The skillful design of air movement through and around buildings is both a technological and experiential enterprise. In other words, buildings construct atmospheres in both the aesthetic and the technoscientific sense. As spatial and material enclosures, buildings are immersive, emanating auras or moods through careful calibration of visual, acoustic, and thermal sensations. As shelters, buildings mediate between erratic exterior and comfortable interior either naturally or mechanically. As notorious energy consumers, buildings generate emissions that alter the chemical composition of the gaseous bubble surrounding the earth. Moving air, driven by a building’s form, orientation, and the location of inlets and outlets, is a primary medium through which these experiential effects, meteorological exchanges, and chemical emissions are transmitted. And these exchanges take place across many scales relevant to the built environment, from the material seam to the geographical territory. However, the architects’ skillset for designing with wind as a complex moving material system is limited to either rule-of-thumb diagrams or complex digital simulations. Neither represent wind as a graspable, experiential, moving material system nor facilitate design thinking across the vast range of scales—from the detail to the planetary—that the built environment impacts.

This essay introduces a third technique for designing buildings as complex environmental mediators across scales—using physical environmental models. To design with wind is to work with a medium that is consequential, resistant to representation, and materially complex, making it particularly challenging to enfold into the architectural design process. This brief essay elaborates on the nature of these three challenges to developing an approach to designing with wind. It introduces how physical environmental models make air movement through and around buildings visible, material, spatial and tangible. Environmental models enable design exploration across scales, revealing technical understandings of fluid dynamics while providing glimpses of immersive environments charged by their atmospheric surroundings.

Air movement has vast consequences across spatial and temporal scales. Phenomena such as turbulence occur at scales as small as several millimetres, changing over the course of seconds or minutes, while global circulation patterns can operate over thousands of kilometres across several years.¹ At the scale of a building, air movement impacts thermal comfort and constructs exterior microclimates. Wind impacts structural performance and material endurance. Wind can reduce or increase dependence on mechanical heating and cooling through natural ventilation. At meso and macro scales of larger settlements, the shifts and courses of wind determine the path and distribution of air-borne particulates such as dust, sand, snow and contaminants. As global weather patterns become more erratic, the wind drives many catastrophic events associated with climate breakdowns, such as the spread of wildfires, the intensity of hurricanes and floods, and the extent of erosion and desertification. Each of these scales has design implications for the built environment.

Architects work representationally, yet air movement is resistant to representation. Architects do not build buildings; they generate representations of them, generally using drawings to develop design concepts and to describe the construction of complex building material assemblies at scale. One of the challenges of representing air movement is that wind is materially and temporally ambiguous, unlike most conventional building materials such as concrete, wood, or steel. Architectural theorist Christopher Hight refers to consequential, trans-scalar environmental conditions such as airflow and thermal exchange as “non-visual phenomena object(s).”² As a “non-visual phenomena object,” wind is complex and resistant to representation because it is invisible, and it follows fluid dynamic principles that are not always intuitive to designers. As a comparison, solar trajectories for any given latitude are visible, legible over time, and entirely predictable, facilitating easy calibration of shading elements. In contrast, wind patterns are invisible and often shift in intensity and direction erratically over a day, season, and year, making their impacts on building orientation and configuration challenging.

There are exemplary projects in which wind actively shapes the form and configuration of buildings. For example, Renzo Piano Building Workshop’s Tjaibou visitors’ center in New Caledonia has a series of large, curving, vertical screens that are shaped to smoothly redirect monsoon winds around the building. In other cases, airflow is encouraged through a building to maximise cooling effects in extreme heat. Australian architect Glenn Murcutt’s residential projects, with expressive roofs, large operable openings, and vertical fins, exemplify this approach, which relies on generating turbulence rather than streamlined airflow. These projects are the exception rather than the norm largely because the complexity of the fluid dynamics of airflow or the lack of an ‘extreme’ airflow parameter driving a project prevent architects from actively foregrounding it in the design process.

One reason for this oversight is that air is materially complex. Wind is not fixed in space or time, nor does it have defining dimensional limits. It is not even clear whether the wind is a material or a thing at all. In a previous issue of Venti, philosopher Tonino Griffero describes the wind as a “quasi-thing,” with particularly ineffable material characteristics. Unlike more conventional architectural materials, “wind is not edged, discrete, cohesive, or solid…nor does it properly possess spatial sides…”³ The temporalities of wind are nebulous. Wind is intermittent and as Griffero notes, it doesn’t age, degrade, nor “show any temporal patina”. Architects value materials for their dimensional stability and temporal longevity; these traits also translate them to static, line-based representational systems. The dispersed, shifting materiality of air movement disrupts these valued attributes.

Architects' two predominant techniques for representing air movement through and around buildings are static environmental diagrams overlaying arrows of anticipated air movement and more complex building performance simulations such as computational fluid dynamics (cfd). Both techniques have limitations. Static diagrams are just that, static, representing a single moment in time at a fixed position in space. As hypothetical projections, they fail to reveal substantive properties of air movement other than directional flow. Instead, they represent the diffuse, generally chaotic, and turbulent medium of air as a swarm of vectors akin to the wind barbs on a meteorological map. Building performance simulations such as cfd have dynamic attributes and can be scaled to some extent within the digital environment. However, they also present air flow as vectors or colour gradients and are designed primarily for engineers, requiring complex inputs to ensure accurate results. For the initiate such as an architectural designer, the software is unwieldy, particularly in the early design stages of a project where numeric precision is not necessary.

I have refined a third technique for visualising and designing with wind, using environmental models, to make the ‘non-visual phenomena object’ of airflow and its consequences across scales visible as complex moving, material systems. Environmental models are physical models that build on the lineage of engineering experimentation devices—such as wind tunnels and water tables—which simulate airflow at scale. Engineers use these devices to derive numerical results about air speed and pressure; these values are then scaled up using appropriate formulas to predict full-scale performance. In my work, environmental models are more qualitative than quantitative. Unlike static diagrams, which are based on rules-of-thumb, and unlike digital simulation, where air behaviour largely remains trapped in the black box of software, environmental models enable direct engagement with fluid materials, appealing to the architect’s inherent spatial and material sensibilities.

The process of designing, constructing, operating, and recording environmental models opens up insights about airflow that aren’t captured through static drawing or digital simulation. They reveal insights at scales that the other conventions neglect, such as the 1:1 scale of the constructed instrument itself and the ambiguous scale captured through strategic photographs of the model testing bed. Moreover, they reveal material tendencies of nonlinear turbulent flow patterns that both the diagram and the simulation, which mostly rely on vectors, reductively abstract.

There are three key concerns that environmental models reveal that the other conventions neglect. First, constructing environmental models entail working directly with the actual material systems (of airflow) at stake, revealing, for example, the capriciousness of air, its sensitivity to disruption, and its tendency to leak. These insights focus on detailed elements of architecture often overlooked in the early design process—the vulnerable seams, joints, and intersections at which two or more material systems come together.

Second, developing model componentry required to direct, contain, straighten, or deviate air movement suggests ways of conceiving buildings composed of elements that do the same. Environmental models incorporate baffles, screens, expansion and contraction cones, twirling rudders, reservoirs, and vessels, each of which plays a particular role in capturing and directing air movement, raising questions: what would an architecture of nozzles, baffles and hoods look like and how would it perform?

Finally, the process of photographing environmental models shifts focus from effects on the testing bed to spatial environments that invite speculative inhabitation. Environmental models generate spatial and material moments, captured through photography, that immerse the viewer into atmospheric worlds, revealing architectural space charged by environmental effects. In some cases, resultant photographs reveal eerie, luminous interiors and watery underworlds rich in atmospheric effect. In other cases, the photographs are more foreboding, presenting buildings as vessels that leak ominously into the atmospheric domains within which they are situated. In this way, environmental models can make the degradative processes of airflow extremely visceral and apparent.

Environmental models attune thinking to atmospheric concerns that, in turn, enable those concerns to drive design decision-making. Designing and constructing environmental models reveals insights about the vessels that hold air and water and the componentry that directs, channels, straightens and speeds up this flow. Constructing environmental models demonstrates the need for lateral stability in response to air movement and for tight tolerance constructional strategies that are air and watertight. Operating the models reveals that architectural space is mutable and activated by shifting environmental effects. Photographs of the models operate in some ways as cautionary tales, revealing the extent to which buildings actively construct their atmospheric surroundings. Rather than reveal a fundamentally new building style through the process, working with environmental models focuses a way of thinking about buildings in the design process: as precise instruments that interface and collude with environmental systems across scales from the seam (that leaks) to the body (that feels) to the building (that mediates) to the world (that immerses).