Now, a team from the US and China has refined its analysis method for creating a map of the world’s urban regions from satellite data. The researchers used the defense meteorological satellite program/operational linescan system (DMSP/OLS) nighttime stable-light (NTL) data, no less.

"The transformation of terrestrial environments by urbanization has been accelerating over the past 30 years, and is likely to continue due to population growth and migration," Yuyu Zhou of the Pacific Northwest National Laboratory, US, told environmentalresearchweb. "Accompanying this process is a range of environmental consequences, with important socio-economic implications. Understanding these interactions is essential to tackle ongoing global changes; the level of understanding hinges greatly on the availability of accurate and consistent information on the distribution and extent of urban areas."

Zhou believes the team’s new method can build a globally consistent and temporally updateable urban extent map from the nightlights data in an efficient and cost-effective way.

The 1 km spatial resolution map showed that globally, urbanization currently stands at about 0.54%. Europe was the most urbanized, at 2.3%, with North America’s 1.2% putting it in second place. Asia was 0.7% urbanized, the figure for Latin America was 0.4%, the former Soviet Union came in at 0.2% and Oceania at just 0.1%. Africa was the least urbanized, with towns and cities covering less than 0.1% of the land area.

By country, urban land area ranged from less than 0.01% to more than 10%. In 70% of countries urbanization stood at less than 1%; it was below 5% in more than 90% of countries. The countries with high urbanization were mainly in Europe.

The latitudes around 30° north had the biggest urban area, largely because of cities in China and the US. The highest urbanized percentage was further from the equator, around 45° north, due to Europe’s influence. Longitudinally, 80° west, in the vicinity of the Boston—New York—Washington corridor, had a large urban area, but there were also several other spikes.

"This information on global urban extent and its spatial pattern is of great value for developing adaptation and risk management measures for urban infrastructure and systems, in the context of global environmental changes and their impacts on natural ecosystems, people and infrastructure," said Zhou.

Compared to other remote sensing systems, the DMSP/OLS NTL data can sometimes over-predict urban extent. So the researchers set thresholds for defining a pixel as showing city lights, altering these thresholds according to the size of the cluster of light in a "cluster-based optimal thresholds method". Now the team has also included parameterization to estimate the key parameters in the cluster-based method. The results showed good agreement with five other urban extent maps.

The DMSP/OLS NTL data is available back to 1992. "We will extend the developed method temporally over 20 years and build a consistent global urban map series, and evaluate the temporal dynamics of global urbanization and its impacts on global environment," said Zhou.

Zhou and colleagues reported their work in Environmental Research Letters (ERL).

Related links

Related stories