earth shelter architecture - elseviermedia.journals.elsevier.com/content/files/design... ·...

T

EARTH SHELTER & ARCHITECTURE _

Figure 1. The Nathan Marsh Pusey Library at Harvard University in Cambridge, Massachusetts, exemplifies an unobtrusive design approach in an historical setting. (Photo courtesy of Hugh Stubbins and Associates, architects)

Design Considerations for Underground Buildings

John Carmody Ray Sterling

University of Minnesota

his article discusses some design concerns unique to under ground buildings or deserving

of special attention when applied to a below-grade design. As with any build ing, a wide range of functional, psy-

John Carmody is an architect and Associate Program Director of the Underground Space Center at the University of Minnesota, Twin Cities. Dr. Raymond Sterling is Director of the Center.

352

chological, and physiological require ments must be resolved in an aesthetically successful manner in an underground structure.

The article is divided into two major parts. In the first part some key prob-

This article is ·a revised version of material first published in Underground Building De sign (New York: Van Nostrand Reinhold, 1983, All Rights Reserved) and is reprinted here with permission.

Underground Space, Vol. 8, pp. 352-362, 1 984.

Printed in the U.S.A. A ll r i ghts reserved.

!ems inherent in underground build ing design are identified. These in clude the functional concerns of building recognition and orientation as well as psychological and physiological problems associated mainly with win dowless environmen ts. Also discussed in this part is the impact of building use on these problems. The second part of the article presents typical design responses to these problems and fo cuses on the exterior image of an un-

0362-0565/84 $3.00 + .00

Copyright © Pergamon Press Ltd.

derground building, entrance design , providing light and view, and interior design.

Key Problems in Underground Building Design

Building Recognition and Image As in conventional buildings, the ex

terior form and character of under ground structures are critical to their image and their acceptance by the pub lic. But, unlike with conventional struc tures, some potentially negative psy chological reactions to subsurface space must be considered in exterior design, and practical problems of recognition and spatial orientation must be ad dressed.

Underground structures offer op portunities to create exterior forms that are difficult, if not impossible, to attain in conventional buildings. For exam ple, although above-grade structures can be designed to be sympathetic with and reflective of natural forms, they have a physical presence on the land scape that creates a clear distinction be tween the man-made and natural en vironments. Placing a building partially or completely below the surface can obscure the mass and the edges of the building, enabling almost complete in tegration of built and natural forms. Not only can this create a more natural image, but it also can provide oppor tunities to place relatively large struc tures in sensitiv.e settings without de stroying the scale, the open space, or the character of the area.

Underground buildings can be de signed in a wide range· of forms, in which the degree of visibility and the character of the building vary greatly. Although creating an unobtrusive, even imperceptible, structure has its advan tages (for buildings such as interpre tive centers, for example), this ap proach may be inappropriate or undesirable for buildings such as those which depend on public recognition for their business.

One important area of concern re lated to the exterior design of an un derground building is the need for a clear understanding of the building size, location, and entry. Most conventional buildings have definite edges, a per ceivable mass, and a clea r entry, so that they can be easily recognized as an ob ject and described as a specific place. An underground building may not provide these visual clues, especially if it is completely below grade with little exposure to the surface. This means that the exterior form and character of a subsurface structure must take into account the special problems of ori entation and recognition of building

Volume 8, Number 5-6, 1984

and entry. For example, in a built- up area, a

structure located under a plaza that is completely surrounded by buildings will likely be easy to describe and locate, since the adjacent buildings define the location clearly. On a more open site with fewer clear boundaries, however, the space above and around the un derground structure must be carefully designed to reflect a sense of place and indicate clearly the entrance to the building.

Exterior space around conventional buildings, when properly designed, can provide orientation, define circulation to the entrance of the building, and serve to discourage vandalism and crime. This is also the case with un derground buildings. For a completely underground building, these aims are usually best accomplished by the use of grade changes, paving patterns, trees, shrubs, and variations in ground cover, along with retaining walls and other building elements.

Just as it is on the exterior, orien tation may be impaired inside a mostly windowless building. Visual cues nor mally provided by exterior views and an awareness of the overall size and shape of the building may not be pres ent. This disorientation may not only present problems in circulating con veniently within the building; it may also contribute to uneasiness and rein force other negative associations with being underground.

An important facet of the practical problem of building recognition on the exterior is the more subjective, aes thetic concern of conveying an image with the building's appearance. Al though entry, lighting, and interior de sign contribute to the image that a building projects, the exterior form is one of the most important elements. Because the exterior profile of an un derground building is certain to be smaller and generally less obtrusive than a comparably -sized above-ground structure, a less visible and less mon umental image will be projected. For some functions- auditoriums, librar ies, museums, parking garages, and factories- this may be an advantage, since a minimal presence on the site is often desirable. Other functions, how ever, may require visibility to the public or may need to serve as a symbol.

Although there are instances where an underground design is simply not an effective means of achieving a highly visible or monumental form , a positive image can still be created in most cases without relying on a large building mass. Partial exposure of the building, berms, and extensive landscaping can be de-

signed to create forms on the site that will draw attention because they are at tractive and provide a contrast with more conventional buildings. Negative Psychological Effects

Underground build i ngs display a wide spectrum of physical character istics and functional uses. In addition, the physiological and psychological re sponses of individuals to the environ ment can be radically different. The combination of these factors makes generalizations about underground buildings in relation to psychological effects oflimited applicability. It is pos sible, however, to list criteria and con siderations that are potentially appli cable to an underground building and to discuss how these factors have been assessed by various investigators of the psychological and physiological re sponses to artificial environments.

One major concern related to un derground buildings is sensory depri vation. Individuals are stimulated by, and many bodily functions are affected by, interaction with the surrounding environment. In addition to physiolog ical effects (discussed below), psycho logical problems can arise from the re duction of external stimuli to an individual (Paulus 1976). This reduc tion of stimuli would occur most sig nificantly in small, enclosed spaces with little interpersonal contact and no con tact with the external environment. It should be noted, however, that a re duction of external stimuli is not always negative; it can provide a more pro ductive work environment and reduce stress in otherwise chaotic surround ings.

Another area of psychological re search which pertains to underground building design is a person's sense of interaction with his environment. It has been theorized that the three pri mary emotional responses to the en vironment are pleasure, arousal, a nd dominance (Mehrabian and Russell 1974). Pleasure is self-explanator y . Arousal refers to the presence or ab sence of stimuli discussed above, and dominance relates to feelings of con trol or influence over the environmen t.

A lowered arousal level is to be a n ticipated in enclosed spaces, but the ef fect on feelings of dominance is not as clear. It has been suggested that, be cause of the practical and environme n tal advantages of an undergroun d building, users may feel they are "bemg

exposed to successful attempts to cop with the demands of the environme n t (Paulus 1976). It has been found by others, however, that practical featu res of a building are not important c nte na

UNDERGROUND SPACE 353

in evaluating user response to working in the space (Hollon and Kendall 1980).

Nevertheless, for an individual choosing to build a certain type of un derground structure , say a n earth sheltered house, for energy conserva tion and environmental reasons, these criteria are often very importa nt and would be expected to create positive feelings about the environment. A poor response to the environment of an un derground building has been lin ked to anxiety, tension, depression, and other mental health problems, although other studies have indicated no measurable differences in achievement , health problems, or absenteeism (Collins 1975,

Hollister 1968, Lutz 1976). Only a few studies have directly ad

d ressed the case of below-grade spaces for working or living environments, and often conflicting data have been gath ered. Assessments of the distinct phys ical characteristics of an undergrou nd space are extremely difficult to sepa rate from other general physical char acteristics of the space and the inter personal environment. In spite of the lack of definitive data, however, the major issues are fairly clear and re peatedly emerge in all studies of un

Figure 2. Earth berms, arched forms, and native landscape materials create a distinctive natural image for the Wildwood School in Aspen, Colorado. (Photo courtesy of David Gibson, architect)

derground and windowless spaces. There follows below a list of conditions that are associated with producing neg ative psychological effects in under ground buildings. Most of the infor mation also pertains to windowless buildings or interior zones inside con ventional buildings.

Lack of natural light. The lack of nat ural light is one of the most often men tioned negative characteristics of un derground space. Access to natural light is important to users of a building even if the proportion of daylight to artifi cial lighting for work tasks is relatively low. The feeling produced by daylight, its variability, and the sense of contact with the outside world are important reasons for its desirability. It has been theorized that horizontal accessibility to daylight provides a closer reference to conventional buildings and is there fore preferable to day lighting from overhead sources (Bennett 1978). In contrast, however, a research study conducted with groups of students viewing slides of building interiors con cluded that spaces with skylights over head were most often selected as the most desirably lit (Hollon and Kendall 1980).

Another important positive psycho logical association of natural lighting is that sunlight connotes warmth. Direct sunlight is not always welcomed by building users, however. Sunlight can cause excessively uncomfortable con ditions in a warm building interior, as

354 UNDERGROUND SPACE

well as personal discomfort to a person who must work in the full sunlight. Ex cessi ve brightness, glare , and large brightness-to-darkness ratios in build ings not completely illuminated by sun light are other problems (Boyer 1980).

Lack of exterior view. The lack of ex terior view from an underground space is another reason for dissatisfaction with this type of building. In addition to providing natural light and sunlight, windows provide a direct view for ob serving weather conditions, creating a sense of contact with the environment, and giving visual relief from immedi

ate surroundings (Collins 1975). Different studies have indicated dif

ferent relative importances for natural light and view (Longmore and Ne'e man 1974). People in work environ ments are more likely to favor a view over direct sunlight, especially if no so lar shading is provided and if they are not free to relocate themselves out of the sun when desired. Occupants of high-density residential developments are more likely to cite the availability of sunlight in the home as more im portant than a good view (Bitter and van Ierland 1967).

Underground location. The location of a building below grade does not pre clude providing the above amenities of natural light and view to the interior of the building. There are, however, psychological barriers to the physical location of a space below grade, even if it has identical physical amenities to

an interior space m a conventional building.

Some individuals may experience claustrophobia or fears related to safety that result in negative reactions to un derground spaces. It is difficult to ex plain the negative association some people have with the concept of being underground - even when interior conditions are identical to those in an aboveground space. In addition to fears related to safety concerns, there may be a generally negative association with death and burial related to under ground space. It is generally assumed , however, that these attitudes are more closely related to a fear of structural collapse, fear of being trapped in a fire in a windowless building, or fear of flooding in a fully below-grade space. Fears for personal safety need not be related to the actual risk experienced, but merely to the perceived risk.

These negative reactions tend to heighten awareness of and exaggerate objects to other physical characteristics of the space that might go unnoticed in a conventional building. For exam ple, small interior spaces, low ceilings, or entry down a narrow, dark stairway may increase these negative associa tions with being underground.

Undesirable internal conditions. Users of underground or windowless build ings frequently complain of poor tem perature and humidity control, and a lack of ventilation and stuffiness (Col lins 197 5) . Generally, none of the


m1 0

problems should be any different for a below-grade or windowless building than they are for a sealed, climate-con trolled conventional building. In fact, temperature and humidity should be easier to control than in an above ground building. Thus, in addition to the actual ventilation air change rate provided, perception of ventilation by occupants is important (Wunderlich 1978). If awareness of the superior in ternal environmental control of an un derground building is clearly apparent to the occupants, some offsetting pos itive attitudes may develop.

Negative Physiological Effects

The physiological effects discussed in this section are those caused directly by the environment of an under ground or windowless structure, and not indirect ailments or reactions caused by psychological stress. The following is a list of conditions that are associated

with producing negative physiological effects in underground buildings.

Lack of natural light. Lighting is prob ably the most important physiological criterion to be considered in designing underground or windowless struc tures. The human body has a direct response to certain spectra of light, in cluding those outside the visible spec trum. For instance, ultraviolet light is known to be important in vitamin D absorption, which is necessary to pre vent disease, aid suntanning, and fight bacteria. In animals, the lighting level and its spectral composition have been shown to be important in reproduc tion, behavior, and physical disposi tion. Lighting also triggers a neuroen docrine function and affects the metabolic state (Wurtman 1968, Ott 1973).

Although much research has been done on the presence or absence of light, little has been done on lighting

levels and spectral compositiOn. Alarming effects of certain light sources have been demonstrated, however, in laboratory animals (Ott 1965). It should be noted that effects are quite differ ent for different animal species.

Ultraviolet light is also not transmit ted by normal window glass; hence, lack of ultraviolet light is common even to buildings with windows. When it is nec essary to rely on artificial lighting en tirely, it is most desirable to replicate the spectral composition of daylight as closely as possible.

Lack of fresh air and indoor air pollu tion. The underground location of a building with only limited opportuni ties for window openings often pre cludes the use of natural ventilation. Adequate ventilation is important to prevent the buildup of indoor air pol lu tant and to remove excess heat from an occupied underground space.

Low air change rates make ventila-

1

1

1

1

1

1

I

1

1

il 11

Figure 3. Because it is set into a sloping site, the National Art Education Center in Reston, Virginia, has the image of a conventional building when viewed from the south, even though it is below grade. (Photo courtesy of the Benham Group)

1

Volume 8, Number 5-6, 1984 UNDERGROUND SPACE 355 1

tion especially important in under ground structures. A pollutant of par ticular concern is radon. Radon is a radioactive gas released in minute quantities by soil and rock materials, including such materials used in build ing construction as concrete and build ing stone. Radon is also absorbed by ground water and then released at a free ground water surface. Normal ventilation rates (in excess of 0.5 air changes per hour) are believed to keep radon levels to below permissible stan dards. This is an active area of re search, and more specific data and guidelines should be produced over the next few years. In addition, it is desir able to prevent the passage of ground water or water vapor from the sur rounding ground to within the build ing envelope, since this water can be a significant source of radon (May 1981).

High humidity. Unless controlled, summertime humidity levels will be higher in underground structures than in above-ground structures, as humid outside air is cooled by the earth-cov ered walls. Humid or damp conditions have not been linked directly with physiological problems, although they may exacerbate certain ailments such as rheumatism. Damp conditions may also encourage the growth of mold and thus increase the potential for allergic reactions.

Excessive noise or lack of noise. Noisy building types such as factories can be placed underground to isolate them from the surface environment. Al though this offers benefits to those above grade, it may create excessive noise levels within the space which have well-known effects on hearing impair ment. On the other hand, under ground spaces can be totally isolated from external sources of noise, creat ing a silent environment. In some cases, however, such total silence may be un nerving and may diminish acoustical privacy within the space.

Impact of Building Use on Problems Before concluding which effects are

most critical and suggesting design ap proaches to ameliorate them, it is use ful to review the various factors related to building use that influence either the psychological or physiological effects. These factors, listed below, can serve to diminish the potentially negative ef fects and help to alleviate building user concerns.

Activity within a building. Internal ac tivity within a building that can offset the lack of external stimuli will nor mally be beneficial in a work environ ment, provided it is not too intrusive in terms of noise or distraction. Al though internal stimuli may help to a!-


leviate the negative psychological ef fects produced by no natural light and exterior view, it has no impact on the physiological effects of a windowless environment.

Occupancy patterns. An individual's reaction to an underground or win dowless environment may be substan tially affected by the length of time he or she expects to spend in that envi ronment. Underground facilities used primarily for short-term activities, such as indoor sports facilities, restaurants, libraries, and shopping centers, will thus normally raise fewer objections than an underground office. Not only are neg ative psychological reactions less of a concern when occupants are in under ground spaces for shorter time pe riods, but negative physiological ef fects are less critical as well.

Need for underground location. Em ployees of underground or windowless facilities appear to be more accepting of their environment if they perceive a rational basis for the location or de sign of the facility. In other words, since windows are detrimental to the oper ation of many sports facilities, mu seums, restaurants, and shops, em ployees and visitors do not focus on the lack of windows as a drawback. Simi larly, windowless laboratory and man ufacturing environments provoke less criticism than windowless office build ings. Although this perception of win dowless environments as appropriate for some functions can reduce some psychological effects, it has no impact on negative physiological effects.

job satisfaction./ Employees who are more involved in their work and derive considerable satisfaction from it may be more tolerant of windowless space than employees who work at repetitive tasks. The extent to which this can ac tually offset the various negative psy chological effects is likely to vary con siderably from individual to individual.

Design Responses to Problems

The various problems of orienta tion, image, psychological and phys iological concerns discussed in the pre vious section focus on a few key elements of building design. These are the def inition of an exterior image, the en trance design, providing natural light and view, and interior design. Ap proaches that have successfully re solved these areas of design are pre sented in this section. Defining an Exterior Image

Beginning with the least visible de sign approach, an underground build ing can have virtually no exterior form at all. Such a windowless chamber would be suitable only for certain functions,

such as auditoriums, classrooms, lab oratories, or various s rvice and stor age spaces. The nonexistent form that results from the lack of visible building mass or edges can be reinforced by not using built forms for entrances, service access, or mechanical systems. Elimi nation of these forms is usually possible only when the underground building is connected to the below-grade levels of an existing above-grade structure, so that access can occur through the above-grade building.

If connecting an underground building to an existing complex of buildings is inappropriate, it is still pos sible to create a building with limited or minimal exposure. A structure can be placed completely below grade, with only entrances, skylights, and court yards exposed to the surface. Although a building beneath the surface is evi dent, the character of a "non-building" can still be achieved. An example of such a design is the Pusey Library at Harvard University (Fig. 1 at head of article). Although the library is simple and unobtrusive in its historical setting, building edges are clearly defined by a narrow recessed courtyard along most of the building perimeter. A larger sunken court at a major circulation crossing provides a clear entrance.

A different approach to the exterior form of an underground structure is to place the building above grade with earth bermed around it. Rather than being as unobtrusive as possible, the berms represent an additional object on the landscape. This can provide a distinct image and can define building edges and exterior space. Entry into a bermed building can be more easily achieved than entry into a completely subgrade structure. Plant materials, re taining walls, building materials, and other landscaping elements may sig nificantly influence the image and character of the building. The use of native plants, wood, stone, and berms that appear to be elements of the sur rounding landscape can produce a nat ural image, as demonstrated by the Wildwood School in Aspen, Colorado (Fig. 2).

Many of the inherent problems of entry, service access, and image are greatly reduced or eliminated if above grade space is combined with below grade space in one building. Even if the amount of space above grade con stitutes a small percentage of the total building, a definite built form appears above grade. The use of an above-grade portion of the structure to resolve ac cess and image problems is certainly appropriate in many settings, although some of the benefits of underground


I

buildings-preservation of the surface and integration with the natural envi ronment, for example-may be com promised. An example of building small above-grade forms combined with larger subsurface spaces is Williamson Hall at the University of Minnesota (see Underground Space 1:4, pp. 325 ff.).

A site with sufficient slope provides design opportunities not available on a flat site. An underground building set into a hillside is clearly exposed on one side while completely underground on the other. This results in a recognizable building form on the exposed site, where entry and service access can oc cur; on the uphill side, the site is not interrupted by any substantial grade changes or built forms. The National Art Education Center in Reston, Virgin ia, exemplifies this approach (Fig. 3).

Entrance Design

In the section above on the exterior form and character of underground buildings, the entrance was discussed as a key element of the exterior image. In terms of the entire site, recognition of and orientation toward the entry is an important concern. The focus of this discussion is the relation between the manner in which underground builq ings are entered and psychological per ceptions.

The degree to which a pedestrian entry is important as a design element is, of course, related to the function of the building. In an underground ware house, the entrance for a few workers is not as important. as the functional needs of delivering-and storing goods. The entrance design is of greater con cern in an underground office or fac tory where larger numbers of people are employed. And, compared to fa cilities where people enter and leave once a day, a public building such as a library, museum, or auditorium with many comings and goings requires even greater attention to entry design.

4a. On a sloping site, an underground structure can be entered without descending outside

or inside the building.

I I

4b. A bermed structure on a flat site can be entered at grade in a manner similar to an above-grade building, although the building appearance may be unconventional.

I

I I 1

4c. A small above-grade form can provide the visual image and serve as the entrance for 1

an underground building. 1

1

1 4d. A completely below-grade structure on a flat site can be entered by first descending 1 into a courtyard outside the building. 1

Figure 4. Entering an underground building can be done in several ways depending on 1 the site and exterior image desired by the designer.

l I

Viewed from the exterior, the en trance may be the dominant image of a building that is mostly below grade. It serves as the transition from the ex terior to the interior. As such, it can reinforce negative associations with underground buildings. The entry area in any building should be a key ele ment in orienting and directing people to the spaces inside, and particularly so in an underground building. In addi tion, the entry often serves as a major area for admitting natural light and ex terior view to a subsurface structure, since it may be one of the few points of exposure to the surface.

In order to minimize the negative feelings associated with entering an underground building, several basic


techniques are used. The most com mon technique attempts to create an entrance that is similar to the entrance of a conventional building. It is most desirable to design this entrance at ground level to avoid the requirement for descending a great number of stairs either immediately outside or inside the entrance. (Descending seems to have negative associations, while ascending is more positive. Historically, for ex ample, large public buildings were de signed with a large exterior staircase leading to a main entrance on the sec ond level of the building.)

In an underground building it is usually not possible to ascend to the main entrance. Under a number of cir cumstances, however , subsurface

structures can be entered on grade, similarly to a conventional building-1 for example, on a sloping site (Fig. 4a). 1 A similar type of entry without descent is possible in a bermed structure on a 1 flat site (Fig. 4b). When a portion of a I building is exposed above grade, a con- 1

ventional entrance can occur in the 1 above-grade mass (Fig. 4c). In this last case the interior spaces must be care- l fully designed to offset negative feel- 1 ings associated with descent inside the building.

In many instances underground structures are additions to existing above-grade buildings. Thus, some of the entry design problems are over come because the mai n entrance oc curs through the conventional build -


ing. It is important, however, to consider the design of the connection between an above- and below-grade building. In the underground addition to the Uris Library at Cornell University, the stair way connecting the two buildings is en closed in glass, transforming a poten tially negative descent into an unusual spatial experience with a panoramic view of the valley below (Fig. 5).

In some cases-a fully below-grade building on a flat site, for example none of the opportunities discussed above for a conventional entry are available. To maintain the low profile of the building while creating an ac ceptable entrance, a sunken courtyard can be used so that one descends out side the building and enters horizon tally at the upper level of the building (Fig. 4d). This approach maintains many of the features of a conventional entrance; negative associations can be partially offset, since the descent oc curs more gradually in a spacious out door area. An example of this ap proach is the spiral-shaped courtyard leading to the main entrance of the Civil and Mineral Engineering Building at the University of Minnesota (see Un

derground Space 8:2, Feature Section). On particularly restrictive sites, cre

ating exterior sunken courtyards as a means of entry may not be possible. The only option may be to enter through a small above-grade mass and descend immediately to the spaces be low. This situation in particular may reinforce negative associations with en tering underground buildings and re quires careful attention to natural light and interior spaciousness in the stair way area.

Entering relatively deep mined space can present design problems · that are difficult to resolve using any of the techniques discussed above. Although horizontal entry to deep spaces can oc cur through tunnels in bluffs, there is little opportunity to introduce natural light or offset the spatial confinement of the tunnels. Entering deep spaces through elevator or escalator shafts provides little opportunity to alleviate negative associations with descent.

Providing Natural Light and View

The lack of natural light and view is, both psychologically and physiologi cally, the single greatest concern re lated to underground space. With the exception of warehouses and buildings with other functions where human habitability and acceptance is a second ary concern, some natural light is de sirable in virtually every type of build

specific function. In a building com posed mainly of small, continuously oc cupied spaces, such as private offices or hospital patient rooms, the entire form of the structure may be shaped by the need for natural light and view. In buildings with functions that are suitable in windowless spaces-class rooms, auditoriums, and exhibition spaces, for example-it may be nec essary or desirable to provide natural light and exterior view only to corri dors and lobby areas, thus giving the designer greater flexibility.

Although designers employ many variations and combinations of tech niques, there are only a few basic ap proaches to providing natural light and exterior view to below-grade spaces. Aside from the functional needs of the spaces, the techniques selected for in troducing natural light and view to a particular building are influenced by the site topography as well as the size and depth of the structure.

On a sloping site, conventional ver tical glazing can be used for the spaces on one side of the building. This ap proach by itself is capable of providing light and view only to the spaces on the building perimeter. If earth berms are placed around a structure on a flat site, conventional vertical glazing can be provided on the building perimeter by creating openings in the berms. For in terior spaces on a sloping or flat site, natural light and view can be provided

by creating courtyards. If a courtyard must be deeper to serve several floors, it must also be larger in area in order to permit sunlight to reach the floor of the courtyard and create the percep tion oflooking outdoors. The view into an exterior courtyard is usually more focused and limited than a view through the building perimeter, making the landscape design of the courtyard a critical concern.

On flat sites, the use of skylights is a common technique for introducing natural light to at least the upper level of an underground structure. In many cases horizontally glazed skylights ac tually provide more natural light than a vertically glazed window, but the same opportunities for exterior views are, of course, not available. Consequently, skylights alone may not be considered an adequate substitute for conven tional windows. Some designers used sloped glazing in courtyards or light wells; this glazing provides natural light from overhead while permitting exte rior views from some angles. This type of glazing is used effectively in Wil liamson Hall at the University of Min nesota (Fig. 6).

An alternative to creating an exte rior courtyard is an interior courtyard with skylights overhead. In large and deep buildings, an interior courtyard provides additional climate-controlled space and, like an exterior courtyard, natural light from above. The extent

ing. The degree to which this is necessary or even desirable for each space in the building depends on the


Figure 5. Designed by architect Gunnar Birkerts, the Uris Library at Cornell University in

Ithaca, New York, is entered by descending a glass-enclosed staircase. (Photo courtesy of

The Arkansas Office)


Figure 6. Sloped glazing provides natural light and view to underground offices in Williamson Hall at the University of Minnesota in Minneapolis. (Photo courtesy of BRW Architects)

to which a view of an interior courtyard is a substitute for an exterior view de pends a great deal on the size and de sign of the space. The use of glass walls between an interior courtyard and the spaces surrounding it is a common technique for transmitting light and view to these spaces. Referred to as "borrowing light," this concept is used in many ways to provide natural light to spaces that are adjacent to spaces that have direct access to light-for ex ample, from a skylight, perimeter win dow , or courtyard.

In addition to the approaches dis cussed above, some other novel tech niques for introducing natural light and exterior view to subsurface spaces have been developed. With the increasing awareness of the advantages of under ground buildings for a variety of func tions, BRW Architects, Inc., of Min neapolis, Minnesota, has experimented with various optical techniques for pro viding or enhancing the effect of nat ural light and view in normally win dowless areas. Mirrors can be used to enhance the light and view from a win dow by reflecting them into spaces not immediately adjacent to the window.


This technique was used in the Walker Community Library in Minneapolis (Fig. 7). Another approach is to pro vide exterior views using mirrors and lenses in a manner similar to a peri scope. In one of the simplest applica tions, a pair of mirrors was used to pro vide views to below-grade offices in the Fort Snelling Visitor Center in Min neapolis.

A more sophisticated periscope-type system using mirrors and lenses pro vides a clear exterior view for a lobby area 110 ft beneath the surface in the Civil and Mineral Engineering Build ing at the University of Minnesota. In the same building, two separate sys tems are used to provide natural light to different areas. One system uses he liostats on the roof to collect sunlight, which mirrors and lenses then project to the offices in deep mined space be low. The other system appears to be a more conventional skylight over the main laboratory space, but reflective surfaces and lenses are used to collect, concentrate, and project the sunlight to the space below more efficiently and accurately. These various optical tech niques not only permit greater design

flexibility for all underground build ings, but also enable light and view to be introduced to deep subsurface spaces or below-grade spaces on very con strained sites where conventional ap proaches do not work. Interior Design

A key element of the impression cre ated by any building, interior design is critical in occupied underground structures. In addition to the normal concerns of creating an attractive in terior environment, special attention must be paid to offsetting the poten tially negative psychological effects dis cussed earlier- claustrophobia, lack of view, loss of orientation, lack of stim ulus, and associations with dark, damp, or cold basements. Interior design, however, has little impact on any of the potentially negative physiological ef fects of underground spaces. Since op portunities to introduce natural light and view are often limited, the interior spaces must be designed to maximize the effect of what light and view there are and to compensate for the lack of light and view in other areas.

A number of techniques are used in


Figure 7. The view of an exterior courtyard is reflected by a large mirror into the interior

of the underground Walker Library in Minneapolis, Minnesota. (Photo courtesy of BRW Architects)

consideration. First, these effects are influenced by the use of the space. Sec ond, not enough information is always available to draw definite conclusions. Finally, many of the negative effects can be alleviated or compensated for with proper design.

It is clear that some underground building functions are more psycho logically acceptable than others. It has been found that an earth-sheltered house with windows was more imme diately accepted as a pleasant place to be than an earth-sheltered work envi ronment (Hollon and Kendall 1980). Manufacturing, storage, and other specialized work settings are in turn more readily accepted than a general office environment. Recreational facil ities, shopping centers, restaurants, and museums appear to be the most readily accepted.

It must be noted, however, that the perceived attributes of a building may not reflect the physical attributes. In a survey of University of Minnesota em ployees (Hollon et al. 1980), occupants

underground buildings to create a feeling of spaciousness that helps to offset claustrophobia and provide more visual stimulus in the absence of ex terior views. Wider corridors and higher ceilings than normal, together with open plan layouts using low partitions, are simple means of creating a spacious feeling. In addition, glass partitions can be used between spaces to provide spa tial relief and variety. Perhaps the most effective approach is the use of large, multi-level central spaces surrounded by smaller spaces. Overlooking large spaces from balconies offsets the feel ing of being below grade and can pro vide vistas nearly equivalent to exterior views, even if the central spaces are not naturally lighted. In addition to pro viding variety within the building, large interior courtyard-type spaces can also help to give building users a reference point in maintaining their orientation.

Although the size and arrangement of interior spaces are critical elements in the perception of an underground building, the more subtle elements of color, texture, lighting, and furnish ings cannot be overlooked. In an en vironment that lacks outside stimuli, variation in lighting and color in par ticular can simulate some of the vari ation that would occur with natural light. Well-lighted, brightly colored surfaces, with an emphasis on warmer tones, can help to offset associations with cold, dark underground spaces. One character istic of windowless space that can be used as an advantage is the lack of dis traction, which enables manipulation of the interior design to focus the oc cupant's attention if desired. Dramatic


lighting can effectively draw attention to artwork, for example.

In buildings with interior courtyards or central spaces, plant materials, fountains, and other landscape ele ments can create a simulated outdoor environment without natural light. Al though this approach has been achieved in windowless spaces with some suc cess, there is always the danger that imitation of conventional above-grade spaces in a superficial way- false win dow frames with painted exterior views, for instance- will leave a negative impression. Rather than working to create an acceptable environment within the limitations of underground space, attention is drawn to what is missing by a poor imitation.

Many of the techniques discussed above are in common use and can serve to create an environment that is reas suringly similar to conventional build ing. On the other hand, it is also pos sible and appropriate in some cases to exploit the uniqueness of being un derground, creating a stimulating pos itive experience. An example of this approach can be found in the design of offices in deep mined space in Kan sas City, Missouri, where rock walls are exposed and often lighted in dramatic ways (Fig. 8).

Conclusion

Although the functional, psycholog ical, and physiological effects discussed above represent drawbacks to under ground or windowless space, a number of modifying factors must be taken into

of an interior windowless zone of a large above-ground building and occupants of a section of the basement of the same building had quite similar evaluations of their work environment as those oc cupants in an above-ground exterior zone (with windows) of the same build ing did. Occupants of the windowless portion of a well-known earth-shel tered building on campus, however, had a much lower evaluation of their work environment.

Assessing the physiological effects of being underground should be some what simpler than assessing the psy chological effects because there are fewer basic concerns, they do not vary as much on an individual basis, and except for length of stay in the build ing-they are less affected by mitigat ing circumstances. But, in spite of the greater facility in clarifying physiolog ical effects, it remains difficult to draw definite conclusions on the physiolog ical impacts of the use of underground buildings. Although the negative ef fects caused by complete lack of sun light and sensory deprivation have been documented , studies of people using underground buildings have not nec essarily concluded that such detrimen tal effects exist.

For example, a fairly comprehensive study was made of workers in an un derground factory in Sweden in the 1940's and 1950's (Hollister 1968). Al though headaches, faintness, and sick ness were reported, much of this was an initial problem that occurred dur ing an acclimatization period to the un derground environment. The study concluded that no m or physiological


Figure 8. Brightly-colored flags, dramatic lighting, glass, and exposed rock walls contribute to the interior design of windowless office space

over one hundred feet underground in Kansas City, Missouri. (Photo courtesy of Great Midwest Corporation)

problems occurred if the proper inte rior climate was maintained. Blood tests on workers after eight years of working underground showed no alteration in the normal blood condition. Two stud ies of children attending an under ground school in New Mexico- the Abo Elementary School- also indicated no evidence of greater absenteeism or health problems in children attending the underground school. In fact, res piratory ailments were reduced be cause of better control and filtration of the air (Lutz 1964 and 1972).

In summary, although there are sev eral issues of concern for people spending extended periods in an un derground or windowless environ ment, a long-term detrimental effect on humans has not been established.

Although scientific proof of negative psychological and physiological effects related to underground buildings is by


no means absolute, the safest course in building design is to alleviate the var ious concerns as much as possible. As evidenced by many of the buildings shown here, competent designers in tuitively respond to creating an ac ceptable environment underground.

D References

Bennett, D. J. 1978. Notes on the under ground. Earth covered buildings and settle

ments. F. Moreland (ed.). Proceedings of a Conference in Ft. Worth, Texas. Gov. Print. Off. Conf-7805138-P2.

Bitter, C., and van Ierland, J.F.A.A. 1967.· Appreciation of sunlight in the home. Proceedings of the Conference on Sunlight in

Buildings . Pp. 27 -37. Rotterdam: Bouwcentrum International.

Boyer, L. L. 1979. Daylighting in subter ranean spaces. Nonresidential applications: Earth

Shelter 2 Conference, Nov. 18-20. Underground Space Center, University of Minnesota.

Collins, B. L. (ed.). 1975. Windows and

peo ple: a literature survey: psychological

reaction to environments with and without

windows. Gaithersburg, MD: U.S. Dept. of Com merce, National Bureau of Standards, Building Science Series 70. June 1975.

Hollister, F. D. 1968. Greater London

Coun cil: a report on the problems of

windowless environments. London: Hobbs the Printers Ltd.

Hollon, S. D. and Kendal, P. C. 1979. Psy

chological responses to underground structures. Nonresidential applications:

Earth Shelter 2 Conference, Nov. 18-20, 1979. Underground Space Center, University

of Minnesota. Hollon, S.D., Kendall, P. C., Norsted,

S., and Watson, D. 1980. Psychological re sponses to earth sheltered, multilevel and aboveground structures with and without windows. Underground Space 5:3, 171-178.

Longmore, J., and Ne'eman, E. 1974. The availability of sunshine and human

re quirements for sunlight in buildings. Journal of Architecture Research

3:2 (May). Lutz, Frank W., and Lutz, S. B. 1964. In-


terim report of the Abo Project. Office of Health, Education and Welfare. Contract #OE-e-99-003 Oanuary).

Lutz, Frank W., Lynch, P. D., and Lutz, S. B. 1972. Abo revisited. Defense Civil Prepar edness Agency. Contract #DAHC20-72-

C-Oll5 Oune). Lutz, Frank W. 1976. Studies of children in

an underground school. Underground Space

5:6, 384-391. Mehrabian, A., and Russell,]. A. 1974. An

approach to environmental psychology. Cam

bridge: MIT Press. Ott,]. N. 1965, Effects of wavelengths of

light on physiological functions of plants and animals. Illuminating Engineering 66,

254-261. Ott, john N. 1973. Health and light , the effects

of natural and artificial light on man and other

living things. Old Greenwich, Conn.: De

vin- Adir Co. Paulus, P. B. 1976. On the psychology of

earth covered buildings. Underground Space

1:2, 127-130.

362 UNDERGROUND SPACE Volume 8, Numbec 5-6, 1984 J

earth shelter architecture - elseviermedia.journals.elsevier.com/content/files/design... ·...

Documents