forested areas pocketed away. And generally here, people don't have lawns like they do in the US with one or two trees and that's it - my backyard has about 10 large trees and many more smaller trees and saplings.
The Netherlands has awarded a 'greenest city award' a couple of times to encourage municipalities to add greenary to their cities. Quite a nice initiative imo
The better way to do this would have been using the pre existing Normalized difference vegetation index that relies on infrared light reflectance of live plants, remote sensing data and satelittle views of this exists, and I would be interested in a comparison between the two.
Hey man! I wanted to make a comment on your previous video, but forgot to finish it, so I'm writing it here. Your analyses are very high quality, which is clearly seen by your eye for detail. Manually handling mismatches in specific instances (not even groups of instances) is a sign of a great data analyst, trust me - I know many analysts who just do basic macro-level data cleaning, run some measures and call it a day. Great to see that you are doing your projects with passion. As a network scientist, I have some topics for you that could be interesting to look into when you do your work: -network filtering: I recall in your previous video filtering cities by geographical radius; I think a worthy approach is to rather extract the so-called network backbone (this is typically for weighted networks, but can be run also on unweighted networks), you can filter it to any node/edge amount too. I know the disparity filter to be most common as choice. -network complexity: there are some entropy based measures for complexity, i.e. "minimum description length". The basic idea is a network is as complex, as many bits of information you need to reconstruct it - very structured cities would not be considered complex, e.g. grid structures. I know some algorithms, like the community detection algorithm Infomap or nested SBMs rely on this. Make a video where you detect communities in cities by network structure! That is called community detection in network science.
Maybe another reason why those US cities are so green is that they are inland? That means no area is taken up by beaches, port infrastructure, etc that generally isn't green. Would also explain why that Kazakh city is so high up the list.
Yeah this is a case where the methodology is just not up to the task at hand. One obvious improvement is to weight the results according to population as right now you end up comparing US suburbia which is definitely very green but has very low density with cities that are way more dense. Nice effort though, keep up the good work.
@@Amechaniaa then well the bigger the city, the more greens it got of course. If we just measure on the green pixels directly then cities like Chongqing would dominate the board since you can't even make out the urban area looking at it's border from above
@@haigiabaonguyen9299you could probably make it an average green instead of a total green, which would be more honest anyway regardless of using the borders or not
wow very unexpected results, didn't expect us to have such green cities. You could have probably somehow crop the satellite footage to the city shape using shp files it would have been also easier to if you have used satellite imaginary in GeoTIF format
from what I could tell, the main reason for northamerican suburbia to dominate the list is the quantity of clusters that are listed/counted because of single homes' lawns and their average one to three trees planted in the back; big city parks or general urban development with a denser presence of trees (typically seen in west/central europe, fe), when viewed from satellite imagery, tends to become big singular clusters. I also support the point that, although not being the purpose of the project (nothing wrong with it, it has its own value and significance), the "greenage" measuring does not necessarily correlate to quality of life or sustainable development, seeing as in these green-dominated american towns, it is strictly caused by inefficient land use, leaving a LOT of empty, practically dead grassy' space behind; this could be traced to many factors, but overall, poor modern city planning philosophy (suburbia), private property inequality, so to say, and, on top of it all, a heavy motormobile oriented infrastructure. My suggestion for a next vid, I might add, would be 'cities with the most uniform buildings', that possible? lol. Great video as always man, can't imagine the work that's been put it sometimes.
It's much easier to separate vegetation in an image if you use multispectral imagery like Landsat/Sentinel - taking an image in thermal infrared easily seperates water, which shows up almost black. There are plenty of established tools like NDVI that are used in industry/science for this - you can also do intereseting things like track the water content of leaves over time, so identify fire-prone regions just from a satellite image (more water in leaves absorbs more infrared light). With enough data and some of the more interesting machine-learning tools, you can identify individual species of vegetation, but even with 'dumb' classification methods, it's simple to split forest from grassland/pasture.
Love your videos! Keep it up man. But could you upload your results somewhere? Sometimes I would really like to browse trough them but sadly there is no way
Yeah the “number of green clusters” condition introduces a sort of perverse incentive. I figure that SE USA pine forest cities are reading as one large cluster of green
I have a question how can I calculate connection of roads like in last video. I downloaded qgis but I don't know how can I get most road intersection nodes count etc?
Aw dang,i wanted to make this 900+ long list but i realized that the ranking number was wrong,so even though i fixed the speed problem (by placing it at 0.25x) there was still possibly some cities that were skipped over unintentionally,massive apologies for anyone who wanted a full list,since i was really willing too :(
ein Beitrag an Montag, 27. Januar 2025 Hello from Germany! I hope, that you enjoyed Australia Day! "the [annual] International Open Tennis Championships of Australia" from Sunday, 12th January 2025 to Sunday, 26th January 2025 Jannik Sinner, is an Austrian, from South Tyrole (Italian Republic). His mother tongue in German. Alexander Swerew, in a Russian, who reside in Hamburg (Federal Republic of Germany). His mother tongue in Russian. the date "27th of January " in history, e. g. ° 1859: Wilhelm II., the last German Emporer, was born. He abdicated 1918! Wilhelm died 1941, with 82 years of age! ° 1945: Soviets conquered the, German, Concretration Camp of Auschwitz-Birkenau (Generalgouvernment, in Poland)! ° 2016: Johannes Rau (Social Democrats), a former Federal President of Germany, died in the age of 85 years!
![Lp. Сердце Вселенной #60 РОЖДЕНИЕ ЛОЛОЛОШКИ [Финал] • Майнкрафт](http://i.ytimg.com/vi/YoR0pAV9FVQ/mqdefault.jpg)
someone make list of these all cities in order by greenness so that i can find where my city is ranked
He did in the video
@@theallduck Not really, each city took a frame and is hard to search through
Yess I bet mine is one of the last
Sure, give me a few hours. You’d speed up that process if I know in what country you live. That way I only have to loop through the cities in your country instead of the entire world. Or you could tell me your city.
@ Kathmandu, Nepal. If u really find that for me, bro ill suck it for free! Full Throttle! Dont worry, that city is there in the video
I feel the issue you mentioned with the coordinates not representing the actual city centres heavily benefitted the US here. Looking at the image you showed for Tulsa, the downtown area isn't even shown. Milwaukee has Dineen Park in it which is 7km+ away from the city center. That's obviously not right as American suburbs are much more green than downtowns
I think the reason that the US dominated the list is because suburban communities are so common here (and an unfortunately large part of the culture); like you said, each house has its own yard and its own tree. It's funny how it's reflective of the individualist culture here -- how many people want their own picturesque home.
And because of that everyone has to drive🤧
@@li_tsz_fung its also the same almost everywhere else lol, its only the big cities and tourist cities in Europe that are walkable, like in the us
Idk if each house having individual yard is bad
I wish there was a way to distinguish gras from trees. A stale green lawn is not quality greenery in my view.
Yeah exactly. I live in Sydney, Australia and while I haven't been, I've had friends that have been to a number of those cities in the US and they've all said Sydney is way more leafy. It also has to do with density, the suburban spaces in the US are so huge that it skews the results, so the city centre can just be a concrete jungle if there's large enough suburbs. Meanwhile here in Sydney you can be in the CBD and have forested areas pocketed away. And generally here, people don't have lawns like they do in the US with one or two trees and that's it - my backyard has about 10 large trees and many more smaller trees and saplings.
The Netherlands has awarded a 'greenest city award' a couple of times to encourage municipalities to add greenary to their cities. Quite a nice initiative imo
1. Tulsa (US) 2. Fresno (US) 3. San Antonio (US) 4. Milwaukee (US) 5. Shymkent (KAZ) 6. Louisville (US) 7. Tampa (US) 8. New York (US) 9. Indianapolis (US) 10. Houston (US)
11. Des Moines (US) 12. Kryvyi Rih (UKR) 13. Hamamatsu (JPN) 14. Concord (US) 15. Adelaide (US) 16. Austin (US) 17. McAllen (US) 18. Essen (GER) 19. Saitama (JPN) 20. Hachioji (JPN)
21. Rochester (US) 22. Palm Bay (US) 23. Raleigh (US) 24. Kitchener (CAN) 25. Düsseldorf (GER) 26. Queens (US) 27. Sacremento (US) 28. Detroit (US) 29. Lodz (POL) 30. Nampula (MOZ)
31. Shizuoka (JPN) 32. Kawaguchi (JPN) 33. Orlando (US) 34. Baton Rouge (US) 35. Nishniy Novgorod (RUS) 36. Columbus (US) 37. Akron (US) 38. Richmond (US) 39. Tashkent (UZB) 40. Jerusalem (ISR)
41. Dayton (US) 42 .Krakow (POL) 43. Portland (US) 44. Munich (GER) 45. Provo (US) 46. Hannover (GER) 47. Lviv (UKR) 48. London (UK) 49. Wroclaw (POL) 50. Nottingham (UK)
51. Glasgow (UK) 52. Stuttgart (GER) 53. Charlotte (US) 54. Frankfurt Main (GER) 55. Vilnius (LIT) 56. Omaha (US) 57. Blantyre (MWI) 58. Berlin (GER) 59. Kottayam (IND) 60. Grand Rapids (US)
61. Bronx (US) 62. Riverside (US) 63. Toronto (CAN) 64. Calgary (CAN) 65. Rotterdam (NLD) 66. Vancouver (CAN) 67. Chelyabinsk (RUS) 68. Ulsan (KOR) 69. Brooklyn (US) 70. Buffalo (US)
71. Saint Petersberg (RUS) 72. Providence (US) 73. Pietermaritzburg (ZAF) 74. Xiangyang (CHN) 75. Stockholm (SWE) 76. Poznan (POL) Chengdu(CHN) 78. Murica (SPA) 79. Albany (US) 80. Cincinnati (US)
81. Winnipeg (CAN) 82. Columbia (US) 83. ST. Louis (US) 84. Johannesburg (ZAF) 85. Novosibirsk (RUS) 86. Bucheon (KOR) 87. Dortmund (GER) 88. Buenos Aires (ARG) 89. Rangoon (MMR) 90. Beijing (CHN)
91. Anyang (KOR) 92. Muntilupa City (PHL) 93. Colorado Springs (US) 94. Barinas (VEN) 95. Moscow (RUS) 96. Boston (US) 97. Auckland (NZ) 98.Knoxville (US) 99. Lubango (ANG) 100. Warsaw (POL)
To be continued...
Come on, you can do it!
Uh... Provo?
No Riga is bullshit
207: Athens, Greece
@MapsCharts he say tree not forest
That one person who is going to write the ranking list of all 933 cities in the comments is going to be a hero!
The better way to do this would have been using the pre existing Normalized difference vegetation index that relies on infrared light reflectance of live plants, remote sensing data and satelittle views of this exists, and I would be interested in a comparison between the two.
Awesome video, this was never something I’d have thought of
4:27 thanks for mentioning my little city :). I enjoy a lot your content.
0:46 JAKARTA MENTIONED!1!!1!!🇮🇩🇮🇩🇮🇩🇮🇩🇮🇩
😂Bahakan 15 km cuma setengah kota jakarta yang terfoto
Hey man! I wanted to make a comment on your previous video, but forgot to finish it, so I'm writing it here.
Your analyses are very high quality, which is clearly seen by your eye for detail. Manually handling mismatches in specific instances (not even groups of instances) is a sign of a great data analyst, trust me - I know many analysts who just do basic macro-level data cleaning, run some measures and call it a day. Great to see that you are doing your projects with passion.
As a network scientist, I have some topics for you that could be interesting to look into when you do your work:
-network filtering: I recall in your previous video filtering cities by geographical radius; I think a worthy approach is to rather extract the so-called network backbone (this is typically for weighted networks, but can be run also on unweighted networks), you can filter it to any node/edge amount too. I know the disparity filter to be most common as choice.
-network complexity: there are some entropy based measures for complexity, i.e. "minimum description length". The basic idea is a network is as complex, as many bits of information you need to reconstruct it - very structured cities would not be considered complex, e.g. grid structures. I know some algorithms, like the community detection algorithm Infomap or nested SBMs rely on this.
Make a video where you detect communities in cities by network structure! That is called community detection in network science.
Ukrainian cities in the list:
8:39 Kryvyi rih (58,80% green)
8:41 Lviv (50,89% green)
8:47 Kyiv (45,43% green)
8:48 Donetsk (44,29% green)
8:49 Kharkiv (43,53% green)
8:49 Zaporizhzhia (43,49% green)
8:58 Dnipro (38,68% green😢)
9:00 Odesa (37,11% green)
I hope i didn’t loose anything.
Maybe another reason why those US cities are so green is that they are inland? That means no area is taken up by beaches, port infrastructure, etc that generally isn't green. Would also explain why that Kazakh city is so high up the list.
Yeah this is a case where the methodology is just not up to the task at hand. One obvious improvement is to weight the results according to population as right now you end up comparing US suburbia which is definitely very green but has very low density with cities that are way more dense.
Nice effort though, keep up the good work.
Love your videos bro. Interesting stuff, and educational.
Can't you get the city borders and then crop the images to only include the parts within the borders?
No
@JosephsJungle8 Elaborate?
Yeah definitely with a software like QGIS. The problem might be with finding a dataset with the borders of all those cities though
@@Amechaniaa then well the bigger the city, the more greens it got of course. If we just measure on the green pixels directly then cities like Chongqing would dominate the board since you can't even make out the urban area looking at it's border from above
@@haigiabaonguyen9299you could probably make it an average green instead of a total green, which would be more honest anyway regardless of using the borders or not
wow very unexpected results, didn't expect us to have such green cities. You could have probably somehow crop the satellite footage to the city shape using shp files it would have been also easier to if you have used satellite imaginary in GeoTIF format
from what I could tell, the main reason for northamerican suburbia to dominate the list is the quantity of clusters that are listed/counted because of single homes' lawns and their average one to three trees planted in the back; big city parks or general urban development with a denser presence of trees (typically seen in west/central europe, fe), when viewed from satellite imagery, tends to become big singular clusters.
I also support the point that, although not being the purpose of the project (nothing wrong with it, it has its own value and significance), the "greenage" measuring does not necessarily correlate to quality of life or sustainable development, seeing as in these green-dominated american towns, it is strictly caused by inefficient land use, leaving a LOT of empty, practically dead grassy' space behind; this could be traced to many factors, but overall, poor modern city planning philosophy (suburbia), private property inequality, so to say, and, on top of it all, a heavy motormobile oriented infrastructure.
My suggestion for a next vid, I might add, would be 'cities with the most uniform buildings', that possible? lol.
Great video as always man, can't imagine the work that's been put it sometimes.
It's much easier to separate vegetation in an image if you use multispectral imagery like Landsat/Sentinel - taking an image in thermal infrared easily seperates water, which shows up almost black. There are plenty of established tools like NDVI that are used in industry/science for this - you can also do intereseting things like track the water content of leaves over time, so identify fire-prone regions just from a satellite image (more water in leaves absorbs more infrared light). With enough data and some of the more interesting machine-learning tools, you can identify individual species of vegetation, but even with 'dumb' classification methods, it's simple to split forest from grassland/pasture.
Your accent is unique. Where are you from?
He is a Cypriot.
Very happy to see Adelaide at number 15 in the world! The massive parklands around the city help
Very good video! I'd love to see what you measure next for a whole bunch of cities
Moscow or Singapore
Was Edmonton included in the list of cities where the green was looked at? If so, I'm REALLY surprised Edmonton did not make the top 10 list
he is gonna be the big youtuber i can say i found when he was still very small
Love your videos! Keep it up man. But could you upload your results somewhere? Sometimes I would really like to browse trough them but sadly there is no way
Really surprised cities in the southeastern US are not on top like Atlanta or Charlotte most of their suburbs look like forests
Yeah the “number of green clusters” condition introduces a sort of perverse incentive. I figure that SE USA pine forest cities are reading as one large cluster of green
I have a question how can I calculate connection of roads like in last video. I downloaded qgis but I don't know how can I get most road intersection nodes count etc?
Woah this probably required a lot of time
Manaus. Mainly because its in the amazon
Which pixel was green the longest in this video?
Could you release the data? The list at 8;25 is incredibly hard to search through
I don't know about that but the most concrete is definitely Istanbul
If you don't have 100k subs by the end of the year I'm flipping my desk
Apparently, according to the UN classification, London is a forest because it has so many trees per square kilometre.
"a city djungle" ;)
Where does Buenos Aires, Argentina rank?
1st Indian city is Chandigarh, at rank 128
My city, Delhi, is at rank 191
Nice video
uhm.. atlanta is forest first
you haev such interesting ideas
As an oklahoman I approve of this message 👍
My town in Estonia is 40% pine forest and probably 80% is green
damn i did not expect my city to be ranked 15th
Aw dang,i wanted to make this 900+ long list but i realized that the ranking number was wrong,so even though i fixed the speed problem (by placing it at 0.25x) there was still possibly some cities that were skipped over unintentionally,massive apologies for anyone who wanted a full list,since i was really willing too :(
Greenest cities are just suburbs
0:05 as of now my prediction is Ljubljana
Nice
I’m a simple man.
I see Indianapolis (my hometown), I click.
This guy is underrated 😢
Atlanta?
this guy is a genious
prediction: Detroit
"SUBSCRIBE!!!" lol
the greenest city is booger city
i live in shymkent yoooo 6th place
Omg what accent is this? 😭
San Francisco is 716
this feels very poorly done
Not everything is so simple, karen
how about perth im interested about my own city
ein Beitrag an Montag, 27. Januar 2025
Hello from Germany!
I hope, that you enjoyed Australia Day!
"the [annual] International Open Tennis Championships of Australia"
from Sunday, 12th January 2025 to Sunday, 26th January 2025
Jannik Sinner, is an Austrian, from South Tyrole (Italian Republic). His mother tongue in German.
Alexander Swerew, in a Russian, who reside in Hamburg (Federal Republic of Germany). His mother tongue in Russian.
the date "27th of January " in history, e. g.
° 1859: Wilhelm II., the last German Emporer, was born. He abdicated 1918! Wilhelm died 1941, with 82 years of age!
° 1945: Soviets conquered the, German, Concretration Camp of Auschwitz-Birkenau (Generalgouvernment, in Poland)!
° 2016: Johannes Rau (Social Democrats), a former Federal President of Germany, died in the age of 85 years!