14. The Impact of Plot Configuration on the Patterns of Spatial Change: A Diachronic Approach to the Urban Redevelopment Processes in New York, Melbourne and Barcelona

Existing physical characteristics of the built environment potentially affect the future (re)development patterns by restricting certain choices and promoting some others. Urban morphology offers an expanding theoretical and methodological framework to investigate this reciprocal relationship between formal preconditions established by buildings, plots and streets and patterns of change in cities. Although plots are widely acknowledged as the most fundamental structural component of urban form (Kropf 1996; Moudon 1997), it remains as the less studied one in the current state of art in urban morphology (Scheer 2018).

In the last years, there is a growing interest in rigorous descriptions of plots and quantitative measurement of plot structures (Bobkova 2019; Berghauser Pont et al. 2019). However, most of these approaches offer ‘synchronic’ comparison between various spatial settings and do not address spatial dynamics and patterns of change, due to the scarcity of longitudinal datasets (see Fig. 14.1). Studies adopting ‘diachronic’ approach are relatively scarce within the discipline and the existing ones offer limited insight as they dominantly rely on conventional measures. This research aims to examine the impact of urban form conditions created by plot structures on the long-term patterns of physical change, by comparing three international case studies: Midtown Manhattan (New York), Central Melbourne (Melbourne) and Eixample (Barcelona). To that end, a number of prevalent hypotheses and recent quantitative measures of plots within the literature will be empirically tested and evaluated.

Full size image

In order to develop a diachronic analysis for the relation between plot structures and physical change, a longitudinal geospatial database has been generated from the cartographic resources, aerial photos and historical records of each case study. The database portrays a series of comparative time frames revealing the physical transformation patterns along with the transition in the configuration of plots. The cross-case analysis of the transformation processes presents empirical findings about the association between the character of plot structures and patterns of change and development.

In the next section (Sect. 14.2), a brief review of urban form studies focusing on plot structures will be presented to identify prevalent hypotheses and common quantitative descriptions of plots. Section 14.3, methodology, presents measures of plots used in the analysis and the key methodological steps, including identification of the case studies, data sources and empirical analysis. In Sect. 14.4, empirical findings of the analysis and cross-case evaluation of the research outputs will be presented. Section 14.5 highlights data-informed insights on the role of plot structures in guiding long-term patterns of physical change and discusses future research directions.

Acquired by the subdivision of land into legal spatial entities, plot is the most fundamental unit of control used in the planning and design of settlements. Apart from its ties to socio-cultural institutions, such as ownership and property (Kropf 2018), plot has a physical dimension which gives it a critical role in structuring the character of urban form and affecting spatial transformation processes. In that sense, from a morphological and spatial-analytical perspective, plot serves as a measurable unit of analysis to unravel the relationship between urban form and spatial change: “a data point consisting only of its measurements, its location, its temporality, its status and its classification” (Scheer 2018).

There are relatively few morphological studies investigating the association between characteristics of plots and patterns of physical change and persistence. Moudon’s (1986) study on Alamo Square is the most known empirical research which demonstrates that plot size has a determinant effect on physical change since the amalgamation of smaller and many plots requires more effort, time and cost. Conzen (1960) demonstrated that size and shape of ‘burgage plots’ and their spatial distribution were determinant on the redevelopment patterns of the medieval town of Alnwick and introduced the concept of ‘burgage cycle’. Burgage cycle describes an evolutionary process in which urban densification proceeds parallel to the progressive subdivision of land into smaller plots. In addition to the relation between plot pattern and urban development, Conzen’s (1960) study presents an intuitive hypothesis arguing that diversity in plot subdivisions could facilitate a more continuous adaptive process and provides a resistance to the completion of burgage cycle. Although it has not received sufficient attention like plot size and shape in the literature, the impact of plot subdivision diversity on the patterns of change and persistence deserves a further empirical inquiry. The most recent empirical studies indicate that while smaller and narrow plots accommodate functional adaptations more successfully and physically persist over time (Vaughan et al. 2015; Tümtürk et al. 2022); larger plots tend to be redeveloped more often in the long term (Whitehand and Carr 1999; Törmä et al. 2017; Hallowel and Baran 2020).

It should be noted that nearly all of these studies have exclusively been conducted in relatively low-dense and monofunctional suburban neighbourhoods. Moudon (1986) warns that some well-known parameters (such as plot size and shape) “may not forcefully curtail change in non-residential areas because of the shorter cycles of business change” (p. 141). Therefore, empirical testing of these prevalent hypotheses and widely used measures within different contexts is an important objective of this research.

In addition to the above-mentioned well-known plot descriptions, this study utilizes a new set of quantitative plot descriptions based on the concept of ‘accessibility’. Although it is not a diachronic research, Bobkova’s (2019) study portrays a new set of plot measures which could be utilized in longitudinal studies to understand temporal transformation and development patterns. Bobkova et al. (2021) demonstrated that accessibility-based plot descriptions (such as accessible plot density and accessible plot frontage) are important variables explaining the concentration and diversification of economic activities in cities.

In the following sections, a methodological framework including conventional and novel plot descriptions will be presented as part of the analytical approach of this research. Accordingly, identified quantitative variables and their relationship with the patterns of physical change will be analysed in a diachronic manner.

The research is based on diachronic and quantitative comparison of multiple case studies through multiple time frames, to understand the association between plot configuration and patterns of physical change. Therefore, it is important to identify both spatial and temporal boundaries of the analytical framework. Three distinctive case studies were selected from different geographies and contexts: Midtown Manhattan (New York), Central Melbourne (Melbourne) and Eixample (Barcelona). Selected study areas are large enough (approximately 2,5 km²) to accommodate a sufficient spatial diversity and a number of different character areas within their boundaries (Davis 2013). Spatial boundaries of the analysis sites were additionally expanded by creating a 400-m buffer—which could be defined as average accessible distance by walking (Gehl 1987). Thus, the surrounding spatial context was also included in the morphological analysis. In terms of temporal boundaries, four different periods were selected to map and compare spatial conditions of each case study. The identified analysis periods correspond to the years when sufficient land surveys and map preparation were conducted before major planning regulations. Although they slightly differ between each case study, selected years successfully represent industrial (the 1880s), modernist (1920s), post-modernist (1960s) and contemporary (2000s) urban development patterns. Availability of the cartographic resources within these periods allowed to generate a comprehensive geospatial database including the elements of plots, buildings and streets across four distinctive time frames.¹

By relying on the existing theoretical discussions on plot structures (Sect. 14.2), the following independent variables were selected to describe and measure the character of plots: plot size, accessible plot density (APD), plot frontage (PF), accessible plot frontage (APF) and accessible plot size diversity (APSD) (see Table 14.1). In addition to the descriptors of plots, patterns of urban redevelopment were conceptualized with respect to the concept of physical change. Physical change describes transformation of the built form either through demolishment of existing buildings or construction of new ones between each analysis period. Identified physical changes between each time frame were projected into the plots containing them and two different populations of plots were categorized: (a) plots accommodating physical change and (b) plots retaining their physical quality. Thus, both numerical values of identified plot variables and categorical information of change or persistence were collected and stored at the scale of individual plots.

Amongst the variables, plot size is the most fundamental geometric description defining the spatial quality of plots and it has been widely discussed as the determinant of spatial change and persistence. The categories of plot sizes were defined by considering the original plot sizes established for each case study at the beginning of nineteenth century. Size categories were identified as extra-small (0–500 m²), small (500–1000 m²), medium (1000–2000 m²), large (2000–5000 m²) and extra-large (above 5000 m²), respectively. Accessible plot density (APD) defines the number of plots accessible within identified catchment (which is 400-m walkable distance in our case (Gehl 1987)) for each individual plot taken as the origin. As different from the plot size which is defining the individual geometric quality of each plot, APD quantifies the intensity of potential destinations (i.e. plots) reachable via street network. Thus, it helps to evaluate the granular character of plot structures in a contextual catchment area, instead of individual size of plots. While the higher values of APD are associated with fine-grain plot compositions defined by many and small plot subdivisions, the lower values indicate coarse-grain compositions of a few and larger plots.

Accessible plot size diversity (APSD) quantifies the degree of the diversity of plot sizes accessible within 400-m walking distance for each individual plot taken as the origin (Feliciotti et al. 2016; Tümtürk 2024). Calculated by the Gini-Simpson Diversity Index, APSD defines the degree of heterogeneity for the distribution of plot sizes within accessible reach.

Plot frontage (PF) is measured by the ratio between plot’s frontage length and its total perimeter; and it defines the narrowness or broadness of plot shape. While lower plot frontage values are associated with narrow plots, higher plot frontage values represent broader plot shapes. Although it is mostly discussed with reference to the conditions of active street life and presence of potential interfaces in the literature, plot frontage could be a critical determinant affecting the patterns of change and persistence as well. Accessible plot frontage (APF) is the cumulative opportunity measure of the same variable and evaluates narrowness/broadness of plot structures within accessible catchment (Bobkova et al. 2017).

Since it is aimed to investigate how ‘existing’ urban form conditions of each time frame affect the patterns of physical change emerging until the next period, calculated measures are represented and mapped for each distinctive time sequence. All quantitative variables, except plot size, are represented by using equal interval discretization for 5 classes: lowest (0–20%), low (20–40%), medium (40–60%), high (60–80%) and highest (80–100%). In the next section, physical transitions between each analysis period for each case study will be presented along with the analysed measures of plot structures. Thus, the hypothetical relationship and alignment between the spatial character of plot structures and long-term patterns of physical change will be discussed critically.

Comparison of each analysis period reveals various transition patterns in plot structures in three distinctive case studies. The number of plots in New York decreases drastically through time, due to amalgamations of plots to accommodate larger building footprints (see Fig. 14.2). Since the analysis areas in Melbourne and Barcelona didn’t reach their spatial extents until the end of the first analysis period, subdivision of the land continued as parallel to urban development processes and the number of plots increased until 1920s. After that period, while Melbourne gradually has lost its fine-grain plot structure like New York, the number of plots in Barcelona remain consistent over years reflecting its enduring urban form characteristics. Similarly, while the average plot size in Barcelona has been stable across years—around approximately 500 m²—average size of plots in New York and Melbourne has substantially increased from 257 m² to 805 m² and from 455 m² to 914 m², respectively (see Fig. 14.3).

Full size image

Full size image

In order to investigate the relation between identified plot variables and the patterns of physical change, plots accommodating physical change and persistence are categorized and mapped for each time frame. The analysis areas in Melbourne and Barcelona were still in urban development process and did not reach their spatial extents until the second time frames. Therefore, their first analysis periods were excluded from the analysis because of the fact that it was theoretically not possible to observe patterns of physical ‘change’, but ‘development’.

Analysis of the relationship between plot size and physical change does not present an obvious and consistent association between them, across case studies and analysis periods. For instance, in Midtown Manhattan (New York), extra-small (XS) plots changed more often than small (S) and medium (M) plots in the second and third analysis periods (see Fig. 14.4). In Eixample (Barcelona), the second analysis period presents a positive relationship between plot size and change and demonstrates that as plot size increases percentage of physical change rises too. However, this relationship could not be find in the third analysis period in a consistent manner. Amongst the case studies, only Melbourne portrays the hypothesized (by existing literature) positive relationship between plot size and physical change. Contrary to widely accepted arguments in the literature, empirical analysis results within this research do not provide support for the relation between plot size and physical change.

Full size image

The empirical analysis of each case study indicates that plots having higher accessible plot density (APD) values are the ones resisting to physical change more through the years (see Fig. 14.5). On the other hand, lower APD values are associated with relatively more physical change. It demonstrates that coarser grain urban fabrics are more prone to physical transitions via frequent building demolitions and new constructions. For instance, in Midtown Manhattan (New York), while 48% of plots in the ‘lowest’ APD class physically changed, only 15% of plots in the ‘highest’ APD class experienced physical transition during the first period (see Fig. 14.5).

Full size image

This relationship is consistent across the second and third periods as well and demonstrates that there is a negative relationship between APD and physical change. Excluding their first analysis periods due to previously mentioned reasons, Central Melbourne and Eixample (Barcelona) portrays the similar relationship across each analysis period (see Fig. 14.5). Though it deserves a systematic and further statistical analysis, it could be argued that the percentages of plots accommodating physical change are comparatively higher in the areas having lower APD values. It shows that the mutual support of many and small plots at particular locations make them more resistant to physical change and retain their urban grain qualities through time. It could be considered that accessible plot density (APD) indirectly measures sizes of plots, as the plot sizes decrease the number of plots accessible within catchment area increases naturally. However, it is important to observe that APD is more successful in capturing patterns of physical change than individual plot size. Especially in New York and Barcelona, there is no obvious relationship between plot size and physical change, as much as that of APD and physical change. Since accessible plot density evaluates the condition of each plot within accessible context, it provides more information in explaining physical change which is arguably affected by the conditions of the surrounding context as well.

Analysis of accessible plot size diversity (APSD) illustrates the degree of heterogeneity in plot sizes within accessible reach. The comparative analysis of each time frame indicates that heterogenous plot structures manifest more and frequent physical change in each case study. On the other hand, homogenous plot structures, identified with lower APSD values, resisted to physical changes more and stayed intact through time (see Fig. 14.6). For instance, in Midtown Manhattan (New York), homogenous character areas—having ‘lowest’ APSD—show 14%, 19% and 12% physical change, respectively, for the consecutive analysis periods. On the other hand, heterogenous areas—having ‘highest’ APSD—show 50, 47 and 38% physical change across these periods. In Central Melbourne, single exceptional peak of 53% physical change in the lowest APSD category could be attributed to large-scale planning interventions to construct coarser superblocks by amalgamating many small plots in the third period. In Eixample (Barcelona), plots accommodating physical change seem to be distributed equally into the categories of APSD ranging from the lowest to the highest. Although there is no obvious relationship between physical change and accessible plot size diversity in Barcelona, the analysis does not indicate contrasting results too. Acknowledging that it presents some exceptional circumstances, accessible plot size diversity (APSD) still provides notable insights into understand physical changes across years for the cases of New York and Melbourne.

Full size image

Plot frontage (PF) defines the degree of being narrow and deep or broad for a plot. According to analysis, individual plot frontage measures do not show a consistent relationship with the patterns of physical change. As shown in Fig. 14.7, the relation between PF and physical change is inconsistent not only between case studies but also within the same cities across analysis periods. The results do not provide empirical support for the widely accepted relation between individual plot shape (narrowness/broadness) and physical change.

Full size image

Contrary to PF, there is a positive and consistent relationship between accessible plot frontage (APF) and physical change for each case study and analysis period (see Fig. 14.8). It could be argued that plot structures composed of deep and narrow plots (with lower APF) are more resistant to physical change. On the other side, areas characterized by broader plot shapes (with higher APF) have accommodated more physical change through time in each case study. For instance, in Midtown Manhattan (New York), broader plot structures—having ‘highest’ APF—show 53, 50 and 35% physical change for consecutive analysis periods. On the other hand, areas characterized by narrow plot structures—having ‘lowest’ APF—show only 21, 35 and 2% physical change across the analysis periods. Similar to the comparison of individual plot size and accessible plot density (APD), accessible plot frontage (APF) seems to be more successful in capturing patterns of physical change than individual plot frontage (PF). It reinforces the argument that accessibility-based measures work better than individual measures (size and shape) in explaining physical change since they evaluate the conditions of plot structures within an accessible context.

Full size image

The empirical analysis indicates that plot configuration and character of plot structures play an important role in guiding long-term physical change in each of selected case studies. Acknowledging some exceptional results in specific periods, the findings illustrate that most long-term patterns of physical change and urban redevelopment could be explained by relying on the identified urban form descriptors. Accessible plot density (APD) has a consistent negative association with physical change in each case study and analysis period. In that regard, it is more successful in capturing patterns of physical change than individual plot size—which has been widely acknowledged as the most fundamental measure defining the rate of change in the built environment. Complementary to that, accessible plot size diversity (APSD) is another important variable having positive relationship with physical change. In that sense, fine-grain and homogenous plot structures are more resistant to the emerging pressures of urban redevelopment and change, comparing to character areas identified with coarser and heterogenous plot structures. In addition, higher values of accessible plot frontage (APF) are associated with higher rates of physical change, indicating a positive association. It could be argued that plot structures composed of deep and narrow plots and identified with lower APF values are more resistant to physical change. On the other hand, the research could not identify a similar relationship between individual version of the measure—plot frontage (PF)—and physical change.

The analysis demonstrates that accessibility based measures perform better in associating with the patterns of physical change, comparing to individual geometric measures of size and shape which are prevalent in the existing literature. It shows that evaluating the character of plot structures within a pre-defined accessible reach leads better results by taking the closer context into consideration (Tümtürk et al. 2024). The analysis should be reproduced at various catchment distances and at different contexts to optimize the findings of this study. Moreover, the analysis showed that the relationship between plot variables and physical change may show some exceptional results at certain periods. These discrepancies could be attributed to large scale and top-down planning interventions, which could be potentially included in future studies. External and non-physical aspects such as socio-economic, political and historical dynamics could be also helpful to interpret discrepancies and contextual circumstances.

The research provides empirical support for the role of plot configuration in guiding long-term patterns of change. As opposed to the prevalent hypotheses focusing on individual geometric measures of ‘size’ and ‘shape’, it demonstrated the success of accessibility based measures in capturing patterns of physical change. These preliminary findings should be questioned with the help of different morphological variables and measures, and supported with a coherent statistical analysis to draw further conclusions. Adopting cluster analysis techniques by relying on multi-variables of plot structures and developing plot typologies to assess their capacity of spatial change could be the next step to extend the outcomes of the research. Critical discussion of these morphological variables and their association with the patterns of change and persistence has a potential to contribute to our understanding of urban redevelopment processes and help to regenerate resilient and adaptable urban spaces by design.