Global and country-level estimates of human population at high altitude

1High altitude introduces environmental stressors that are distinct from sea level. Hypobaric hypoxia, colder and drier climates, and increased radiation challenge reproductive success. In addition to the direct effects of high altitude, the health of high-altitude residents (particularly in rural communities) is compounded by vulnerability to food and water insecurity associated with land degradation and the concurrence of conflict and climate change. Population estimates for the number of humans residing at high altitude varies widely. Older estimates were either guesses or based upon assigning a percentage of select countries’ populations residing at high altitude. Cohen and Small produced gridded population estimates based on censuses and coregistered these demographic data with elevation data, estimating that 389.4 million live at >1,500 m. The prevailing assumption is that over 140 million humans live above 2,500 m. This is based on the 139 million estimated to be residing above 2,500 m in 1995 which was calculated by estimating the percentage of populations residing above 2,500 m in select countries (table 1 in ref. ). However, studies using geographic information system (GIS)-based approaches with updated georeferenced data have estimated that, in 2010, there were 83 million residents at >2,500 m, and, in 2017, 74.9 million lived at >2,500 m. These reports do not provide country-level estimates and provide population estimates only for select elevation ranges/thresholds. To understand the impact of life at high altitude on human physiology, adaptation, health, and disease, it is imperative to know how many humans live at high altitude and where they live. Herein, we estimate that 81.6 million humans live above 2500 m, and we provide population estimates at 500-m elevation intervals for every country.

2Over 500 million humans live at ≥1,500 m (6.58% of the total population), 219 million at ≥2,000 m (2.88% of the total population), 81.6 million at ≥2,500 m (1.07% of the total population), 25.2 million at ≥3,000 m (0.33% of the total population), 14.4 million at ≥3,500 m (0.19% of the total population), 6.4 million at ≥4,000 m (0.084% of the total population), 2 million at ≥4,500 m (0.027% of the total population), and 0.31 million at ≥5,000 m (0.004% of the total population). Table 1 presents the standard and model atmosphere-calculated barometric pressure, partial pressure of inspired oxygen, and global populations at 500-m elevation intervals.

3Fig. 1 illustrates that high-altitude regions are present in all continents and shows the 10 most populated countries with altitudes of ≥1,500 m, ≥2,500 m, and ≥3,500 m, and the 10 countries with the greatest percentage of national population residing at ≥1,500 m, ≥2,500 m, and ≥3,500 m. Dataset S1 contains the estimated population at 500-m elevation intervals for every country.

4We used georeferenced population and elevation data to estimate the global and country-level population of humans at high altitude. We estimate that 81.6 million humans live at ≥2,500 m, and that these humans are primarily from non-Western countries. This figure is considerably less than non-GIS−based estimates and expands upon GIS-based estimates by providing country-level estimates at 500-m elevation intervals. We present population for 500-m intervals to provide flexible interpretation of “high altitude,” as the threshold for high altitude to elicit a physiological response varies between individuals and populations. Research is encouraged in countries with considerable high-altitude populations, to understand how the environmental stress (physiological and social) of high altitude impacts physiology, adaptation, health, and disease.

5Humans reside at high altitude in each continent; notably, the majority of humans at high altitude live in non-Western countries. As with medical research generally, high-altitude research has focused on the lowlander of European descent ascending to high altitude. Select permanent high-altitude populations have been studied, with considerable differences in physiological measures reported; however, the populations studied represent just a fraction of humans at high altitude. Among the studied populations, differences in candidate genetic variants that may be adaptive and their phenotypic associations have been identified. The distribution of where humans live at high altitude, presented herein, and their duration of high-altitude residence affords a broad natural experiment to better define the relationships between genotype and phenotype and provide insight into the evolutionary and mechanistic bases of physiological adaptations to high altitude. Further, an expanded understanding of the potential physiological differences arising from life at high altitude will lead to improvements in the diagnosis and treatment of disease, ultimately improving the health and well-being of high-altitude populations.

6Mobile, nomadic, or pastoralist populations and those with lowest income and most at risk are underrepresented in census data. Therefore, we have likely underestimated the most vulnerable populations and may have disproportionately undercounted rural high-altitude populations. The LandScan Global datasets are not recommended to be used as a change or migration tool, as the LandScan database is constantly improving its input data. Further, the LandScan dataset provides only a population count and not demographics (e.g., age, sex, ethnicity, etc.). Nevertheless, LandScan is the community standard for global population distribution.

7Population estimates were calculated using QGIS 3.4.0-Madeira software. Population data were acquired from Oak Ridge National Laboratory’s LandScan Global 2019 dataset. LandScan provides an ambient population count for cells with a spatial resolution of 30 arc-seconds (∼1 km2). LandScan uses the best available census counts to estimate population and applies a spatial distribution model that includes land cover, roads, slope, urban areas, village locations, and high-resolution imagery analysis. The mean elevation data from the Global Multiresolution Terrain Elevation Data (GMTED2010) were also acquired at a resolution of 30 arc-seconds (∼1 km2). Country borders were acquired from the Database of Global Administrative Areas (GADM, version 3.6; https://gadm.org/download_country_v3.html). Further details on the methodology are provided in SI Appendix.