Urban land use reflects the location and level of spatial accumulation of activities such as retailing, management, manufacturing, or residence. They generate flows supported by transport systems.
Urban areas are characterized by social, cultural, and economic activities taking place at separate locations forming an activity system. Some are routine activities because they occur regularly and are thus predictable, such as commuting and shopping. Some activities tend to be irregular and shaped by lifestyle (e.g. sports and leisure) or specific needs (e.g. healthcare and education). Such activities are usually related to the mobility of passengers. In addition, there are production activities related to manufacturing and distribution, whose linkages may be local, regional, or global. Such activities are usually associated with the mobility of freight. Since activities have different locations, their separation generates passenger and freight movements. Therefore, transportation and land use are interrelated because of the locational and interactional nature of urban activities.
Most economic, social, or cultural activities imply a multitude of functions, such as production, consumption, and distribution. Urban land use is a highly heterogeneous space, and this heterogeneity is in part shaped by the transport system. There is a hierarchy in the distribution of urban activities where central areas have emerged because of economic (management and retail), political (seats of government), institutional (universities), or cultural factors (religious institutions). Central areas have a high level of spatial accumulation and the corresponding land uses, such as retail. In contrast, peripheral areas have lower levels of accumulation attributed to residential and warehousing areas.
The preferences of individuals, institutions, and firms have an imprint on land use in terms of their locational choice. The representation of this imprint requires a typology of land use, which can be formal or functional:
Formal land use. Representations concerned with qualitative attributes of space such as its form, pattern, and aspect and are descriptive in nature.
Functional land use. Representations concerned with the economic nature of activities such as production, consumption, residence, and transport, are mainly a socioeconomic description of space.
At the global level, cities consume about 3% of the total landmass. Although the land use composition can vary considerably depending on the function of a city, residential land use is the most common, occupying between 65 and 75% of the footprint of a city, excluding transportation. Commercial and industrial land uses occupy 5-15% and 15-25% of the footprint, respectively. There are also variations in the built-up areas that are commonly a function of density, automobile use, and planning practices. In automobile-dependent cities, roads and parking lots account for 35 to 50% of the land-use footprint. Within a parking lot, about 40% of the surface is devoted to parking vehicles, while the remaining 60% is for circulation and access to individual parking spaces. These variations are the outcome of a combination of factors that reflect the unique geography, history, economy, and planning of each city.
Land use, both informal and functional representations, implies a set of relationships with other land uses. For instance, commercial land use involves relationships with its suppliers and customers. While relationships with suppliers will dominantly be related to the mobility of freight, relationships with customers will also include the mobility of people. Thus, a level of accessibility to both systems of circulation must be present for a functional transportation/land use system. Since each land use type has specific mobility requirements, transportation is a factor of activity location.
Within an urban system, each activity occupies a suitable, but not necessarily optimal, location from which it derives rent. Transportation and land use interactions mostly consider the retroactive relationships between activities, which are related to land use, and accessibility, which is transportation-related. These relationships have often been described as a classic “chicken-and-egg” problem since it is difficult to identify the cause of change; do transportation changes precede land-use changes or vice-versa? There is a scale effect at play in this relationship as large infrastructure projects tend to precede and trigger land-use changes. In contrast, small-scale transportation projects tend to complement the existing land use pattern. Further, the expansion of urban land use takes place over various circumstances such as infilling (near the city center) or sprawl (far from the city center) and where transportation plays a different role. For infilling, land value becomes high enough to justify developments despite potential congestion, while for sprawl, accessibility has improved enough.
Urban transportation aims to support transport demands generated by the diversity of urban activities in diverse urban contexts. A key to understanding urban entities thus lies in analyzing patterns and processes of the transport-land use system since the same processes may result in a different outcome. This system is highly complex and involves several relationships among the transport system, spatial interactions, and land use:
A conundrum concerns the difficulties of linking a specific transportation mode with specific land use patterns. While public transit systems tend to be associated with higher densities of residential and commercial activities and highways with lower densities, the diversity of modes available in urban areas, including freight distribution, conveys an unclear and complex relationship. Further, land use is commonly subject to zoning restrictions in terms of the type of activities that can be built as well as their density. Therefore, land use dynamics are influenced by planning restrictions and the urban governance structure.
The relationships between transportation and land use are rich in theoretical representations that have significantly contributed to regional sciences. They can be investigated empirically by observing and analyzing real-world changes in the urban spatial structure. However, empirical investigations cannot readily be used for simulation and forecasting. For that purpose, the relationships between transportation and land use can also be investigated through models trying to synthesize the spatial structure through a series of assumptions about urban dynamics. Since cities have diverse sites and political, economic, historical, and cultural contexts, there are no absolute rules concerning their spatial organization.
Since transportation is a distance-decay-altering technology, the spatial organization of cities is assumed to be strongly influenced by the concepts of location and distance. Several descriptive and analytical urban land use models have been developed over time, with increased levels of complexity. All involve some consideration of the effect of transport in the explanations of urban land use structures. Changes commonly result from locational decisions such as building a facility (residential building, warehouse, store, office tower, etc.) or a transportation infrastructure (road, transit line, port, airport, etc.).
Von Thunen’s regional land use model is the oldest representation based on a central place, the market town, and its concentric impacts on surrounding agricultural land use. The model was initially developed in the early 19th century (1826) to analyze agricultural land use patterns observed in Germany. The concept of economic rent is used to explain a spatial organization where different agricultural activities are competing for the usage of the available land. The closer a location is to the market, the lower the transportation cost and availability of land. The underlying principles of this model have been the foundation of many others, where economic considerations, namely land rent and distance decay, are incorporated. The core assumption of the model is that agricultural land use is patterned in the form of concentric circles around a market that consumes all the surplus production, which must be transported. It is this transportation cost that bears the most influence on the purpose of the land. The closer the market, the higher the intensity and productivity of agricultural land use, such as dairy products and vegetables, while the further away, less intensive uses, such as grain and livestock, dominate. Many empirical concordances of this model have been found, notably in contemporary North America.
Another range of early models, such as Weber’s industrial location model developed in 1909, dealt with industrial location to minimize the total transportation costs of accessing raw materials and moving the output to the market, indicating an optimal location for the activity. The main principle explored by early models is that transportation costs primarily influence locational choice and the resulting land uses. This assumption is not surprising since, in the late 19th century and the early 20th century, land transportation options were limited and of a relatively high cost.
The Burgess concentric model was among the first attempts to investigate spatial patterns at the urban level in the first quarter of the 20th century. Although the purpose of the model was to analyze social classes, it recognized that transportation and mobility were important factors shaping the spatial organization of urban areas and the distribution of residential choices. This model’s formal land use representation is derived from commuting distance from the central business district, creating concentric circles, and each circle represents a specific socioeconomic urban landscape. This model is conceptually a direct adaptation of the Von Thunen model to the distribution of urban land use since it deals with a concentric representation, which considers a transportation trade-off between commuting and renting housing. Therefore, if the cost of commuting declines due to improvements (e.g. new transit lines), the outcome is that more people can afford to live further away, which results in urban sprawl. Even close to one century after the concentric urban model was developed, spatial changes in cities such as Chicago still reflect this process.
Sector and multiple nuclei land use models were developed to consider numerous factors overlooked by concentric models, namely the influence of transport corridors (Hoyt, 1939) and multiple nuclei (Harris and Ullman, 1945) on land use and growth. Both representations consider the emerging impacts of motorization on the urban spatial structure, particularly through the beginning of suburbanization and the setting of polycentric cities. Cities could be structured by several subcenters of different importance and function with mixed land uses in between.
Such representations also consider that transportation infrastructures, particularly terminals such as rail stations or ports, occupy specific locations and are also land uses. In the second half of the 20th century, the construction of airport and container port complexes, including logistics zones, created new nodes around which urban land uses developed. Further, urban sprawl became apparent as a force shaping urban land use development, characterized by automobile dependency and low-density suburban locations.
Hybrid models attempt to include the concentric, sector, and nuclei behavior of different processes in explaining urban land use. They try to integrate the strengths of each approach since none of these appear to provide a completely satisfactory explanation. Thus, hybrid models, such as those developed by Isard (1956), consider the concentric effect of central locations (CBDs and sub-centers) and the radial effect of transport corridors, all overlaid to form a land use pattern. Hybrid representations are also suitable to explain the evolution of the urban spatial structure as they combine different spatial and temporal impacts of transportation on urban land use, such as concentric and radial impacts. It recognizes the evolution of technological influences on urban mobility, from walking and cycling, to the setting of modern urban transit systems along corridors and the network of urban highways.
Land rent theory was also developed to explain land use as an outcome of a market where different urban activities compete to secure a footprint at a location. The theory is strongly based on the market principle of spatial competition, where actors bid to secure and maintain their presence at a specific location. The more desirable a location is, the higher its rent value and the intensity of activities. Transportation, through accessibility and distance decay, is a strong explanatory factor on land rent and its impacts on land use. Conventional representations of land rent leaning on the concentric paradigm are challenged by structural modifications of contemporary cities identified by hybrid models.
The applicability and dynamics of land use models are related to issues such as the history, size, and locational setting of a city. For instance, concentric cities are generally older and of smaller size, while polycentric cities are larger and relate to urban developments that have taken place more recently. This also includes the impacts of public transit systems that can vary according to the level of automobile dependence.
Another issue is the representativeness and applicability of a global urban landscape. While most conceptual approaches related to the relationships between transportation and land use have been developed using empirical evidence related to North America and Western Europe, this perspective does not necessarily apply to other parts of the world. The most important distinguishing factors include:
To what extent globalization will favor a convergence of land use patterns across the world’s cities remains to be seen. Irrespective of the urban context, standard technologies such as the automobile, construction techniques and materials, information technologies, and managerial practices (e.g. urban planning or supply chain management) are likely to homogenize the land use structure of global cities.
Both land use and transportation are part of a dynamic system subject to external influences and internal changes. Each component of the system is continuously evolving due to changes in technology, policy, economics, demographics, and even culture or values. Since transportation infrastructure and real estate development require significant capital investments, understanding their dynamics is highly relevant for investors, developers, planners, and policymakers. As a result, the interactions between land use and transportation are played out as the outcome of the many decisions made by residents, businesses, and governments.
The field of urban dynamics has expanded the scope of conventional land use models, which tended to be descriptive, by considering the relationships behind the evolution of the urban spatial structure. This focus has led to a complex modeling framework, including a wide variety of components such as the transportation network, housing locations, and workplaces. Among the concepts supporting urban dynamics are retroactions, whereby changes in one component influence other associated components. As these related components change, there is a feedback effect on the initial component, which is either positive or negative. The most significant components of urban dynamics are:
Several transportation land use models have been developed to represent complex urban dynamics, with the Lowry model among the first (1964). Its core assumption is that regional and urban growth (or decline) is a function of the expansion (or contraction) of the basic sector, which is represented as export-based employment that meets non-local demand. An urban area produces goods and services, which are exported. This employment is, in turn, impacting the employment of two other sectors; retail and residential. Its premises were expanded by several other models, known as “Lowry-type” models, applied to various cities. The core of these models relies on a regional economic forecast that predicts and assigns the location of the basic employment sector. They depend on the reliability and accuracy of macroeconomic and micro-economic indicators and forecasting. Such forecasting tends not to be very accurate as it does not capture well the impacts of economic, social, and technological changes, which also change the relevance of indicators.
Another line of models emerged in the 1990s with the rise of computing power. Cellular automata are dynamic land use models developed to represent space as a grid where each cell is a discrete land use unit. Cell states thus symbolize land uses, and transition rules express the likelihood of a change from one land use state to another. Because cells are symbolically connected and interrelated (e.g. adjacency), models can be used to investigate the dynamics, evolution, and self-organization of cellar automata land use systems. The cellular approach allows for achieving high spatial detail (resolution) and realism, and linking the simulation directly to visible outcomes on the regional spatial structure. They are also readily implementable since Geographic Information Systems are designed to work effectively with grid-based (raster) spatial representations. Cellular automata improve upon most transportation – land use models that are essentially static as they explain land use patterns. Still, they do not explicitly consider the processes of creating or changing them.
The issue of articulating transportation and land use interactions remains, particularly in the current context of interdependence between local, regional, and global processes. There is also the risk of unintended consequences (unaccounted feedback) where a change may not result in an expected outcome. For instance, improving road transportation infrastructure can have the potential to create even more congestion as new users are attracted by the additional capacity. Globalization has substantially blurred the relationships between transportation, land use, and its dynamics. The primary paradigm is concerned with some factors once endogenous to a regional setting that has become exogenous.
Many economic activities that provide employment and multiplying effects, such as manufacturing, are driven by forces that are global in scope and may have little to do with regional dynamics. For instance, capital investment in infrastructures and facilities could come from external sources, and the bulk of the output could be bound to international markets. In such a context, it would be challenging to explain urban development processes in coastal Chinese cities, such as the Pearl River Delta, since export-oriented strategies are among the most significant driving forces. Looking at the urban dynamics of such a system from an endogenous perspective would fail to capture dominantly exogenous driving forces.
The relationships between transportation and land use have been the focus of a long line of geographical representations, including models, and are mainly driven by economic and technological changes. It is expected that ongoing changes related to digitalization, such as e-commerce and automation in manufacturing and distribution, will continue to shape the urban spatial structure in the 21st century.
