Global Statistics of “Mountain” and “Alpine” Research

Using the Web of ScienceH, a bibliometric analysis of the worldwide research activities associated with mountains—or the alpine part of mountains—is presented, according to country, institution, and subject. Half of the human population depends on mountains in one way or another, and mountains cover (depending on the definition) between 12 and 26% of the ice-free terrestrial area. About 20% of the human population lives in mountains or their immediate forelands. One third of all protected areas are in mountains, and they supply water to nearly one half of the world population (Korner and Ohsawa 2005). Which are the countries that are contributing most to scientific research in mountains? Which are the leading institutions? How much are various fields of science contributing to the international scientific mountain literature?

Using the Web of ScienceH, a bibliometric analysis of the worldwide research activities associated with mountains-or the alpine part of mountains-is presented, according to country, institution, and subject.
Half of the human population depends on mountains in one way or another, and mountains cover (depending on the definition) between 12 and 26% of the ice-free terrestrial area. About 20% of the human population lives in mountains or their immediate forelands. One third of all protected areas are in mountains, and they supply water to nearly one half of the world population (Kö rner and Ohsawa 2005). Which are the countries that are contributing most to scientific research in mountains? Which are the leading institutions? How much are various fields of science contributing to the international scientific mountain literature?

Bibliometric data
This analysis is based on scientific publications (articles, proceedings, and reviews) in peer-reviewed journals that are listed in the ISI Web of Science (Science Citation Index Expanded, 1900until October 18, 2008; 95% of all hits date after 1980, 1% of hits are older than 1970) and that use either the word ''mountain'' (including all variants of it, such as mountainous) or the term ''alpine'' in either the title, the abstract, or keywords (instances in the main text are not included). In other words, it is the authors' definition of what they considered a mountain or what they rated as alpine that determined the sum of articles obtained for this assessment. In its strict biogeographic meaning, the term ''alpine'' refers to the vegetationcovered land area above the climatic tree line (ie at elevations where trees cannot grow [Kö rner 2003]). The public uses the term in a much wider sense. In the scientific literature a wider meaning is also often applied, but the term certainly refers to ''high mountain terrain'' worldwide, not just in the Alps. It became customary in some disciplines to replace ''alpine'' by ''Andean'' or ''Afro-alpine'' in the respective regions. In this bibliometric survey, most of such cases may be covered by the term ''mountain,'' but certainly not exhaustively. With the two terms ''mountain-'' and ''alpine'' most of the relevant literature should be covered. With the term ''mountain-'' many more fields of science are included than with ''alpine'' (eg medicine and engineering) so that a ranking by scientific fields became rather obscure and was thus restricted to ''alpine.'' ''Elevation'' and ''altitude'' are terms so widely used (eg in aeronautics) that using them in a search would make it nearly impossible to identify mountain-or alpinespecific works among the results (Table 1).
Country and institution affiliation is provided by ISI for each individual author in all works including multiauthor papers. Hence, a single paper may reflect contributions from different countries, which, in fact, makes the country statistics more robust. We assume that any remaining uncertainty is random, not affecting the results. Institutions are commonly universities or defined (by one address) research laboratories. Multilocus institutions such as the University of Colorado system, or the Chinese Academy of Sciences complex, or Centre National de la Recherche Scientifique (CNRS) in France do not match this criterion and thus do not permit a fair comparison of institutions; hence, these unspecific institutions are not ranked. The potential bias introduced by the size of universities

MountainNotes Mountain Research and Development (MRD)
An international, peer-reviewed open access journal published by the International Mountain Society (IMS) www.mrd-journal.org was not accounted for. As the results will show, all this hardly affected the outcome, because all leading universities in the concerned fields are among the smaller guild, by international standards, and thus, large universities are found at low ranks, despite their size. In the following analysis, results obtained with the search term ''mountain-'' and ''alpine'' are treated separately.

Results of bibliometric analysis
Of all 14,226 publications that matched the search term ''alpine,'' 56% were authored or coauthored by researchers situated in either the United States (20%), Switzerland (15%), France (11%), or Italy (10%) ( Table 2). On a proportional basis (ie per number of inhabitants [ Table 3]), Switzerland is leading, with 282 published works per one million inhabitants, followed by New Zealand (151), Austria (145), and Norway (94), with the nations coleading in absolute terms falling back to only 24 publications per million in France and Italy and 10 in the United States. The absolute numbers of publications found under ''mountain-'' are dominated by engineering, mining, and medicine and show that the United States hold a 40% overall share (Table 4). The per capita ranking of ''mountain-'' publications is similar to the result for ''alpine-'' (Table 5), with Switzerland, New Zealand, Norway, Scotland, Canada, and Austria ranking as the top six. The results of this bibliometric analysis most likely reflect the combined effects of scientific dedication to mountain or alpine research and the overall culture of publishing scientific research in the various countries.
In terms of numbers of publications per institution, a rating is rather difficult, given the very different sizes and broad geographical distribution of institutions, and meaningful data appeared to result from the search term ''alpine'' only (   In terms of scientific fields, the problem is that papers often belong to more than one subject group. Consequently, the sum of the percentages does not yield 100% if the ranking is done by specific disciplines (journal profiles). Therefore, the analysis based on 10,760 articles in the 324 journals with .10 hits for ''alpine'' is presented in bulk for major sectors only. The two main blocks are geosciences with 45% of all papers across subfields (geology, geochemistry, mineralogy, physical geography, hydrology, meteorology, and atmospheric sciences) and biology with 40% of the papers (plant or animal sciences and ecology, both terrestrial and aquatic). Multidisciplinary works constitute 9% of the papers, and those that cannot be allocated make up 6%. If ranked by the search term ''mountain*'', the hit list is dominated by geosciences in the widest sense.

Conclusions
Overall, this analysis illustrates that an overproportional fraction of mountain research is produced in a handful of rather small countries and institutions. By all criteria, Switzerland emerges as a leader in mountain research. The small population of an otherwise scientifically very active society may contribute to the high per capita ranking; however, even in absolute terms Switzerland contributes nearly as much as the United States to the scientific output in alpine research. A remarkable outcome of this analysis is the high ranking of Austria, New Zealand, Scotland, and Norway, all very small countries each producing an overproportional contribution to both ''mountain'' and ''alpine'' research per capita.
The explanation for these rankings may lie in the importance of mountains in certain countries and their culture and history, including the status of economic development. However, it does not seem that the mountain area per se is a key criterion. In Switzerland, ca. 70% of the country (whose area is only 40,000 km 2 ) consists of mountains. Both in terms of mountain land area (5.7 million km 2 , or two-thirds of the country) and number of people living in mountains (ca. 350 million), China ranks at the top, but it is only very recently that published contributions of China to mountain research have started to increase rapidly. It seems, rather, that the major drivers of mountain and alpine research are geopolitical position, scientific tradition, and economic welfare, which explains the global patterns of scientific leadership in mountain research as reflected in this bibliometric analysis.
The exceptional position of Swiss mountain research is rooted in a long tradition. The first description of the elevational changes of vegetation in mountains was published by Conradus Gessner in 1555 in Zü rich. The foundation of atmospheric sciences (that in mountains in particular) can be attributed to the works by Horace-Benedict de Saussure around 1800 in Geneva. The first textbook on alpine plant ecology was published by Carl Schrö ter in Zü rich in 1906. Switzerland was not directly involved in the major political crises of the last century and thus experienced a continuous development of the scientific endeavor of mountain research over more than a century. Such dedication and commitment to mountain-related research are reflected in today's hosting of international scientific networks in Switzerland such as the World Glacier Monitoring Service (in Zü rich), the Mountain Research Initiative (MRI, currently in Bern), the Global Mountain Biodiversity Assessment Given the critical role mountains play worldwide for the supply of major goods and benefits but also for nature conservation, it is time that more countries, the larger and/or economically strong ones in particular, take over responsibility for mountain research at home and abroad, where support for regional research capacity is needed.