Leftover Under the Deck: Why Indian Undercrofts Stay Lost Space Until We Score Them
Elevated roads and metros keep arriving as throughput machines. The ground beneath them is still treated as leftover right-of-way—pier bays that become dumps, parking lots or fenced pockets. Drawing on Trancik’s lost-space theory, Kuala Lumpur and Indian undercroft studies, and programmed precedents from Mumbai and Delhi, this article argues that undercrofts become shaded civic rooms only when appraisals score continuity, light and safety, programme fit, soft edges and stewardship—not beautification alone.
Sujeet S. Thakare
Urban Design & Planning
Indian cities have spent a generation lifting traffic into the air. Flyovers, elevated metros and interchange ramps relieve junctions on paper. On the ground they leave a second city: the undercroft—the linear residual volume between piers, under decks and beside ramps. Walk beneath almost any urban flyover and the pattern repeats. Construction debris. Charged parking. Blank grills. A pocket garden that stops at a fence. After dark, fear. The deck above is scored for vehicles per hour. The room below is rarely scored at all.
This article treats that residual volume as Roger Trancik’s lost space—leftover urban land that fails figure-ground continuity, linkage and place (Trancik, 1986)—and asks a design question distinct from EDGE-5’s concern with crossing a hard linear barrier. EDGE-5 diagnosed transversal severance. UNDER-5 diagnoses the undercroft as public realm: whether pier bays can become continuous, lit, programmed civic rooms rather than dumps and car yards. The argument is organised as LOST-5: continuity (L1), light and perceived safety (L2), programme fit to adjacent land use (L3), soft edge and frontage (L4), and stewardship (L5).
The sections that follow summarise the research base, note method and limits, analyse Indian undercrofts through LOST-5 with six original drawings, discuss transfer limits from overseas cases, and close with implications for municipal and urban-design appraisal.
What the research says
Trancik’s Finding Lost Space remains the clearest theoretical frame for leftover ground around object infrastructure. Lost spaces are unused or poorly defined urban areas that undermine the continuity of figure and ground, break pedestrian linkage, and fail to become places people can claim (Trancik, 1986). Elevated decks and their pier fields are textbook generators of such residuals: the structure is an object; the undercroft is the leftover.
Kevin Lynch’s image elements help explain why the undercroft matters to memory as well as use. Edges and paths organise how people orient themselves; a long shaded strip under a flyover can become either a memorable path with soft edges or a dead zone that people avoid after dusk (Lynch, 1960). Jane Jacobs’s account of border vacuums—edges that dampen street life beside large single-use or infrastructural breaks—is a useful companion where undercrofts sterilise the adjacent frontage instead of feeding it with eyes on the street (Jacobs, 1961).
Comparative research on elevated highways and flyovers has been denser in Southeast Asia than in most Indian municipal manuals. Anuar and Ahmad’s observational study of interstitial spaces under the DUKE and AKLEH elevated highways in Greater Kuala Lumpur reports wasted land beneath structures that otherwise improve regional connectivity; the undercroft typologies they map tend to be small, irregular, enclosed, poorly lit and fragmented (Anuar & Ahmad, 2018). Qamaruz-Zaman, Samadi and Azhari document user-centred activities that already occupy leftover space under Kuala Lumpur flyovers—evidence that informal appropriation precedes formal programming (Qamaruz-Zaman et al., 2013). In the infrastructural-urbanism literature, Franck frames the edge and the underneath as “other” urban public spaces that cities can occupy rather than ignore (Franck, 2011). Ignasi de Solà-Morales’s terrain vague essay, often cited in this lineage, describes residual and indeterminate urban ground as culturally charged emptiness rather than mere vacancy—useful as metaphor, not as an Indian statutory category (Solà-Morales, 1995).
Indian and India-facing work is catching up. Manibala and Shanmuga Priya review place-making literature and eight international and Indian cases—including Matunga’s under-flyover garden in Mumbai, Underpass Park and the Bentway in Toronto, and Kathipara in Chennai—then frame landscape guidelines for voids beneath flyovers and apply them to a Padi flyover pilot in Chennai. Their issues table lists access barriers, traffic surround, water seepage, exposed services, illegal parking, rubbish dumps and criminal-activity risk; their activity matrix matches surrounding land use to programme (markets and seating near commercial fabric; play and walking tracks near residential and mixed edges) (Manibala & Shanmuga Priya, 2022). Longani’s observational study of six Pune flyovers reinforces the same logic: surrounding land use strongly shapes what the undercroft will support—industrial edges tilting toward bus and employee parking, commercial edges toward markets, residential edges toward parks—while access, location relative to pedestrian flow, noise, scale and luminance decide whether a designed space is used or abandoned (Longani, 2025).
Outside peer-reviewed journals, practice toolkits and built projects supply transferable craft. The Design Trust for Public Space and New York City Department of Transportation El-Space Toolkit treats residual space under elevated transit as a reclaimable public layer, with particular attention to lighting quality, green infrastructure and surface finish rather than fixture counts alone (Design Trust for Public Space & New York City Department of Transportation, n.d.). In Mumbai, StudioPOD with MVRDV’s One Green Mile / Parel Baug converts roughly 200 metres under the Senapati Bapat Marg flyover into a sequence of public “rooms”—lounge, gym, shaded seating, performance and reading—within a wider 1.8 km streetscape that adds walk, cycle, planting and lighting (MVRDV, n.d.; StudioPOD, n.d.). In Delhi, Creators Architects’ Neighbourhood Cultural Centre below Raja Garden Flyover treats the undercroft as a linear civic spine with amphitheatre seating, exhibition walkways, tensile capsules, porous perimeter security and zoned lighting—an early Indian attempt to programme residual mobility land at neighbourhood scale (Creators Architects, n.d.).
Read together, the literature does not say that every pier bay should become a park. It says that treating the undercroft as leftover ROW guarantees lost space; that programme must fit adjacent use; and that light, continuity and maintenance decide whether “activation” survives the first monsoon.
Method and sources
This article is a synthesis of published urban-design theory, peer-reviewed and conference studies on undercroft and interstitial space, publicly available project documentation, and original analytical drawings based on OpenStreetMap extracts (2026) for a dense Indian arterial corridor with elevated road geometry—Lower Parel / Senapati Bapat Marg and a comparable intact-fabric window. No new household survey or illuminance meter campaign was conducted. Overseas findings are treated as transferable hypotheses, not as Indian statutes. Project pages are cited as design documentation, not as evaluated post-occupancy evidence.
Analysis: LOST-5 as a field diagnostic
Score each axis Strong / Partial / Weak on a one-page sheet before pier grids, fence lines and “beautification” packages freeze. A corridor that moves vehicles well while leaving a dark, discontinuous dump under the deck is not a complete public-realm outcome.
L1 Continuity — can people walk through the pier field?
Lost space, in Trancik’s linkage sense, begins when the undercroft is chopped into fenced cul-de-sac pockets that never connect longitudinally along the deck or laterally to the footpath and adjacent streets (Trancik, 1986). Manibala and Shanmuga Priya’s guidelines emphasise defined entrances, zebra crossings into the void, and pedestrian routes of roughly two metres that reconnect areas across the residual strip (Manibala & Shanmuga Priya, 2022). Anuar and Ahmad’s KL observations show how elevated structures can fragment interstitial ground into leftover cells that do not read as a continuous public layer (Anuar & Ahmad, 2018).
Figure 1 contrasts continuous urban fabric with the residual strip that forms under an elevated arterial: the solids remain dense, yet a linear lost band appears where the deck shadows the ground. Figure 3 zooms to the pier-bay module: continuous pedestrian paths through bays versus fenced pocket gardens that stop the walk. Continuity is not an aesthetic preference; it is the difference between a shaded civic corridor and a series of dead ends.
Source: Author's drawing based on OpenStreetMap data (© OpenStreetMap contributors, ODbL), 2026.
Source: Author's drawing, 2026.
L2 Light and perceived safety — can people see, and be seen, after dark?
Undercrofts fail when daylight is inadequate at the centre of the span, artificial light is absent or broken, sight lines are blocked by parked trucks and solid grills, and active edges disappear after shop hours. Manibala and Shanmuga Priya list safety—bollards, a slight rise from carriageway, layered planting buffers, and controlled perimeter—as a physical design factor precisely because dump yards and dark corners invite misuse (Manibala & Shanmuga Priya, 2022). Longani’s Pune observations link luminance and scale to perceived safety: well-lit bays attract use; dark corners at flyover ends attract misuse (Longani, 2025). The El-Space Toolkit’s emphasis on illuminating structure, surfaces and pedestrian routes—not only mounting a few fixtures—is a useful craft transfer from New York’s elevated lines, with the obvious caveat that Indian maintenance regimes differ (Design Trust for Public Space & New York City Department of Transportation, n.d.).
Figure 4 pairs a blank-grill, poorly lit bay with a soft-edge bay that keeps eyes on the undercroft through planting, seating and open sight lines. L2 is where Jacobs’s eyes-on-the-street argument meets infrastructural shade (Jacobs, 1961).
Source: Author's indicative section, 2026.
L3 Programme fit — does the use match what surrounds the flyover?
This is the axis Indian studies make hardest to ignore. Manibala and Shanmuga Priya’s activity matrix ties surrounding land use to programme: shaded seating, book stalls and play near commercial–residential edges; informal markets and bike repair near commercial–industrial edges; multi-functional space where industrial and residential meet (Manibala & Shanmuga Priya, 2022). Longani finds the same coupling in Pune: industrial surroundings support bus termini and employee parking; commercial surroundings support markets; residential surroundings support parks—and a park designed without reading its surroundings can sit unused while a market thrives a few kilometres away (Longani, 2025). Qamaruz-Zaman and colleagues show that informal user activities already invent programme under KL decks; formal design that ignores those attractors fights the grain of the city (Qamaruz-Zaman et al., 2013).
One Green Mile’s Parel Baug rooms and Raja Garden’s cultural spine are not universal templates; they are examples of programme that answers a dense mixed corridor rather than a generic “green” sticker (MVRDV, n.d.; Creators Architects, n.d.). L3 fails when every undercroft is given the same lawn and the same fence, regardless of who walks past.
L4 Soft edge and frontage — what does the bay present to the street?
A soft edge—planted buffers, seating niches, active frontage, porous screens—turns the pier bay into a room people can occupy. A hard edge—bumper-to-bumper parking, debris piles, blank metal grills—turns it into a border vacuum beside the footpath (Jacobs, 1961; Trancik, 1986). Manibala and Shanmuga Priya’s visual guidelines call for layered native planting as noise and pollution buffer, treatment of columns and service runs, and bioswales at the periphery (Manibala & Shanmuga Priya, 2022). Raja Garden’s project documentation describes landscape buffers, porous perimeter security and material activation that break the monotony of the underside (Creators Architects, n.d.). Figure 2 holds a shared ground datum and compares a dead undercroft (parking and dump) with a programmed shaded room—poche, people, planting—under the same indicative deck.
Source: Author's indicative section, 2026.
L5 Stewardship — who owns the lights after inauguration day?
Undercroft projects fail quietly when jurisdiction is unclear among road agency, municipal corporation, traffic police and private sponsor; when lighting and cleaning have no operating budget; and when “beautify then abandon” leaves a photo opportunity and a locked gate six months later. Longani notes spaces that are designed yet feel unsafe during construction or remain unused because access and maintenance are unresolved (Longani, 2025). Manibala and Shanmuga Priya’s Kathipara case entry records maintenance failure, dump conditions and illegal activity before redesign narratives (Manibala & Shanmuga Priya, 2022). L5 is the axis that converts L1–L4 from drawing-board claims into a public asset. Without it, LOST-5 collapses back into lost space.
Figure 5 sets LOST-5 in an open field—no UI cards—so a corridor team can score the five verbs against the pier spine. Figure 6 sequences the familiar Indian trajectory: dump → parking yard → programmed civic undercroft. The sequence is not inevitable progress; it is a policy choice about what the appraisal sheet values.
Source: Author's diagram, 2026.
Source: Author's small-multiples drawing, 2026.
Discussion: limits of transfer
Three hedges are required. First, Kuala Lumpur highway studies and New York el-space toolkits emerge from different property regimes, lighting standards and informal-economy bargains than a typical Indian municipal flyover. Their value is diagnostic vocabulary—interstitial typology, lighting craft, green infrastructure—not copy-paste sections (Anuar & Ahmad, 2018; Design Trust for Public Space & New York City Department of Transportation, n.d.). Second, Conference and student papers such as Manibala and Shanmuga Priya (2022) and Longani (2025) are useful for Indian pattern recognition and guideline logic; they are not multi-city randomised evaluations. Third, firm project pages for One Green Mile and Raja Garden document design intent and programme; they are not independent post-occupancy evaluations of equity, maintenance cost or night-time safety for every user group (MVRDV, n.d.; StudioPOD, n.d.; Creators Architects, n.d.).
EDGE-5 and UNDER-5 can also be confused. A foot-over-bridge that stitches two sides of a railway answers severance. It does not automatically redeem the residual volume under an elevated road. Conversely, a beautiful undercroft park that remains a fenced cul-de-sac may score on L3 aesthetics while failing L1 continuity and L5 stewardship. The two diagnostics should sit on the same corridor sheet: one for crossing the edge, one for occupying the underneath.
Conclusion / Implications for practice
Indian elevated infrastructure will keep arriving. The question is whether the ground it creates stays leftover ROW or becomes a shaded civic layer. This article has argued that undercrofts remain Trancik’s lost space until appraisals score continuity through pier bays, light and perceived safety, programme fit to adjacent land use, soft edges rather than parking and debris, and stewardship that outlasts the inauguration photograph.
For practice, a one-page LOST-5 scorecard belongs beside the traffic chapter on flyover, elevated metro and interchange packages. Corridor teams should publish undercroft continuity plans and lighting/ops agreements with the same seriousness as lane configurations. Programme should be read from surrounding land use—markets where commercial edges already pull people; play and seating where residential fabric needs shade—rather than from a single beautification kit. Soft edges should replace blank grills where safety allows porosity. And stewardship—agency, budget, night lighting—should be named before the first planter is installed.
The deck can move vehicles. The room beneath can still be a public room. Lost space is a design and governance failure, not an inevitable shadow of concrete.
Notes
Data. Figure-ground drawings use OpenStreetMap building and highway extracts (© OpenStreetMap contributors, ODbL), redrawn by the author, 2026. Sections and framework diagrams are indicative analytical drawings, not surveyed construction documents.
Interests. No competing interests.
Series. Continues the morphology series after EDGE-5 / SEVER-5. UNDER-5 addresses residual undercroft as public realm; it does not replace EDGE-5’s transversal severance diagnostic.
References
Anuar, M. I. N. M., & Ahmad, R. (2018). Exploring possible usage for elevated highway interstitial spaces: A case study of DUKE and AKLEH, Kuala Lumpur. Planning Malaysia, 16(3), 208–220. https://doi.org/10.21837/pmjournal.v16.i7.512
Creators Architects. (n.d.). Neighbourhood cultural centre below Raja Garden Flyover. https://creatorsarchitects.com/projects/neighbourhood-cultural-centre-below-raja-garden-flyover/
Design Trust for Public Space & New York City Department of Transportation. (n.d.). El-Space Toolkit. https://designtrust-prod.s3.amazonaws.com/media/files/El-Space_Toolkit.pdf
Franck, K. A. (2011). Occupying the edge and the underneath—“other” urban public spaces. In T. Hauck, R. Keller, & V. Kleinekort (Eds.), Infrastructural urbanism: Addressing the in-between (pp. 117–129). DOM Publishers.
Jacobs, J. (1961). The death and life of great American cities. Random House.
Longani, S. M. (2025). Examining the influence of surrounding on the use of space under the flyover. International Journal of Engineering Research & Technology, 13(4). https://www.ijert.org/examining-the-influence-of-surrounding-on-the-use-of-space-under-the-flyover
Lynch, K. (1960). The image of the city. MIT Press.
Manibala, R. T., & Shanmuga Priya, G. (2022). Effective use of urban voids environment and infrastructure. International Journal of Engineering Research & Technology, NCA-2022, 10(3). https://www.ijert.org/effective-use-of-urban-voids-environment-and-infrastructure
MVRDV. (n.d.). One Green Mile. https://mvrdv.com/projects/787/mumbai-green-mile
Qamaruz-Zaman, N., Samadi, Z., & Azhari, N. F. N. (2013). Under the flyovers of Kuala Lumpur: User centred activities in leftover spaces. Journal of Asian Behavioural Studies, 3(10), 89–103.
Solà-Morales, I. (1995). Terrain vague. In C. C. Davidson (Ed.), Anyplace (pp. 118–123). MIT Press.
StudioPOD. (n.d.). One Green Mile: Under flyover park — Parel Baug. https://studiopoddesign.com/portfolio-item/one-green-mile-underflyover/
Trancik, R. (1986). Finding lost space: Theories of urban design. Van Nostrand Reinhold.
Comments
Post a Comment