Cranes in the DMZ

The Korean Demilitarized Zone (DMZ), a 4-kilometer-wide and 248-kilometer-long territory across the Korean peninsula, was established as a buffer zone at the end of the Korean War (1950-1953) to deter military conflicts between North and South Koreas. Remained as a no-man’s land for over 70 years, the DMZ is now transformed into a nature reserve, the “accidental” habitat for many endangered species, including several migratory crane species.

The Korean DMZ

Red-crowned cranes (Grus japonensis) and White-naped cranes (Grus vipio) fly from Siberia to spend winter (November to March) in the DMZ and surrounding areas. The grains from rice field provide enough food resources, while the wetlands inside of the DMZ offer safe resting places. The number of cranes increased from several dozens in the 1990s to several thousand by now, making the DMZ and the vicinity as one of the most important wintering sites for these internationally endangered species.

Red-crowned cranes (c) Seunghwa Yoo

White-naped cranes (c) Seunghwa Yo

Cranes have been highly appreciated in the Korean traditional culture, viewed as a symbol of long life and good fortune. They are big birds, which can measure up to 1.5 meters in length and weigh 4 – 7 kilograms. Red-crowned cranes are distinguished by the red patch on their head, while white-naped cranes have a beautiful dark grey body. Juveniles of both species have brownish necks and bodies. We used to have a good population of wintering cranes, but unfortunately lost most of them during the past century through urbanization, industrialization, and the Korean War (1950–1953).

It was a welcome surprise when the cranes started to come again to the DMZ and surrounding areas for the past few decades.
However, the cranes in and around the DMZ are now faced with the new challenges presented by human encroachment into their habitats – relaxed border controls, greenhouses, roads, and other infrastructure – alongside the changing climate. Still, ecological monitoring of the cranes in this area is fairly restricted due to security reasons alongside landmines inside of the DMZ.

What we do:

We have set up a dozen of trail cameras at rice fields of Cheorwon, Gangwon Province. The cameras are located outside of the DMZ to comply with the security restrictions. Our local collaborators, who are farmers and have experiences with monitoring devices in their rice fields, have kindly allowed and helped us to install trail cameras within their properties to monitor cranes and other animals. While we cannot, the cranes fly over the fortified lines, making use of the wetlands in the DMZ and rice fields just outside of it. By collecting and studying remotely-captured crane images, we hope to have a better understanding of the crane behaviors and population dynamics. In doing so, we hope to illustrate the utility of, and the need for, wildlife monitoring assisted by remote-sensing devices and artificial intelligence, especially for the DMZ where human access is restricted.

The trail cams set up to monitor crane activities.

The study area – Northern Cheorwon, Gangwon Province

Research goals

The interdisciplinary research team, which consists of not just ecologists but social scientists, computer engineers, and local residents as well, aims at:

• Examining the ways in which cranes make use of rice fields in winter
• Exploring conservation potential of rice fields to support the crane populations
• Identifying changes in the crane populations in relation to the site-specific changes in the environment
• Developing an AI algorithm that can provide species-specific numbers of the cranes

We need your help

Our trail cameras have been capturing exciting crane images including those of nighttime, which have not been seen much before. We need your help to convert these images to useful datasets to be ready for ecological analyses and the training of eco AI. You can help us by identifying and tagging the cranes on the photographs.

Why not make computers do this task? Yes, machines will, but not yet. Existing algorithms are too rough to identify locally-abundant species – We tried and the machine thinks white-naped cranes are hippopotamus! We also tried the crane counting algorithm, which we have been developing with photographs taken by ecologists. But the machine trained with human-produced images was not competent with device-produced images. This means that we need a specific dataset to improve the AI to work better with trail cam images. In addition to AI training, the dataset will also be used by our ecologist collaborators to better understand crane behaviours regarding their use of watered rice fields and roosting patterns. The findings of this research can help us convince policymakers and developmental agents of the urgent need for protecting crane habitats and improve crane surveillance in and around the DMZ.