{"id":4640,"date":"2015-03-25T07:11:12","date_gmt":"2015-03-25T14:11:12","guid":{"rendered":"http:\/\/www.esri.com\/about\/newsroom\/insider\/?p=4640"},"modified":"2015-03-25T07:11:12","modified_gmt":"2015-03-25T14:11:12","slug":"map-gives-new-insights-into-global-population","status":"publish","type":"insider","link":"https:\/\/www.esri.com\/about\/newsroom\/insider\/map-gives-new-insights-into-global-population","title":{"rendered":"Map Gives New Insights into Global Population"},"content":{"rendered":"

Esri\u2019s World Population Estimate, a new probability surface that estimates the location and count of people throughout the world, is now available in ArcGIS Online.<\/em><\/strong>
\nEsri has been producing a global population estimate in ArcGIS Online for several years. This data is in the form of point features with population counts and characteristics assigned to each point; and it is used behind the scenes in apps such as Business Analyst Online<\/a>, Community Analyst<\/a>, and Esri Maps Apps<\/a>. That may sound simple, but those points are big data; with nearly a billion locations represented. The Geographic Data Enrichment tools depend on those points as the basis for describing the characteristics of local populations in countries lacking a census or countries that do not make detailed census data available.
\n

\"\"
World Population Estimate map for Jakarta, Indonesia.<\/figcaption><\/figure>\nBased on this earlier point data work, Esri released the World Population Estimate (WPE)<\/a> in December 2014. WPE takes the form of a raster surface, which is far easier to make available in ArcGIS Online and use in analysis models than the previous point data.
\n
\nHow the WPE was made<\/strong>
\nEarl Nordstrand, a senior researcher at Esri, has been working on and refining the method for producing the WPE for several years. Nordstrand\u2019s method<\/a> uses data rather than the traditional approaches of sending census enumerators out to count people or estimates of economic activity as a surrogate for counting households or people. Nordstrand\u2019s data includes satellite imagery for land cover from MDA\u2019s NaturalVue<\/a>, and he used a texture-detecting model to find the areas that are likely to have people. Because land cover data inherently includes information about where surface water, agriculture, and permanent ice, these can be eliminated, because there is a high level of certainty that people as a rule do not live in these locations.
\n
\"\"
World Population Estimate map for Mexico City.<\/figcaption><\/figure>\nBy itself, this model does not produce a perfect result, so Nordstrand used other data like road intersections and extreme climates to increase the certainty of some locations while eliminating others. Place name points from GeoNames.org<\/a> were also used; though this definitely improves the result, the hard part, according to Nordstrand, is these points do not indicate how large or spread out these places are. Nordstrand also used census data where it existed to calibrate and further refine where people were likely to be located. In some cases, this census data also inherently contained information about where people do not<\/em> live, including locations of airports, industrial complexes, and the like.
\n
\nBy the Numbers<\/strong>
\nThe WPE is based on input data that ranged in date from 1990 to 2013. The estimated number of people is 6.89 billion which is consistent with other estimates of the world\u2019s population in 2013. Based on World Bank estimates, there are just over 129.7million square kilometers of land (this does not include inland water bodies or Antarctica, but does include extreme terrain such as mountains). The WPE shows that humans live on about 7.9% of this land area. The WPE shows that 55.48% of people live in urban areas, with concentrations of people over 2,500 persons per square kilometer, and that these urban areas only occupy about 0.56% of the earth\u2019s surface.
\nYes, that means that over half the world\u2019s population lives on less than 1% of the land area. Such proportions have been documented and available from the World Bank and other organizations for many years, and Nordstrand used them to verify that his model was producing valid results.
\n
\"\"
World Population Estimate map for Tokyo, Japan.<\/figcaption><\/figure>\n
\nUsing the World Population Estimate<\/strong>
\nKnowing where people are likely to be living is important for many areas of inquiry, including:<\/p>\n