1Periglacial geomorphology is the sub-discipline of geomorphology concerned with the landforms and processes of the cold non-glacial regions of the world (French, 1996). However, while this definition captures the traditional essence of the sub-discipline, it does not reflect, nor address, two important emergent issues within periglacial geomorphology. The more fundamental is the relationship between the older sub-discipline of periglacial geomorphology with its roots in climatic geomorphology, and more broadly geography; and the younger, presently more vigorous discipline of geocryology, here defined simply as permafrost science. The latter discipline is one of the cryospheric sciences and has a sharply different disciplinary heritage to that of periglacial geomorphology. These overlapping, even conflicting, interests can be viewed as an illustration of M. Church’s (2005) recent observation of the drift of geomorphology away from geography. Probably, the situation also reflects the normal ecological development of any scientific discipline that sees it invading new territory opportunistically, dying back when the prevailing paradigm appears stale, and dying out when it has been exhausted. At the disciplinary scale these steps seem traumatic, but for science overall they invariably represent healthy growth.
2The second issue is the origin and development of periglacial landscapes as traditionally defined. Here, the debate is not among disciplines, but rather reflects maturation within periglacial geomorphology itself. It is seen in H.M. French’s (2000) discussion of the nature and variety of contemporary periglacial environments and M-F. André’s (2003) scrutiny of the possible temporal development within periglacial regimes. These concerns are ultimately interwoven with the first issue because, whether the matter is considered in its spatial and/or temporal form, evaluation must embrace a functional definition of what is, and what is not, periglacial geomorphology.
3In an attempt to address these problems, our paper identifies current trends in periglacial geomorphology and discusses the disciplinary considerations that are involved. We focus upon the fact that (i) geocryology is now a recognized member of the cold-climate earth-science disciplines, (ii) the nature of geomorphology and Quaternary science is changing, and (iii) traditional Pleistocene periglacial studies are in decline.
4The early development of geocryology occurred in Russia where, as early as 1924, an Institute of Permafrost was established at Yakutsk, central Siberia, by the Soviet Academy of Sciences. By 1940, the first edition of what was to become a standard text in the Soviet Union, Obshcheye Merzlotovedeniya (General Permafrostology) was published (Sumgin et al., 1940) and by the mid 1960s the first of many undergraduate textbooks had emerged. By comparison, North American geocryology is of relatively recent origin with interest in permafrost only becoming important during and immediately after the Second World War (Muller, 1943) and geocryology emerging initially from within geophysics (Lachenbruch, 1957). In China, geocryology developed even more recently than North America but in a Soviet-style context (Academia Sinica, 1975). Expansion of permafrost studies into alpine regions is also relatively new and has centred largely upon rock glaciers and the creep and stability of frozen rock masses, especially in Europe (Haeberli, 1985).
5For several reasons, the relations between geocryology and geomorphology are complex. First, for many years, permafrost studies were conducted in North America and the Soviet Union not only in relative isolation to each other, but also in isolation from mainstream (geographical) geomorphology. Second, both Russian and Chinese geocryology adopt a holistic, all-encompassing approach whereas North American permafrost studies are usually characterized as being either ‘science’ or ‘engineering’. Thus, there is no North American text that equals the breadth and depth presented by the most recent Russian and Chinese texts, General Geocryology (Yershov, 1990) and Geocryology in China (Zhou Youwu et al., 2000). Third, permafrost studies sit awkwardly between the disciplines of geology and geography. For example, in North America, periglacial geomorphology is taught usually in geography departments while permafrost is within geology, geophysics or Earth sciences departments. In Europe, many geography departments exist in Faculties of Science, others in Faculties of Arts. Fractionation also occurs in Russia, where, at Moscow State University, a Department of Cryolithology exists within the Faculty of Geography while a Department of Geocryology exists within the Faculty of Geology. Finally, Chinese universities exhibit comparable divisions with periglacial geomorphology found within departments of geography while permafrost studies and geocryology are usually undertaken within engineering and science contexts.
6As M. Church (2005) laments, geographical geomorphologists are slowly losing their discipline to geophysicists and to programs located in refocused Earth sciences departments. This trend, ongoing since the 1960s, came first as the result of quantification, then of increasingly rigorous process studies founded on Newtonian principles, and finally, as the inevitable product of the all-embracing theory of plate tectonics which ultimately led geophysicists to be interested in topics previously held to be largely geomorphological. M. Church (2005) goes even further to suggest that the Newtonian approach of geographic geomorphologists is actually a shriven one in comparison to that undertaken by those entering geomorphology from other disciplines. Within periglacial geomorphology this approach produced a sub-discipline focused largely upon quantitative process studies. By the early 1990s process measurement had clearly demonstrated not only the serious shortcomings inherent to traditional versions of periglacial processes (and especially of weathering processes) but also that the azonal processes operating in cold environments, such as running water, wind, waves, and gravity-controlled mass movements, differ little, if at all, from similar processes in other climatic environments. As such, the study of these processes rightly constitutes a sub-set of mainstream process geomorphology because they vary only in their magnitude and frequency rather than in any so-called ‘unique’ manner.
7Thus, it might appear that a distinct sub-branch of geomorphology devoted specifically to the operation of ‘mainstream’ processes in cold climates is now unnecessary. For example, in the introduction to a volume of collected papers recently published, H.M. French (2004) has explicitly characterized modern periglacial geomorphology as a branch of geocryology, leaving the common azonal processes associated with running water and wind action to companion volumes in the series. However, this operational definition is not without criticism because it neglects the important and unusual role played by snow in controlling ground temperatures and in influencing soil moisture conditions and slope runoff regimes. It also fails to consider the enhanced action of wind in high latitudes (Seppälä, 2004) and the role of sea ice in Arctic and Antarctic coastal processes. Furthermore, it is telling indeed that no papers dealing with the operation of these azonal processes in cold climates were included in the paper selections made by the editors of the other volumes. An earlier concern with such issues led C.E. Thorn (1978) to emphasize the role of snow as a unifying concept within periglacial geomorphology and subsequently (Thorn, 1992) to attempt a definition of periglacial geomorphology in purely process terms. Finally, the recent trend that recognizes the importance of ‘non-periglacial’ contributions to cold region landscape development (André, 1999) serves both to further undermine traditional climatic geomorphology precepts while simultaneously sharpening the question ‘what precisely is periglacial geomorphology?’ D. Barsch (1993) took C.E. Thorn to task for producing an overly narrow definition of periglacial geomorphology that would create a sub-discipline akin to glacial, eolian, or fluvial geomorphology. The counter to D. Barsch’s perspective is clear. If contemporary researchers in cold regions find many important contributing factors to be ‘non-periglacial’ in the traditional sense, then it makes infinite sense to identify a suite of periglacial processes, but to refrain from calling cold regions ‘periglacial’ out of recognition of their hybrid genesis.
8The nature of Quaternary science in the last 50 years has also changed due largely to the expansion and proliferation of sophisticated dating techniques. Today, studies involving paleo-environmental and paleo-geographical reconstruction no longer rely solely upon descriptions of the morphological and stratigraphic evidence of cold-climate conditions that was typical in traditional Pleistocene periglacial studies. Instead, a broader range of features and organisms is now utilized. For example, the study of ground ice, in the form of thaw unconformities, truncated ice bodies, and cryostructures can now be used to infer previous freezing and thawing events, and permafrost history (Burn, 1997; Murton and French, 1994; Melnikov and Spesitvsev, 2000; Murton et al., 2004; 2005). Thus, traditional Pleistocene periglacial geomorphology has been largely replaced by the appropriate application of Quaternary science and cryostratigraphic studies.
9The growth of periglacial geomorphology occurred mainly in Europe in the two decades following 1945 (French, 2003). Most dominant were Pleistocene studies dealing with paleo-geographic reconstruction in the mid-latitudes. The ‘periglacial fever’ of the time (André, 2003) fostered a trendy sub-discipline of climatic geomorphology, but one whose underlying tenets were dubious. Two widely-held assumptions or interpretations fuelled this intense disciplinary growth. First was the uncritical acceptance of the importance of mechanical (frost) weathering in cold regions and of rapid cold-climate landscape modification. In Europe, a sequence of widely-regarded texts by J. Tricart and A. Cailleux (1967) and J.Tricart (1970) promoted these ideas. At the same time, an IGU Periglacial Commission under the leadership of J. Dylik was especially active between 1952 and 1972 and an international journal, Biuletyn Peryglacjalny, was started in Lódz, Poland.
10As early as the mid 1970s, this first assumption was being seriously challenged. Initially, air climates were shown to be poor indicators of the relevant ground climates. Ground observations in both high latitudes and at high elevation (Thorn and Hall, 1980) failed to record the numerous freeze-thaw cycles that were thought responsible, a shortcoming also compounded by a lack of moisture in many situations. It was quickly realized that mechanical weathering involved not only frost action but also other mechanisms such as thermal stress and hydration shattering. These processes were not unknown and are not unique to periglacial environments. In recent years, K. Hall (1997) has taken the lead in promoting the role of thermal stress as a mechanism of rock disintegration in cold climates. Today, it is increasingly understood that a variety of chemical, biochemical, physical and mechanical processes operate, and often interact, in cold regions (Pope et al., 1995; Dixon et al., 2002; Hall et al., 2002; Etienne, 2002).
11The second weakness was that insufficient consideration was given to, first, the influence of lithology upon so-called ‘periglacial’ landscapes and, second, the variability, duration and efficacy of cold-climate conditions. This second weakness is still largely neglected in most regional periglacial studies and in standard periglacial texts (Washburn, 1980; French 1996). The possible exception is the Antarctic periglacial literature which, by necessity, focuses upon the bedrock terrain exposed in the relatively small areas that are ice and vegetation-free. In hindsight, it can be seen that the process assumptions underpinning traditional Pleistocene periglacial geomorphology were erroneous. Not only were the necessary observational data lacking but also the conceptual or theoretical framework was flawed. While such judgement may seem harsh, virtually all science ultimately falls victim to such an evaluation.
12We believe that modern periglacial geomorphology should be viewed as a process sub-discipline of geomorphology that is distinct from both geocryology and Quaternary science. It should be similar to the better known process sub-disciplines of, for example, fluvial, hillslope, eolian, glacial, coastal and karst geomorphology. For example, in the case of karst geomorphology, the central process is that of solution, in the case of hillslopes it is gravity. In the case of periglacial geomorphology, the key processes are those associated with seasonal and perennial frost. So, while geocryologists prefer to focus upon the thermal implications of the terrain and the presence of ice within the ground, periglacial geomorphologists prefer to emphasize the associated landforms and their growth and modification through time. Obviously, there is considerable overlap between the two. For example, the recognition of anti-syngenetic wedges on hillslopes (Mackay, 1990, 1995) is a classic illustration of the overlap between landscape evolution (i.e. geomorphology) and permafrost-related processes (i.e. geocryology). Likewise, there is an even more complex overlap between periglacial geomorphology and Quaternary science because this occurs via geocryology. This is best illustrated by the geocryological sub-discipline of cryostratigraphy, where ice and sand wedges, truncated ice bodies, thaw unconformities and the nature of cryostructures and cryotextures allow one to infer past permafrost history. It is now common to see cryostratigraphy applied to problems within the more traditional fields of both geomorphology and Quaternary science (Burn, 1997; Shur and Jorgenson, 1998; Murton et al., 2005).
13We also believe that permafrost, being a purely thermal concept, cannot be the only diagnostic criterion for periglacial geomorphology. This is because geomorphology is concerned primarily with landforms and periglacial landforms that are not primarily controlled by ground temperature alone. Instead, the broad features of cold-climate terrain are largely influenced by lithological variability, the nature and distribution of any ice contained within the bedrock or surficial materials, and the action of azonal processes. A classic example is provided by the structurally-controlled landscape of Ellef Ringnes Island in the Canadian High Arctic that is currently being fashioned by running water, wind and frost (St-Onge, 1965). On the other hand, permafrost, ground ice, and the thawing and refreezing of permafrost, must be central concepts in periglacial geomorphology in the same way that hillslopes and running water are central components to geomorphology at large. It follows therefore, that there is some overlap between periglacial geomorphology and geocryology.
14Periglacial geomorphology must also be viewed as one of the group of sciences that concern the cryosphere. The latter is defined as that part of the Earth’s crust, hydrosphere and atmosphere subject to temperatures below 0 ºC for at least part of the year. Obviously, the cryolithosphere (i.e. perennially and seasonally cryotic ground) is central, and the cryohydrosphere (i.e. snow cover, glaciers, and river, lake and sea ice) slightly less central, to periglacial geomorphology. Periglacial geomorphology has a special interest in the thawing and freezing of ground. A schematic summary of the interactions between geomorphology, periglacial geomorphology, geocryology and the cryospheric sciences appears in Figure 1.
Fig. 1 – Schematic diagram illustrating the disciplinary interactions and overlap of periglacial geomorphology.
Fig. 1 – Diagramme illustrant les interactions disciplinaires et les chevauchements au sein de la géomorphologie périglaciaire.
A: relations between physical geography, geomorphology and periglacial geomorphology. B: relations between periglacial geomorphology and geocryology and their interactions with Quaternary science and other natural sciences. C: periglacial geomorphology and its overlap with certain cryospheric sciences. Note: All disciplinary boundaries are porous and those marked by dashed lines are particularly so. For example, in C, the broken line between perennial and seasonal frost indicates that periglacial geomorphology extends into geocryology and that geocryology extends, but to a more limited extent, onto periglacial geomorphology. Similarly, the two unbroken lines between seasonal frost and snow, and between perennial frost and glaciers, must also be regarded as porous boundaries.
A : la relation entre géographie physique, géomorphologie et géomorphologie périglaciaire. B : la relation entre la géomorphologie périglaciaire et la géocryologie et leurs interactions avec les sciences du Quaternaire et d’autres sciences naturelles. C : la géomorphologie périglaciaire et ses chevauchements avec certaines sciences de la cryosphère. Remarques : toutes les frontières disciplinaires sont poreuses, en particulier celles soulignées par un trait en pointillés. Par exemple, en C, la ligne brisée entre le gel saisonnier et le gel permanent indique que la géomorphologie périglaciaire s’étend dans le domaine de la géocryologie et que cette dernière s’étend également, mais à un moindre degré, dans la géomorphologie périglaciaire. Pareillement, les deux lignes continues entre la neige et le gel saisonnier et entre le gel permanent et les glaciers doivent être considérées comme des frontières poreuses.
15Here, we outline what we perceive to be the salient components of modern periglacial geomorphology. We do so in the following sequence: i) the nature of permafrost-related processes, ground ice, and associated landforms; ii) the azonal processes that operate in cold non-glacial environments; iii) the ice-marginal (proglacial) environment and associated paraglacial transitions; iv) the alpine (montane) environment; v) Pleistocene cold-climate paleo-environmental reconstructions; vi) environmental and geotechnical studies associated with frozen ground, ground freezing and global climate change.
16Processes that are clearly unique to periglacial environments relate to ground freezing. These include the growth of segregated ice and associated frost heaving, the formation of permafrost, the development of cryostructures and cryotextures in perennially-frozen soil and/or rock, the occurrence of thermal-contraction cracking, and the growth of frost mounds of various sorts. The stratigraphic study of frozen earth material, especially the amount, distribution, and origin of the ice that is contained within it, constitutes the geocryological sub-discipline of cryolithology. Although not strictly geomorphological in nature, cryolithology is relevant to understanding permafrost history, to the interpretation of permafrost-related landforms, to describing periglacial sediments, and when one undertakes inferences as to past climates or Pleistocene paleo-geographic reconstruction.
17A number of frost action processes operate in the near-surface layer subject to seasonal thaw (the active layer), the near-surface permafrost located above the depth of zero annual amplitude, and the zone of seasonal freezing and thawing in non-permafrost regions. These processes include moisture migration within frozen ground and numerous processes associated with repeated freezing and thawing (e.g. soil churning or cryoturbation, frost creep, solifluction and gelifluction, the upfreezing of stones and particle-size sorting). Many of these processes give rise to distinct small-scale forms of patterned ground that complement the large-scale polygons that result from thermal-contraction cracking.
18Slopes that are frozen, or are thawing, experience relatively unusual conditions associated with pore-water expulsion and thaw consolidation. These can promote rapid mass failures that are relatively distinct from other failures that might occur on slopes that evolve under non-frozen conditions in either temperate or warm climates. It is in the study of all these processes, either in the field, the laboratory, or by modeling and simulation, that periglacial geomorphology has become a sub-branch of the broader discipline of geocryology. An understanding of all these processes constitutes the essential underpinning to modern periglacial geomorphology.
19Some processes, not necessarily restricted to environments traditionally labelled periglacial, are important in cold non-glacial regions on account of either their high magnitude or frequency, or their widespread occurrence. These often centre upon the seasonal freezing of soil and bedrock, including the disintegration of exposed rock by either mechanical (frost) wedging or the poorly understood complex of physical, biochemical or physico-chemical processes that are sometimes referred to as ‘cryotic’. Slightly better understood are the azonal processes associated with running water, wind, snow, and waves. These may assume distinctive characteristics under traditionally-defined periglacial conditions but do not require either an excessively cold climate or a peripheral ice-marginal location for their effective operation. Nevertheless, their study complements many other aspects of geomorphology, permits comparison and, as a result, better comprehension of the process involved.
20Some azonal processes are of special interest to periglacial geomorphologists. For example, the role of snow as an important source of moisture must be highlighted because, by definition, most cold non-glacial regions are also arid. Snow is also important as an abrasive agent at low temperatures and acts as a local source of soil moisture and hence ground heaving and frost action. However, the normal effects of aridity in cold environments are lessened by low evaporation and/or evapotranspiration rates. For example, the boreal forest is not usually regarded as typically ‘periglacial’ in nature, yet it experiences cold-climate conditions and is commonly underlain by permafrost. Wind action is also of interest to periglacial geomorphology because the limited vegetation north of the tree line in higher latitudes, and above the timberline in montane environments, allows wind to have important erosional and transportational effects (Seppälä, 2004). Finally, sea ice and river ice (‘ice-infested waters’), by restricting the time duration of wave action and/or open channel flow, and through ice-pushing, ice-jams and other impacts, can produce relatively distinct coastal, river channel and lake conditions. The study of all these processes constitute a part of geomorphology, irrespective of the direction in which geomorphology is currently moving. Yet an understanding of such processes is also essential to an understanding of cold region landscapes.
21While the majority of periglacial environments do not demand an ice-marginal location for their cold-climate characteristics, those areas that are immediately adjacent to major ice sheets and/or glaciers experience cold-climate conditions that fluctuate in time and space. In the early beginnings of periglacial geomorphology, the ice-marginal environments of Svalbard and Greenland were regarded as the typical high-latitude analogs for the periglacial conditions that probably existed in the mid-latitudes of Europe during the cold periods of the Pleistocene. While this is clearly not the case, and Lozinzki’s so-called ‘periglacial realm’ finds no modern counterpart, ice-marginal environments, by virtue of their proximity to glaciers and the cold-climate conditions that they experience, provide an instructive example of specific and specialized periglacial terrain undergoing paraglacial transition.
22Like the proglacial environment, the alpine environments that occur above timberline were initially regarded as typical ‘periglacial’ environments. However, they also represent site-specific and specialized terrains in which steep slopes and gravity-controlled processes play a critical role together with cold-climate processes aided by wind and snow. As such, they are also instructive for periglacial geomorphology. Many recent studies focus upon the occurrence of mountain permafrost, the controls over its distribution, the creep of ice-rich debris (rock glaciers), and the instability of frozen rock masses as bedrock temperatures warm. These examples of alpine geomorphology are equally an integral component of geocryology.
23The growth of cryostratigraphy, together with the increasing sophistication of Quaternary science, means that traditional Pleistocene periglacial paleo-geographic studies have become obsolete. Morphological and stratigraphic evidence must now be interpreted within the context of a more realistic appreciation of permafrost and the climatic and other controls over related cold-climate processes, augmented by isotopic and other dating techniques. In other words, advances in cryostratigraphy and Quaternary science have now encroached upon, and largely superseded, the traditional Pleistocene and climatically-based ‘periglacial’ studies.
24All disciplines thrive best when they have societal relevance, and periglacial geomorphology is no exception. Here, the opportunities are many and potential growth is large. For example, the development and utilization of the natural resources of the northern regions of North America and Eurasia will require sound engineering and infrastructure design. Man-induced thermokarst and other disturbances need to be minimized through management and regulatory practice. The fact that global climate change may first become apparent at high latitudes, and that permafrost can be regarded as a temperature archive, both in terms of past and future temperatures, has promoted recent monitoring studies such as Circumpolar Active Layer Monitoring (CALM) and Permafrost and Climate in Europe (PACE). In alpine regions, the increased utilization of upper slopes for recreation activities, and the potential for slope instability consequent upon permafrost thaw, has promoted the study of mountain permafrost. Likewise, the thinning of sea ice and the potential expansion of arctic shipping lanes has prompted international attention towards cold-climate coasts, as illustrated by the Arctic Coastal Dynamics (ACD) initiative of the International Arctic Science Committee (IASC). We can think of many other examples where periglacial geomorphology should play a useful societal role.
25A major aim of geomorphology is to create models of landscape evolution. Such models embody assumptions as to the processes involved, their speed of operation and their associated rates of erosion, transport and deposition, and the manner in which surface morphology changes through time. While various cyclic and non-cyclic (equilibrium) models characterize the history of geomorphology, only the descriptive and non-quantitative cyclic models of Boch and Krashnov (1943), Hopkins (1949), and Peltier (1950) explicitly apply to cold-climate terrain. The reality is that the nature of landscape evolution under cold non-glacial conditions remains largely neglected in cold region geomorphology. The task is daunting but notable exceptions already exist such as the application of thaw-consolidation theory to creep and deformation of permafrost (McRoberts and Morgenstern, 1974; Morgenstern, 1985), the evolution and stability of thawing slopes (Harris and Lewkowicz, 2000; Davies et al., 2001), and recent laboratory attempts that simulate solifluction (Harris et al., 2001).
26One must reluctantly conclude that much geographically-based periglacial geomorphology lacks a rigorous theoretical base. It remains characterized by flawed thinking that reflects its qualitative and climatic heritage. Unless a research agenda that rectifies this weakness is put in place, so-called ‘periglacial geomorphology’, as presently configured, runs the risk of falling by the wayside. Currently, more progress is being made in geocryology, Quaternary science, and geomorphology, even as the latter is losing its geographical base to ‘reductionist’ science, than in periglacial geomorphology because these disciplines are seen to be more ‘scientific’ and to have more pragmatic and pressing research agendas.
27We argue, therefore, that a strong periglacial geomorphology must be firmly process-based. It must clearly reject its climatic geomorphology underpinnings. For example, so-called ‘periglacial regions’ exist only as cold-climate zones in which seasonal and perennial frost, snow and azonal processes operate. There is no ‘core’, or typical, or definitive periglacial region, and Lozinski’s ‘periglacial realm’ is a largely academic concept that does not exist today. Instead, it is more realistic to think in terms of areas of the world that are of interest to geomorphologists wishing to understand cold-climate, but not glacial, landscapes. While most of these areas contain permafrost as a central element, there are no clear-cut boundaries. In short, periglacial geomorphology needs to sharpen its scientific rigor before it is either consumed or by-passed, as the case may be, by either geocryology, Quaternary science or the other sub-disciplines of geomorphology.
28We conclude that the core of modern periglacial geomorphology should concern the study of both perennial and seasonal ground ice and related landscape development. These subjects lie at the heart of the discipline and must supply the solid scientific base upon which periglacial geomorphology relies for its credibility. Such a perspective makes permafrost a central, but not defining, element of periglacial geomorphology. Therefore, much that constitutes modern periglacial geomorphology must also be regarded as a component of geocryology. However, other components of periglacial geomorphology include the impact of seasonal freezing and the roles of seasonal snow, and of fluvial, lacustrine, and sea-ice covers. Furthermore, the geomorphology of cold non-glacial regions must embrace the study not only of geocryology and periglacial geomorphology as defined above, but also the azonal processes that exhibit distinct behaviour and/or magnitude and frequency distributions. Perhaps more so than permafrost-related and other periglacial processes, the contribution of azonal processes to cold-climate landscapes is still poorly understood.
29Inevitably, geocryologists must have a vested interest in the landforms and landscapes that provide them with a context for permafrost; equally, periglacial geomorphologists must be committed to understanding the role of permafrost in shaping a large portion of their landscape of interest. These mutual interactions are similar to those between hydrologists, hydraulic engineers, and fluvial geomorphologists. Analogous areas of overlap exist in other branches of geomorphology. Quaternary scientists also have a legitimate interest in periglacial geomorphology because the fluctuating environments of the Quaternary included significant periods that experienced cold-climate conditions. It appears therefore that the challenge for periglacial geomorphology is to maintain a bridging position between the changing nature of geomorphology, the emerging discipline of geocryology, and the increasing sophistication of Quaternary science. Societal relevance, in addition to scientific excellence, must also play an increasing role.