Definition and description of the Montane Regions of the World

By Rahbek, C. Borregaard, M.K., Hermansen, B. Nogues-Bravo, D and J. Fjeldså. 2019. Center for Macroecology, Evolution and Climate.

We present here a map of the mountain regions of the world, accompanying our publication Rahbek et al. 2019a and Rahbek et al. 2019b.

The different definitions of mountains (refs to Korner 2017) and its global inventory have been for long an active area of research. Previous efforts of global mapping of mountain ranges or regions (Kapos et al. 2000; Korner et al. 2011; Korner et al. 2017; Sayre et al. 2018), have used elevation, ruggedness, or combination of them both, to map mountain regions, and supported also the use of climatic, geological or biogeographic units to subdivide mountain ranges in operational research units.

We approach, in accordance to those mapping efforts, the mapping of montane regions based on key topographic features including elevation and slope. Those are geomorphological aspects that control key parameters defining the unique nature of mountain regions, including variation in climatic conditions (elevation, aspect, slope), the effect of gravity in geoecological dynamics (slope) and the atmospheric pressure and the oxygen level (elevation).

These basic principles to define and map mountain regions, and the consideration of climatic conditions and geological context, is custom-made to better explore biogeographical patterns (i.e., species richness) of animals and plants, at global scale, across mountain regions of the world. It should therefore be used with caution for applications when using only fine-scale resolution biological or when used in other disciplines or research contexts, like climatology, meteorology, hydrology, or cryology.

Exploring previous attempts to map mounatin areas

To circumscribe montane regions we firstly explored the aforementioned previous attempts to map globally mountain areas (Kapos et al. 2000; Korner et al. 2011). We mainly relied upon Kapos et al. (2000) classification. Kapos proposed a classification scheme of seven classes to map mountains: the upper three classes are delimited purely by elevation: 2500-3499 meters; 3500-4499 meters and 4500 meters.

Land between 1500 and 2499 m is classed as mountain if it slopes more than 2 degrees; this threshold proved to be appropriate for eliminating mid-elevation plateaux.  Between 1000 and 1499 m, land that either surpasses a steeper slope threshold of 5 degrees, or has a local elevation range of 300 m or more, is classified as mountain.

Between 300 and 999 meters, land was classifying as mountainous if the local elevation range was 300 m or more. However, this fine resolution means that transitions between montane areas and surrounding lowlands will in some cases appear as broad habitat mosaics, hundreds of isolated pixels which marginally qualify as “montane”, so we opted to delineate the polygons, a solution that has been also recently adopted in Korner et al. 2017.

Using Kapos classification and map as a basis, we drawn polygons, digitized in ArcGIS 10.5, to circumscribe grid-cells classified as mountains by Kapos et al. 2000 when 1) an aggrupation of mountain grid-cells represents an identifiable mountain unit recognized in the mountain literature (i.e. Pyrenees), 2) an aggrupation of mountain grid-cells contains approximately an minimum area of 2500 square kilometers.

Because of polygons may include a) grid-cells do not define by mountains under Kapos classification, b) grid-cells of high elevation but of low slope (i.e, parts of the Tibetan Plateu), or c) to exclude rugged terrain that may be considered as mountainous, we checked for each polygon those aspects through visual inspection of hundreds of satellite pictures in Google Earth, and modify the first version of the polygons accordingly.

Creating more coherent units

Given that the emphasis of our reviews (Rahbek et al. 2019a; Rahbek et al. 2019b) focuses on the biogeographic and abiotic aspects (i.e., climate or geology) of mountains, and for operational purposes, some mayor mountain ranges were divided in smaller units constituting more coherent units from a climatic and biogeographic point of view.

The Andes, for example, is subdivided in 5 polygons: Northern Andes, Western Central Andes, Andean Yungas, Southern Andes and Patagonian Andes. Several highlands are separated by disjunctions of tens of kilometers, but should still be treated as one region, based on biogeographic uniformity. Other areas, although abutting, have been recognized as different biogeographic units, and could therefore be recognized as different montane regions, if possible.

This reservation refers to the cases where recognized biogeographic units in fact represent a jigsaw puzzle of interlocking ecological and elevational zones, which would be difficult to tease apart, considering the coarse resolution of most of the environmental and biodiversity datasets that are available for analysis.

This is for instance the case in the Sino-Malayan mountain world and within the tropical Andes region, where we adopted a simplified scheme. Our main source for separation of broadly recognized biogeographic regions was Udvardy’s World Biogeographic Regions (Udvardy 1975). This is an expert assessment based on qualitative evidence. A modern, hierarchical cluster analysis of mammal distributions (Kreft & Jetz 2010) generally confirms the boundaries of broad biogeographic regions, but this kind of analysis generally fails to recognize montane areas, as the range-restricted montane species provide less connectivity than the widespread species inhabiting the lowland matrix habitat.

However, as more detailed distributional data become available it is possible now to define areas of endemism in greater detail than in the past, and we therefore supplemented Udvardy’s broad regionalization with boundary placements based on the most important endemic bird areas (Stattersfield et al. 1998).

Presenting a global inventory of the mountain regions

Our resulting map presents a global inventory of the mountain regions of the world encompassing 136 unique polygons, and covering approximately the 25% of the emerged lands of the planet (32.9 millions of square kilometers).

This is in overall agreement with previous efforts, for which mountain areas are estimated to be roughly between a 20% to a 30% of the emerged lands (see Korner et al. 2017 and Sayre et al. 2018), and similar to the 24% in Kapos et al. 2001. We have opted for coherent recognizable mountain units, favoring broad and large polygons when possible, to facilitate its use to analyze global patterns of biodiversity.

The majority of global datasets for vertebrate and plant distributions range between 1 to 4 degrees (grid cells from 10,000 to 160,000 square kilometers), and therefore map constituted mainly by small and fragmented polygons with areas smaller than the grain resolution of global biodiversity datasets would not facilitate an adequate use of our map of mountain regions.

Moreover, the high spatial similarity of the geographical extent on mountain regions across the world and the high coincidence on the square kilometers covered globally, and by continent, by mountain regions between our map and the previous classification (Korner et al, 2017), enhance the comparability of these two mountain maps. In Korner et al. 2017, the area in millions of square kilometers of mountain polygons for Europe is 1.4, for Asia 14.0, for North America 4.7 or 3.3 for South America.

In our map, the area in millions of square kilometers for Europe is 1.6 and for Asia is 14.8, resembling previous attempts. Our estimates for North America and South America, 6.7 and 4.89 respectively, are larger than in previous attempts.

References

• Kapos V, Rhind J, Edwards M, Price M, Ravilious C. 2000. Developing a map of the world's mountain forests. In: Price M, Butt N, editors. Forests in Sustainable Mountain Development: A State of Knowledge Report for 2000. IUFRO [International Union of Forest Research Organizations] Research Series 5. Oxon, NY: CAB International Publishing, pp 4–9.
  http://dx.doi.org/10.1007/1-4020-3508-X_52
  
• Körner C, Paulsen J, Spehn E. 2011. A definition of mountains and their bioclimatic belts for global comparisons of biodiversity data. Alpine Botany 121:73–78.
  
  
• Körner C, Jetz W, Paulsen J, Payne D, Rudmann-Maurer K, Spehn E. 2017. A global inventory of mountains for bio-geographical applications. Alpine Botany 127:1–15.
  
  
• Sayre R, Frye C, Karagulle D, Krauer J, Beyer S, Aniello P, Wright D, Payne D, Adler C, Warner H, Van Sistine P, Cress J. 2018. [This issue.] A new high-resolution map of world mountains and an online tool for visualizing and comparing characterizations of global mountain distributions. Mountain Research and Development 38(3):240–249.