• More from M-W
  • To save this word, you'll need to log in. Log In

Definition of ghost town

Examples of ghost town in a sentence.

These examples are programmatically compiled from various online sources to illustrate current usage of the word 'ghost town.' Any opinions expressed in the examples do not represent those of Merriam-Webster or its editors. Send us feedback about these examples.

Word History

1894, in the meaning defined above

Dictionary Entries Near ghost town

ghost story

Cite this Entry

“Ghost town.” Merriam-Webster.com Dictionary , Merriam-Webster, https://www.merriam-webster.com/dictionary/ghost%20town. Accessed 20 Jan. 2024.

Kids Definition

Kids definition of ghost town, more from merriam-webster on ghost town.

Nglish: Translation of ghost town for Spanish Speakers

Subscribe to America's largest dictionary and get thousands more definitions and advanced search—ad free!

Play Quordle: Guess all four words in a limited number of tries.  Each of your guesses must be a real 5-letter word.

Can you solve 4 words at once?

Word of the day.

See Definitions and Examples »

Get Word of the Day daily email!

Popular in Grammar & Usage

8 grammar terms you used to know, but forgot, homophones, homographs, and homonyms, commonly misspelled words, a guide to em dashes, en dashes, and hyphens, absent letters that are heard anyway, popular in wordplay, the words of the week - jan. 19, 10 words for lesser-known games and sports, your favorite band is in the dictionary, etymologies for every day of the week, 7 common idioms that come from chickens, games & quizzes.

Play Blossom: Solve today's spelling word game by finding as many words as you can using just 7 letters. Longer words score more points.

ghost cities definition

  • History Classics
  • Your Profile
  • Find History on Facebook (Opens in a new window)
  • Find History on Twitter (Opens in a new window)
  • Find History on YouTube (Opens in a new window)
  • Find History on Instagram (Opens in a new window)
  • Find History on TikTok (Opens in a new window)
  • This Day In History
  • History Podcasts
  • History Vault

6 Famous Ghost Towns and Abandoned Cities

By: Evan Andrews

Updated: September 14, 2023 | Original: March 10, 2015

Ghost Towns and Abandoned Cities

1. Pripyat, Ukraine

ghost cities definition

At 1:23 a.m. on April 26, 1986, a catastrophic meltdown took place inside reactor number four at the Soviet nuclear power plant at Chernobyl. The explosion that followed sent flames and radioactive material soaring into the skies over Pripyat, a nearby city built to house the plant’s scientists and workers. It took 36 hours before the town’s 49,000 residents were evacuated, and many later suffered severe health effects as a result of their brief exposure to the fallout.

Soviet authorities later sealed off an 18-mile exclusion zone surrounding Chernobyl, leaving Pripyat an abandoned ghost town. The city has since languished for nearly three decades as a chilling reminder of the disaster. Its buildings have decayed and been partially reclaimed by the elements, and wild animals roam through what were once bustling apartments, sports complexes and an amusement park.

In the town post office, hundreds of letters from 1986 still sit waiting to be mailed. While radiation levels in Pripyat have dropped enough in recent years to allow urban explorers and former residents to make brief visits, scientists estimate that it could take several centuries before the town is once again safe for habitation.

2. Oradour-sur-Glane, France

Oradour-sur-Glane, France

On the afternoon of June 10, 1944, the village of Oradour-sur-Glane was the scene of one of the worst massacres of French civilians during World War II. In what is believed to have been an act of revenge for the town’s supposed support of the French Resistance, a Nazi Waffen SS detachment rounded up and murdered 642 of its residents and burned most of their houses to the ground. The men were taken to barns and machine-gunned, and the women and children were locked in a church and killed with explosives and incendiary grenades. Only a handful of people managed to survive by playing dead and later fleeing to the forest.

A new Oradour-sur-Glane was built nearby after the war ended, but French President Charles de Gaulle ordered that the burned-out ruins of the old town be left untouched as a monument to the victims. The facades of dozens of brick buildings and charred storefronts still remain, as well as graveyards of rusted cars and bicycles, scattered sewing machines and unused tram tracks. The site is also home to a museum, which holds a collection of relics and mementos recovered from the rubble.

3. Hashima Island, Japan

Hashima Island, Japan

Today, Hashima Island is a vacant labyrinth of crumbling concrete, sea walls and deserted buildings, yet it was once among the most densely populated places on the planet. The small island off the coast of Nagasaki was first settled in 1887 as a coal mining colony. It was later purchased by Mitsubishi, which built some of the world’s first multi-story, reinforced concrete buildings to house its bursting population.

Hashima remained a hive of activity for the next several decades, especially during World War II when the Japanese forced thousands of Korean laborers and Chinese POWs to toil in its mines. By the 1950s, the 16-acre rock was packed to the gills with more than 5,200 residents. Most workers found the cramped conditions unlivable, and the city was promptly abandoned after the mine closed in 1974.

Forty years of neglect have left Hashima a dilapidated ruin of collapsed staircases and condemned apartments. Many of its high-rises are still filled with old televisions and other relics from the mid-20th century, and its once-teeming swimming pools, barbershops and school classrooms now sit in shambles. The island was officially opened to tourists in 2009, and it has since served as the inspiration for the villain’s hideout in the 2012 James Bond film “Skyfall.”

4. Varosha, Cyprus

Varosha, Cyprus

In the early 1970s, the immaculate beaches of Varosha, Cyprus served as one of the most popular millionaires’ playgrounds in the Mediterranean. The suburb boasted a thriving tourism economy, and celebrities such as Elizabeth Taylor and Brigitte Bardot were known to take in the sand and sun at its high-class beachfront hotels.

All that changed in August 1974, when Turkey invaded Cyprus and occupied its northern third in response to a Greek nationalist-led coup. Varosha’s 15,000 residents fled the city in terror, leaving their valuables and livelihoods behind. Most assumed they would return once the fighting stopped, but ongoing political strife has seen Varosha waste away behind a heavily-guarded barrier ever since.

The few intrepid explorers who have ventured into the no man’s land describe the resort as a crumbling ghost town. Trees have grown through the floors of restaurants and homes, and most of the former residents’ belongings have been looted or destroyed. What is left stands as a spooky time capsule of the 1970s, including bellbottoms in shop windows and 40-year-old vehicles still parked at car dealerships. In recent years, Greek and Turkish Cypriots have held talks regarding reopening the former jet-setters’ haven, but experts estimate that it would take upwards of $12 billion to make its decrepit buildings livable again.

5. Bodie, California

Bodie, California

Bodie, California was officially founded in 1876 after miners stumbled upon rich deposits of gold and silver in its hillsides. Gold-crazed prospectors flocked to the settlement at a rate of more than two dozen per day in the late-1870s, and its population eventually soared to some 10,000 people. Thanks to larger-than-life accounts of whiskey-fueled shootouts, the outpost soon earned a reputation as a “sea of sin” filled with rough men, prostitutes and opium dens.

Like most boomtowns, Bodie eventually went bust. By the 1880s, it had outgrown its meager infrastructure, and a succession of harsh and deadly winters convinced many of its prospectors to move to more profitable locales. The population dwindled until the 1940s when the last residents finally shipped out.

Since then, Bodie has become known as one of the nation’s most well-preserved ghost towns. Its 200 ramshackle buildings are kept in a state of “arrested decay” by park rangers, and tourists flock to the site to explore its 1880s Methodist church, saloons and post office as well as the ruins of a burned-out bank vault.

6. Fordlandia, Brazil


In 1927, Henry Ford began work on “Fordlandia,” a massive rubber plantation in the jungles along Brazil’s Tapajós River. The automotive magnate needed the town as a steady source of rubber for his car tires and hoses, but he also saw the venture as a chance to bring small-town American values to the Amazon. Having already left his mark on cities like Dearborn, Michigan, he designed a company town complete with swimming pools, a golf course, suburban-style bungalows and weekly square dancing sessions.

Unfortunately for Ford, his experiment was doomed almost from the start. Fordlandia’s rubber trees fell victim to leaf fungus, and its employees chafed under the town’s strict regulations, which included a ban on alcohol. Clashes between Brazilian laborers and American managers soon became a common occurrence. During one riot over cafeteria rules, Fordlandia’s employees destroyed most of their mess hall with machetes and pushed the town’s trucks into the river.

Henry Ford eventually sank $20 million into his would-be workers’ paradise, but the town failed to produce any latex for his automobiles. Having never visited the city himself, he finally sold it to the Brazilian government in 1945 for pennies on the dollar. The wilderness has reclaimed large portions of Fordlandia’s campus in the years since, but many of its buildings are still standing, and the town has become a minor tourist destination for backpackers and curiosity seekers.

ghost cities definition

Sign up for Inside History

Get HISTORY’s most fascinating stories delivered to your inbox three times a week.

By submitting your information, you agree to receive emails from HISTORY and A+E Networks. You can opt out at any time. You must be 16 years or older and a resident of the United States.

More details : Privacy Notice | Terms of Use | Contact Us

We sent you SMS, for complete subscription please reply.

You are using an outdated browser. Please upgrade your browser or activate Google Chrome Frame to improve your experience.


Wonder of the Day #1074

What Is a Ghost Town?



Have You Ever Wondered...

  • What is a ghost town?
  • What causes ghost towns?
  • Where are some famous ghost towns?
  • radiation ,
  • transportation ,
  • Radiation ,
  • Transportation

Today’s Wonder of the Day was inspired by Anniston. Anniston Wonders , “ What is a ghost town? ” Thanks for WONDERing with us, Anniston!

There you are, riding your horse across the hard-packed dirt and sand. You’re hungry and thirsty. Suddenly, you see something in the distance! What is it? Are those buildings? Could it be a town?

You spur your horse to move faster in the sweltering heat. You must reach that town. There, you’ll find the food and water you’re craving. You can almost taste the bacon cheeseburger and fries you’ll order. You can feel the coolness of the sweet lemonade that’ll quench your thirst .

You ride faster and faster. The buildings get larger and larger until finally you’re there. But something’s wrong. These buildings are empty. The entire town is desolate . The only occupants are ghosts of the past. What is it? A ghost town, of course!

You may have seen them in movies, but do ghost towns exist in real life? Believe it or not, they do indeed exist and can be found all over the world. Any abandoned city, town, or village can be considered a ghost town. They usually also have visible remains, such as empty buildings.

Formerly bustling towns can become ghost towns for a variety of reasons. One example is settlements that spring up due to a particular economic activity, such as the discovery of a natural resource. They can become ghost towns when that commodity runs out.

In the past, such towns—often called boomtowns—were settled and quickly came to life. People there built mines or mills to harness natural resources, such as gold or coal. After these were taken, the workers often moved on. They went to another town to pursue similar work. This left the once-busy boomtown a shell of its former self.

Ghost towns can also be created by changes in access . For example, historic Route 66 encountered many changes during its lifetime. Occasionally, new interstate highways would be built that would lead to the closure of old roads. If a town depended upon that road’s traffic for its livelihood, its closure could mean the death of the settlement.

In a similar way, ghost towns have been created when railroads are abandoned or re-routed to different places. The creation of dams across the country has also occasionally resulted in the creation of ghost towns. This happens because of the flooding of previously occupied lands.

Disasters like repeated flooding can also create ghost towns. Fire can do the same . The town of Centralia, Pennsylvania, was abandoned in 1984 due to a mine fire. That fire has burned continuously underneath the town since 1962!

Disasters at nuclear power plants have created many ghost towns, especially in Ukraine, Belarus, and Japan. Due to contamination from nuclear radiation, hundreds of towns in these countries have been abandoned .

Today, ghost towns still receive visitors. They come to see the remnants of the past. Some of these famous tourist destinations include Bannack, Montana; Calico, California; Oatman, Arizona; Bodie, California; and Thurmon, West Virginia.

Have you ever visited a ghost town? Would you like to? Some people think they’re a bit spooky, but others find them fascinating. Maybe you can stop in at an abandoned town on your next vacation!

Common Core , Next Generation Science Standards , and National Council for the Social Studies ."> Standards : C3.D2.His.2, CCRA.L.3, CCRA.L.6, CCRA.R.1, CCRA.R.2, CCRA.R.4, CCRA.R.10, CCRA.SL.1, CCRA.SL.2, CCRA.W.2, CCRA.W.7, CCRA.L.1, CCRA.L.2

Wonder What's Next?

We realLEE think you’ll enjoy tomorrow’s Wonder of the Day!

Find an adult who can help you with the activities below:

  • Are there any ghost towns near where you live? If not, don't worry! You can grab a friend or family member and travel online to explore ghost towns virtually. Just check out this Ghost Town Gallery ! Which of these ghost towns would you most like to visit? Discuss with your friend or family member.
  • Many of today’s towns and cities started for economic reasons, much like boomtowns. Do you know why your town was founded? Is it near a natural resource? Does it have easy access to a highway or railroad? Ask an adult to help you do some research into the history of where you live. Then, write a summary of what you find out. 
  • Take a video tour of the ghost town of Bodie, California . Then, summarize this town’s story for a friend or family member. How was the town founded? What caused it to become a ghost town? Include any other details you think are important.

Wonder Sources

  • http://en.wikipedia.org/wiki/Ghost_town (accessed 08 May 2020)
  • http://www.ghosttowns.com/ (accessed 08 May 2020) 
  • http://www.travelandleisure.com/articles/americas-coolest-ghost-towns (accessed 08 May 2020)

Did you get it?

Wonder contributors.

We’d like to thank:

Christah , Vihaan , Brandon , Jack and Angel for contributing questions about today’s Wonder topic!

Keep WONDERing with us!

Wonder Words


That's a great question, Daichi! You might like our Wonder "Are Ghosts Real?"



Hey, Marissa&ASHLEY--

We get it. That video had a Johnny Cash song in it, and they can be pretty powerful. We've added this WONDER to our list for refreshing. Thanks for your comment. 


We're glad that you learned something new, Zariam!


Hi, bob! What specific questions do you have about this Wonder?


Since this is just an introduction to the topic, we encourage you to take a Wonder Journey to learn more about ghost towns!


Hi, Meme! Do you live in a small town? We think that you might also be interested in  Wonder 1557: Where Is the Quietest Town In America?

I don’t like your website


weel i love this website

We love that you're WONDERing with us, bob!

Not even  Wonder 1226: What Is an Internet Meme? ??


We appreciate that you are WONDERing with us and looking for more information here at Wonderopolis, Jaran! Have you checked out the links under our "Try It Out" section?  Ghost Town Gallery  and  Ghost Town Video Tours  should help keep you WONDERing more about this topic!


rhylan and noelle

Luckily, we currently have  1959 Wonders for you to explore!! Thanks for being great Wonder Friends, rhylan and noelle!


Same here!  We would love to visit one sometime.  DEFINITELY during the day, though. ?


Thanks, Lucy!  Our questions come from Wonder Friends like you and our sources are listed at the bottom of the Wonder if you ever want to check them out.  Thanks for WONDERing with us!


I've been to some of the ghost towns that are around the world its really creepy

That's cool, Savannah! We're sure that was fun and scary at the same time! ??


Great question, Daleylynn!  Did you find your answer in the article?


Thank you so much, Sidney!  We are so thrilled that you enjoy WONDERing with us!  We hope you get a chance to visit a ghost town one day!  We think that would be exciting.


We think it would be very cool to visit a ghost town, Mike! There is something eerie and mysterious about it, yet fascinating. Glad this Wonder has got you thinking!


Thanks for joining the conversation, Hayes!  We are glad you checked out this Wonder! ?


Thanks for sharing your thoughts with us, Jose! Super Mario Run sounds like a fun game!



It's great to hear from you, DangerDenis! Thanks for sharing your connection to this Wonder of the Day!


Jason voorhes


Oh hey, Jason. Wait, Jason Voorhes or a different Jason? If you're going to ask us to go camping with you, we think we will politely decline. Thanks for WONDERing with us!

Thanks for visiting Wonderopolis, Jason! We hope you stick around and explore the website some more! :)


Thanks for sharing that interesting fact, Miah! We're glad you learned something new! :)


Great question, Ashley! We encourage you to keep researching about Western towns and ghost towns at the library and online! We're glad this Wonder sparked your interest! :)


I like it !!!!!!!!!!!!! I don't think it is

We love your enthusiasm about WONDERing, Kaitlyn! We're glad you also liked the video! :)

Hi, matthew! We hope you had fun exploring this Wonder! Thanks for visiting Wonderopolis! :)

Hi, Patrick! We're glad you liked this video. Videos and images are great to pair with text to keep WONDERing and learning! :)

Hi, marcuc! Thanks for stopping by Wonderopolis for a quick hello! We hope you have an awesome day, Wonder Friend! :)


We're happy to help, jjjr! The Wonder text above is a good source of information and don't forget to check out the links found in the Try It Out section to learn more about ghost towns! Have a nice day, Wonder Friend! :)


Wonder Friend

Thanks for joining the conversation, markus! We're glad you're here! :)



Hi Wonder Friend! Thanks for WONDERing with us! Wow, that's a great WONDER! Have you seen a ghost town before? Keep WONDERing! :)


Tyler Jarvis

You might be right, Tyler! Welcome back! :)


Hi Makayla! Thanks for WONDERing with us today! We can tell you really read the Ghost Town WONDER! What was your favorite part? Can you remember a movie you've seen that had a Ghost Town in it? That's fun to WONDER about! :)


Daquan Room6

Hi Daquan Room6! We wish we could, but we haven't spotted any ghosts - have you? Halloween is coming up, perhaps there will be ghosts to Wonder about then! :)

Trevor Room6

Great question, Trevor Room6! If there are people living in Towanda, PA, that means it's an active town, not a ghost town. :)


Hey Zoey1005, thanks for sharing your comment with us! Ghost towns don't have ghosts in them, they are just deserted towns where people used to live and have since moved away. We Wonder if you have ever seen a ghost before? We Wonder if there will be ghosts to Wonder about as Halloween approaches?! :)


Thanks for WONDERing with us today, Nickia! :)


That's WONDERful, Pinkie Pie! We Wonder if you have a pet at home? :)



Hi Kai! We hope you'll read our Wonder to learn more about ghost towns! :)


girl or boy

Thanks for joining the discussion and sharing your opinion! We understand everyone likes different Wonders! :)

Thanks for WONDERing with us today, Cat! :)


We're so glad you learned something new with us today, Hannah! Keep up the great work! :)


Hi there, 4R Rats! Thanks for WONDERing with us today! :) Ghost towns can be anywhere - we're glad you're thinking about the societies that live below us - just like ants! :) We bet most of the animals left with the people, but perhaps some animals have made ghost towns their new homes! We hope you'll continue to Wonder about this fun topic! :)


Hi Dori, thanks for sharing your comment! We're glad you're here, and we hope you like our Wonder tomorrow, too! We've got a few dog Wonders to share with you: https://www.wonderopolis.org/wonder/why-do-dogs-have-wet-noses/ https://www.wonderopolis.org/wonder/do-prairie-dogs-bark/ :)


Hi Max! Thanks for coming back to Wonder with us today! :) We Wonder if you have ever visited a ghost town? We think it would be cool to walk back into history! :) Thanks for sharing your predictions with us... we think you're on the right path! :)

Mekhi Room6

Hi there, Mekhi Room6! What a great question! We learned about the "boomtowns" that were full of people who were searching for gold and wealth. Once all the gold was discovered, they moved along to new towns to find occupations and homes. When they left, the towns became ghostly and deserted. We Wonder if you have ever visited a ghost town? :)

Ciani Room6

Hi Ciani Room6! Thanks for WONDERing with us today! We don't know for sure if ghosts exist! We have never seen one, but that doesn't mean they're not real! What do you think, Wonder Friends? :)

Josey Room6

Hi Josey Room 6, great question! While some people might say there are, others use the term "ghost town" to describe the feeling of a town. It's empty, almost as if only ghosts live there! We Wonder if you have ever visited a ghost town before? :)

We are undergoing some spring clearing site maintenance and need to temporarily disable the commenting feature. Thanks for your patience.

Related Wonders for You to Explore


Who Was Lou Castro?


Who Was Dalip Singh Saund?


Who Was the Surfer of the Century?


Who Was Grace Lee Boggs?


How Did the British Save Children From the Nazis?

Drag a word to its definition

Select a Wonder Word:

Match its definition:


You’ve matched all of the definitions correctly.

Share results

Question 1 of 3

What is the best description of a ghost town?

  • a A cemetery, or place where dead bodies are kept Not Quite!
  • b A city or neighborhood where paranormal activity has been witnessed or is believed to take place Not Quite!
  • c A mirage, or something that isn't really there Not Quite!
  • d A community that has been abandoned by its inhabitants Correct!

Question 2 of 3

What is the main reason why boomtowns are created?

  • a dynamite Not Quite!
  • b a particular economic activity Correct!
  • c a fireworks celebration Not Quite!
  • d natural and man-made disasters Not Quite!

Question 3 of 3

Which of the following are not conditions that can cause a ghost town?

  • a the loss of a natural resource and changes in access Not Quite!
  • b flooding and fire Not Quite!
  • c rerouted railroads and the creation of dams Not Quite!
  • d a full moon and gravitational forces Correct!

Quiz Results

Share Results

Spread the joy of wonder, get your wonder daily.

Subscribe to Wonderopolis and receive the Wonder of the Day® via email or SMS

Join the Buzz

Don’t miss our special deals, gifts and promotions. Be the first to know!

Share with the World

Tell everybody about Wonderopolis and its wonders.

Share Wonderopolis

Wonderopolis widget.

Interested in sharing Wonderopolis® every day? Want to add a little wonder to your website? Help spread the wonder of families learning together.

You Got It!



© National Center for Families Learning (NCFL)

Go to the homepage

Definition of 'ghost town'

Ghost town in british english, ghost town in american english, examples of 'ghost town' in a sentence ghost town, trends of ghost town.

View usage for: All Years Last 10 years Last 50 years Last 100 years Last 300 years

Browse alphabetically ghost town

  • ghost shrimp
  • ghost story
  • ghost train
  • All ENGLISH words that begin with 'G'

Quick word challenge

Quiz Review

Score: 0 / 5


Wordle Helper


Scrabble Tools


Why Is Ghost Town Called Ghost Town?

Why Is Ghost Town Called Ghost Town, United States Ghost Towns

Ghost towns are abandoned or semi-abandoned settlements that once thrived but have since been deserted, often due to economic shifts, natural disasters, or human-caused events. The term “ghost town” evokes a sense of mystery and intrigue, as these places seem frozen in time, leaving behind only the faint echoes of their once vibrant past.

In this article, we will explore why ghost towns are called ghost towns and explore the fascinating stories behind some of the world’s most famous examples.

The Definition of a Ghost Town

The term “ghost town” describes a deserted or abandoned settlement with visible remnants of buildings, infrastructure, or other structures. Historian T. Lindsey Baker states that a ghost town is “a town for which the reason for being no longer exists.”

This definition implies that a town must have lost its economic or social purpose to be considered a ghost town. While the exact criteria for classifying a town as a ghost town may vary, there are some common characteristics that most ghost towns share:

  • A significant decrease in population
  • Abandoned or deteriorating buildings and infrastructure
  • A declining or non-existent local economy

Ghost towns can be found all over the world , from mining communities in the American West to ancient cities in the Middle East. They serve as reminders of the ever-changing nature of human settlements and the forces that can lead to their decline and eventual abandonment.

Causes of Ghost Town Formation

There are several reasons why a town might become a ghost town. Some of the most common causes include:

Economic Shifts

Many ghost towns were once thriving communities that experienced a sudden decline in economic activity. This could be due to the depletion of a natural resource, such as the exhaustion of a mine or the deforestation of a logging town, or the collapse of a major industry that supported the local economy.

When the economic activity that sustained a town disappears, so too does its population, leaving behind a ghost town.

Natural Disasters

Natural disasters, such as floods, earthquakes, or volcanic eruptions, can also lead to the formation of ghost towns. These events can cause widespread destruction and make a town uninhabitable, forcing residents to abandon their homes and seek shelter elsewhere.

In some cases, towns may be rebuilt nearby, leaving the original settlement to decay and become a ghost town .

Human-Caused Events

Human actions, such as wars, government policies, or environmental contamination, can also contribute to the creation of ghost towns. In some cases, entire towns may be forcibly evacuated due to threats to public safety, such as radioactive contamination from a nuclear disaster or dam construction that will flood the area.

Other towns may be abandoned due to lawlessness, political unrest, or other human-caused events, making it difficult or impossible for residents to continue living there.

Ghost Towns Around the World

Ghost towns can be found in every corner of the globe, each with its own unique history and story to tell. Some of the most famous ghost towns include:

United States

Ghost towns are particularly prevalent in the western United States, where the boom-and-bust cycle of mining towns and the rapid expansion of the railroad led to the creation of many short-lived settlements. Some notable examples include:

  • Bodie, California: A gold-mining town that was abandoned when the mines were depleted, Bodie is now a well-preserved state park and popular tourist destination.
  • Virginia City, Montana: Once a thriving gold rush town, Virginia City is now a National Historic Landmark and a popular tourist attraction.
  • Centralia, Pennsylvania: This town was abandoned due to an underground coal mine fire that has been burning since 1962, making it uninhabitable.

Canada also has its share of ghost towns, many of which were once mining or logging communities that declined when the resources they depended on were exhausted. Some notable Canadian ghost towns include:

  • Barkerville, British Columbia: A restored gold rush town that now serves as a tourist attraction and heritage site.
  • Val-Jalbert, Quebec: A former pulp mill town that was abandoned when the mill closed down, Val-Jalbert is now a popular tourist destination.

Ghost towns can also be found throughout Europe, with many abandoned due to economic shifts, natural disasters, or human-caused events. Some examples include:

  • Oradour-sur-Glane, France: A destroyed village and its inhabitants massacred by German troops during World War II, Oradour-sur-Glane has been left as a memorial to the victims.
  • Craco, Italy: This medieval village was abandoned due to landslides and has since become a popular filming location for movies.

Asia and the Middle East

In Asia and the Middle East, ghost towns are often the result of wars, political upheaval, or natural disasters. Some well-known examples include:

  • Pripyat, Ukraine: A city that was evacuated following the Chernobyl nuclear disaster in 1986, Pripyat remains abandoned and off-limits to the public due to high radiation levels.
  • Aghdam, Azerbaijan: This city was destroyed during the First Nagorno-Karabakh War and has been left in ruins ever since.

The Revival of Ghost Towns

There has been growing interest in preserving and restoring ghost towns as tourist attractions, heritage sites, and even as new communities in recent years. Some ghost towns, such as Barkerville in British Columbia or Bodie in California, have been transformed into popular tourist destinations, allowing visitors to step back in time and experience these historic places firsthand.

Other ghost towns, such as the former mining town of Black Hawk, Colorado, have been revitalized and repurposed as resort communities, featuring hotels, restaurants, and casinos housed in restored historic buildings.

Whether they are preserved as relics of the past or given new life through adaptive reuse, ghost towns serve as a testament to the ever-changing nature of human settlements and the many forces that can shape their rise and fall.

The Appeal of Ghost Towns

There is a certain allure to ghost towns that has captivated the imagination of many people throughout history. These abandoned settlements, with their decaying buildings and silent streets, provide a glimpse into a bygone era and the lives of the people who once called them home. They also serve as a reminder of the impermanence of human endeavors and the ever-present possibility of change.

Ghost towns are often the subject of legends, folklore, and ghost stories, adding to their mystique and appeal. They also provide unique opportunities for exploration, photography, and even paranormal investigation.

Whether you are drawn to ghost towns for their historical significance, their eerie atmosphere, or the stories they have to tell, these fascinating places offer a unique window into the past and a haunting reminder of the passage of time.

Ghost Town Tourism

As interest in ghost towns has grown, so too has the popularity of ghost town tourism. Many ghost towns have been transformed into tourist attractions, with visitors flocking to these sites to learn about their history, explore the remaining structures, and immerse themselves in the eerie atmosphere of these abandoned places.

Ghost town tourism can take many forms, from guided tours of well-preserved historical sites like Bodie, California, and Barkerville, British Columbia, to off-the-beaten-path adventures in more remote and lesser-known locations. Some ghost towns even offer overnight accommodations, allowing visitors to fully immerse themselves in the experience of staying in a deserted town.

Whether you’re a history buff, an amateur photographer, or simply someone who enjoys exploring the world’s hidden corners, ghost town tourism offers a unique and unforgettable experience that will leave a lasting impression.

The Future of Ghost Towns

As more and more people become interested in the preservation and restoration of ghost towns, these once-forgotten places are being given a new lease on life. In some cases, this means transforming them into tourist attractions or adaptive reuse projects, while in others, it may involve efforts to repopulate and revitalize these communities.

Regardless of their ultimate fate, ghost towns will continue to serve as a reminder of the ever-changing nature of human settlements and the forces that can lead to their rise and fall. They also provide valuable opportunities for education, research, and exploration, allowing us to learn from the past and better understand the complex interplay of factors that shape the world we live in today.

So, why is a ghost town called a ghost town? The term “ghost town” refers to the eerie atmosphere and sense of abandonment that pervades these deserted settlements, where the only remaining “residents” are the ghosts of the past.

These towns have been left behind for various reasons, including economic shifts, natural disasters, and human-caused events, leaving behind a haunting reminder of the impermanence of human endeavors and the ever-present possibility of change. As interest in ghost towns continues to grow, so too does the opportunity to preserve, restore, and learn from these fascinating and haunting places.

Related Posts

Ghost towns and abandoned cities of the world.

Ghost Towns And Abandoned Cities Of The World, United States Ghost Towns

Forgotten Towns of America – Time Stands Still

Forgotten Towns Of America Time Stands Still, United States Ghost Towns

Creepy Ghost Towns In America

Creepy Ghost Towns In America, United States Ghost Towns

What is a Ghost Town? WMH Town Classifications Explained

While there is not one official definition of what a ghost town is, there has been much discussion on the subject. Wikipedia’s basic definition is as follows:

A ghost town is an abandoned village, town or city. A town often becomes a ghost town because the economic activity that supported it has failed, or due to natural or human-caused disasters such as floods, government actions, uncontrolled lawlessness, war, or nuclear disasters. The term is sometimes used to refer to cities, towns, and neighborhoods which are still populated, but significantly less so than in years past.

Berlin Nevada

To elaborate on that definition, Wikipedia quotes the work of T. Lindsay Baker, Chair of Industrial History at Tarleton State University in Stephenville, Texas:

The town’s reason for being must no longer exist.  Although abandoned sites clearly meet this requirement, this does not exclude certain semi-abandoned sites from being classified as ghost towns, provided they maintain only a skeleton population.

There must be tangible remains of the town for visitors to see.  These can range from a dilapidated main street populated with abandoned mercantile/government buildings to only a town cemetery (Baker also stipulated that townsites included in his own books must have public access and be evenly distributed throughout Texas).

Burke Idaho

Mining Town Classifications Used At Western Mining History

Western Mining History is a bit different than other ghost town websites in that many of the towns profiled are not ghost towns at all. The only criteria for being profiled at WMH was that the town had a history that was related somehow to mining.

Related to mining could mean that the town literally sits among and near the mines themselves, or it could mean that the town was a railroad hub, supply hub, or smelter center for nearby mining districts. These “supply hub” towns were often located in more hospitable locations than the nearby mining towns and were less likely to be abandoned as the mining industry declined. Examples of these hub towns are  Baker City ,  Durango , and  Salida .

Some of these towns have even become large cities like Boulder, Denver, or San Francisco – towns that will be profiled here at some point and are certainly not ghost towns by any stretch of the imagination.

Custer Idaho

My challenge during the course of building the site was to somehow classify towns in a way that gave some meaning to their current state of preservation and population. I also wanted to keep it as simple as possible so I came up with three general classifications: Active town, Near Ghost Town, and Ghost Town.

Active Town: Up until now the focus on WMH was to profile and photograph mining towns that still had a significant of intact buildings and a sustainable year-round population. Therefore, many of the towns you will at the site now are active towns.

Near Ghost Town:  This is the classification that could probably be debated by many but I felt that something was needed to distinguish those towns that weren’t really ghost towns yet but had undergone a significant reduction of population and the amount of intact structures since the town’s peak. Examples of these near ghost towns are  Goldfield  and  Sumpter .

Bourne Oregon

Ghost Town:  For the most part a ghost town at Western Mining History is defined by the criteria described by T. Lindsay Baker (see above), but with some small differences. A ghost town at Western Mining History is a place that is mostly or all abandoned. It may have no residents at all or a relative handful of part-time or year-round residents.

Generally a ghost town has no active businesses. Where WMH parts with Baker is with the idea that “There must be tangible remains of the town for visitors to see.”. In the future WMH will profile most of the mining camps of the West, including those that have been wiped off the landscape completely by time and neglect, or more dramatically, by open-pit mining.

Therefore, it will be useful to classify some towns as ghost towns for the purpose of organizing them on this site, even if they no longer exist at all.

Bodie California

Nothing is Perfect.  Reality is that most things, including towns, have a wide range of characteristics that don’t always fall into a few neat classifications. However, it is useful to determine a classification for a town even it the fit doesn’t seem quite right.

Classification allows for a website that is highly organized, easy to navigate, and provides the best user experience. I will continue to examine these classifications and update them as needed. Any feedback on the subject would be appreciated (you can contact me through the contact link at the bottom of this page or through the  WMH Facebook page ).

Planning Tank

What are ghost towns?

What are ghost towns .

Ghost towns can be perceived as an image of burning or creating abandoned buildings. It can be alternatively called as an abandoned village, city or town that might contain substantial visible remaining buildings and infrastructure. Ghost towns are important when they get absorbed, especially for the use of prospector. But they are also remembered as historic places too.

While the term ghost town is used in a different way across cultures and nations, debate goes on as to what can be considered as a ghost town. Some argue that a ghost town should contain the tangible remains of buildings that existed there while others describe it as a town for which the reason for being no longer exists. Generally, the term is used loosely in the absence of a common definition.

Ghost Town

What are the reasons of abandonment of a city or town ?

A town or a city often becomes a ghost town because of various reasons. For instance, due to natural or man-made disasters such as prolonged droughts, floods, extreme heat or cold, war, pollution, nuclear disasters etc. It may also happen due to failure of economic activity that supported the town once.

  • Natural and Man-made Disasters – Natural and man-made disasters can lead to production of ghost towns. There are various instances in which the residents of towns and villages have evacuated the original site and relocated some distance apart due to continuous events of flooding, landslides, mine fire, nuclear contamination etc. Chernobyl disaster is a well-known example of massive evacuation.
  • Failure of Economic Activity – Ghost towns may result when there is a single economic activity on which the town is dependent gets depleted or collapsed. This might be a planned or an unplanned activity. For instance, precious metal mining in an area would bring intensive but short-lived economic activity to a remote area which would turn into a ghost town once the resource gets depleted.
  • Poor connectivity – Improper planning of roads and other infrastructure can also lead to creation of a ghost town because in such cases, the town loses its connectivity with other parts of the area.
  • Wars and Massacres – During earlier times, some towns became deserted when their populations were massacred. During wars, heavy fighting forced the entire population to flee and so the area became entirely ruined and uninhabited.

Why people visit ghost towns ?

Just like any other popular tourist spot that attracts huge number of tourists every year, people visit such abandoned and broken cities and toppled monuments for ruin gazing which describes people’s fascination with empty places. Frozen and lost in time, these abandoned places draw a number of curious travelers and explorers in search of reminders of our own hubris and the power of time. For some, these towns are spooky while for others these are fascinating.

Famous ghost towns

There are various ghost towns from infamous nuclear disaster zone near Chernobyl to Henry Ford’s doomed jungle paradise in Brazil that are widely known across the world for their derelict buildings, silent streets and haunting views which offers a glimpse into the lives of once thriving communities.

Hashima island, Japan

Hashima is a small island off the coast of Nagasaki which was first settled as a coal mining colony in 1887 and was later purchased by Mitsubishi, which built some of the world’s first multistorey, reinforced concrete buildings to house its bursting population. The place remained a hive of activity for several decades, especially during World War II, when the Japanese forced thousands of Korean laborers and Chinese prisoners to toil its mines. By the 1950s, the sixteen-acre island was packed to the gills with more than 5,200 residents. The city was promptly abandoned after the mine closed in 1974 when most workers found the cramped conditions unliveable. Forty years of neglect has turned the island into a dilapidated ruin of collapsed staircases and condemned apartments. The island was officially opened for tourists in the year 2009 and it has also been a shooting location for various Hollywood movies.

Bodie, California

Bodie was founded after miners found rich deposits of gold and silver in its hillsides in 1876. Gold crazed prospectors started settling in the area at a rate of more than two dozen per day and its population soared to around 10,000 people. But few years later by the 1880s, Bodie eventually went bust as it had outgrown its meagre infrastructure which was followed by harsh and deadly winters. By 1940s, all the people finally shipped out and since then the place has become known as one of the nation’s most well-preserved ghost towns.

Oradour-sur-Glane, France

The village of Oradour-sur-Glane witnessed one of the worst massacres of French civilians during World War II in 1944. It is believed that around 642 residents of the village were murdered and most of their houses were burned to the ground. The men were machine gunned and the women and children were locked in church and killed with explosives and grenades. Only a few people managed to survive by playing dead and fleeing to the forest later on. While a new village was built nearby after the war ended, the burned-out ruins of the old town remain untouched as a monument to the victims.

  • Everything about Dark Tourism | Concept, Types, Examples
  • 6 Reasons Why People Pursue Dark Tourism

About The Author

ghost cities definition

Leave a Comment Cancel Reply

Your email address will not be published. Required fields are marked *

Save my name, email, and website in this browser for the next time I comment.

Notify me of follow-up comments by email.

Notify me of new posts by email.

ghost cities definition

Download GPX file for this article

Ghost towns

ghost cities definition

  • 1 Understand
  • 2 Natural disasters
  • 3.1 Climate change
  • 3.2 Environmental
  • 3.3 Nuclear
  • 3.4 War and forced relocation
  • 4.1 Deserted medieval villages
  • 4.2 Fisheries, islands and outports
  • 4.3 Gold rush towns
  • 4.4 Abandoned mining communities
  • 4.5 Railway and highway abandonment
  • 4.6 Abandoned military installations
  • 4.7 Industrial abandonment
  • 4.8 Abandoned resorts
  • 4.9 Casualties of urbanization
  • 4.10 Failed economic developments
  • 5 Stay safe

A ghost town is a place where physical evidence remains to mark the site of a once-active human settlement which has been abandoned, leaving few or no inhabitants.

Understand [ edit ]

ghost cities definition

There is no commonly accepted definition of a ghost town. It is usually implied to have enough remaining or partially remaining buildings to look like a town. Some ghost towns might have a handful of permanent inhabitants; hospitality staff, researchers, or inhabitants who never left.

A few ghost towns are part of exclusion zones due to man-made or natural disasters. More commonly, ghost towns quietly appear when the reason for the town's creation no longer exists. A mining town is abandoned once too little ore remains to be profitable, a railway town is abandoned once the train no longer stops, a manufacturing town is abandoned when its last factory closes. Occasionally a village can avoid becoming a ghost town by finding a new vocation to replace a dying industry, but this becomes substantially more difficult if the town site is far off the beaten path.

While some ghost towns have been partially restored and commercialised as tourist traps , many more are in remote or awkward locations where the abandoned buildings are left to be slowly reclaimed by the elements. While legal consequences for trespassing are improbable in many of these locations, the leave-no-trace principle remains vital so that subsequent travellers may view these sites without key pieces being damaged, removed or buried in rubbish. Regardless of location, some ghost towns (or older parts of them) are archaeological sites .

Once no physical evidence remains, a settlement is typically removed from lists of ghost towns. Examples would include towns entirely flooded by hydroelectric development or wilfully demolished, if no traces remain of the former village.

Some ghost towns have been used as motion picture sets, and become destinations of fiction tourism , in particular for horror fiction .

Natural disasters [ edit ]

ghost cities definition

  • The original town of Poggioreale (nowadays known as Poggioreale Vecchia, Old Poggioreale) in the Trapani province in western Sicily was largely destroyed in the 1968 Belice earthquake. After the quake, a new Poggioreale was built at a location supposedly safer from future earthquakes about 4 km to the south, leaving the old one as a ghost town. Similarly, many other towns in the Belice Valley including Gibellina, Vita, Santa Margherita di Belice and Salaparuta were rebuilt at locations some distance away from the original towns.
  • Qushan, the former county seat of Beichuan County in China's Sichuan Province was abandoned following the 2008 Sichuan earthquake. The ruins have been preserved as a memorial park and a museum has been set up nearby.

Man-made disaster [ edit ]

ghost cities definition

Climate change [ edit ]

Environmental [ edit ], nuclear [ edit ].

ghost cities definition

War and forced relocation [ edit ]

ghost cities definition

Economic abandonment [ edit ]

Deserted medieval villages [ edit ].

ghost cities definition

An abandoned village, in archaeology, is an abandoned settlement with few visible remains. Some are solely archaeological sites — in others, some remnants of a town site can still be seen, but on a smaller scale than in a ghost town. The Dutch and German languages refer to a deserted medieval village as a Wüstung . In some, a drop in population due to the Black Death of 1348–49 caused the few remaining residents in a marginal location to move to a more viable settlement, in others, crop failures on marginal lands or the enclosure of formerly-common arable farmland by lords of feudal manors caused peasant farmers to relocate in search of a livelihood. Wars and general "bad times" were also often a reason for villages to become deserted. In Central Europe a lot of villages became deserted in the course of the 1618-1648 " Thirty Years War " that killed more than half the population in some areas.

  • Gainsthorpe medieval village , south of Hibaldstow in Lincolnshire , England . A clearly-marked but desolate archaeological site now in the care of English Heritage.
  • Old Wolverton, between Bletchley and Milton Keynes in the United Kingdom , slightly northwest of the modern Wolverton settlement. Deserted 1654 as common arable land was enclosed and converted to pasture by local land lords, eliminating the livelihood of feudal peasants who once farmed the land. Two village ponds and field patterns marking a deserted village are all that remain.
  • Wharram Percy, on the western edge of the chalk Wolds of North Yorkshire , is visible as a ruined parish church at the side of a small pond. The rest of the former village, demolished by a feudal lord for conversion into sheep pasture, is now merely an archaeological site.

Fisheries, islands and outports [ edit ]

  • Garden Island, in the Thousand Islands (Ontario, Canada). The Calvin Shipyard's (1836-1914) proprietors owned the entire island, which included an incorporated village, public library and company store. The business relied on plentiful, inexpensive local timber which eventually became scarce. The island lost its ferry service in 1976; the former machine shop was destroyed by fire in the 1980s. Little remains except some private cottages, a road network and a few ruins.
  • Grand Bruit , east of Port aux Basques and Rose Blanche in Newfoundland, abandoned 2010 and now silent. Accessible only by boat and economically dependent on fisheries. Atlantic cod stocks collapsed in the 1990s, the schoolhouse closed in 2007, the coastal ferry last stopped here July 8, 2010. Petites, a similar Newfoundland outport near Rose Blanche , was abandoned in 2003 and Great Harbour Deep, on the eastern side of the Great Northern Peninsula , was abandoned in 2002. The Rock was originally colonised as a string of tiny outports, coastal fishing villages accessible by boat in an era before highways and motorcars; provincial government paid residents to abandon three hundred of these tiny villages from 1954-1975 to avoid the cost of extending services to small, isolated populations. In most, houses were loaded onto barges and moved to other outports by sea. Grand Bruit's 31 residents left houses and furnishings behind; some are used seasonally as cottages.
  • Grytviken in the British Antarctic Territory operated as a whaling station until 1966. The last events in the area were two minor battles during the Falklands War. Today it's a ghost town and a popular stop on cruises to Antarctica.
  • Kirovsky in Kamchatka was a Soviet town with catching and processing fish being its economic base. It could support as much as 4,000 inhabitants before it was abandoned in 1964 as the fish stocks got depleted due to the Japanese drift netting in the area. An iconic concrete building right on the coast, half eaten by the sea and wave action and often likened to a scene from a post-apocalyptic film, is the only remain of the town.

Gold rush towns [ edit ]

ghost cities definition

Common in North America as colonisation pushed settlements westward in the 1800s, a gold or silver rush typically involved towns of as many as a few thousand people constructed in remote wilderness almost overnight once word was out that prospectors had spotted precious metals. Most of these mining towns disappeared as quickly as they had formed, their original purpose ended as soon as valuable minerals had been depleted.

  • Barkerville (BC), Canada — An 1861 gold rush town, once with a population as high as 5,000, was abandoned by the turn of the century.
  • Bodie (California), USA — Now part of Bodie State Historic Park, between Bridgeport (California) and Mono Lake . The "Deadwood" of the Eastern Sierra, Bodie is preserved in a state of arrested decay. A small portion of town (about 110 structures) is still standing, including one of the many gold mills. Interiors remain as they were left, visitors walk the deserted streets of a once-bustling town where shelves were left still stocked with goods.
  • Custer (Idaho), USA — A gold mining boom town (1896-1910) abandoned on resource depletion; adjacent mining town Bonanza was destroyed by fires in 1889 and 1897. Now state parkland with picnic area.
  • Goldfield, near Apache Junction (Arizona), USA — An 1892 gold mining town abandoned five years later when the gold was depleted, now rebuilt as a tourist attraction due to its proximity to Phoenix .
  • Molson , near Oroville (Washington), USA — An 1896 gold rush on the Canadian border brought 300 people; the town's founders never registered the land and were forced to relocate the settlement a half-mile away in 1909. A railroad ran from 1906 to 1935. The principal mining and farming industries began to die in the 1920s, the border station closed in 1941 and the post office was abandoned in 1967. A small collection of empty buildings and a schoolhouse museum remain today.
  • Oatman (Arizona), USA — Defunct western gold town on bypassed highway, established early 1900s and abandoned in the 1930s. Wild burros and Route 66 tourists still roam the streets.
  • Rhyolite, near Beatty (Nevada), USA — Founded as a mine town in 1905, Rhyolite quickly became the third largest city in Nevada. After a little more than a decade, the gold was depleted and the inhabitants gone. Today the Cook Bank building is Nevada's most frequently photographed ruin.
  • South Pass City , near Lander (Wyoming) USA on the Oregon Trail , is a 19th-century gold mining town abandoned in the 1950s. Now a tourist ghost town.
  • Walhalla ( Gippsland , Victoria) Australia — An 1863 gold rush town, the last mine closed in 1914. Portions of the town were rebuilt after 1977 for tourism and cottages.

Railway and highway abandonment [ edit ]

ghost cities definition

  • Amboy (California), USA — created as one in a series of alphabetically-named rail towns at which steam trains once stopped to take on water. The train no longer stops.
  • Cisco, near Moab (Utah), USA — old west rail town, a saloon and water-refilling station for the Denver and Rio Grande Western Railroad mentioned in Johnny Cash song "Cisco Clifton's Fillin Station". The town's decline coincided with the demise of the steam locomotive.
  • Cooladdi (Queensland), Australia — pre-dated the arrival of the railway, but died after the rails were rerouted away from the town.
  • Depot Harbour (Ontario), Canada — Great Lakes cargo port serving a rail line which crossed Algonquin Provincial Park to Ottawa . The rail line fell into disuse as competing railways were consolidated and was abandoned.
  • Glenrio (New Mexico and Texas), USA — Former railway town (the tracks are now gone) and Route 66 rest stop (bypassed by a freeway, now I-40). Texola , a rail town on Route 66 just across the border from Texas to Oklahoma, is almost as dead; as of 2010, only three dozen people remained.
  • Hackberry (Arizona), USA — An 1875 mining town west of Peach Springs ; the silver mine closed in 1919 amid legal infighting between its owners. Route 66 came to town in 1926; Arizona 66 from Kingman to Seligman (82 miles) was bypassed by I-40 on a more direct 69-mile route in the 1970s. Hackberry was abandoned from 1978 to 1992, Valentine and Truxton also became highway ghost towns while Peach Springs was kept marginally alive by the Hualapai nation.
  • Lyndhurst ( South Australia , Australia) and Farina are just a fraction of its original size after the Ghan railway (earlier called the Great Northern Railway) did its last run via Lyndhurst in 1980. The Ghan now runs more a northwestern alignment towards Coober Pedy these days rather than a northeastern.

Abandoned military installations [ edit ]

ghost cities definition

  • Jussarö, an island near Raseborg , Finland — Former iron ore mining site used by the army for urban war simulations (1967-2005), then abandoned. A lighthouse still stands on the island.
  • Peenemünde , in northeastern Germany near the Polish border — V1 and V2 rockets were built and launched here by the Germans during WW2. The abandoned sites are today an open air museum.
  • Skrunda-1 radar base, near Kuldīga , Latvia — Soviet over-the-horizon radar installation, dismantled 1998 and abandoned. Sixty buildings included apartment blocks, a school, barracks and an officers club; effectively, a former village of 5000 people. A private Latvian company Iniciative Europa purchased the site for 170,000 Latvian lats in 2010 but, as of 2012, the property remains abandoned with a lone guard blocking the main entrance to visitors.

Industrial abandonment [ edit ]

ghost cities definition

  • -3.83 -55.4964 29 Fordlândia . was established as a rubber plantation for the Ford Motor Company in the middle of the Brazilian jungle in 1928. However as the management didn't know anything about tropical agriculture and the indigenous workforce was badly treated (which even led to an uprising in 1930), the project turned into a fiasco pretty quickly. Ford made a new attempt in nearby Belterra (where rubber is still produced); however, the invention of synthetic rubber made Henry Ford's grandson sell everything to the Brazilian government in 1945 for about 1% of what they had invested in the projects over the years. Today you can visit the ruins of the plantation by car or tour from Santarém (Brazil) .  
  • Val-Jalbert, near Roberval (Québec) - Industrial town built around a mechanical pulp and paper mill, powered by a waterfall. Obsolete once pulp for paper was manufactured using chemical (not mechanical) process, now a commercial tourism site with a small modern hydroelectric generating station.
  • Misnebalam, Yucatán (3 miles north of the Dzibilchaltún archaelogical site) once had a thriving hacienda but when it closed, the town's population drifted away, with the last two residents moving out in 2010.

Abandoned resorts [ edit ]

  • Arlington (Missouri), opposite Jerome on Gasconade River. Originally served by the Pacific Railroad, Stony Dell Resort was a popular pre-World War 2 Route 66 rest stop in the Ozarks , with a pool fed from underground streams. Portions were destroyed when the highway was realigned and widened, the rest is ruins. The original bridge across Little Piney Creek was removed, forcing traffic to bypass the village on what is now I-44. A few miles northeast, a deteriorating ghost tourist court (John's Modern Cabins, near Vernelle's Motel in Newburg) rests abandoned since the 1970s.
  • Elkmont (Tennessee), founded in 1908 by the Little River Lumber Company as a logging town, later a resort with a hotel. When Great Smoky Mountains National Park was created, former cottage owners in Elkmont were initially permitted to lease their properties back. This ended in 1992 as the government refused to renew the leases. The former Wonderland Park Hotel structure collapsed in 2005; some other abandoned buildings survive as the Elkmont Historic District, listed on the National Register of Historic Places.
  • Prora in northeastern Germany was projected as a monstrous Baltic Sea resort for 20,000 travellers by the Nazis in the 1930s. Construction was almost finished when the Second World War broke out, but the resort never did open. Part of it was later used as barracks by the East German military. Today a small part of it is an official museum and another part of it has been refurbished and is used as a hotel, but most of the buildings are empty.
  • Salton Sea , a shallow expanse of water in the Californian desert accidentally created in 1905 when the Colorado River was diverted to the basin it occupies for irrigation, had a booming tourist trade in the 1950s and '60s in the towns that line its shores. However, it all came to an end in the 1980s as the lake's salinity and pollution levels increased to such an extent that there were massive fish die-offs. The former resorts that surround the lake are now semi-ghost towns; Bombay Beach in particular is (in)famous for its former beachside trailer park with abandoned structures encrusted by salt.
  • Tskaltubo was a premier Soviet spa resort in the Caucasus, featuring a gorgeous Stalinist architecture in a beautifully landscaped setting, now mostly derelict and overgrown. After the dissolution of the Soviet Union, the number of visitors reduced to a trickle, so some of the buildings were converted to the use of ethnic Georgians fleeing the conflict in nearby Abkhazia, and those that weren't, simply got abandoned.
  • Yashima, Kagawa Prefecture (屋島) near Takamatsu , Shikoku, Japan. Resort with six hotels, a cable car and a few shops built during a 1980s real estate boom, now abandoned.

Casualties of urbanization [ edit ]

Abandoned villages out in the countryside, left to deteriorate as their inhabitants all moved to the cities.

  • Village without People (无人村) in Kaiping , China . A tall watchtower and a couple of small neighborhoods of traditional Chinese houses, some with remnants of furniture, pottery, and clothes inside, slowly being reclaimed by the trees and vines since its last residents moved out in 1998.
  • In 2011 the proposal of Movimento Libero Perseo Roveraia Eco-lab, based on sustainability;
  • In 2019 there was a proposal aiming to recover the village with a mix of functions called Ecomuseum of Pratomagno .

Failed economic developments [ edit ]

ghost cities definition

Cities have been built as planned communities and never occupied:

  • A China International Trust and Investment Corporation development, Kilamba New City (30km/18miles from Luanda , Angola) was designed to house a half-million people but (as of 2013) had less than a tenth that due to the lack of a middle class able to afford mortgage loans. One school remains open, serving primarily students arriving by bus from other towns.
  • Helsinki was first founded by the Swedish king in 1550 at the mouth of Vantaa river . Intended as a competitor to the Hanseatic city of Tallinn , the new city failed to attract inhabitants, and after fires and disease outbreaks, the village disappeared in a little more than a century. While Helsinki eventually started growing in the late 18th century at the southern tip of Helsinki peninsula (about 5 kilometers further southwest), a few stones marking the site of Helsinki's first church is the only thing left of the original city.

Stay safe [ edit ]

As these sites are mostly abandoned, their condition is deteriorating rapidly. Roads are often unmaintained. Bridges and structures, if in poor condition, may not be able to bear your weight. The floorboards of abandoned buildings may be rotten and ready to break; buildings may be close to roof collapse. Sites may also be contaminated with anything from broken glass to asbestos to disease-ridden animal droppings .

If a site was abandoned due to man-made environmental disaster, it may still be heavily contaminated. Chernobyl and Fukushima are prime examples, with actively enforced exclusion zones with high levels of radioactive contamination.

See also [ edit ]

  • Archaeological sites
  • Old towns , historic sites which are still inhabited
  • Pioneer villages , living history museums or open-air museums which reconstruct an entire working town
  • Architecture and Urbex

ghost cities definition

  • Has custom banner
  • Has mapframe
  • Has map markers
  • Historical travel
  • Topic articles
  • Usable topics
  • Usable articles
  • Pages with maps

Navigation menu

  • Pop culture
  • Writing tips
  • Daily Crossword
  • Word Puzzle
  • Word Finder
  • Word of the Day
  • Synonym of the Day
  • Word of the Year
  • Language stories
  • All featured
  • Gender and sexuality
  • All pop culture
  • Grammar Coach ™
  • Writing hub
  • Grammar essentials
  • Commonly confused
  • All writing tips

a town permanently abandoned by its inhabitants, as because of a business decline or because a nearby mine has been worked out.

Origin of ghost town

Words nearby ghost town.

  • ghost prisoner
  • ghost runner
  • ghost shrimp
  • ghost story
  • ghostwriter

Dictionary.com Unabridged Based on the Random House Unabridged Dictionary, © Random House, Inc. 2024

How to use ghost town in a sentence

It’s a fascinating look at how ghost towns aren’t just dusty relics but can in fact teach us a thing or two about what a vibrant community could be in the 21st century.

Downtown centers became ghost towns, and revenue dropped sharply for coworking spaces as clients ditched short-term leases.

The sudden absence of these workplace cues presents a challenge for anyone whose office stands like a veritable ghost town due to Covid-19.

Visiting the Smithsonian during the pandemic can feel like passing through a ghost town .

The pandemic has turned the world’s financial capitals into ghost towns as nervous workers avoid mass commuting.

As Monday turned to Tuesday morning, five hostages had escaped and the Central Business District had turned into a ghost town .

The sun will set in less than an hour and a hubbub will emerge from the ghost-town houses and farms.

And the Ukrainian army, slowly, uncertainly, but ineluctably, is closing in on this besieged ghost-town of a city.

By the afternoon it was a ghost town inside the campus gates.

It's a ghost town now, just a handful of weathered wooden buildings sagging beneath snow.

According to the map, the ghost town was in a valley next to a dry lake bed.

A ghost town should have ghosts but none walk at the present time.

She said no more about it just then, as they had reached the old ghost town of Gleeson.

The tall, well-built cowboy star swung into his saddle and they trotted away between two tumbledown houses of the ghost town .

The trail led toward the hills back of the ghost town and it was evident that the man they were trailing had rested frequently.

British Dictionary definitions for ghost town

a deserted town, esp one in the western US that was formerly a boom town

Collins English Dictionary - Complete & Unabridged 2012 Digital Edition © William Collins Sons & Co. Ltd. 1979, 1986 © HarperCollins Publishers 1998, 2000, 2003, 2005, 2006, 2007, 2009, 2012

Cultural definitions for ghost town

A town, especially a boomtown in the old American West, that has been completely abandoned and deserted: “If you drive through the desert, you can still see the main street of Dry Gulch, a ghost town.”

The New Dictionary of Cultural Literacy, Third Edition Copyright © 2005 by Houghton Mifflin Harcourt Publishing Company. Published by Houghton Mifflin Harcourt Publishing Company. All rights reserved.

Other Idioms and Phrases with ghost town

A once thriving town that has been completely abandoned, as in Many of the old mining communities are ghost towns now . This idiom implies that there are no living people left in town. [First half of 1900s]

The American Heritage® Idioms Dictionary Copyright © 2002, 2001, 1995 by Houghton Mifflin Harcourt Publishing Company. Published by Houghton Mifflin Harcourt Publishing Company.

Cambridge Dictionary

  • Cambridge Dictionary +Plus

Meaning of ghost town in English

Your browser doesn't support HTML5 audio

  • aerotropolis
  • Cantabrigian
  • central city
  • conurbation
  • garden city
  • metropolitan elite
  • native place
  • non-municipal
  • As is true with many ghost towns, people moved to Bodie originally because of gold.  
  • Some American ghost towns still have a little life in them.  
  • You can find small, forgotten towns, called ghost towns, in many places in the western part of the United States.  

ghost town | Intermediate English

Examples of ghost town, translations of ghost town.

Get a quick, free translation!


Word of the Day


a board on which cheese is cut or served

It’ll have to do: talking about things that aren’t very good or very bad

It’ll have to do: talking about things that aren’t very good or very bad

ghost cities definition

Learn more with +Plus

  • Recent and Recommended {{#preferredDictionaries}} {{name}} {{/preferredDictionaries}}
  • Definitions Clear explanations of natural written and spoken English English Learner’s Dictionary Essential British English Essential American English
  • Grammar and thesaurus Usage explanations of natural written and spoken English Grammar Thesaurus
  • Pronunciation British and American pronunciations with audio English Pronunciation
  • English–Chinese (Simplified) Chinese (Simplified)–English
  • English–Chinese (Traditional) Chinese (Traditional)–English
  • English–Dutch Dutch–English
  • English–French French–English
  • English–German German–English
  • English–Indonesian Indonesian–English
  • English–Italian Italian–English
  • English–Japanese Japanese–English
  • English–Norwegian Norwegian–English
  • English–Polish Polish–English
  • English–Portuguese Portuguese–English
  • English–Spanish Spanish–English
  • English–Swedish Swedish–English
  • Dictionary +Plus Word Lists
  • English    Noun
  • Intermediate    Noun
  • Translations
  • All translations

Add ghost town to one of your lists below, or create a new one.


Something went wrong.

There was a problem sending your report.

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • View all journals
  • My Account Login
  • Explore content
  • About the journal
  • Publish with us
  • Sign up for alerts
  • Open access
  • Published: 17 June 2022

Underload city conceptual approach extending ghost city studies

  • Xiuyuan Zhang 1 ,
  • Shihong Du   ORCID: orcid.org/0000-0003-0321-8972 1 ,
  • Hannes Taubenböck 2 , 3 ,
  • Yi-Chen Wang 4 ,
  • Shouhang Du 1 ,
  • Bo Liu 1 &
  • Yuning Feng   ORCID: orcid.org/0000-0001-6439-596X 1  

npj Urban Sustainability volume  2 , Article number:  15 ( 2022 ) Cite this article

2527 Accesses

1 Citations

1 Altmetric

Metrics details

  • Developing world
  • Sustainability

Global population growth and land development are highly imbalanced, marked by 43% of population increase but 150% of built-up area expansion from 1990 to 2018. This results in the widely concerned ghost city phenomenon and runs against the sustainable development goals. Existing studies identify ghost cities by population densities, but ignore the spatial heterogeneity of land carrying capacities (LCC). Accordingly, this study proposes a general concept termed underload city to define cities carrying fewer people and lower economic strength than their LCC. The underload city essentially describes imbalanced human-land relationship and is understood in a broader context than the usually applied ghost city. In this study, very high-resolution satellite images are analyzed to obtain land functional structures, and further combined with population and GDP data to derive LCC. We empirically identify eight underload cities among 81 major Chinese cities, differing from previous findings of ghost cities. Accordingly, the proposed underload city considers heterogeneous human-land relationships when assessing city loads and contributes to sustainable city developments.


The United Nation sustainable development goals (SDGs; UNDP, 2015) 1 closely concern sustainable cities and communities (SDG 11), requiring inclusive and sustainable urbanization with benign human-land relationships 2 . Particularly the SDG 11.3.1 indicator that compares population growth to land consumption is addressed; thus, balanced human-land relationship plays an important role in sustainable city development. However, governments worldwide have difficulties in balancing urban land expansion and population growth rates, as land development and population growth are not always synchronized, especially during the rapid global urbanization 3 , 4 ; additionally, global urban agglomeration development leads to population concentrations in metropolises but population outflows in other cities 5 . Both can cause imbalanced human-land supply-demand and result in overload or underload of land resources, which deviates from SDG 11. Prior work on urbanization related consequences has mainly focused on land overload issues, such as overlarge population, heat island, traffic congestion, or pollution 6 , but scant attention has been paid to land underload particularly in relation to ghost cities 7 , 8 , 9 .

Different from the traditional image of ghost towns that are abandoned because of economic or natural resource hardship 10 , ghost cities/neighborhoods are generally understood as vacant areas that were newly built but uninhabited 11 , 12 , 13 . A number of studies have identified ghost cities based on population-related measures, using nightlight satellite images 14 , 15 , mobile phone positioning 16 , social media vitality 7 , or point-of-interest data 17 . The conceptual approaches towards ghost cities in these studies mostly consider population density, but ignore the land’s spatial heterogeneity, and thus the specific land carrying capacities (LCC) of diverse cities for people and economy remain undifferentiated 18 , 19 , 20 , 21 , 22 . In fact, cities contain distinct functional zones with varying spatial structures, i.e., heterogeneous land functional structures (LFS), that support variant human socioeconomic activities and have different LCC for people and economy 23 , 24 , 25 . Employing a unified threshold to extract ghost cities hence abstracts the varying LCC across different cities 21 , 26 . Besides these issues, ghost city studies have encountered another conceptual bottleneck. Wade Shepard who proposed the widely recognized definition of ghost cities suggested that there could be no ghost city but only temporary ghost stages in China 27 . From the perspective, ghost cities would disappear with urbanization development and population influx 5 ; thus, ghost city might lose significance and partly lose popularity 7 . Whether or not this will happen, we believe a more general concept rather than the extremes of ghost cities should be developed and relevant in the long term.

To account for the unknown effects raised above, we extend the ghost cities and propose a concept of underload cities. These underload cities are defined to carry fewer people and lower economic strength than their LCC which are featured by LFS. There are three reasons for proposing the concept: 1) underload cities as a synthetical geographic concept consider not only population, but also economy and LCC, which meet the requirement of SDG 11 that highlights human-land relationship 2 and are different from ghost cities that solely measure population density; 2) underload cities as an imbalanced phenomenon always exist in the process of land construction and human development, which can be more common and last even longer than ghost cities 27 , and thus have more profound impact on sustainable city development; and 3) underload cities are the opposite to overload cities which have been widely studied and represent another imbalanced phenomenon in urbanization 5 , 6 ; thus, investigating underload cities can complement existing researches of overload cities. Based on this concept, underload cities essentially represent imbalanced human-land relationships 17 , and it can be imperative to identify underload cities for re-integration of land resources, coordination of human-land relationship, and promotion of sustainable city development 28 .

However, two methodological research gaps need to be bridged for studying underload cities. On the one hand, LFS need to be characterized. LFS represent the land’s spatial heterogeneity and refer to land functions and their spatial structures. These are the basic units where people undertake different socio-economic activities, and thus are fundamental for measuring LCC 26 . Three issues should be considered for extracting LFS: 1) high spatial resolution is needed to enable detailed representation of LFS 28 , 29 , but current uses of low and medium-resolution land cover/use data are not sufficient for this purpose; 2) considering land functions’ diverse representations and abstract categories, traditional satellite image classification methods cannot recognize land functions accurately 26 ; and 3) considering land structures’ spatial complexity and heterogeneity, existing image textures and landscape matrices cannot measure land structures comprehensively and robustly 29 . To address these challenges, this study derives very high-resolution land functional zone maps, based on which we quantify LFS and cluster different types to express LFS robustly.

On the other hand, LCC needs to be evaluated. LCC refers to the population and economy that a city can carry considering its LFS. It measures the equilibrium state of human-land relationship and thus serves as a benchmark for identifying underload cities. Existing LCC evaluation methods are limited, because they consider either local land functions 30 , 31 , or mean population and GDP as the standard 17 , but few consider both 32 , 33 , 34 . To address the limitation, this study presents an evaluation method for LCC, measuring the correlation between LFS and population/GDP in a relational system across cities for a better benchmarking. Based on this technique, we identify China’s underload cities by comparing their actual populations and GDP to the estimated LCC.

The study will resolve the two methodological challenges raised above and identify underload cities from 81 major cities in China (Fig. 1a ), including all province-level municipalities, provincial capitals, and top 50 cities in GDP, covering a total area of 983,215 km 2 . First, we map land functional zones using very high-resolution (VHR) satellite images. Second, we characterize and cluster their LFS. Third, we evaluate LCC by combining LFS and demographic/economic data. And finally, we identify underload cities, discuss their general patterns, and compare the results with ghost cities and the SDG 11.3.1 indicator, respectively.

figure 1

a Distributions of the considered 81 cities whose names and detailed information are provided in the Supplementary Fig. 6 . b Land functional zone map exemplified for Shanghai. The legend illustrates the 12 functional classes derived. c – f , Four selected regions in Shanghai to illustrate the land functional zones in details, where the maps are set at 60% transparency and overlapped with satellite images.

Mapping VHR land functional zones

We use VHR satellite images at 2.4 m resolution fused by multiple sensors (see Data in Methods) and multilevel semantic segmentation (see Methods) to map land functional zones. According to the codes of urban land-use classification and planning published by the Ministry of Housing and Urban-rural Development of China, we divided the land of 81 cities into 12 nonoverlapping categories of land functional zones (three residential classes, commercial, industrial, institutional, transportation, as well as undeveloped, water, open space, farmlands, and forest and grass), and generate very-high-resolution land functional zone maps (see Fig. 1 and Supplementary Fig. 7 ). The overall mapping accuracy is 85.0%, and a Kappa index is measured at 0.82, which are evaluated based on 25,419 test samples (see Supplementary Method ). Compared to existing land cover/use data, the generated land functional zone map has two advantages. First, the map has relatively high resolution so that functional zone boundaries and categories can be more accurately represented. Second, the functional categories are closely related to socioeconomic activities, i.e., living, working, etc., which can be more applicable to human-land relationship analysis and LCC evaluation.

Characterizing and clustering land functional structures

For quantitative representations of heterogeneous land functional structures (LFS), we measure LFS of each city by 64 indices (including functional, structural, and thematic ones) which can be derived from the land functional zone maps (see Methods). These indices are defined to characterize land functional services, spatial structures, and socioeconomic attributes. This is done in reference to existing studies as they find land functional and structural indices can directly influence population distribution and economic strength 26 , 30 , 31 . In this study, we also presume that the socio-economic attributes, e.g., living environments and industrial types 29 , influence or even define LCC for population and economy. Then, we cluster LFS types at city level using these indices to robustly express land functions and structures. As the result, we group 81 cities into six types (T 1 -T 6 in Fig. 2 ) using the density peaks clustering algorithm (DPCA), where the type number of 6 is detected by the DPCA considering samples’ densities and distances in feature spaces (see Methods and Supplementary Fig. 3 ).

figure 2

a Spatial distributions of the six types. b Mean population, GDP, and representative cities of the six types. c Cities’ distributions in the LFS feature space, with structural aggregation as the x-axis and structural fractal dimension as the y-axis to represent the 2D feature space, because of their highest information entropy. A larger font size is used to highlight the six representative cities in the feature space. d – i Normalized land functional proportions for the six types. The normalized proportion of a functional category in a specific type refers to its ratio to the largest proportion of the category among all types, and their standard deviations are labeled.

These six types differ significantly in land functions and structures (see Supplementary Table 2 ): T 1 Comprehensively-developed cities have the largest proportions for commercial, industrial, L1-residential, L2-residential, open spaces, institutional, and transportation lands. These cities are mainly distributed in developed areas, e.g., the Yangtze river delta, the Pearl river delta, and the Shandong peninsula megalopolis. T 2 Administrative-centered cities feature higher proportions of institutional and residential lands. They mainly are province-level municipalities and provincial capitals. T 3 Coastal cities have the smallest undeveloped lands and are mainly distributed in the eastern coastal areas. T 4 Cities with development potentials have the smallest farmland proportion but the highest proportion of undeveloped lands, indicating large developmental potentials. They are all distributed in southern China. T 5 Rural-agricultural cities have the largest proportions of the categories of L3-residential and farmland, indicating a relatively developed first industry and a backward urbanization level, and they are mainly distributed in northern China. T 6 Ecological cities have the smallest proportions for most kinds of built-up lands, but the largest proportion of relatively natural lands, e.g., forest and grass, and provide important ecological services. Most T 6 cities are located in northwest China. The clustered six types represent LFS of diverse cities robustly, as LFS indices are relatively homogeneous in each type, so that we can describe LFS’s correlation with population/GDP type by type and accurately evaluate LCC.

Evaluating LCC based on LFS

We evaluate cities’ LCC based on LFS indices, type by type, where six LFS types generated in the prior section are considered to reduce LFS indices’ heterogeneity and improve the robustness of LCC estimation results 29 . Specifically, we employ stacked autoencoder (see Methods) to integrate and dimensionally reduce LFS indices, and use ordinary least squares to estimate LCC by measuring the correlation between LFS indices and population/GDP (see Methods) in each type of cities (Eq. 1 and 2 ),

where \(LCC_c^{{{{\mathrm{Pop}}}}}\) and \(LCC_c^{{{{\mathrm{GDP}}}}}\) refer to the city c ’s LCC for population and GDP respectively, and the city c belongs to u - th LFS type ( type u ). \(\overrightarrow {{{{\boldsymbol{FS}}}}} _{{{\boldsymbol{c}}}}\) denotes the LFS indices of the city c which are integrated and dimensionally reduced by stacked auto-encoder, \(LR_u^{{{{\mathrm{Pop}}}}}\) and \(LR_u^{{{{\mathrm{GDP}}}}}\) are linear regression models which are trained by cities in type u . As presented in Fig. 3b, e , the LCC results are highly heterogeneous. This proves that different cities have different LCC, and thus ghost city analysis considering a unified population threshold are limited due to aggregation effects. According to LCC, we evaluate the cities’ land carrying conditions by two indicators, i.e., shortage in land carrying (SLC) and land carrying rate (LCR) (see Supplementary Table 3 ).

figure 3

a – c the actual population of the cities, and their expected LCC for and SLC of populations. d – f the actual GDP of the cities, and their expected LCC for and SLC of GDP. g , h LCR for population and GDP of the cities. i overall LCR of the cities where dark brown represents the overload (high carrying rate), and light yellow the underload (low carrying rate).

SLC of population and GDP are defined as differences from the actual population and GDP respectively to their expected LCC. We applied the SLC to the 81 major cities (see Supplementary Fig. 6 ) and their spatial distributions are shown in Fig. 3 . In terms of population, we identify Zhuhai as the most underloaded in China. It features an extremely low SLC, i.e., −6.76 million people, as it has a large LCC of 8.78 million people but only 2.02 million residents. Two further eastern cities also have large LCC but carry much smaller populations: Taizhou in the Jiangsu Province (SLC of population = −4.57 million) and Jiaxing in the Zhejiang Province (−4.12 million). Even regional central cities such as Guangzhou (−3.06 million) and Xi-an (−2.63 million) are found to be underloaded for the population. The former is the capital city of the Guangdong Province, which can bear 18.36 million people considering its LCC but only carries 15.31 million residents. The latter is the capital city of the Shaanxi Province, whose LCC is 12.83 million, but it has only 10.20 million residents. The underpopulation of these cities indicates a serious waste of land resources and also imbalanced human-land relations. In terms of economy proxied by GDP, Zhuhai is found again the most underloaded city of China (SLC of GDP = −124.71 billion RMB). Some other cities are also weak in SLC with respect to economy, e.g., Jiaxing (−69.22 billion), Qinhuangdao (−42.36 billion), Tianjin (−39.53 billion), and Weifang (−36.83 billion). The negative values relate to the lower GDP than their inherent LCC, and thus the results for these cities signify inefficient land utilizations.

With LCR, we aim to measure a city’s carrying rate for population ( LCR P ) and GDP ( LCR G ), respectively. They are calculated as follows: \(LCR^{{{\mathrm{P}}}} = \frac{{Actual^{{{{\mathrm{Pop}}}}}}}{{LCC^{{{{\mathrm{Pop}}}}}}}\) and \(LCR^{{{\mathrm{G}}}} = \frac{{Actual^{{{{\mathrm{GDP}}}}}}}{{LCC^{{{{\mathrm{GDP}}}}}}}\) , where Actual Pop and Actual GDP refer to the actual population and GDP of the city. Then, we calculate the overall LCR as LCR = LCR P  ×  LCR G , where LCR > 100% means relative overload and LCR < 100% denotes underload. LCR P and LCR G are calculated for 81 cities. To analyze cities’ land carrying conditions and to express the relationship between LCR P and LCR G , a LCR feature space is drawn (Fig. 4 ). We find a generally linear correlation of cities’ LCR P and LCR G (Fig. 4a ), with a correlation coefficient of 0.68. This indicates that most cities are over- or underloaded for population and GDP simultaneously, except for 17 cites (Fig. 4b ). The cities with underload population but overload GDP are located in the UO quadrant, including Taizhou (Jiangsu), Lanzhou, Changzhou, Nanjing, Wuxi, Yangzhou, and Qingdao. Conversely, the cities with overload population but underload GDP are in the OU quadrant, including Harbin, Tongling, Zhoushan, Quanzhou, Yancheng, Taizhou (Zhejiang), Jining, Foshan, Chuzhou, and Shijiazhuang. The results demonstrate that different cities have different LCR and even a city can have different carrying conditions for population and economy ( LCR P vs. LCR G ).

figure 4

a Points represent the 81 cities in the LCR feature space, x-axis shows LCR for population ( LCR P ), and y-axis is LCR for GDP ( LCR G ). UO means underload population but overload GDP, while OU means the opposite; OO denotes overload for both, while UU is underload for both. b A close-up view of cities distributed in the UO and OU quadrants with city names, LCR P , and LCR G labeled.

Identifying underload cities

We identify underload cities based on LCR. Cities whose \(LCR \,<\, \overline {LCR} - Std\left( {LCR} \right)\) , i.e., LCR < 71.6% in this study, are defined as underload, where \(\overline {LCR}\) denotes the average LCR of 81 cities and Std ( LCR ) the standard deviation of LCR. Cities whose \(\overline {LCR} - Std\left( {LCR} \right) \,\le\, LCR \,<\, \overline {LCR} - \frac{1}{2}Std\left( {LCR} \right)\) , i.e., 71.6% ≤  LCR < 85.8%, are considered as slightly underload; \(\overline {LCR} - \frac{1}{2}Std\left( {LCR} \right) \,\le\, LCR_c \,<\, \overline {LCR} + \frac{1}{2}Std\left( {LCR} \right)\) , i.e., 85.8% ≤  LCR  < 114.2%, as well-balanced; \(\overline {LCR} + \frac{1}{2}Std\left( {LCR} \right) \,\le\, LCR_c \,<\, \overline {LCR} + Std\left( {LCR} \right)\) , i.e., 114.2% ≤  LCR  < 128.4%, as slightly overload; \(LCR_c \ge \overline {LCR} + Std\left( {LCR} \right)\) , i.e., LCR  > 128.4%, as overload. This analysis intends to identify underload or overload cities in a relational system across China for a better benchmarking by comparisons. Consequently, the 81 cities are classified into five types (Fig. 5 ). Eight underload and 15 slightly underload cities are found, and identifying other three types is the additional contribution of the study (see Supplementary Fig. 5 ).

figure 5

a Cities’ LCR which are ranked in descending order. b Spatial distributions of the five types of cities.

In our analysis, we identify 15 cities as slightly underload: Huizhou, Shaoxing, Guangzhou, Jinhua, Xuancheng, Nanjing, Lanzhou, Xi-an, Changzhou, Dalian, Huzhou, Luoyang, Zhangjiakou, Chizhou, and Guiyang. They consist of the first to fifth tiers of cities and are scattered across the country (Fig. 5b ), indicating a lack of association between the slightly underload cities and their locations/city tiers. This finding is different from the common perception about cities’ loads and development levels. For example, Guangzhou has a large population and developed economy, but it is found to be slightly underloaded for population and GDP when its large LCC is taken into consideration. It is similar for Nanjing and Xi-an, both capital cities for previous dynasties, and now serving as capitals of Jiangsu and Shaanxi Provinces being population hubs with advanced economies. However, they are found to be underloaded when comparing their populations and GDP to LCC.

We further identify eight underload cities: Zhuzhou (LCR = 67.4%), Dongying (65.4%), Haikou (57.9%), Qinhuangdao (54.1%), Taizhou (Jiangsu) (52.4%), Jiaxing (47.3%), Ma-anshan (46.2%), and Zhuhai (16.91%). Among these cities, Dongying, Ma-anshan, and Jiaxing have been recognized as ghost cities in previous studies 8 , 15 , 35 . The main economic activities in Dongying and Ma-anshan are petroleum and steel industries. Due to the structural crisis of their industries, they have experienced slow economic development and low population growth recently 8 . Leichtle et al. (2019) revealed the ghost city phenomenon in Dongying with a considerable mismatch between its estimated population capacity and residential number 8 ; Jin et al. (2017) profiled Ma-anshan with low urban vitality as a ghost city 15 . Additionally, an influential report “Ranking list of ghost city index in China mainland (2014)” ranked Jiaxing high as a potential ghost city 35 . Our findings corroborate this, as Jiaxing has a low LCR P of 53.4% and a low LCR G of 88.6%, indicating that the city is underloaded for both population and GDP. These comparisons reflect the close relationship between underload and ghost cities, and verify the rationality of the detected underload cities. The remaining five underload cities however have not been noticed in prior work, and they are identified because our approach accounts for heterogeneous LCC.

Analyzing general patterns of the underload cities

We find three general patterns of these underload cities. Firstly, the cities whose original industries are experiencing structural crisis can have slow development or even declines in population and GDP, resulting in LCC surpluses and underload cities, such as Dongying and Ma-anshan. The phenomenon is very common worldwide and occurred in many international cities, e.g., Detroit in the U.S., Hashima in Japan, and Fordlandia in Brazil 15 . This underload condition is often irreversible, unless new industries develop.

Secondly, the cities on the edge of urban agglomerations can be influenced by neighboring metropolises, as their population and GDP are affected by the siphon effect from surrounding cities. In this case, we find Qinghuangdao in Beijing-Tianjin-Hebei 36 , Zhuzhou in Changsha-Zhuzhou-Xiangtan 37 , Jiaxing and Taizhou (Jiangsu) in Yangtze River Delta as underload cities. Similarly, small or medium-sized cities around metropolises, e.g., New York, London, and Tokyo, can also face the same problem 38 .

Thirdly, the cities supported by major national policies can have fast land development with high urban expansion rates; thus, their LCC skyrocket, but their populations and economy will not rise rapidly in the short term. In this case, two Special Economic Zones of China including Zhuhai and Haikou 39 are found as underload cities. This underload condition is essentially a temporary product of the construction phase, as Brasilia in 1960s and Sejong in 2010s, and can be alleviated with population influx and economic growth. Taking Zhuhai as an example, the population increase in Zhuhai has speeded up at a rate of 7% annually since 2017 and may soon catch up with that of built-up expansion 40 , so that the extremely underload phenomenon in Zhuhai may thus be alleviated in the future, according to the current development trend.

Comparing underload cities with ghost cities

We compare the alternative concept of underload cities with the often-used concept of ghost cities 41 , 42 , to demonstrate the significance of the broader conceptual approach. First of all, these two concepts have significant differences in definition, as ghost cities mostly consider population density, but underload cities measure population, GDP, and LCC; thus, underload cities cannot be directly related to ghost cities. If we compare both concepts in an empirical manner, we find the following: we calculate population density (PD; the ratio of population to built-up area) to identify the top ten ghost cities (see Supplementary Table 4 ), and examine the differences between these ten ghost cities and the findings of this study (Fig. 6 ). Consequently, the ten ghost cities based on low PD are: Lhasa (PD = 0.29 thousand people km −2 ), Dongying (0.97), Urumqi (1.00), Chizhou (1.11), Zhangjiakou (1.14), Xuancheng (1.33), Nanning (1.60), Chongqing (1.60), Qianghuangdao (1.92), and Jinhua (1.94). Six out of these ten ghost cities are detected in this study as underload or slightly underload cities, including Dongying, Qinghuangdao, Chizhou, Zhangjiakou, Xuancheng, and Jinhua.

figure 6

a , b LCR and PD of 81 cities. c – f Urumqi, Lhasa, Nanning, and Chongqing which are identified as ghost cities according to PD but nonunderload by LCR.

Conversely, the remaining four ghost cities, i.e., Lhasa, Urumqi, Nanning, and Chongqing, are not identified as underload. These cities are located in outlying areas in western China. Due to rugged terrains and harsh natural environments, their LCC are limited. Although their low PD characterize them as ghost cities, the populations of these cities may have depleted their limited LCC, resulting in the categorization of them as non-underload. Identifying ghost or underload cities thus has implications and divergences in city planning. Let us take Nanning as an example: based on the underload city concept, Nanning has a high LCR at 133.8% (Fig. 6e ), and could be considered overloaded, needing population outflow and incentives for land developments; from the ghost city viewpoint, however, a different suggestion such as more population influx and reduced land constructions could be made due to the low population density of Nanning.

In addition, among the eight underload cities identified in this study, only two are found among the ten ghost cities; while the other six have ordinary PD and are not recognized as ghost cities. These six cities have large LCC which greatly exceed their population and GDP, and thus are classified as underload. Based on the underload city concept, these six cities should increase population and limit land constructions. In contrast, from the ghost city viewpoint, the six cities can continue land constructions, which potentially exacerbates the cities’ underload conditions. The comparisons above underscore the conceptual and in consequence differences for underload and ghost cities. Employing the ghost city viewpoint could have adverse impacts on human-land relationship, for its negligence in LCC. Therefore, more conceptual approaches besides ghost cities are needed to assess land carrying conditions for sustainable city planning, and thus underload cities in synopsis allow a complementary perspective and improved analysis.

Comparing LCR to the SDG 11.3.1 indicator

SDG 11.3.1 refers to the ratio of land consumption rate to population growth rate, which measures land use efficiency 5 and also concerns the human-land relationship 43 , 44 . Here, we compare LCR with SDG 11.3.1 to demonstrate the significance of this study to SDG.

Existing studies on SDG 11.3.1 measure land consumptions by impervious surface or artificial land 45 , 46 , but they mostly ignore heterogeneous land functions and structures inside impervious surface which however can greatly influence population growth 47 . In consequence, previous measurements of land consumption rates are limited, resulting in aggregated information on SDG 11.3.1. Differently, this study models diverse land functions and their structures (LFS) within cities to measure land consumption from socioeconomic and spatial-structural perspectives, based on which LCR can be estimated to quantify the relationship between land consumption and population with more comprehensive and detailed information. Although SDG 11.3.1 and LCR both produce aggregated evaluations at the city scale, LCR however considers intracity land-function heterogeneity and generates more meaningful results to explain the human-land relationship. Accordingly, LCR can be regarded as a complement to SDG 11.3.1 and an important indicator for evaluating sustainable city development. Furthermore, using LCR can identify two unsustainable city development modes, i.e., underload and overload cities (see Supplementary Fig. 5 ); thus, the concept, approaches, and findings of this study provide a different perspective for understanding the imbalanced human-land relationship and unsustainable development modes, potentially contributing to SDG.

This study demonstrates the use of VHR satellite data and derived LFS (including land functions and structural types) to evaluate the LCC of cities, and the importance of considering LCC to identify underload cities. This study is distinct from prior works on ghost cities and the SDG 11.3.1 which mainly considers population density but does not relate to the LCC.

This study identifies eight major underload cities across China, among which five cities would not have been detected had the conventional ghost city concept been applied. We also find three general patterns of these underload cities, whose original industries are experiencing structural crisis, whose population and GDP are affected by the siphon effect from surrounding cities, and whose land development is too fast supported by major national policies. Among those, the first pattern of underload cities can be irreversible, but the third is basically a temporary construction phase and can be alleviated with population influx and economic growth.

This study also finds 15 slightly underload cities. Different from the common perceptions that big cities are over developed and congested by large populations, some big cities can be recognized as slightly underload in the study, because they could carry more people and GDP when their LFS and LCC are considered. It is suggested that these underload cities need population and GDP influx and limit further land developments, so as to best use their LCC. The findings of this study provide new insights into human-land relationships, potentially contributing to a more sustainable city development.

However, the empirical verification in this study has two limitations. Firstly, the study focuses on evaluating land loads at the scale of prefecture-level cities but ignores the heterogeneity within them. Prefecture-level cities are essentially administrative units and are usually composed of multiple county-level cities which however may have different land load conditions 11 . Accordingly, a local evaluation, such as underload counties/neighborhoods, should be developed to reveal heterogeneous human-land relations within cities. Secondly, this study considers a part of Chinese cities, but not include international cities in other countries, owing to the limitation of acquiring very high-resolution satellite data. In the future, international evaluations on underload cities can be conducted to find the differences in human-land relation among different nations and international cities.

In summary, we hope to work together with other scholars to resolve these limitations, and enrich the conceptual approach by studying underload cities/counties/neighborhoods and supplementing the international verification cases.

For demographic and economic data , population and GDP data are retrieved from the “China city construction statistical yearbook in 2019” ( https://www.mohurd.gov.cn/gongkai/fdzdgknr/sjfb/tjxx/jstjnj/index.html ), where GDP represents the gross domestic product, and population refers to the resident population including both permanent (registered) and temporary residents who live there for more than six months in one year.

For land observation , satellite images covering 81 major cities are acquired by splicing and resampling images from different sensors, including SPOT-6, GF-1, and ZY-3. The image product was acquired in 2019 and basically contained three visible bands (i.e., red, green, and blue) with a very high resolution of 2.4 meters. The images have been ortho-rectified and thus provide accurate image features for mapping land functional zones. In addition, built-up area of each city is obtained from “China city construction statistical yearbook in 2019”, where built-up area refers to the developed land that carries intensive human activities.

Multilevel semantic segmentation to map land functional zone

We employ a multilevel semantic segmentation (MSS) proposed by our previous study 48 to map land functional zones. MSS consists of a semantic segmentation at object level and a neighborhood optimization at block level. For the object-level semantic segmentation, it excels in classifying complex structures with strong internal heterogeneity, confusing boundaries, and multiscale representations 48 , 49 ; thus, it can recognize diverse land functions. It generally has three processes, i.e., object segmentation, deep feature encoding, and semantic prediction (see Supplementary Method ). For object segmentation, a multiresolution segmentation approach 50 is used to segment satellite images into objects. For deep feature encoding, objects are processed by a densely connected convolutional network (DenseNet) 51 and four dilated convolutions 52 with different convolution kernels (1 × 1, 3 × 3, 3 × 3, 3 × 3) and dilated rates (1, 6, 12, 18). For semantic prediction, multiple deep feature maps generated by DenseNet and dilated convolutions are stacked, convoluted, and up-sampled to predict land function categories for objects. In addition, MSS considers a neighborhood optimization at block level, which uses conditional random field to model the neighboring relationship among land functional zones and eliminates the wrong classification results within roadblocks. Technical details of MSS see the reference 48 .

For training the model, we manually delineated and labeled 84,730 samples of land functional zones based on visual interpretations and field investigations, 70% of which are clipped into 94,306 non-overlapped patches of 512 × 512 pixels and fed into the model, where the pixels hit by manually labeled samples are set as their categories, and others are null values. The hyperparameters set in the training process are illustrated as follows. The TensorFlow system is generally used for implementation; the segmentation scale of multiresolution segmentation is set to 50; a L2-regularization is considered to avoid overfitting; an Adam optimizer with a learning rate of 2 × 10 −4 is employed for training 53 ; and cross entropy is used as loss function, i.e., \(CELoss = \frac{1}{N}\mathop {\sum }\nolimits_{i = 1}^N \mathop {\sum }\nolimits_{j = 1}^M - \widehat {y_{ij}}{{{\mathrm{log}}}}\left( {y_{ij}} \right)\) , where N denotes the number of samples, M the number of land function categories, y ij represents the probability that i -th sample belonging to the j -th category, and \(\widehat {y_{ij}}\) represents the real class with one-hot encoding; the semantic segmentation structure will be trained for 50,000 iterations. The hyperparameter tuning is demonstrated in our previous study 48 .

For evaluating mapping results, we utilize the remaining 30% samples (25,419) and measure the pixel-wise confusion matrix 54 . Overall accuracy, OA = N T / N , is calculated to assess mapping accuracy, where N denotes the pixel number of test samples and N T the number of accurately recognized pixels. In addition, Kappa index of mapping result is also measured \(Ka{{{\mathrm{pppa}}}} = \frac{{OA - P}}{{1 - P}}\) , where \(P = \frac{{\mathop {\sum }\nolimits_{i = 1}^M A_i \times B_i}}{{N \times N}}\) , M refers to the number of functional categories, A i the pixel number of the i -th category in test samples, B i the pixel number of the i -th category in mapping results 54 .

Index system for characterizing LFS

For quantitatively expressing LFS per city, we integrate indices related to land use, urban planning, and landscape pattern to propose an index system 55 , 56 , 57 , which includes functional, structural, and thematic indices (see Supplementary Table 1 ). Functional indices describe functional types and services, e.g., functional proportion, area, abundance and dominance; structural indices measure spatial structures of diverse land functions, e.g., structural fragmentation, minimum distance, aggregation; and thematic indices characterize socio-economic attributes, e.g., industrial type, living environment, public and ecological service. These indices have been verified effective to characterize land functions and structures 55 , 56 and closely related to LCC evaluation 29 , 30 , 31 . All these indices can be derived from the generated land functional zone maps.

Density peaks clustering algorithm to detect robust LFS type

To reduce the influence of LFS’s heterogeneity on modeling human-land relationships, we extract robust LFS types by a state-of-the-art clustering model. We apply the density peaks clustering algorithm (DPCA) 58 , as DPCA excels in overcoming outliers, explicitly selecting clustering centers, and processing aspheric distributions 59 . DPCA recognizes samples with large local densities and large distances to other high-density samples as clustering centers, thus it mainly considers two indicators in clustering, i.e., local density and distance to high-density samples. On the one hand, local density of the i -th sample \(\rho _i = \mathop {\sum }\nolimits_{j = 1:N,j \ne i} \tau \left( {d_{ij} - d_c} \right)\) , where N denotes the number of samples, d ij the distance from sample i to sample j in the feature space, d c is a threshold defining the neighborhood range, and τ ( x ) = 1 when x < 0; on the other hand, i -th sample’s minimum distance to the samples with higher density is formulated by \(\delta _i = \mathop {{\min }}\limits_{j:\rho _j > \rho _i} (d_{ij})\) or \(\delta _i = \max (d_{ij})\) when ρ i > ρ j ( j = 1: N , j  ≠  i ). Accordingly, the d ij is core to DPCA, and we define d ij as \(d_{ij} = {{{\mathrm{max}}}}\left( {d_{ij}^{{{\mathrm{f}}}},d_{ij}^{{{\mathrm{s}}}},d_{ij}^{{{\mathrm{t}}}}} \right)\) , where \(d_{ij}^{{{\mathrm{f}}}}\) represents the distance from sample i to j in functional-index space, \(d_{ij}^{{{\mathrm{s}}}}\) that in structural-index space, \(d_{ij}^{{{\mathrm{t}}}}\) that in thematic-index space, and distances are measured by the Euclidean distance. This definition of d ij ensures clusters’ homogeneity in all feature spaces.

As demonstrated above, we calculate the distances between cities in feature spaces, measure ρ i and δ i of each city, and select six cluster centers by ρ i  ≥ 7 and δ i  ≥ 0.5 in this study with considering the cities’ distributions in ρ - δ space (see Supplementary Fig. 3 ). Other cities are assigned to the clusters of their nearest neighbor with higher density.

Combining SAE and OLS to estimate LCC

Estimating LCC needs modeling the human-land relationship which is essentially to measure the correlation between LFS indices and socio-economic attributes, i.e., population and GDP in our case. However, two technical issues arise: 1) how to fuse diverse LFS indices, and 2) how to avoid the “curse of dimensionality“ 60 affecting correlation modeling. We use a stacked auto-encoder (SAE) to resolve these two issues. SAE has been widely used for feature fusion and dimensionality reduction 61 . SAE is composed of multiple auto-encoders (AEs) 62 , and each AE consists of three layers: input, hidden and output layers, where the input and hidden layers compose encoder, hidden, and output layers compose decoder (see Supplementary Fig. 4 ). For encoding, the input features \({{{\vec{\boldsymbol x}}}} \in R^N\) are dimensionally reduced into hidden features \({{{\vec{\mathbf y}}}} \in R^K\) , where N refers to the dimension number of input features and K that of hidden features; for decoding, the \({{{\vec{\mathbf y}}}} \in R^K\) reconstructs \({{{\vec{\boldsymbol x}}}}^\prime \in R^N\) . Accordingly, AE is formulated as:

where \(\overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{x}}}}}\) , \(\overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{y}}}}}\) , b x , and b y are parameters of encoder and decoder respectively, and f (*) represents a sigmoid activation function 63 . Training AE is essentially to minimize the reconstruction loss between \({{{\vec{\boldsymbol x}}}}\) and \({{{\vec{\boldsymbol x}}}}^\prime\) , \(RecLoss = \Vert{{{\vec{\boldsymbol x}}}} - {{{\vec{\boldsymbol x}}}}^\prime\Vert\) , and the hidden features are extracted as the features after dimension reduction. Similarly, SAE is constructed by stacking multiple AEs, with each AE trained layer by layer, and only the encoder part is kept after training. The whole network is finally fine-tuned with minimizing global reconstruction loss. The hyperparameters in the training process are described as follows: three hidden layers are considered with their dimensional numbers set as 32, 16, and 8; the batch size is set as 16; every AE is trained for 150 iterations; and the whole SAE is fine-tuned for 500 iterations. As a result, the LFS indices can be nonlinearly integrated with dimensions reduced from 64 to 8, resolving the two issues demonstrated above.

We finally use original least squares (OLS) to model the human-land relationship and predict the land carrying capacity (LCC) of each city. For a city c i (1 ≤  i  ≤ 81), it belongs to the u - th LFS type type u (1 ≤  u  ≤ 6), and other cities of type u , c j   ∈   type u (1 ≤  j  ≤ 81, j  ≠  i ), are considered to train the regression model, i.e., \(y_{c_j}^{{{{\mathrm{Pop}}}}} = \overrightarrow {{{{\boldsymbol{FS}}}}_{{{{\boldsymbol{c}}}}_{{{\boldsymbol{j}}}}}} \cdot \overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{u}}}}^{{{{\mathbf{Pop}}}}}} + b_u^{{{{\mathrm{Pop}}}}}\) and \(y_{c_j}^{{{{\mathrm{GDP}}}}} = \overrightarrow {{{{\boldsymbol{FS}}}}_{{{{\boldsymbol{c}}}}_{{{\boldsymbol{j}}}}}} \cdot \overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{u}}}}^{{{{\mathbf{GDP}}}}}} + b_u^{{{{\mathrm{GDP}}}}}\) , where \(y_{c_j}^{{{{\mathrm{Pop}}}}}\) and \(y_{c_j}^{{{{\mathrm{GDP}}}}}\) denote the c j ’‘ population and GDP respectively, \(\overrightarrow {{{{\boldsymbol{FS}}}}_{{{{\boldsymbol{c}}}}_{{{\boldsymbol{j}}}}}}\) the LFS indices of c j processed by SAE, and \(\overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{u}}}}^{{{{\mathbf{Pop}}}}}}\) , \(b_u^{{{{\mathrm{Pop}}}}}\) , \(\overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{u}}}}^{{{{\mathbf{GDP}}}}}}\) and \(b_u^{{{{\mathrm{GDP}}}}}\) are parameters of the two regression models, which can be trained by cities in type u and estimated by OLS 64 ; thus, the LCC of c i for population and GDP can be predicted by \(LCC_{c_i}^{{{{\mathrm{Pop}}}}} = \overrightarrow {{{{\boldsymbol{FS}}}}_{{{{\boldsymbol{c}}}}_{{{\boldsymbol{i}}}}}} \cdot \overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{u}}}}^{{{{\mathbf{Pop}}}}}} + b_u^{{{{\mathrm{Pop}}}}}\) and \(LCC_{c_i}^{{{{\mathrm{GDP}}}}} = \overrightarrow {{{{\boldsymbol{FS}}}}_{{{{\boldsymbol{c}}}}_{{{\boldsymbol{i}}}}}} \cdot \overrightarrow {{{{\boldsymbol{w}}}}_{{{\boldsymbol{u}}}}^{{{{\mathbf{GDP}}}}}} + b_u^{{{{\mathrm{GDP}}}}}\) .

Data availability

All data shown in the figures and tables are publicly available, and the source data are provided with this paper. Furthermore, the produced land functional zone maps of 81 major cities in China, and the estimated land carrying capacities of these cities (see Supplementary Fig. 7 ) are open to download at https://geoscape.pku.edu.cn/ .

Code availability

We acknowledge Rodriguez A. and Laio A. for their density peak clustering algorithm which is publicly available at https://people.sissa.it/~laio/Research/Res_clustering.php . We acknowledge Arcmap (v10.2) for providing secondary development and operation interface. Our code of multilevel semantic segmentation for mapping land functional zones, feature extraction for measuring land functional structure indices, and stacked auto-encoder (SAE) for processing features are maintained centrally by DoLab, Peking University ( https://geoscape.pku.edu.cn/ ) and is available on request.

United Nations Development Programme (UNDP). Sustainable development goals (SDG). Retrieved from https://www.undp.org/content/undp/en/home/sustainable-development-goals.html (2015).

Kaur, H. & Garg, P. Urban sustainability assessment tools: a review. J. Clean. Prod. 210 , 146–158 (2019).

Article   Google Scholar  

Liu, X. et al. High-spatiotemporal-resolution mapping of global urban change from 1985 to 2015. Nat. Sustain. 3 , 564–570 (2020).

Wang, L. et al. China’s urban expansion from 1990 to 2010 determined with satellite remote sensing. Sci. Bull. 57 , 2802–2812 (2012).

Article   CAS   Google Scholar  

Fang, C. & Yu, D. Urban agglomeration: an evolving concept of an emerging phenomenon. Landsc Urban Plan 162 , 126–136 (2017).

Hu, J., Wang, Y., Taubenböck, H. & Zhu, X. X. Land consumption in cities: a comparative study across the globe. Cities 113 , 103163 (2021).

Williams, S., Xu, W., Tan, S. B., Foster, M. J. & Chen, C. Ghost cities of China: identifying urban vacancy through social media data. Cities 94 , 275–285 (2019).

Leichtle, T., Lakes, T., Zhu, X. X. & Taubenböck, H. Has Dongying developed to a ghost city? - evidence from multi-temporal population estimation based on VHR remote sensing and census counts. Comput. Environ. Urban 78 , 101372 (2019).

Woodworth, M. D. & Wallace, J. L. Seeing ghosts: parsing China’s “ghost city” controversy. Urban Geogr 38 , 1270–1281 (2017).

Lee, J., Park, Y. & Kim, H. W. Evaluation of local comprehensive plans to vacancy issue in a growing and shrinking city. Sustainability-basel 11 , 4966 (2019).

Shi, L. et al. Urbanization that hides in the dark – spotting china’s “ghost neighborhoods” from space. Landsc Urban Plan 200 , 103822 (2020).

Roger S. A. The Empty City of Ordos, China: A Modern Ghost Town. Retrieved from https://weburbanist.com/2011/01/10/the-empty-city-of-ordos-China-a-modern-ghost-town/ (2011).

Shepard, W. Ghost Cities of China, Vol. 1 ( Zed Books Asia Studies Archive, 2015). ISBN: 978-1-78360-220-9

Ge, W., Yang, H., Zhu, X., Ma, M. & Yang, Y. Ghost city extraction and rate estimation in China based on NPP-VIIRS night-time light data. ISPRS Int. J. Geoinf 7 , 1–17 (2018).

CAS   Google Scholar  

Ma, X., Tong, X., Lu, S., Li, C. & Ma, Z. A multisource remotely sensed data oriented method for “ghost city” phenomenon identification. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 11 , 2310–2319 (2018).

Chi, G., Liu, Y. & Wu, H. Ghost cities analysis based on positioning data in China. Comput. Sci. 68 , 1150–1156 (2015).

Google Scholar  

Jin, X. et al. Evaluating cities’ vitality and identifying ghost cities in China with emerging geographical data. Cities 63 , 98–109 (2017).

Xie, S., Gong, J., Huang, X. & Xia, J. A study on coordination of urbanization in the pearl river delta between 1990 and 2015. Guangdong Land Science 16 , 18–26 (2017).

Lane, M. The carrying capacity imperative: assessing regional carrying capacity methodologies for sustainable land-use planning. Land Use Policy 27 , 1038–1045 (2010).

Yan, H., Liu, F., Liu, J., Xiao, X. & Qin, Y. Status of land use intensity in China and its impacts on land carrying capacity. J. Geogr. Sci. 27 , 387–402 (2017).

Agustizar, A., Muryani, C. & Sarwono, S. Agricultural land carrying capacity and shift of land use in upstream of Grompol watershed, central java province. IOP C. Ser. Earth Env. 145 , 012070 (2018).

Li, M. Evolution of Chinese ghost cities. China Perspect. 1 , 69–78 (2017).

Zhao, M., Xu, G., Jong, M. D., Li, X. & Zhang, P. Examining the density and diversity of human activity in the built environment: the case of the pearl river delta, China. Sustainability-basel 12 , 1–21 (2020).

Yan, S., Wang, X., Chen, J., Zeng, M. & Wang, L. Quantitative research on regional carrying capacity in eastern China’s economic zone based on GIS. Res. Ind. 15 , 120–130 (2013). In Chinese.

Tang, H. & Jiang, J. Research on the population carrying capacity of the land resources in the economic area of Zhujiang Delta. Chin. Geogr. Sci. 11 , 174–180 (2001).

Zhang, X., Du, S. & Zhang, J. How do people understand convenience-of-living in cities? A multiscale geographic investigation in Beijing. ISPRS J. Photogramm. Remote Sens. 148 , 87–102 (2019).

Shepard W. The myth of China’s ghost cities. Reuters , Retrieved from https://www.reuters.com/article/idUS170445800220150422 (2015).

Ferris, J., Norman, C. & Sempik, J. People, land and sustainability: community gardens and the social dimension of sustainable development. Soc. Policy Adm. 35 , 559–568 (2010).

Zhang, X., Du, S., Du, S. & Liu, B. How do land-use patterns influence residential environment quality? a multiscale geographic survey in Beijing. Remote Sens. Environ 249 , 112014 (2020).

Christianawati, A., Prima, F. R., Aryati, S., Salsabila, G. & Thilfatantil, M. H. Analysis of the relationship of land carrying capacity and building area towards the development of Kualanamu International Airport in 2010 and 2017. J. Degrade. Min. Land Manage. 7 , 2095–2103 (2020).

Yao, Q. et al. A comparative analysis on assessment of land carrying capacity with ecological footprint analysis and index system method. Plos One 10 , e0130315 (2015).

Holden, S. T. & Otsuka, K. The roles of land tenure reforms and land markets in the context of population growth and land use intensification in Africa. Food Policy 48 , 88–97 (2014).

Sun, J., Di, L., Sun, Z., Wang, J. & Wu, Y. Estimation of GDP using deep learning with NPP-VIIRS imagery and land cover data at the county-level in conus. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 13 , 1400–1415 (2020).

Chen, Y., Chen, X., Liu, Z. & Li, X. Understanding the spatial organization of urban functions based on co-location patterns mining: a comparative analysis for 25 Chinese cities. Cities 97 , 102563 (2020).

Investment Times. Ranking list of ghost city index in China mainland . Retrieved from https://km.loupan.com/html/news/201410/1464288.html (2014).

Wang, W. Study on the act of government in the urbanization process of Hebei province—taking Qinghuangdao as an example. In Proceedings of the 2013 International Academic Workshop on Social Science , 466–469 (2013).

Zou, Z. et al. Valuing natural capital amidst rapid urbanization: assessing the gross ecosystem product (GEP) of China’s ‘Chang-Zhu-Tan’ megacity. Environ. Res. Lett. 15 , 124019 (2020).

Fujita, K. A world city and flexible specialization: restructuring of the Tokyo metropolis. Int. J. Urban Reg. Res. 15 , 269–284 (2010).

Xu, D. & Liu, J. The analysis of real estate investment value based on prospect theory-with an example in Hainan province. Rev. de la Fac. de Ing. 32 , 63–69 (2017).

Xian, S., Li, L. & Qi, Z. Toward a sustainable urban expansion: a case study of Zhuhai, China. J. Clean. Prod. 230 , 276–285 (2019).

Cooke, P. After the contagion. ghost city centres: closed “smart” or open greener? Sustainability-basel 13 , 3071 (2021).

Wang, Q., Li, R. & Cheong, K. C. Shandong’s Yintan town and China’s “ghost city” phenomenon. Sustainability-basel 11 , 4584 (2019).

Melchiorri, M., Pesaresi, M., Florczyk, A. J., Corbane, C. & Kemper, T. Principles and applications of the global human settlement layer as baseline for the land use efficiency indicator—SDG 11.3.1. ISPRS Int. J. Geoinf 8 , 96–98 (2019).

Schiavina, M. et al. Multi-scale estimation of land use efficiency (SDG 11.3.1) across 25 years using global open and free data. Sustainability-basel 11 , 5674 (2019).

Jiang, H. et al. An assessment of urbanization sustainability in China between 1990 and 2015 using land use efficiency indicators. npj Urban Sustainability 1 , 1–13 (2021).

Mohapatra, R. & Wu, C. Modeling Urban Growth at a Micro Level: A Panel Data Analysis. Int. J. Appl. Geospat. Res. 6 , 36–52 (2018).

Mudau, N. et al. Assessment of SDG Indicator 11.3.1 and Urban Growth Trends of Major and Small Cities in South Africa. Sustainability-basel 12 , 7063 (2020).

Du, S., Du, S., Liu, B. & Zhang, X. Mapping large-scale and fine-grained urban functional zones from VHR images using a multi-scale semantic segmentation network and object based approach. Remote Sens. Environ. 261 , 112480 (2021).

Chen, L., Papandreou, G., Kokkinos, I., Murphy, K. & Yuille, A. L. DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs. IEEE Trans. Pattern Anal. Mach. Intell. 40 , 834–848 (2018).

Baatz, M., & Arno Schape, A. Multiresolution segmentation: An optimization approach for high quality multi-scale image segmentation, Vol. 1 ( Angewandte Geographische Informationsverarbeitung , 2020).

Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition , 4700–4708 (2017). https://ieeexplore.ieee.org/document/8099726

Yu, F., & Koltun, V. Multi-scale context aggregation by dilated convolutions. arXiv : 1511 .07122 (2015). https://arxiv.org/abs/1511.07122

Kingma, D. P., & Ba, J. Adam: A Method for Stochastic Optimization. arXiv: Learning , 1412 .6980 (2014). https://arxiv.org/abs/1412.6980

Silván-Cárdenas, J. & Wang, L. Sub-pixel confusion-uncertainty matrix for assessing soft classifications. Remote Sens. Environ. 112 , 1081–1095 (2008).

O’Neill, R. V. et al. Indices of landscape pattern. Landsc Ecol 3 , 153–162 (1988).

Pontius, R. G. Jr, Cornell, J. D. & Hall, C. A. Modeling the spatial pattern of land-use change with GEOMOD2: application and validation for Costa Rica. Agric Ecosyst Environ 85 , 191–203 (2001).

Gavrilidis, A. A., Ciocănea, C. M., Niţă, M. R., Onose, D. A. & Năstase, I. I. Urban landscape quality index–planning tool for evaluating urban landscapes and improving the quality of life. Procedia Environ. Sci. 32 , 155–167 (2016).

Rodriguez, A. & Laio, A. Clustering by fast search and find of density peaks. Science 344 , 1492 (2014).

Ding, S., Du, M., Sun, T., Xu, X. & Xue, Y. An entropy-based density peaks clustering algorithm for mixed type data employing fuzzy neighborhood. Knowl. Based Syst. 133 , 294–313 (2017).

E, W., Ma, C., Wu, L. & Wojtowytsch, S. Towards a Mathematical Understanding of Neural Network-Based Machine Learning: What We Know and What We Don’t. CSIAM Trans. Appl. Math. 4 , 561–615 (2020).

Li, W. et al. Stacked autoencoder-based deep learning for remote-sensing image classification: a case study of African land-cover mapping. Int. J. Remote Sens. 37 , 5632–5646 (2016).

Vincent, P., Larochelle, H., Bengio, Y., & Manzagol, P. A. Extracting and Composing Robust Features with Denoising Autoencoders. Machine Learning, Proceedings of the Twenty-Fifth International Conference (ICML 2008) , Helsinki, Finland, 1096–1103 (2008). https://dl.acm.org/doi/10.1145/1390156.1390294

Yin, X., Jan, G., Egbert, A. L., Jan, V. & Huub, J. S. A flexible sigmoid function of determinate growth. Ann. Bot. 91 , 753–753 (2003).

Efron, B. Discussion of least square regression, least angle regression. Ann. Stat. 32 , 465–469 (2004).

Download references


The work presented in this paper is funded by the National Key Research and Development Program of China (No. 2021YFE0117100), National Natural Science Foundation of China (No. 42001327) and the China Postdoctoral Science Foundation (No. 2019M660003 and No.2020T130005).

Author information

Authors and affiliations.

Institute of Remote Sensing and GIS, Peking University, Beijing, 100871, China

Xiuyuan Zhang, Shihong Du, Shouhang Du, Bo Liu & Yuning Feng

German Remote Sensing Data Center, German Aerospace Center, Wessling, 82234, Germany

Hannes Taubenböck

Institute for Geography and Geology, Julius-Maximilians-Universität Würzburg, Würzburg, 97074, Germany

Department of Geography, National University of Singapore, Singapore, 117570, Singapore

Yi-Chen Wang

You can also search for this author in PubMed   Google Scholar


Z.X. and D.S. designed the research and conduct the experiments, H.T. provided conceptual refinements, D.S., F.Y., and L.B. collected the land observation data, Z.X. and D.S. led writing of the paper, W.Y. and H.T. revised the paper, and all authors contributed to interpreting the results and writing the paper.

Corresponding author

Correspondence to Shihong Du .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Zhang, X., Du, S., Taubenböck, H. et al. Underload city conceptual approach extending ghost city studies. npj Urban Sustain 2 , 15 (2022). https://doi.org/10.1038/s42949-022-00057-x

Download citation

Received : 23 September 2021

Accepted : 24 May 2022

Published : 17 June 2022

DOI : https://doi.org/10.1038/s42949-022-00057-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

  • Explore articles by subject
  • Guide to authors
  • Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

ghost cities definition

  • Skip to main content
  • Skip to primary sidebar

Writing Tips Oasis

Writing Tips Oasis

How to Describe a Ghost Town in a Story

By Isobel Coughlan

how to describe a ghost town in a story

Do you need some tips on how to describe a ghost town in a story? Use the 10 words featured in this post as guide to help you.

Somewhere with a scary atmosphere that could be haunted.

“He didn’t want to visit the spooky ghost town, but he was worried what the group would call him if he said no.”

“The spooky ghost town was real and scary, unlike those kitsch fairground rides that can’t even scare children.”

How it Adds Description

The word “spooky” is a perfect pairing for a ghost town as it implies an area is scary or even haunted. If your ghost town is literally home to spirits or ghosts, this word can signify their presence. However, it can also point to a general unpleasant atmosphere and portray that your characters are creeped out .

2. Intimidating

Someone or somewhere that’s frightening to the point you lose confidence .

“She took one step towards the intimidating ghost town and changed her mind. She wasn’t going to face her fears today.”

“They looked at the intimidating ghost town and it looked back, taunting them with its presence.”

If your ghost town scares your characters, the word “intimidating” can show the effect it’s having on them. “Intimidating” shows someone is feeling nervous or frightened, and this is a perfect way to show the intensity of your ghost-like setting. It can also foreshadow future plot points in the town, ideal if you want to hint at the future.

Somewhere that’s home to ghosts or spirits.

“But the haunted ghost town is just an old tale… Isn’t it?”

“She flat-out refused to talk about the haunted ghost town, and everyone had to respect her decision.”

The adjective “haunted” clearly implies that the location is home to ghosts or spirits. This is a powerful word to use if you’re writing a horror novel, as it helps to build a scary setting. It can also hint at the ghost town’s past, and you can use this word to build up curiosity surrounding your fictional world’s history.

Somewhere very quiet and almost silent.

“The hushed ghost town didn’t bother her. It was the people back home that got on her nerves.”

“He was shocked by the hushed ghost town. He expected hustle and bustle in all the streets.”

The word “hushed” conveys a place is very quiet or silent. If your ghost town is uninhabited or home to a scare population, “hushed” can portray the atmosphere there. “Hushed” can also build suspense, and you can pair this adjective with creepy action to scare your reader and characters.

Something or somewhere not being used by anyone.

“Don’t turn left off the freeway, there’s an old vacant ghost town over there. People haven’t lived there in years.”

“He crept through the vacant ghost town as if someone was watching him, but no one had lived here since the accident.”

The word “vacant” describes a place that’s completely empty, which is perfect when describing a ghost town. This word lets your reader know there’s no inhabitants. It can also be used to build an image of a neglected place, for example a run-down town that has bad infrastructure.

6. Disgraced

Somewhere that has lost the respect of the authorities of people.

“The locals had left the disgraced ghost town after the accident, and they had no intentions of coming back.”

“The disgraced ghost town never regained respect, and it has been left to rot.”

If your ghost town has been abandoned because of an incident or stigma, the word “disgraced” can help explain the situation to your reader. “Disgraced” describes somewhere that’s fallen out of favor with local opinion, and this can hint that something bad happened in the town. It also implies the town is a bad place, and the inhabitants questionable.

7. Chilling

Somewhere very scary.

“Even the thought of the chilling ghost town made her hair stand up on end.”

“The chilling ghost town made him question his courage; he did not feel safe there at all.”

If your ghost town is unpleasant and scary, “chilling” is a helpful word to use. This adjective shows that the location has a physical effect on the characters, as “chilling” refers to a type of fear that resonates in the body.

Somewhere far away from urban areas or cities.

“She didn’t want to leave the comfort of the city for a remote ghost town, but she had to honor her manager’s instructions.”

“Don’t talk to me about community, you live in a remote ghost town!”

Ghost towns with few inhabitants are common as you move further away from urban areas. Therefore, “remote” is a good adjective to use if you want to illustrate more about the ghost town’s location. In a horror story, “remote” can create a sense of helplessness, as there are no nearby authorities to help the characters.

9. Disturbing

Somewhere that evokes feelings of sadness or worry.

“It was a disturbing ghost town. All the houses looked as if they were frozen in time.”

“She awoke in the disturbing ghost town, and her stomach instantly sank.”

If you simply want to illustrate how horrible your ghost town is, the word “disturbing” can help. This adjective points to a location that’s scary or physically unpleasant, which is great for building a clear mental image of the settlement.

Somewhere that makes you feel nervous or is slightly strange .

“She couldn’t take the eerie ghost town anymore; it was too quiet and uncanny.”

“Together, they explored the eerie ghost town, but they were shocked at the reason it was so quiet.”

“Eerie” is linked to places that are odd or scary, and this is a great way to insight fear in your reader. The word “eerie” can also help to portray your character’s anxiety, especially when you pair it with more negative descriptive language.


  1. 15 Fascinating Ghost Towns to See Across America

    ghost cities definition

  2. 20 Scariest Ghost Towns Around the World

    ghost cities definition

  3. 14 Famous Ghost Towns Around The Globe

    ghost cities definition

  4. 20 Scariest Ghost Towns Around the World

    ghost cities definition

  5. 35 Creepy Abandoned Cities & Ghost Towns That Will Give You the Chills

    ghost cities definition

  6. 6 Ghost Towns Around The World

    ghost cities definition


  1. Ghosts From Different Countries


  3. Ghost cities

  4. Why is China full of Ghost Cities?

  5. China's Ghost Cities Shock Everyone

  6. Ghost Cities in the World! #ghostcities #ghost #traveldestinations


  1. Ghost town

    Category: Geography & Travel Related Topics: town See all related content → ghost town, town that was once an active community but has since been abandoned by all or nearly all of its residents. Ghost towns are found on every continent.

  2. Ghost town

    A ghost town, deserted city, extinct town or abandoned city is an abandoned settlement, usually one that contains substantial visible remaining buildings and infrastructure such as roads.

  3. Ghost town Definition & Meaning

    : a once-flourishing town wholly or nearly deserted usually as a result of the exhaustion of some natural resource Examples of ghost town in a Sentence After all the gold was mined, the place became a ghost town.


    a town where few or no people now live SMART Vocabulary: related words and phrases Towns & regions: towns, cities & villages (general) aerotropolis anti-city anti-urban backwater burgher Cantabrigian central city civic conurbation garden city metro multi-city municipal native place non-municipal open city retirement community small-town smoke

  5. State of Oregon: Oregon Ghost Towns

    Ghost towns defy easy explanation. Many are abandoned villages or cities, often with substantial visible remains. Crumbling buildings, lonely cemeteries, and rusting industrial equipment mark the graves of these communities. Some have no residents besides crows, coyotes, and rattlesnakes.

  6. 6 Famous Ghost Towns and Abandoned Cities

    1. Pripyat, Ukraine Getty Images / Sean Gallup Students chairs stand on rotting floorboards in an auditorium of an abandoned school on September 30, 2015 in Pripyat, Ukraine. The city lies in the...

  7. What Is a Ghost Town?

    What Is a Ghost Town? Listen Print 28 Comments SOCIAL STUDIES — Geography Have You Ever Wondered... What is a ghost town? What causes ghost towns? Where are some famous ghost towns? Tags: See All Tags activity, boomtown, dam, disaster, economic, fire, flood, ghost, highway, natural, radiation, resource, Route 66, tourism, town, transportation,

  8. Ghost Towns

    GHOST TOWNS, the term used to identify communities that once prospered but later declined and were deserted, usually due to economic shifts and reversals.

  9. Ghost town

    A ghost town, deserted city, extinct town or abandoned city is an abandoned settlement, usually one that contains substantial visible remaining buildings and infrastructure such as roads. A town often becomes a ghost town because the economic activity that supported it has failed or ended for any reason . The town may also have declined because of natural or human-caused disasters such as ...

  10. GHOST TOWN definition and meaning

    A deserted town, esp one in the western US that was formerly a boom town.... Click for English pronunciations, examples sentences, video.

  11. Why Is Ghost Town Called Ghost Town?

    The term "ghost town" describes a deserted or abandoned settlement with visible remnants of buildings, infrastructure, or other structures. Historian T. Lindsey Baker states that a ghost town is "a town for which the reason for being no longer exists."

  12. What is a Ghost Town? WMH Town Classifications Explained

    A ghost town is an abandoned village, town or city. A town often becomes a ghost town because the economic activity that supported it has failed, or due to natural or human-caused disasters such as floods, government actions, uncontrolled lawlessness, war, or nuclear disasters.

  13. What are ghost towns?

    Ghost towns can be perceived as an image of burning or creating abandoned buildings. It can be alternatively called as an abandoned village, city or town that might contain substantial visible remaining buildings and infrastructure. Ghost towns are important when they get absorbed, especially for the use of prospector.

  14. Ghost towns

    South Pass City, near Lander (Wyoming) USA on the Oregon Trail, is a 19th-century gold mining town abandoned in the 1950s. Now a tourist ghost town. Walhalla ( Gippsland, Victoria) Australia — An 1863 gold rush town, the last mine closed in 1914. Portions of the town were rebuilt after 1977 for tourism and cottages.

  15. GHOST TOWN Definition & Usage Examples

    Ghost town definition: . See examples of GHOST TOWN used in a sentence.


    ghost town meaning: 1. a town where few or no people now live 2. a town where few or no people now live 3. a town that…. Learn more.

  17. Underload city conceptual approach extending ghost city studies

    First of all, these two concepts have significant differences in definition, as ghost cities mostly consider population density, but underload cities measure population, GDP, and LCC; thus ...

  18. Evolution of Chinese Ghost Cities

    2 Despite the above reports, there are methodological difficulties in assessing the scope of the ghost city phenomenon, in particular the lack of a standard definition of a "ghost city." Shepard Wade listed several definitions of ghost cities in his book. 8 One point is clear: it is necessary to redefine the ghost city, which was originally ...

  19. Exploring Lideco Hanoi: An Abandoned Ghost Town in Vietnam

    A privately owned Vietnamese construction company, Tu Liem Urban Development JSC - AKA Lideco - broke ground on the eponymous 38-hectare estate about a decade ago. Between 2007 and 2012, more than 600 French-style villas were erected, the beginnings of what was supposed to be a model housing project. Today, all but a few of the homes remain ...

  20. How to Describe a Ghost Town in a Story

    Definition. Somewhere with a scary atmosphere that could be haunted. ... Ghost towns with few inhabitants are common as you move further away from urban areas. Therefore, "remote" is a good adjective to use if you want to illustrate more about the ghost town's location. In a horror story, "remote" can create a sense of helplessness ...

  21. Opportunity for a Paradigm Shift? The Case of Changzhou

    ing the scope of the ghost city phenomenon, in particular the lack of a stan-dard definition of a "ghost city." Shepard Wade listed several definitions of ghost cities in his book.(8) One point is clear: it is necessary to redefine the ghost city, which was originally considered "a place that has died" or "an abandoned city."

  22. What are Ghost Spaces?

    Ghost spaces are underutilized public spaces in an urban area. The primary cause of ghost spaces is a poor design that lacks interesting features or a pleasant atmosphere that draws people. Secondary factors in some cities are safety, perceived safety, pollution and inadequate transportation. In some cases, committees of experts such as ...