Have a personal or library account? Click to login
Horizontal building extensions: potential in Finnish blocks of flats Cover

Horizontal building extensions: potential in Finnish blocks of flats

By: Jyrki Tarpio and  Panu Lehtovuori  
Open Access
|Mar 2026

Full Article

1. INTRODUCTION

The growth of suburban housing estates in Europe during the 1960s and 1970s addressed modernising societies’ housing demand. Often new areas were designed as self-contained neighbourhood units. Today, these boom-year housing estates require renewal and retrofitting for sustainability. In growing cities and metropolitan regions, renewal typically involves densification that often takes the form of demolition and redevelopment (Belmessous et al. 2005; Stjernberg 2022). In some cases, vertical extension is used (Soikkeli et al. 2015). Other design concerns in socially sustainable renewal include modifications to flat distribution to meet changing market demand and accessibility to address the needs of the ageing population (Smas et al. 2013).

In addition to the social concerns, circularity in the built environment calls for designing and constructing buildings and infrastructure to maximise resource efficiency and minimise waste throughout their entire life-cycle. In the context of suburban estates’ renewal and retrofitting, this involves an effort to postpone usable buildings’ demolition. Especially important is to find alternatives for demolition and redevelopment, where existing structures’ embodied carbon is lost. As Huuhka (2023: 929) argues, extending the life of existing buildings outperforms new buildings in terms of carbon emissions, especially in the short term—right now when rapid climate change mitigation is the top priority.

In Finland, the historic peak of new housing construction was in 1974, with 73,000 new units finalised in that year. Today, flats built during the 1970s only represent one-fifth of the national housing stock (Statistics Finland n.d.). To address the large and imminent task of renewing the boom-year housing estates, the Finnish Ministry of Environment organised a research programme1 for evidence-based policies and designs. This article is based on work conducted in Tampere University in the period 2020–24. The key question of two related research projects2 was to find new solutions to densify the urban structures built between the 1960s and the 1980s, without compromising the environmental and social qualities achieved in the original neighbourhood unit designs (Tarpio & Norvasuo 2024; Tarpio et al. 2024). In these projects, horizontal extendibility of suburban multifamily residential buildings was studied for the first time in Finland.

1.1 KNOWLEDGE GAP AND RESEARCH QUESTIONS

Horizontal extensions of blocks of flats are a promising but relatively rare approach for circular and socially sustainable urban densification. To add knowledge to this evolving field, this study investigates the potential of blocks of flats’ horizontal extensions in selected Finnish cases. The following research questions are addressed:

  • What are the most essential block and building-level morphological features that affect the possibilities for horizontal extensions?

  • How can the studied extension modes diversify flat distribution in buildings and qualities of flats?

Five basic modes of horizontal extension were identified previously (Tarpio & Norvasuo 2024; Tarpio et al. 2024). Section 2 explains them along with the research method and case study selection. Section 3 explores the results in detail. Sections 4 and 5 draw the lessons together and generalise key outcomes for the Nordic and international audiences.

1.2 CONTEXT

In horizontal extensions (sometimes called lateral extensions), the existing load-bearing structure is preserved and stairwells stay in use. Often also facade components can be either kept or reused locally. Streets, parks and yards that characterise the social programme of neighbourhood units may remain. Individual dwellings partly stay intact and are partly modified by either enlarging and adding qualities or by subdividing to smaller units. In all, horizontal building extensions may provide a socially balanced and environmentally sound new approach to the renewal of suburban housing estates, complementing demolition and redevelopment, and vertical extension.

Despite a clear promise for sustainable practice, academia has little research on horizontal extensions of blocks of flats. A literature review provides technical, economic and process analyses (Choi & Choi 2021; Dimkov & Papasterevski 2021; Moshaver 2020), work on low-rise buildings (Douglas 2006), reviews on Lacaton & Vassal’s projects in France (Panzini & Quadrato 2022), and some studies on both horizontal and vertical extension possibilities of Swedish Million Programme buildings of the 1960–70s (e.g. Ghafoori & Javadian 2024). In the Swedish context, Chhaya (2017) focused on two buildings in Umeå and explored how different depths of horizontal extension create variety in the flats’ distribution and quality.

In Finland, some studies have been conducted on renewing suburban estate buildings internally (e.g. Kaasalainen & Huuhka 2020), as well as on extending buildings vertically, i.e. constructing one or two additional storeys on top of the existing buildings (e.g. Hilli-Lukkarinen 2019). Reflecting the international situation, Finnish research and practice on renewing the post-war building stock with horizontal extensions is lacking, though. Recognising the knowledge gap on horizontal building extensions, this paper focuses on the potential of horizontal extension in the context of Finnish suburban housing estates of the 1960s–80s.

In Finland, housing estates were deliberately constructed outside city centres (Hankonen 1994). The early post-war estates are characterised by layouts where buildings are situated to respect the existing topography and vegetation. The main volume of housing production in the 1960–80s, however, took place in rationalised gridiron layouts that in Finland are called ‘neo grid plans’ (uusruutukaava) (Figure 1): a reference to the historic gridiron wooden towns. Planners understood the ‘neo grid plans’ as principally open, allowing for flexibility in design (Tarpio & Norvasuo 2024: 54–55). Even though buildings’ horizontal extension was not originally thought about, the open-form principle is an interesting starting point with which to study that option currently.

Figure 1

Matinkylä suburban housing estate in Espoo in the late 1960s: a ‘neo grid plan’ in the making.

Source: Finnish Labour Archives (Työväen arkisto), CC BY-NC-ND 4.0; https://tyark.fi/en/home/.

Since the late 1960s, housing estates’ production was standardised, which meant not only standardising building parts but also using standardised designs based on modular grids, resulting in serial building types and repetitive flat layouts. This kind of standardisation was typical in Finland and also in most European countries (Monclús et al. 2018). Precast concrete was the main construction material, adding to a certain uniformity of scale, detailing and appearance. This resulted in neighbourhoods that resemble each other and which have been criticised for their placelessness (Roivainen 1999). Recently, Finnish suburban housing estates have experienced a minor renaissance as the means of public transport has improved while the moderate-density, easy-access-to-greenery and functional flats are experienced as positive factors (Uimonen 2022). Taken together, the above observations and developments form the context of this study, pointing to its relevance and impact.

2. METHODS

This study employs research-by-design (de Jong & van der Voordt 2002) applied to real-life cases. Research-by-design in architecture and planning involves using the design process itself as a method of enquiry to generate new knowledge and insights into complex problems. While not a new approach, it is gaining traction as a way to address the challenges of the built environment by integrating reflective design practice with replicable research methods.

Following Creswell (2009), Lenzholzer et al. (2013) distinguish four knowledge claims in research-by-design: (post)positivist, constructivist, advocacy/participatory and pragmatic.3 The present study employs the (post)positivist methodology, focusing on spatial configurations, dimensions and functionalities in horizontal extensions of blocks of flats. It sets the results in pragmatic claim to understand better their context and applicability. By studying housing architecture in Finland, research-by-design has been lately applied in varying forms by, for example, Saarimaa & Pelsmakers (2020), Pelsmakers et al. (2022), Tarpio & Huuhka (2022), Tarpio & Norvasuo (2024) and Meriläinen & Tervo (2025). The method can produce new and detailed knowledge that could not be obtained by any other method.

2.1 RESEARCH BY DESIGN ON SELECTED CASES: A FOCUS ON MOST TYPICAL BUILDINGS

In this study, research-by-design involved experimental designs to selected housing blocks and buildings. Original drawings of the blocks and buildings were acquired from the building inspection archives of the city of Espoo. Based on their information, they were redrawn using AutoCad software. Through iterations between three-dimensional design in block scale (footprint and volume) and two-dimensional design in building scale (drafting floor plans and analysing dwellings) it was uncovered how the feasible horizontal extensions affect block layouts and existing dwellings. Drawing alternative options for extensions constituted a major part of the research.

The study concentrated on horizontal, low-rise residential buildings that contain two or more cores (stairwells) with direct access to dwellings. In Finnish housing architecture, this type is called ‘lamella building’ (lamellitalo). A ‘lamella’ is the segment of a building that consists of dwellings and other spaces grouped around one stairwell (i.e. no horizontal corridors). In lamella buildings, the segments are often similar or relatively similar. In the international context of architectural research, the phrase ‘lamella’ has been occasionally employed with this meaning (Meriläinen & Tervo 2025). It will be used in this paper hereafter.

Lamella buildings are typical in Finnish neighbourhood units. This study focused on blocks consisting of lamella buildings constructed between the 1960s and the early 1980s. Point buildings, i.e. those where dwellings are clustered around a single staircase, are relatively common as well. They were excluded from this study as their general dimensions, geometry and logic of flat distribution deviate from lamella buildings to the extent that same modes of extension cannot be applied. For the same reason, blocks of flats with deck access or central corridor access were also excluded.

2.2. SPECIFYING THE CASE: LAMELLA BUILDING WITH FIVE ROOM ZONES

Lamella buildings of the studied era usually employ a distinct division of stairwell and habitable spaces into transverse room zones extending through the building frame. The stairwell is daylit and located in the middle of the lamella within one room zone. Habitable spaces usually surround it in five room zones. This arrangement is common, although variations exist, and larger lamellas with six and occasionally even seven room zones can be found. Figure 2 shows a floor plan of a lamella building where one lamella contains five room zones and the other six on the stairwell side of it. This study concentrated on the horizontal extendibility of residential buildings with five room-zone lamellas.

Figure 2

Floor plan of a lamella building with two lamellas and particularly noticeable transverse room zones (highlighted in one lamella).

In the research-by-design, five fundamental principles were applied at the dwelling scale:

  • the maximum width of a dwelling should not exceed three room zones to avoid unusually large units

  • flats with two or more bedrooms must have windows to more than one direction

  • studio flats and one-bedroom flats may have windows to one direction only

  • extensions may change only one existing habitable room per flat into windowless secondary space, such as a utility room, vestibule, closet or bathroom

  • entrances and one bathroom with a toilet should be accessible for wheelchair users in new flats and preferably also in modified existing flats.

Additionally, horizontal extensions examined in this study contained at least two habitable rooms. This means that solely balcony extensions and small extensions containing only one habitable room were not studied.

In elongated lamella buildings, there is a clear contrast between narrow end facades (gable facades) and long facades. Applying the above-mentioned principles, buildings can be extended horizontally in both gables and the long stairwell sides (Figure 3). Hence, rectangular lamella buildings can be extended in three directions, while one direction remains unextendible. Five distinct basic modes of horizontal extension can be defined (Tarpio & Norvasuo 2024).

Figure 3

Five basic modes of horizontal extendibility of a lamella building.

Note: Extensions are highlighted; modifications to the original building and new spaces are shown in black.

In Figure 3, gable extensions (A) can be accomplished with or without a stairwell renewal. As stairwells are placed apart from gable ends, gable extensions add new rooms to existing flats. To avoid unusually large dwellings, gable extensions were designed as an additional room zone. To maintain the sizes of rooms in this zone proportionate to the existing rooms, its maximum width was considered to be 4 m. Other extensions (B–E) are to be located on the long side of the building next to stairwells or between them. They all require stairwell renewals. A side wing extension (B) is built as a continuation of the gable facade. It may contain habitable space in one, two or three room zones, and thus its width may vary approximately between 4 and 11 m. A side wing with deck access (C) resembles the side wing extension, but it may be longer, as several flats can form a row along the deck. In Finland, the maximum length of the deck is determined by the permitted maximum route length of 30 m from the entrance door of the furthermost flat to the compartmented stairwell (Ministry of the Environment 2017: 32 §). A middle wing extension (D) resembles a side wing extension but is built as a projecting addition on the long side of the building in front of a staircase and one or two original habitable rooms. For middle wings, the deck access arrangement was not considered possible as the deck cannot be a straight continuation of the stairwell without causing the need to convert more than one habitable room of a flat into secondary space. A frame depth increase can be implemented partially, e.g. between two stairwells (E), or in some cases to the whole length of the building. Frame depth increases enlarge existing habitable rooms or allow the division of one flat into two, and occasionally add new rooms to existing flats.

2.3. CASE STUDIES: FIVE BLOCKS AND FIVE BUILDINGS

From three neighbourhood units that represent ‘neo grid plans’, Matinkylä, Soukka and Kuitinmäki in the city of Espoo, five blocks were selected for case studies. Blocks are composed of multistorey lamella buildings with two or more stairwells. One building from each block, consisting of five room-zone lamellas, was selected for the building and dwelling-scale layout study. All case blocks and buildings represent typical spatial, compositional and structural qualities and include a typical flat distribution described in the following. The selected blocks and buildings are presented in Figure 4. Additionally, aerial photographs of the blocks are provided in File 1 in the supplemental data online.

Figure 4

Five selected case blocks and selected case study buildings.

Two cases (Matinraitti 14 and Avaruuskatu 1) constitute so-called ‘windmill blocks’ (tuulimyllykortteli in Finnish). In them, buildings are positioned at right angles to each other on four sides of the plot, creating a rectangular composition with gaps between all four buildings. After its appearance in an urban renewal competition in 1962 (Hankonen 1994: 183), this block type soon became a common element in Finnish suburban housing estate design. In two case blocks, buildings are grouped to form three dominant rows, with a perpendicular complementing building in one case. Tiistiläntörmä (Figure 4, upper right corner) represents a different block type where buildings are placed in a half-open chain.

The cases indicate how the open form principle of ‘neo grid plans’ has facilitated different block layouts. In gridiron-based Finnish neighbourhood units, most grid cells are typically reserved for housing, while car parking is located to some adjacent cells. Thus, in four of the selected cases, car parking is located outside of the studied block.

In the case study blocks, there are 23 buildings and 55 lamellas in total. Most buildings (n = 18) consist of two lamellas. Of the 55 lamellas, 40 contain three flats and 15 two flats per storey. Stairwells are usually located in the north-western or north-eastern side of the buildings.

In the building and dwelling-scale layout study, all five basic modes of horizontal extension were applied to all selected buildings. The buildings were originally designed between 1968 and 1983. Four of them are made up of two lamellas, and one of four (Figures 5 and 6). They are three or four storeys high and originally without lifts (elevators). Modifying the stairwells to include lifts when constructing horizontal extensions would improve the accessibility of the flats significantly.

Figure 5

Dwelling distribution in the five case study buildings.

Figure 6

Load-bearing structure: positions of wet spaces (blue) and basic dimensions of the five studied buildings.

In Figure 5, the dwelling types are relatively similar in all cases. The lamellas usually have three flats per storey. However, in Soukankaari 11–13, one lamella contains two flats per storey (lower right corner). Typically, the three flats comprise a studio, a one-bedroom flat and a two-bedroom flat, the single-aspect studio being located between the two larger flats. A common variation is that where all three flats are one-bedroom flats (Soukankaari 11–13, second lamella from the left). In total, this variation exists in 10 of the 55 lamellas. Among the case buildings in Avaruuskatu 1, three flats are grouped unusually as the two smallest are positioned in the corners of the building.

In Figure 5, although the studied buildings are relatively similar, the composition of lamellas within one building and the position of wet spaces within flats vary. The lamellas can be attached repetitively so that the flat configuration around the stairwell stays the same (Tiistiläntie 6 and Matinraitti 14), or the lamellas can be mirrored within the building (Avaruuskatu 1 and Yläkartanonkuja 5). While the position of wet spaces in studio flats is similar in all buildings, the larger flats exhibit three alternatives (Figure 6): wet spaces in the middle of the building frame (Avaruuskatu 1 and Tiistiläntie 6) next to stairwells and reaching to the facade (Yläkartanonkuja 5 and Soukankaari 11–13), or between the entrance space and kitchen (Matinraitti 14).

Equal to spatial distribution, buildings’ structural features share many similarities with small variations (Figure 6). In four buildings, the load-bearing structure consists of the so-called cross-wall construction, but the positioning and dimensioning of load-bearing walls varies slightly. In three cases (Matinraitti 14, Yläkartanonkuja 5 and Soukankaari 11–13), the walls between flats, the stairwell walls, the walls between habitable rooms and gable walls are all load-bearing. In Tiistiläntie 6, gable walls, walls between flats and stairwell walls are load-bearing. In the exceptional case Avaruuskatu 1, the load-bearing structure consists of a mix of walls and columns.

3. RESULTS

The block-scale findings of the case studies will be presented first, followed by observations in building and dwelling scales. For the full set of drafted floor plans of the five studied buildings, using all five basic modes of extension, see File 2 in the supplemental data online. Findings in building and dwelling scales will be discussed referring to the full set of floor plans. Selected plans are presented in this section as representative examples, with added graphic features and design details to highlight important observations.

3.1 BLOCK SCALE: APPLICABILITY OF EXTENSIONS AND IMPACTS ON BLOCK MORPHOLOGY

Designs to the five case blocks show how block layouts can be modified with horizontal building extensions and how the basic modes of extension can be applied in varying rectangular building compositions. In their original state, all five blocks are relatively open in character. Hence, one objective was to examine to what extent the blocks can be closed, i.e. how the gaps between buildings can be filled and buildings joined with horizontal extensions. The block scale designs (Figure 7) do not exhibit all possible alternatives but show a variety of examples on how extensions can be applied.

Figure 7

Examples of horizontal building extensions in five case study blocks.

Note: Original buildings are presented in dark grey; their extensions in black.

The designs in two cases at the top of Figure 7 demonstrate that ‘windmill blocks’ can be modified with horizontal extensions in multiple ways, and all basic modes of extension can be applied. Importantly, the designs show that the extensions can generate clear changes to the block form, but full closure is not possible. They also demonstrate that in some situations, different modes can be used in one building, creating alternatives and diversity. In a two-lamella building, for example, a gable extension can be introduced to one lamella and a side wing extension to the other.

The design examples reveal clear limitations to extensions in ‘windmill blocks’. They show that two gaps between buildings can be filled and buildings joined successfully with A+B- or C-type extensions (Figure 7). However, due to stairwell locations, two other gaps remain open because in both cases all stairwells are located in buildings’ north-west or north-east side. This means that in two buildings in both blocks, the stairwells are placed in the outer sides of the block opposite the gaps, and none of the basic modes of extension can be applied to close the gaps opposite stairwells (Figure 8, Avaruuskatu 1). Additionally, plot boundaries are located very close to the same buildings (Figure 7), which makes it unfeasible to implement B–E extensions within the plot area.

Figure 8

Gaps between buildings that cannot be filled with basic modes of extension (red highlight).

Note: Stairwells are shown in black.

The block extension designs in the middle of Figure 7 demonstrate that in both row-formed blocks all basic modes of extension can be utilised, as was the case with ‘windmill blocks’. Another similarity is that the row-formed blocks cannot be fully closed with the basic modes of extension. Despite this, Yläkartanonkuja 3–7 provides many modification alternatives. In this block, three buildings are exceptional as their stairwells are positioned unusually in the south-western side of the building (Figure 4, stairwells marked black) and due to this, some block modifications appear quite organically (Figure 7). Additionally, the boundaries of the three different plots within the block pose only one limitation to extensions, as one gable extension cannot be accomplished in the southernmost building.

Unlike other blocks of this study, Soukankaari 11–13 consists solely of long parallel buildings (Figure 7). Accordingly, there are fewer places for gable, side wing and side wing with deck access extensions (A–C), leading to more frequent employment of middle wing extensions and frame depth increases (D–E). All stairwells are positioned on buildings’ north-western sides. As they determine the place of most extensions, some challenges with window directions occur. However, with B–D extensions, most spaces in the new wings can still be designed with windows facing north-east or south-west. Proposed E-type extensions with windows facing north-west are partially located in places where the original building frame depth is relatively small (inner free dimension of 9.5 m) (Figure 6). Hence, there are relatively good opportunities to design the resulting flat layouts to secure reasonable daylight, although this is subject to more detailed design and may require some special solutions. One example of such a solution is to shape middle wing extensions (D) narrowly (Figure 7, D–E block extension; see also the layout at the bottom of Figure 10).

The designs at the bottom of Figure 7 demonstrate horizontal extendibility alternatives in a block in which buildings are placed in a half-open chain-like manner. This block has fewer extension alternatives than the other case study blocks due to the overall building configuration and original spatial design of some lamellas.

In Tiistiläntörmä 1 and Tiistiläntie 6, the buildings are located relatively close to the boundaries of the two plots. Moreover, three lamellas of three buildings on the south-western sides of the plots were originally designed to contain a small side wing. Hence, these lamellas were considered unextendible in the study. Furthermore, the gap between three buildings in the middle of the block creates a significant challenge to extensions. This gap between three gables (Figure 8) cannot be filled by using solely gable extensions with the maximum length of 4 m each. There are also very limited options to place windows to an extension between the three buildings. The overall configuration of this block, combined with the fact that most of the stairwells are located on the block’s perimeter, leaves most suitable places for extensions between gables of two or three buildings. Accordingly, there are fewer extension options than in other case study blocks.

3.2 BUILDING SCALE: IMPACTS TO FLAT DISTRIBUTION

Alternative floor plans using all basic modes of extension were drafted to all five case study buildings. The results (see File 2 in the supplemental data online) demonstrate that despite certain spatial and structural differences between the studied buildings (Figures 5 and 6), each can be horizontally extended using all basic modes of extension, with several feasible layout variations within each mode.

Horizontal extensions to a lamella building affect and alter the size distribution and number of its flats in varying ways. Figure 9 shows examples of the layouts generated using the basic modes of horizontal extension to Matinraitti 14. With gable extensions (A), original gable flats can be enlarged. A gable extension may add one, two or three new habitable rooms to the original flat (two topmost A versions and A+D). Alternatively, a gable extension with a stairwell renewal may convert an original gable flat to two small flats (lowest A version and A+B; see also Figure 10, Tiistiläntie 6). In Figure 9, small side wing extensions (B) may enlarge original gable flats and larger extensions add a new flat. Examples of new one- to three-bedroom flats added to the original building by side wing extensions are shown. Side wing extensions with deck access (C) typically generate small new flats. Middle wing extensions may add new rooms to original flats or modify some existing and generate additional new flats (A+D and D versions).

Figure 9

Examples of flat types generated by the five basic modes of extension.

Note: Modifications to the original building are marked in red; other colours indicate the original flat types (the layout is in the top-left corner) and new flats as well as modified original flats (other floor plans). Numbers indicate the number of flats per storey.

Figure 10

Examples of the modified original flats and new flats in the five buildings generated by the basic modes of extension.

Note: Modifications to the original buildings are marked red; selected modified original flats and new flats are coloured.

In Figure 9, frame depth increases (E) add space to a lamella building in a different manner than other extensions. In places other than corners of the building, the frame depth increase deepens the existing habitable rooms without increasing their number. The additional floor area gained by most frame depth increases is, therefore, dark space which cannot function as habitable rooms but must be used for secondary domestic purposes or as access area. In the E design examples, the increase in frame depth is partially exploited to form a central corridor between the stairwells. With this intervention, new places for flat entrances were introduced, and at the same time, two original flats in the middle of the building were transformed into four smaller ones, increasing the share of small flats. Another frame depth increase example (Figure 10, Avaruuskatu 1) shows that original two-bedroom mid-frame flats can be subdivided into two studios with a frame depth increase without creating a central corridor.

The design examples in Figure 9 show that horizontal extensions affect the original flats. They also prove that the typical flat distribution of a Finnish lamella building built in the 1960s–80s, i.e. an equal number of studio flats, one-bedroom flats and two-bedroom flats, can be converted to include large dwellings (three- to four-bedroom flats) or solely small flats (studios and one-bedroom flats).

To analyse the impacts of the basic modes of extension to lamella buildings more precisely, one layout from each case study building was selected and drawn with more detail. Selected dwellings that deviate considerably from the original standardised flat types are highlighted (Figure 10).

The buildings in Avaruuskatu 1 contain unusual studio flats located in two corners of the buildings (Figure 5). In Figure 10, a frame depth increase (E) is employed, and the original studio flats are converted into compact two-bedroom flats (56 m2) with an open living–dining–kitchen area. In this extension example, the frame depth increase generates eight relatively small flats (42–56 m2) per storey.

In Tiistiläntie 6 (Figure 10, upper right corner), gable extensions (A) are employed to both gables. In one gable, a two-bedroom flat is converted into a four-bedroom flat with the bedrooms located distinctively in four corners, providing a layout with good privacy in all bedrooms. The new gable wall is windowless, which allows the extension to be connected to another building or extension. In the other gable, the original one-bedroom flat is converted into two small flats. The resulting one-bedroom flats’ and studios’ kitchens are positioned at the gable. This requires windows to the new gable wall and, due to Finnish legislation (Ministry of the Environment 2024: 5 §), a minimum free distance of 8 m to the next building. The elongated studio flat has windows to two directions and hence good daylight conditions.

The design examples of Yläkartanonkuja 5 and Matinraitti 14 (Figure 10) demonstrate the difference between a side wing and a side wing with deck access extensions (B–C). In both examples, the original gable flat is reduced slightly in size. By the side wing extension (B), a new three-bedroom flat with windows to two opposite directions can be introduced. By the side wing with deck access extension (C), two new studios and a new one-bedroom flat are introduced. In this example, the new studio in the end of the extension wing is the only new flat that open to two opposite directions. The gables of the side wings in both examples are windowless, enabling the extensions to be connected to another building or extension.

Narrow middle wing extensions (D) are used in Soukankaari 11–13 to transform two one-bedroom flats into three-bedroom ones and one two-bedroom flat into a four-bedroom one. Like the new three-bedroom flat in Yläkartanonkuja 5 in Figure 10, they have two entrances. This double-access arrangement provides options to use the flats for three-generational living, or in the case of Soukankaari 11–13, to create a formal entrance to the common areas of the flats and a more private one to the private room sections.

The design examples shown in Figures 9 and 10 demonstrate that new flats can be introduced in various sizes, and original flats can be enlarged or shrunk. They also show that the basic modes of extension can be used creatively and adapted to meet with some special features of the buildings in ways that are spatially and functionally advantageous. Moreover, they show that the extensions densify the plots moderately: the examples in Figure 10 provide 22.0–36.5% additional gross floor area to the buildings.

3.3 DWELLING SCALE: IMPACTS ON THE QUALITIES OF INDIVIDUAL FLATS

The above examples show that horizontal extensions inevitably affect the original domestic spaces of lamella buildings. Their impact is, however, strongly dependent on (1) the mode of extension employed and (2) the place or location of the dwelling within the building frame. These aspects will be discussed separately in this section.

3.3.1 Impacts of the extension modes to original habitable rooms

Figure 11 demonstrates how the extensions affect habitable rooms and other spaces of the original building. As stated in Section 2.2, gable extensions (A) do not necessarily require any modifications to stairwells, but all other extensions necessitate stairwell transformations. Equally, gable extensions usually do not block windows from habitable rooms, but other extensions do. Thus, most extensions create a need to transform one or several original habitable rooms into secondary domestic space. Side wing and side wing with deck access extensions (B–C) usually take place in front of two habitable rooms. The original corner room may remain as a habitable room if a new window can be placed to the original gable wall, but the other room must be transformed into secondary space. The design examples presented in the previous section show that this space can often be utilised as wet space for one or two flats. A middle wing extension (D) usually blocks the windows of two original habitable rooms and requires transforming them into secondary spaces. Extension E illustrates how a frame depth increase makes many original rooms several metres deeper, and changes the nature of their floor area from habitable rooms with daylight to dark or indirectly lit space.

Figure 11

Original habitable rooms and other spaces affected by the basic modes of extension.

Note: Affected rooms and spaces are hatched.

3.3.2 Impacts of the extension modes on gable and mid-frame flats

In Figure 12, three basic modes of extension (A–C) are positioned either to a gable of a lamella building or as a continuation of its gable wall, and they necessarily affect the original gable flats next to the extension. A middle wing extension (D) impacts a gable flat when is placed centrally in front of the stairwell and the rooms on both sides of it. A frame depth increase (E) affects gable flats in those cases when applied to the whole length of the building. This means that all basic modes of extension may impact gable flats, but with some narrow versions of D–E extensions they may remain unchanged.

Figure 12

Flats affected by the basic modes of extension.

Again in Figure 12, flats located in the mid-frame and extending through the frame as well as small flats located opposite stairwells are affected differently. Fundamentally, three modes of extension (A–C) do not have any direct impact on them but two modes (D–E) affect flats located in the mid-frame in the usual cases. The small flats opposite stairwells (the flats left unnamed in E), however, usually remain untouched by extensions. Some of them are altered in those rare cases when stairwells are connected by a central corridor (Figure 9).

To compare the impacts of horizontal extensions to gable flats and mid-frame flats more precisely, three flats from different case buildings were selected. A representative layout of each different size-version appearing in the full set of floor plans (see File 2 in the supplemental data online) was then chosen. Finally, the chosen layouts were drawn in more detail. The resulting sets of varying size-versions are presented in the following. Two-bedroom flats serve as examples of the impacts of extensions to gable flats and mid-frame flats. Additionally, a one-bedroom flat of which the very same layout exists in the gable and mid-frame locations is presented to compare the influence of location on a similar flat.

Horizontal extensions may enlarge or shrink gable flats in various ways. The extensions shown in the case studies provide several options on how gable flats’ floor area and number of habitable rooms can be grown or shrunk. The floor area of the two-bedroom flat example presented in Figure 13 is originally 80.5 m2. With B or D extensions, three options to shrink the flat appeared, in the smallest of which the floor area is 58.5 m2 and the flat contains one bedroom. With A, B and D extensions, five options to grow the flat appear, and the largest of them has a floor area of 119.5 m2 and has four bedrooms. Hence, several modes of extension provide various alternatives to modify the size and the number of rooms of a gable flat.

Figure 13

Example of gable unit size-modification options.

A two-bedroom mid-frame flat can also be enlarged or shrunk. However, clearly fewer options are available. The example in Figure 14 shows one alternative where the original flat is subdivided into two small flats, and two examples where the floor area of the flat has grown but the number of bedrooms remained the same. The size-modification alternatives are limited compared with the gable flat.

Figure 14

Example of mid-frame unit size-modification options.

In an example derived from Matinkylä 14 (Figure 15), options for similar flats in different locations within the building frame are compared. There are five options to grow the gable version of the original 56.5 m2 one-bedroom flat (including one where the flat is subdivided), and two options to grow the mid-frame flat. For both gable and mid-frame versions, two options to shrink exist. In the shrunk versions of the mid-frame flat, all windows are located on the same side of the building, whereas shrunk gable flats have windows to two directions with different views. As in the previous comparison of the two-bedroom flats (Figures 13 and 14), this example offers fewer options to size alterations in the mid-frame than in the gable location.

Figure 15

Comparison of size-modification options for the same unit when located in the gable and mid-frame.

4. DISCUSSION

The horizontal extension of block of flats is a little-studied and rarely used approach in urban renewal. This study can be characterised as a pilot. In it, five horizontal extension options to selected case blocks and buildings were studied in the (post-)positivist analytical frame, and the results were evaluated based on their practical value and potential impact.

The study shows that horizontal extensions can change neighbourhood units’ urban morphology, but within certain limits. Using the five basic modes of horizontal extension, suburban estates’ open layout, characterised by an abundance of green spaces and walking routes through blocks, cannot be converted to fully closed urban blocks. In terms of building intensity (floor area ratio), horizontal extension can provide moderate increases, typically well below 50%. This means that to make a renewal project viable, horizontal extension may be a complementary approach to vertical extension, leading together to marked increases in building intensity and economic efficiency. Thus, in urban densification processes, horizontal extensions provide an additional tool for planners and developers to move away from the currently dominant demolition and redevelopment model which is negative both socially and in terms of carbon balance.

Although it is a somewhat limited approach for urban densification, horizontal extensions are a versatile means for modifying flat distribution and diversifying originally highly standardised dwelling types of lamella buildings. The study suggests good potential in both enlarging and shrinking existing flats, especially those in a gable position, as well as in creating a broad array of new and modified dwellings that can address current demographic and cultural needs. The number of flats per lamella can grow from the typical three to four in one storey, and a deck access extension can introduce several more. Accessibility can be improved significantly by installing lifts when renovating stairwells, which is required with all basic modes of extension except gable extensions (Figure 3, A). It can also be improved by renewing or enlarging wet spaces in flats, which proved to be spatially feasible in all case study buildings and with all modes of extension. However, as this study concentrated on layouts and horizontal dimensions of space, the influence of construction details to the accessibility of wet spaces was not considered.

Being able to maintain existing building frames in proper use and update their functionality is crucial for circularity. The layout designs of this study utilising all basic modes of extension (Figure 3, A–E) are done so that the whole floor area within the existing building frame stays fully functional. The modified floor plans also fulfil Finnish fire regulations. Hence, the study shows that horizontal extensions to Finnish blocks of flats built in the 1960–80s have good potential in extending the buildings’ usable life and maintain most of their embodied carbon. Previous research has proved that even without extensions, Finnish lamella buildings of the era can be transformed internally in a versatile way (Huuhka & Saarimaa 2018; Kaasalainen & Huuhka 2020). Taken together, this pilot study expands on those results and confirms that lamella buildings have good value and a large potential for modification.

The aim of the study was to explore the spatial potential for horizontal extensions in block, building and dwelling scales using a research-by-design method. The study provides new typological and morphological knowledge that could not have been obtained with other methods. As the study was design based, broader social and policy issues of urban renewal, such as fighting socio-economic segregation or addressing the financialisation of (social) housing, were not included. Additionally, resident participation and the study of residents’ housing preferences in the case study buildings were not part of this research. Moreover, construction cost calculations or thermal benefit and energy consumption estimates of the different extension solutions were not realised. These aspects open possibilities for further research.

Five spatial and functional principles of horizontal extension (see Section 2.2) laid the foundation for the research-by-design method of this study. These design criteria ensured that all conducted designs were spatially feasible and followed Finnish regulations. The case blocks and buildings are all located in the city of Espoo. However, they represent rather well typical suburban estate planning and design solutions from the late 1960s to the early 1980s realised broadly in Finland. The ‘neo grid plan’ layouts, populated with buildings with short five transverse room-zone lamellas are common across the country, opening the possibility to generalise the results. To learn how generalisable the findings are internationally, further comparative research is needed.

This study recognised that the open blocks formed by relatively short buildings and the use of relatively short lamellas favour horizontal extendibility. While typical in the studied suburban estates, these features were not employed originally for extendibility. The open block forms were used to realise city planning ideals of the time, whereas the short lamellas with five transverse room zones fulfilled the functional and technical aims of rapid construction. Interestingly and surprisingly, they appear as good features for horizontal extendibility today.

5. CONCLUSIONS

A general conclusion of this research-by-design study is that Finnish suburban estates built between the 1960s and 1980s have high potential for horizontal building extensions. Findings on the most essential block and building-level morphological features which affect the possibilities for horizontal extensions are discussed below, followed by observations about how the studied extension modes diversify flat distribution in buildings and qualities of flats.

5.1 MORPHOLOGICAL FEATURES AFFECTING HORIZONTAL EXTENDIBILITY

The five selected cases exhibit three distinct block layouts: ‘windmill blocks’, parallel building rows and open chain-like configurations. In all cases, the individual buildings are relatively short, often consisting of only two lamellas. The most important morphological features affecting horizontal extendibility are:

  • stairwells’ location in relation to the inner part of the block

  • buildings’ distance from plot boundaries

  • the frequency and spatial configuration of gaps between buildings.

The placement of stairwells in block’s buildings is a major determinant of the potential for horizontal extensions. In 20 of the 23 buildings in the case study blocks, stairwells were placed in buildings’ north-western or north-eastern side, leaving the sunny south-eastern and south-western directions for habitable rooms and balconies. This space allocation is beneficial to flat design, but somewhat disadvantageous to horizontal extensions. The basic modes of extension could be best employed if all stairwells were located facing the inner yards of the blocks. In that case, side wings, middle wings and frame depth increases (Figure 3, B–E extensions) could be placed there, and the blocks would gain density by growing inwards. In cases where stairwells are located towards block’s perimeter, B–E extensions need to be placed on the outer edge, thus increasing blocks’ footprints.

In ‘neo grid plans’ (discussed in Section 1.2), buildings follow a rectangular pattern, but their location vis-à-vis plot boundaries may vary substantially. In this study, extensions over the plot boundary were not considered possible. Thus, in cases where extensions need to be placed on blocks’ outer side, buildings’ distance from the plot boundary may pose limitations or even exclude implementing extensions. This is especially typical in cases where stairwells face the plot border.

Gaps between two or three buildings are typical features of the semi-open residential blocks of the 1960s–80s. The gaps between two gables or one gable and one stairwell-side of a building are viable for the basic modes of extension. However, gaps between three buildings are challenging. As those gaps are between three gables, only gable extensions (Figure 3, A) can be used. Furthermore, dwelling design is hampered by limited options to place windows as many directions are blocked by existing buildings.

5.2 DIVERSIFICATION OF FLAT DISTRIBUTION AND QUALITIES OF FLATS

The case studies show that by employing the basic modes of horizontal extension (gable, side wing, side wing with deck access, middle wing extensions and frame depth increase) (Figure 3), the typical flat distribution of lamella buildings with five room zones can be altered in several ways. The most important building- and flat-level observations about horizontal extendibility are:

  • the importance of a flat’s location in the building frame

  • the possibility to influence both flat distribution and dwelling typology in a concerted manner

  • the relative ease to introduce lifts (elevators) as part of horizontal extension.

Alternatives to modify existing flats with basic modes of horizontal extension vary considerably based on their location within the building frame. There are more options to convert gable flats than mid-frame flats, and only few options exist to convert the original small flats opposite stairwells. This is a key finding of this study.

Existing gable flats and mid-frame flats can be enlarged and the building’s flat distribution converted to include mainly large flats (three- to four-bedroom flats). Equally, existing gable flats and mid-frame flats can be shrunk or subdivided to convert a building to include solely studios and one-bedroom flats. Additionally, new flats of various sizes can be introduced with extensions. Besides diversifying flat sizes, flat types can be diversified with horizontal extensions, too. This can be achieved by converting existing flats or constructing new flats to side or middle wing extensions. Case studies provide several examples of converted and new three- or four-bedroom flats with two entrances which can be used for three-generational living or partially as home offices.

Lifts can easily be fitted while conducting horizontal extensions. The case studies proved that converting stairwells of lamella buildings to include new lift cores in the extensions can be achieved without major spatial difficulties or structural barriers. Adding lifts markedly improves the accessibility of the flats and provides a good foundation to extend the useful life of these buildings.

Notes

[1] Lähiöohjelma 2008–22; for more in Finnish, see https://ym.fi/lahioiden-kehittaminen/.

[2] Muuttuva käsitys lähiöstä tiivistyvässä kaupungissa and Holamallit. In English: ‘The changing concept of neighbourhood unit in a densifying city, https://projects.tuni.fi/muuttuvalahio/esittely/in-english/ and ‘Horizontal extension of suburban blocks of flats’.

[3] According to Creswell (2009), the (post)positivist worldview represents the traditional form of scientific research, focusing on empirical observation and theory verification. While constructivism refers to research that aims to understand participants’ multiple meanings in their social context, the advocacy/participatory approach aims for their empowerment and change. Pragmatism, for Creswell, is a practice-oriented methodological choice, where research interest is in situations and real-life consequences.

AUTHOR CONTRIBUTIONS

Both authors contributed to the conception and writing of the article. J.T. conducted the data collection and undertook the research-by-design.

COMPETING INTERESTS

The authors have no competing interests to declare.

DATA ACCESSIBILITY

The data supporting this study are available in the supplemental data online. Additional data are available from the corresponding author upon reasonable request.

ETHICAL APPROVAL

According to the guidelines of the Ethics Committee of Tampere University, the used research protocol does not require an ethical review.

SUPPLEMENTAL DATA

Two supplemental data files can be accessed at: https://doi.org/10.5334/bc.783.s1

File 1: Aerial images of the case study blocks.

File 2: Floor plans of horizontal extensions drafted to case study buildings.

DOI: https://doi.org/10.5334/bc.783 | Journal eISSN: 2632-6655
Language: English
Submitted on: Jan 19, 2026
|
Accepted on: Mar 16, 2026
|
Published on: Mar 31, 2026
Published by: Ubiquity Press
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2026 Jyrki Tarpio, Panu Lehtovuori, published by Ubiquity Press
This work is licensed under the Creative Commons Attribution 4.0 License.