Architecture as an ecosystem: A ten-year review of biodiversity integration in a building

Abstract

As cities densify, integrating biodiversity into architecture has become a growing concern for design and ecological practice. Yet few buildings place biodiversity at the core of conception rather than treating it as an environmental add-on, and even fewer are evaluated ecologically over extended periods. The School of Sciences and Biodiversity in Boulogne-Billancourt, France, combines an intensive green roof conceived as a mosaic of habitats for local species, inspired by the woodlands of île-de-France, with a bioreceptive façade intended for spontaneous colonization with minimal maintenance. This article presents a ten-year evaluation of the project, combining species surveys with post-occupancy observations of use and maintenance. It examines how ecological, design, and user data collected over time can be interpreted to extract adaptive management strategies within architectural practice. Over the decade, 345 species were recorded, including 207 plants, 100 invertebrates, 20 birds and a few reptiles and bats, at levels comparable to nearby parks. The results show that ecological persistence on buildings depends strongly on design choices, user practices, and maintenance. Conflicts between access and conservation required spatial reorganization. The study offers practical insights for future ecosystemic buildings, emphasizing habitat heterogeneity, long-term monitoring, and adaptive management as key design and operational principles.

Graphical abstract

Keywords

Green buildings / Biodiversity / Green roofs / Ecosystemic architecture / Urban ecology / Bioreceptive facade / Post-occupancy evaluation

1 Introduction

In contemporary architecture, ecological concerns have primarily been addressed through strategies aimed at reducing energy consumption, minimizing carbon footprints, and incorporating bio-based materials (Vale and Vale, 1991; Steele, 2005). This orientation has led to the widespread adoption of passive design principles, low-carbon construction methods, and sustainability certifications. In parallel, green architecture―including green roofs and green walls (Dunnett and Kingsbury, 2004), also referred to as vertical greening systems (Medl et al., 2017)―has gained prominence for its ability to provide ecosystem services such as thermal regulation, stormwater management, and urban heat-island mitigation (Alexandri and Jones, 2008; Charoenkit and Yiemwattana, 2016; Bustami et al., 2018; Auer et al., 2019; Barriuso and Urbano, 2021), while contributing to more biophilic cities (Beatley, 2011; Wilson, 2017).

Although these approaches have transformed architectural and urban design, they have largely prioritized energy and environmental performance over ecological complexity. Only recently have research and design practices begun to explore how buildings can actively host and sustain biodiversity, positioning architecture as part of urban ecological networks (Francis, 2011; Mayrand et al., 2018; Mayrand and Clergeau, 2018; Chayaamor-Heil and Vitalis, 2021). In the context of accelerating biodiversity loss (IPBES et al., 2019), caused by rapid urbanization (McKinney, 2002; Kondratyeva et al., 2020), buildings have the potential to provide analogue habitats (Lundholm, 2006; Francis and Lorimer, 2011; Werner and Kelcey, 2017) for local flora and fauna, contributing to ecological continuity in dense cities.

Extensive ecological research has documented the biodiversity performance of green roofs and, increasingly, living walls. These studies quantify species richness, colonization dynamics, and ecosystem services, demonstrating the ecological value of vegetated systems (Barra and Johan, 2021; Muratet et al., 2024). In parallel, material scientists and ecological engineers have advanced the concept of bioreceptivity―the capacity of construction materials to support spontaneous colonization without intensive maintenance (Guillitte, 1995; Jim and Chen, 2011).

Together, these bodies of work have expanded architecture’s ecological potential. Yet within architectural research, biodiversity is still most often introduced as an add-on, through standardized systems or prefabricated habitats integrated after design completion. Such strategies rarely influence the building’s morphology, material articulation, or use, leading to what has been described as blandscaping (McKinney, 2006; Fortel et al., 2016; Connop, 2018)―a simplification of urban ecological design that favors ornamental uniformity over spontaneous, site-specific processes.

While the biodiversity of individual systems has been widely studied, fewer inquiries consider buildings as integrated ecological assemblages that host multiple and interacting habitats. There remains limited understanding of how these habitats evolve together over time, how their trajectories relate to design intentions and maintenance, and how their observation can inform future projects. Within ecological or architectural research, there remains a notable lack of long-term studies on intensive green roofs conceived as low-maintenance habitats composed of native species. Moreover, longitudinal, building-scale studies that combine ecological data and post-occupancy observations are rare. This gap limits our ability to translate ecological monitoring into adaptive design and management strategies for real projects.

Addressing this gap, the present study focuses on a building conceived from the outset as a fragment of regional woodland and cliff habitat―an evolving ecological system monitored over a decade by its original interdisciplinary design and research team. It addresses two complementary questions: (1) how does biodiversity and habitat composition evolve over time on such a building, within a practice-based research context; and (2) how can the knowledge produced through this long-term observation contribute to developing adaptive management approaches for ecosystemic architecture?

The term ecosystemic architecture refers here to a design approach that conceives a building as a living assemblage of habitats―one that interacts with ecological processes, supports spontaneous colonization, and evolves through time. By analyzing a decade of monitoring, the research seeks to understand how architectural and ecological decisions influence the coexistence of human and non-human life within a single project.

The case study is the School of Sciences and Biodiversity in Boulogne-Billancourt, France, located in the dense ZAC du Trapèze eco-district―a former industrial site redeveloped into a mixed-use neighborhood (Fig. 1). Completed in 2014, the project integrates two complementary strategies: a green roof planted with native species inspired by regional woodland communities of île-de-France, and bioreceptive façades designed to encourage spontaneous colonization with minimal maintenance. Together, these systems create a mosaic of habitats that transform the building into a dynamic ecological environment embedded in the urban fabric.

The research is led by an interdisciplinary design and research team composed of architects and an ecologist who collaboratively conceived the project and have followed its evolution voluntarily for over a decade. This sustained observation has produced a unique dataset that combines naturalist surveys, maintenance and management records detailing interventions and adaptive strategies, and post-occupancy audits describing how the building’s users experience and interact with its ecological systems.

The study adopts a mixed empirical and practice-based framework, integrating ecological monitoring with operational and user observations. Ecological data provide evidence of species colonization and persistence, while management records and user audits reveal how ecological processes and building use have evolved together. This combined approach bridges scientific monitoring and architectural practice, offering insight into how long-term observation can guide adaptive ecological management and future design strategies.

By considering the building as an integrated ecosystem, the research highlights how architecture can participate in ecological processes and support biodiversity within urban contexts. The decade-long monitoring of the project offers rare insights into how living systems establish and persist within an architectural framework, revealing the intertwined roles of materiality, design, and adaptive maintenance in sustaining biodiversity. More broadly, this study contributes to an expanding field that redefines architecture as a dynamic component of urban ecosystems (Riley et al., 2019; Canepa et al., 2022), capable of coevolving with its ecological and human inhabitants through mutual ecological, practical, and psychological benefits (Fuller et al., 2007) and aligning with contemporary nature-based solutions frameworks (Raymond et al., 2017).

The paper is structured as follows: Section 1 presents the materials and methods, the design and ecological concept, interdisciplinary framework, datasets, and methods. Section 2 summarizes the ecological and operational results, showing how different habitats evolved over time. Section 3 discusses their implications for adaptive design and management in architecture. Section 4 concludes with broader reflections on the potential of ecosystemic approaches to support biodiversity in dense urban environments.

2 Materials and methods

2.1 Design and development

2.1.1 Architectural and ecological concept

The School of Sciences and Biodiversity was conceived as a living building that integrates ecological processes within its architecture. The project combines a preschool, an elementary school with eighteen classrooms, and a gymnasium within a 4000 m² site. From the outset, the interdisciplinary team of architects and ecologists designed the building as a biodiverse and functional system mimicking and integrating multiple ecological habitats (Blanco et al., 2021). The goal was to create an architecture that would evolve through time with minimal maintenance while serving as an educational landscape.

The building’s volumetric composition―terraces, roofs, and circulation routes―links interior programs to their ecological surroundings (Fig. 2). The concrete envelope, conceived as a thick and uneven crust, provides both structural enclosure and habitat potential. By varying the inclination, depth, and exposure of the façade elements, the architects created a mosaic of microhabitats favorable to spontaneous colonization. Irrigation of the green roof was limited to the first two years, allowing autonomous diversification of plant communities. Early post-occupancy observations revealed strong educational engagement but also soil compaction and plant removal, leading to the first informal follow-ups.

2.1.1.1 Roof and terrace.

The rooftop and terraces form the largest planted areas, connected to the ground via a grass ramp that facilitates fauna movement. These spaces include designated gardens for pedagogical activities, allowing children to engage directly with nature or observe seasonal changes, integrating environmental interaction into education.

Imagined in collaboration with ecologists,^† the rooftop drew inspiration from the oak-hornbeam forest typical of the Greater Paris Region. Adaptations for the rooftop context included excluding oak trees to avoid root system issues, and dividing the 2000 m² roof into three zones based on soil thickness. These variations support species growth while controlling tree height and protecting the roof structure, creating conditions that replicate local forest ecosystems.

2.1.1.2 Wall.

The concrete façade serves as a mineral envelope linking the school’s programs, green spaces, and biodiversity. Inspired by traditional stone walls, its design incorporates gaps and interstices within and in-between the elements, creating niches for biodiversity (Fig. 3). Concrete was chosen for its similarity to stone in texture and durability, as it erodes on the surface while maintaining structural integrity, promoting vegetation growth over time with a decrease of its pH with the influence weathering (Sanmartín et al., 2021). Its malleability also allowed for the integration of birdhouses, grooves, and drainage systems through the casting of the blocks.

2.1.2 The bioreceptive façade

The bioreceptive façade was designed as a complete bio-receptive envelope. Its lower part is completely vertical on the first 3 m. Then, the wall angles inward and offers even more different situations and thicknesses, notably profound cavities, which are favorable to the largest bird nests (Hostetler and Holling, 2004).

The façade comprises 1436 blocks, including 1039 standard and 397 special blocks, dimensioned to meet fabrication and construction constraints. The wall’s design incorporates varying thicknesses, cavities, and notches, providing diverse microhabitats (Jim and Chen, 2011) for targeted species such as pipistrelles, chickadees, and swallows (Gunnell et al., 2012, 2013). Furthermore, diverse situations such as exposed or shadow zones, with more or less humidity, emerge from these blocks’ combinations and provide a series of niches fulfilling the needs of different species of plants, insects and small animals such as wild bees (Fortel et al., 2016). Upper grooves encourage pioneer vegetation development through the gradual accumulation of micro-soils over years, carried by birds and winds. This façade reinterprets the ecological and morphological features of old stone walls (Jim and Chen, 2011), creating a dynamic, evolving structure that supports biodiversity and integrates technical and ecological considerations.

The façade generates diverse microclimatic conditions that further support ecological differentiation. Variations in orientation, inclination, and cavity depth create gradients of sun exposure, temperature, and humidity, offering niches suitable for both xerophilous and shade-tolerant species. The thermal inertia of the concrete moderates daily temperature fluctuations, while capillary water retention within cavities maintains localized humidity favorable to bryophytes, lichens, and invertebrates. These microclimatic contrasts―between south-facing, sun-exposed surfaces and cooler, shaded north-facing―enhance the façade’s overall bioreceptivity and resilience, encouraging a mosaic of colonization dynamics over time.

To support diverse taxa, the façade integrates specific design features.

● Birdhouses, embedded within blocks, are dimensioned and oriented according to the nesting preferences of local species such as pipistrelles, chickadees, and swallows (Gunnell et al., 2012, 2013). Their integration influences the façade’s morphology, creating a dynamic pattern of thicknesses.

● Micro-cavities (3—15 mm in diameter, 10 cm deep) drilled at downward angles prevent water stagnation and serve as nesting sites for solitary bees, damselflies, and small arachnids (Fortel et al., 2016).

● Built-in planters (47 cm × 16 cm × 6 cm), distributed across orientations, host native plant species adapted to varying exposure and humidity.

● Grooves and vertical cables direct rainwater from the roof to the ground, providing passive irrigation and ensuring ecological continuity along the façade.

Together, the rooftop and façade form an interconnected gradient of microhabitats designed to evolve over time with minimal intervention (Fig. 4).

2.1.3 The green roof ecosystem

The 2000 m² rooftop ecosystem, developed in collaboration with ecologists, reinterprets the oak—hornbeam forest typical of the île-de-France region. It is an intensive green roof, structured in three strata of soil depth (Fig. 5).

● Forest Core (550 m², 1 m depth)―Dense planting of 220 trees (hornbeams, lindens, maples) forms a closed canopy habitat for birds and insects.

● Shrub and Woodland Ring (500 m², 1 m depth)―Intermediate vegetation of 18 shrub species, including hazel and berry-bearing plants (Rubus fruticosus, Ribes), supports pollinators and small mammals.

● Mesophilic Prairie (850 m², 0.5 m depth)―A mix of perennials, grasses, and legumes provides nectar and ground cover.

A 1100 m² ramp (0.3 m depth) connects the ground and roof, improving ecological connectivity for small fauna. A 100 m² pedagogical garden enables outdoor learning while respecting the roof’s ecological processes. Maintenance was intentionally limited to the first two years; afterward, plant communities were allowed to adapt naturally to local conditions.

2.2 Data collection methods

Data collection combined ecological surveys, post-occupancy audits, with ecological engineering and design-led interventions conducted over a ten-year period. Figure 6 synthesizes the chronology of monitoring phases, survey protocols, and adaptive management actions carried out between 2014 and 2023.

2.2.1 GROOVES baseline dataset (2016—2019)

The School of Sciences and Biodiversity was part of the GROOVES (Green ROOfs Verified Ecosystem Services) program led by the Agence Régionale de la Biodiversité d’île-de-France (Barra and Johan, 2021). This regional study assessed the ecological value of green roofs through standardized multi-year field protocols. Although conducted independently from the design—research team, it provides a critical ecological baseline preceding later monitoring phases.

Between 2016 and 2019, floristic and faunal surveys were carried out following Inventaire National du Patrimoine Naturel (INPN, 2018) standards (Fig. 6). Floristic inventories (2016—2019) combined free surveys and Florilège-Prairie quadrats across the entire roof, bases, and selected façades, yielding 525 records of 186 species. Faunal surveys (2017—2019) targeted terrestrial invertebrates and pollinators using sweep-net transects and aspirator sampling, complemented by SPIPOLL citizen-science data, resulting in 146 database records, representing 107 taxa dominated by Hymenoptera, Diptera, and Coleoptera.

These data document the roof’s early ecological establishment, recording pioneer flora and invertebrate colonization prior to the extended surveys conducted by the design—research team.

2.2.2 Design-research team ecological surveys (2018—2022)

Between 2018 and 2021, the design-research team―comprising the project’s architects and the collaborating ecologist―conducted a series of monitoring campaigns to document biodiversity dynamics and habitat evolution on the building (Fig. 6). This work combined systematic field surveys with long-term observational follow-up, providing a coherent ecological record complementary to the initial GROOVES dataset.

2.2.2.1 Systematic surveys.

Monitoring focused on five main taxa―vascular flora, invertebrates, birds, bats, and reptiles―using standardized naturalist protocols adapted to the building’s configuration.

● Vascular flora were inventoried through Florilège-Prairie quadrat surveys on the three rooftop habitats (forest core, shrubland ring, mesophilic prairie) and the connecting ramp, in July 2019, May 2020, and May—July 2021. Gardened zones were excluded. Species were classified as planted or spontaneously colonized, and relative abundance was visually estimated within 1 m² quadrats.

● Invertebrates were surveyed on roofs and façades between May and September 2021, under sunny conditions (10:00—17:00), using sweep-net transects and visual identification following SPIPOLL and STERF guidelines. Lepidoptera, Orthoptera, Coleoptera, and Odonata were the target groups. Specimens were briefly examined with a 10 × magnifier and released unharmed.

● Avifauna were monitored through 10-min listening points and binocular observations in July 2019 and from March to November 2021, using the STOC-EPS protocol. Surveys assessed breeding, fledging, and migratory activity on the roof and façades, complemented by nest-box checks in 2020—2021.

● Chiroptera (bats) were surveyed using the Vigie-Chiro POINT FIXE protocol with automatic recorders (Pettersson D240X) on the roof and at the building base in July 2019, June 2021, and September 2021. Recordings (.wav) were analyzed via Kaleidoscope and TADARIDA, with additional walking transects using Pettersson D240X and M500-384 detectors for real-time sound analysis.

● Reptiles, mainly Podarcis muralis, were occasionally observed during façade inspections in 2020—2021.

● Soil biodiversity (earthworms and surface macrofauna) was assessed through three participatory diagnostics involving school children in November 2021, following simplified sampling protocols under the ecologist’s supervision.

2.2.2.2 Ongoing opportunistic surveys.

From 2018 to 2022, the ecologist within the design-research team also carried out repeated field observations to track ecological change between formal survey periods (Fig. 6). Conducted during maintenance operations and educational sessions (typically from April to November), these observations documented.

● Spontaneous colonization by new plant species.

● The progression of vegetation strata and soil accumulation.

● Behavioral activity of pollinators and birds (foraging, nesting, and flight paths).

● Microclimatic variations influencing habitat use.

Notes were recorded through field sheets and photographic documentation during each visit, generating complementary qualitative data on the site’s ecological succession and the resilience of designed habitats.

2.3 Ecological engineering interventions (2020—2021)

Following the first monitoring results and early user feedback, the design research team implemented a series of targeted ecological interventions between 2020 and 2021 to improve habitat quality and reinforce the building’s ecological performance. Actions focused on three main systems―the grove, roof meadow, and façades―combining restoration ecology with educational and participatory engagement involving teachers and children.

2.3.1 Grove ecosystem

After five years, the grove had matured into a small, wooded habitat (trees 7—8 m high) but showed soil compaction and limited understory growth. Manual decompaction and the addition of oak—hornbeam leaf litter from nearby woodlands were used to restore soil fertility. Low wooden fences were installed to protect regenerating areas, and approximately 400 native understory plants (e.g., Carex sylvatica, Veronica chamaedrys, Fragaria vesca) were introduced during collaborative workshops with schoolchildren and educators, linking ecological restoration to environmental learning.

2.3.2 Roof meadow ecosystem

A restoration experiment in 2020 reintroduced seeds and “flower hay” from a dry grassland at Domaine National de Marly-le-Roi. Using a false-sowing technique on a 200 m² test plot, seeds of Salvia pratensis and Bromus erectus were spread with children’s participation. Selective mowing and the planting of 200 native perennials in 2021 enhanced species richness and tested the feasibility of low-maintenance, biodiverse urban meadows.

2.3.3 Façade ecosystem

By 2020, concrete carbonation had created favorable conditions for pioneer colonization. The team introduced native wall flora (Cymbalaria muralis, Centranthus ruber, Sedum acre) mixed with a custom substrate (crushed brick, sand, lime, compost). Around 600 L of this mix were applied manually to façade joints, followed by 100 climbing ivy (Hedera helix) plantings. The intervention sought to accelerate natural colonization and diversify habitats for wall-dwelling flora and invertebrates.

These participatory measures laid the groundwork for long-term ecological succession and informed the adaptive management strategies discussed in Section 3.2.

2.4 Post-occupancy audits (2015—2021)

2.4.1 Informal follow-up (2015—2020)

In the years following completion, the design—research team maintained informal exchanges with school staff and city services to observe how outdoor spaces were appropriated and maintained. Between 2015 and 2020, several spontaneous audits and meetings addressed emerging issues on the rooftop, terraces, and façades. In 2015, three coordination meetings were held as the school pursued E3D certification, involving teachers, maintenance staff, and local partners (Maison de la Nature, GPSO). These revealed strong pedagogical engagement―vegetable gardens and nature-based activities―but also conflicts with ecological intent, such as trampling and removal of organic debris. A 2016 follow-up with the Agence Régionale de la Biodiversité highlighted coordination gaps among user groups. The 2018 drought exposed maintenance vulnerabilities, prompting renewed discussions on management. Throughout this period, the follow-up remained practice-based and voluntary, generating valuable qualitative insights but without a formalized protocol. By 2020, recurring conflicts in use and ecological degradation underscored the need for a structured long-term audit.

2.4.2 Formal audits (2020—2021)

In 2021, the City of Boulogne-Billancourt officially commissioned the design-research team to conduct a structured post-occupancy audit of building use and biodiversity management. The objective was to document daily practices, identify conflicts, and propose adaptive management measures linking human and ecological requirements. A sequence of stakeholder workshops and working sessions formalized the process.

● March 31, 2021: Initial meeting with the school director, a teacher, municipal representatives (Deputy Mayor for Education; members of the ecological transition mission), after-school coordinators, the developer SPL Val de Seine Aménagement, GPSO representatives, an ecologist, and an architect. Aim: document patterns of use and emerging conflicts.

● September 28 and November 17, 2021: Work sessions with the school administration, city officials, and SPL Val de Seine Aménagement to refine proposed interventions.

● December 17, 2021: Final review meeting with the school director, two parent representatives, municipal ecological transition delegates, SPL representatives, an ecologist, and an architect. Aim: present and validate the illustrated audit report and finalize an operational action plan.

The resulting operational plan synthesized findings from user observations and ecological monitoring, informing the adaptive ecological interventions described in Section 2.3. Rather than following a rigid experimental protocol, the audits evolved as a practice-based, iterative process through which architects, ecologists, and users collaboratively generated empirical insights into the co-evolution of human practices and biodiversity within the building.

2.5 Data integration and analysis

The analytical phase aimed to link ecological monitoring, user audits, and adaptive management within a unified interpretive framework. Ecological data were first consolidated from the different monitoring campaigns (GROOVES, design research team surveys, and post-intervention observations), yielding a comprehensive inventory of plant, invertebrate, bird, reptile, and bat species recorded on site. Each record was georeferenced by habitat unit―roof meadow, grove, façade, or ground edge―and associated with its status (spontaneous, sown, or planted) and biogeographic origin (native or non-native).

Because the protocols and sampling frequency varied across campaigns, the analysis emphasized presence and persistence and diversity of species rather than abundance. Temporal comparisons were structured around three key phases: the post-construction phase (2016—2018), the intermediate period preceding ecological interventions (2019—2020), and the post-intervention phase (2021). This framework allowed the team to trace the long-term evolution of biodiversity across habitats and to identify processes of spontaneous colonization and habitat stabilization.

The quantitative results―such as total species counts per taxon and per habitat―were complemented by qualitative ecological assessments, focusing on microhabitat conditions (substrate depth, exposure, moisture) and observed ecological interactions. The analysis of bird presence per habitat (wall, grove, edge, meadow) provided a proxy for faunal use of architectural elements, while additional synthesis tables linked these ecological observations with findings from post-occupancy audits.

Finally, the integration of ecological and user data supported the definition of adaptive ecological strategies, presented in the Results and Discussion sections. These strategies translate observed patterns of species persistence, spontaneous establishment, and user practices into operational guidance for long-term management of ecosystemic architecture.

3 Results

3.1 Naturalist surveys

Across 2016—2021, 856 observations documenting 345 species were compiled (plants: 207; animals: 138) (Table 1).

3.1.1 Flora and habitat evolution

3.1.1.1 Flora.

Across all surveys conducted since 2016, a total of 207 plant species were recorded on the site, of which 180 species (87%) are indigenous (Appendix A). This proportion is consistent with other urban contexts, although most woody plantings were intentionally composed of native species. Approximately 42% of all recorded species were deliberately introduced, either during the initial establishment in 2014 or during the renaturation operations undertaken in 2020. The remaining spontaneous species, representing nearly half of the total flora, indicate a notable colonization dynamic for a site as isolated from surrounding biological flows as a green roof. Over time, however, this dynamic appears to decline, suggesting a stabilization process likely linked to habitat saturation by competitive species.

Spontaneous taxa originated either from the topsoil seed bank or through later aerial dispersal. They consist mainly of annual ruderal species associated with cultivated or disturbed soils (Avena fatua, Kickxia elatine), urbanophilic species (Hordeum murinum, Ervilia hirsuta), and species from dry, open habitats (Vulpia myuros, Trifolium arvense), along with a few less common perennial fallow species (Artemisia vulgaris, Jacobaea vulgaris).

In terms of rarity, 10.6% of all recorded taxa were classified as rare to extremely rare, including 1.5% extremely rare, 1.5% very rare, 4.5% rare, and 3% fairly rare species. The spontaneous flora followed a similar pattern, with 4% rare to fairly rare taxa and 58% extremely common species. Across the monitoring period, 31.9% of planted or sown species and 35.9% of early spontaneous species persisted until the final survey; overall, 63.6% of all species were spontaneous, of which 87.8% were regionally native (Appendix A).

By 2021, 89 species were recorded, indicating a relatively high turnover rate among spontaneous taxa―primarily annual ruderal species. This comparison remains partial, as adjacent gardens and ground-level areas were excluded from the 2021 inventory. Several hygrophilic species initially present (Epilobium hirsutum, Lysimachia vulgaris) were no longer observed, likely reflecting the limited availability of moist microhabitats on the rooftop.

3.1.1.2 Heritage value.

Among the 118 spontaneous species, 101 are indigenous (85.6%), underscoring the site’s ecological specificity. Nine species of regional heritage value were identified, all classified as at least “fairly common” in île-de-France (Conservatoire botanique national du Bassin parisien). Most of these are annuals, except for Tragopogon dubius (Common Salsify), a perennial characteristic of meso-xerophilic meadows. Notably, Filago germanica (Common Cudweed), a rare species typical of dry alluvial habitats, dominates certain south- and southwest-facing balconies where mineral substrates limit perennial competition. Conversely, Laphangium luteoalbum (Jersey Cudweed), a hygrophilic and very rare species, has not been recorded since 2016, likely due to insufficient moisture on the rooftop. Other characteristic dry-habitat taxa include Medicago minima (Bur Medick), Trifolium arvense (Hare’s-foot Clover), and Blackstonia perfoliata (Yellow-wort), a pioneer annual of marl soils often found in association with the Common Salsify.

3.1.1.3 Habitat evolution.

Over the decade of observation, the building’s vegetated envelope evolved into a mosaic of distinct yet interconnected habitats (Fig. 7). Each has followed a specific ecological trajectory shaped by initial design intentions, spontaneous colonization, and successive management interventions.

Wall habitats (2640 m²): The bioreceptive façades, structured by concrete megaliths with cavities and recesses, were designed to host chasmophytic vegetation characteristic of regional walls (Parietarietea judaicae, Asplenietea trichomanis). Re-seeding and natural dispersal progressively increased floristic diversity. Sedums (Sedum rupestre, S. acris, S. album) have become established as dominant species, while ferns (Asplenium trichomanes, A. adiantum-nigrum) and companions such as Geranium robertianum occasionally appear. Over time, these vertical assemblages have stabilized and now provide structural continuity between the roof and surrounding vegetation, supporting microhabitats for invertebrates and birds.

Grove (826 m²): Initially planted with species representative of Oak—Hornbeam forests―Field Maple (Acer campestre), Hornbeam (Carpinus betulus), Small-leaved Lime (Tilia cordata), and Wild Cherry (Prunus avium)―the grove has gradually matured. The development of a dense shrub layer (Crataegus orientalis, Cornus sanguinea, Prunus spinosa, Viburnumspp.) has enhanced shelter and food resources for fauna while reducing wind exposure. Yet canopy closure limited understory regeneration, prompting enrichment in 2020—2021 with locally sourced, shade-tolerant herbs (Glechoma hederacea, Carex sylvatica, Prunella vulgaris). Indicators such as increasing organic matter and the appearance of soil macrofauna point to progressive pedogenesis and ecological stabilization.

Meadow edge (360 m²): Bordering the grove, the meadow edge underwent moderate succession. Initially dominated by social grasses (Dactylis glomerata, Arrhenatherum elatius), it gradually incorporated ruderal and fallow species (Malva sylvestris, Artemisia vulgaris), maintaining a transition zone that connects woodland and open habitats.

Gardened spaces (345 m²): Perennial and biennial beds (sage, rosemary, artichoke) and small vegetable plots form cultivated areas dedicated to pedagogical use. Although excluded from the 2021 ecological inventory, their continued maintenance and human use influence adjacent spontaneous assemblages through local disturbance and seed dispersal.

Basophilic meadow (205 m²): Originally dominated by annuals such as Anisantha sterilis and Bromus hordeaceus, this area has progressively transitioned into a basophilic grassland resembling regional calcareous meadows. Reseeding and the 2021 introduction of 200 locally sourced perennials (Achillea millefolium, Briza media, Centaurea jacea, Scabiosa columbaria) accelerated this shift. The community is now increasingly stable and floristically rich, dominated by Bromus erectus and Salvia pratensis, with dense flowering observed since spring 2022.

Meadow grasslands (395 m²): These plots, established during the initial planting phase, remain more dynamic and less structured. They are still largely dominated by competitive grasses (Lolium perenne) and annual species but show early signs of perennial establishment, suggesting a slow trajectory toward stabilization under limited management.

Annual fallow land (347 m²): Developing on shallow, mineral soils, these meso-xerophilic fallows have maintained pioneer communities dominated by Trifolium arvense, Medicago minima, and Filago germanica. Their persistence, particularly on south- and southwest-facing sections, demonstrates the ecological value of open, low-maintenance zones. As they host several rare species, these areas are preserved as reference habitats.

Perennial fallow land (93 m²): This area shows a gradual transition toward greater perenniality and structural complexity. Species typical of the Dauco—Melilotion alliance (Daucus carota, Helminthotheca echioides, Cota tinctoria, Achillea millefolium) have become established under a regime of late mowing that promotes both flowering and overwintering niches for insects. In 2021, this habitat was experimentally enriched with fifteen Anacamptis pyramidalis (Pyramidal Orchid) individuals rescued from a nearby site; follow-up monitoring in 2022 will assess their establishment.

3.1.2 Fauna

From 2016 to 2021, 265 wildlife observation records were collected, bringing the inventory to 138 animal species: 7 Orthoptera, 8 Lepidoptera, 16 Hymenoptera, 19 Coleoptera, 5 Odonata, 23 Hemiptera, 10 Diptera, 25 other invertebrates, 1 reptile, 20 birds, and 4 bats (Table 3).

3.1.2.1 Birds.

Bird monitoring showed a strong increase in species richness over time: from only two chickadee species in 2017 to twenty species recorded between 2019 and 2021 (Table 2). Several species nest on the façades or in the rooftop woodland, others feed or stop during migration, and a few are observed only in flight. Two confirmed or probable breeders―the House Sparrow (Passer-domesticus, île-de-France) and the Eurasian Goldfinch (Carduelis carduelis, France)―are listed as Vulnerable and represent local conservation priorities. This relatively high diversity, given the dense urban context, reflects the variety of habitats created on the building, providing resources for nesting and foraging in meadows, fallows, shrubs, trees, litter, and façade cavities.

Cavity-nesting and façade species: Tits (Cyanistes caeruleus, Parus major) were the most frequent residents. One or two breeding pairs of each species occupied nest boxes integrated into the façades annually since 2018. The Black Redstart (Phoenicurus ochruros) has been present each breeding season since 2019, using façade cavities or semi-open nest boxes for nesting and the roof habitats for foraging. The House Sparrow (Passer domesticus), listed as Vulnerable in île-de-France, was regularly observed on the roof and at the base of the building, with fledglings recorded in 2020 and 2021.

Woodland and edge assemblages: The roof grove and adjacent shrub layer supported typical woodland and edge species. The Common Blackbird (Turdus merula) was observed year-round, feeding on soil invertebrates and fruits of Sorbus and Prunus. The Common Wood Pigeon (Columba palumbus) nested each year since 2019 in the grove. The Chaffinch (Fringilla coelebs), European Robin (Erithacus rubecula), and Eurasian Goldfinch (Carduelis carduelis, Vulnerable in France) were regularly present in the woodland and meadow edges, where they fed on seeds or fruits. Occasional records included the Blackcap (Sylvia atricapilla) and Hedge Accentor (Prunella modularis).

Passage and opportunistic species: Several migratory or opportunistic species were observed in transit or foraging on site, including Long-tailed Tit (Aegithalos caudatus), Common Swift (Apus apus), Common Starling (Sturnus vulgaris), and Common House-martin (Delichon urbicum). Predatory and scavenger species such as the Common Kestrel (Falco tinnunculus), Magpie (Pica pica), and Carrion Crow (Corvus corone) were regularly recorded, while the invasive Rose-ringed Parakeet (Psittacula krameri) was occasionally seen flying over the building.

Together, these records indicate the presence of both breeding and transient bird species associated with the variety of habitats developed on the building’s façades and roof.

3.1.2.2 Invertebrates.

A total of 112 invertebrate species were recorded across the roof, façades, and gardens, reflecting the structural and microclimatic diversity of the building (Table 3).

Orthoptera: 7 species were identified, including 4 grasshoppers (Caelifera) and 3 bush-crickets (Ensifera). Chorthippus biguttulus and C. brunneus dominated dry meadows, while moisture-tolerant taxa such as Gomphocerippus rufus, Pseudochorthippus parallelus, and Tettigonia viridissima occurred near shaded woodland edges.

Lepidoptera: 8 butterfly species and 1 diurnal moth were observed. Grassland species (Coenonympha pamphilus, Maniola jurtina) and Aricia agestis reproduced on site. Migratory and garden visitors included Vanessa cardui, V. atalanta, Pieris rapae, and Macroglossum stellatarum. Lasiommata maera, typically found in rocky habitats, was repeatedly seen on façades, suggesting potential use of mineral surfaces as substitute basking sites.

Hymenoptera: 16 species were recorded. The honeybee (Apis mellifera) was abundant, alongside solitary bees (Lasioglossum, Megachile, Anthidium, Xylocopa violacea) foraging in the garden and meadow. Perforated concrete blocks on façades served as insect hotels, occupied by overwintering individuals at various heights. Parasitic wasps (Holopyga amoenula, Chrysis, Eupelmus, Gasteruption) and the bee-hunting Philanthus triangulum were also noted. Ants included Lasius niger, L. platythorax, Myrmica scabrinodis, and the invasive L. neglectus.

Coleoptera: 19 beetle species were found, among them native Coccinella septempunctata and invasive Harmonia axyridis. Fungal-feeding Vibidia duodecimguttata and Psyllobora vigintiduopunctata occurred in humus-rich edges, while Tytthaspis sedecimpunctata preferred dry meadows. Chrysomelids such as Altica, Oulema, and Chrysolina americana occupied the aromatic beds. Ground beetles (Calathus, Acupalpus, Notiophilus) and flower visitors (Valgus hemipterus, Cetonia aurata) were also observed. Tree age still limits cavity-dependent larval development.

Hemiptera: 23 species were identified, primarily leafhoppers and mirid bugs, along with shield bugs (Palomena prasina, Dolycoris baccarum, Graphosoma italicum) and the firebug (Pyrrhocoris apterus) on Tilia trunks.

Odonata: 5 dragonfly and damselfly species were occasionally observed hunting over the roof (Platycnemis pennipes, Erythromma lindenii, Orthetrum cancellatum, Sympetrum striolatum, S. fonscolombii). The latter, rare in île-de-France, was recorded once in 2019.

Diptera and Others: 10 dipteran species were noted, including Toxoneura muliebris on flowering edges. Litter samples contained 18 arachnids, 4 crustaceans, 2 myriapods, and 3 mollusks. On façades, Helix aspersa and Cepaea hortensis sheltered in concrete cavities.

3.1.2.3 Bats and reptiles.

Acoustic surveys conducted in 2019 and 2021 identified 4 bat species. Pipistrellus kuhlii, P. nathusii, and Nyctalus noctula showed very low activity, suggesting occasional transit between the site and the Seine corridor. In contrast, the Common Pipistrelle (P. pipistrellus) was recorded at all listening points during both campaigns, with continuous nocturnal foraging activity in September. Repeated social calls indicate the likely presence of a nearby roost. The façades’ non-jointed concrete blocks, with interstices a few centimeters wide, provide potential microcavities for this opportunistic species, which is protected in France and listed under Annex IV of the EU Habitats Directive.

One reptile species, the Common Wall Lizard (Podarcis muralis), was observed basking on the façade. Also listed under Annex IV, it is a regional conservation priority due to the loss of rough masonry in the inner suburbs of île-de-France. The façade’s cavities and offsets offer suitable refuges, but the sealed lower blocks restrict access and limit long-term establishment.

3.2 Contribution of the ecological engineering interventions (2020—2021)

Ecological engineering interventions implemented in 2020—2021 aimed to enhance biodiversity by shifting from horticultural maintenance toward restoration ecology techniques. They addressed soil compaction, user disturbance, and limited vegetation resilience identified during earlier monitoring.

Of the 28 species recorded in the hay meadow at Domaine National de Marly-le-Roi―the source of the restoration material―11 were subsequently found on the school’s roof in 2021 (see Appendix B). Direct seeding trials (2019—2021) failed due to trampling and soil disturbance, even under irrigated and decompacted conditions, confirming that germination was incompatible with daily use of the site.

Transplanting nursery-grown individuals proved more effective. Among 250 plants installed in autumn 2020, mature specimens such as Glechoma hederacea and Carex sylvatica displayed high survival, while young seedlings declined under competition and foot traffic. A second trial in spring 2021, involving pupils in planting 150 seedlings, achieved lower establishment rates, underscoring the advantage of autumn plantings.

In the prairie restoration zone, a 200 m² false-seeding test plot showed that hay mulch reduced annual weed competition and favored native establishment. The addition of 200 locally sourced perennials in October 2020 accelerated flowering and stabilized the herbaceous cover.

3.3 Building post-occupancy audits

Post-occupancy audits conducted alongside ecological monitoring revealed how users’ daily activities interacted with the building’s evolving ecological habitats. While naturalistic surveys identified key ecological pressures, the audits emphasized reconciling educational and recreational uses with biodiversity goals.

Recurring conflicts were observed between access and conservation―particularly trampling in restored meadows, soil compaction in the woodland, and unsupervised movement near façades. In response, spatial adjustments were developed to balance these pressures: half of the woodland was designated as open-access for outdoor classes and half as protected biodiversity refuge, while the meadows were reorganized with late-mow zones and fenced perimeters near guardrails.

Audits also underscored the need for clearer programmatic zoning of shared spaces. Separate garden areas were defined for the preschool, elementary, and after-school programs, and a large aromatic garden was created to support pollinators and outdoor learning activities. On façades and balconies, users expressed interest in localized planting and stronger visual connection to vegetation, prompting the installation of planters, water points, and climbing species.

Together, these findings guided the spatial reorganization implemented in 2023, integrating safety, educational engagement, and ecological restoration (Table 4). They mark a transition toward adaptive management strategies developed through continuous feedback between ecological monitoring and use.

4 Discussion: Lessons learned and future directions

4.1 Architecture as a natural ecosystem

From its starting point, the School of Sciences and Biodiversity in Boulogne-Billancourt was designed with biodiversity integration as a central objective. The selection of ecosystems―meadows, woodland edges, deciduous groves, and wall habitats―shaped the project’s spatial and ecological framework, distinguishing it from conventional green roofs. Unlike extensive or semi-intensive green roofing systems, which often prioritize aesthetic greenery, this project aimed to establish complex, self-sustaining ecosystems with high species diversity and long-term ecological value. However, this approach required new architectural adaptations, ecological engineering techniques, and an ongoing commitment from the design team.

Biodiversity monitoring recorded 345 species (207 plants and 138 animals), levels comparable to an urban park. Bird richness reached 20 species in 2021, close to the average of 9.6 ± 0.7 species per park reported across 37 Paris parks (Mennill et al., 2025). Survival rates aligned with those observed on low-maintenance green roofs. Although long-term comparisons for the same types of green roof and climate remain scarce, Schneider et al. (2021) reported 36% species persistence over 12 years in a semi-arid climate, while Catalano et al. (2016) observed that, after 30 years, spontaneous stress-tolerant and ruderal species replaced most sown species on a Berlin green roof.

The high turnover in species composition reflects the staged ecological characteristic of human-made ecosystems. Initially, the site was dominated by planted vegetation, followed by the spontaneous arrival of pioneer and opportunistic species colonizing vacant ecological niches. Faunal colonization followed a similar trajectory, with aerially mobile species―birds and insects―establishing first. Over time, competition led to species selection, stabilizing the plant communities.

Despite the increasing biodiversity, monitoring data indicate fragility in these ecosystems, with many species present in low densities and requiring continued support. Recent ecological management efforts, such as prairie restoration and woodland enhancement, have started to reinforce key habitats, improving conditions for species such as the Common Blue butterfly (Polyommatus icarus) by increasing populations of its host plants (Lotus corniculatus). Similarly, woodland restoration through selective thinning and the introduction of woody debris and Ramial Chipped Wood (RCW) has aimed to support soil biodiversity.

Another critical architectural feature has been the building’s living wall, which acts as transitional zones between the building basis and the rooftop habitats. These façades are subject to extreme environmental conditions―including drought stress, temperature fluctuations, and low substrate availability (Pech, 2013)―which only specialized species can tolerate. Rather than functioning as a continuum of green space, the façades provide microhabitats for cliff-dwelling insects and birds, mimicking natural rock formations (Francis and Lorimer, 2011). Over time, weathering processes created surface heterogeneity and carbonation of concrete, improving conditions for cryptic biodiversity such as fungal and microbial communities (Miller et al., 2012; Manso et al., 2015). To accelerate this process on the school’s facades and to stimulate aging processes, experiments have been conducted using degraded mortar substrates with native rock-dwelling plant species with first positive results. This observation aligns with the principles of bioreceptivity (Sanmartín et al., 2021), emphasizing the role of 3D surface characteristics―such as roughness, porosity, and cavities―and environmental factors like pH, moisture, and light in fostering biological colonization.

The advantage of cementous living walls over other types of green walls is that these are better integrated into the architecture (structurally and aesthetically), have a lifespan equal to that of the building―longer than that of traditional living walls―and are self-supporting (Lewandowski et al., 2023). However, the positive bioreceptivity of the school’s walls must be put in perspective with the amount of concrete used in the project, as concrete—cement in particular—has a negative impact on the environmental (Zabalza Bribián et al., 2011; Barcelo et al., 2014). Other architecture researchers are developing living walls based on the principle of bioreceptivity with cementitious materials with a pH between 5.8 and 7 (Manso et al., 2014, 2015; Lewandowski et al., 2023); porous concrete (Hitti et al., 2021; Mustafa et al., 2021); or low-pH and bioreceptive concrete (Riley et al., 2019; Veeger et al., 2021). Other studies have focused on morphology and envisioned façades not as skins but as barks, similar to the concrete bark-like walls of the School of Science and Biodiversity. Cruz and Beckett (2016) have developed a series of living walls to advance bioreceptive architecture. As in the school, these walls use a "hydrophilic" design to support more discreet and autonomous vegetation compared to traditional green walls, featuring tree-bark-inspired geometries. In 2021, the same team introduced "cork-crete", a concrete using cork, as an alternative material.

The greening of facades also influences the potential for biodiversity hosting, with the development of ivy becoming a key structural element: its late blooming extends the availability of resources for pollinators, its autumn berries are consumed by many birds, and its evergreen foliage provides shelter for a wide variety of insects (Madre et al., 2015). 235 species have been recorded in France on ivy (De Flores and Loïs, 2017). Despite these ecological advancements, the high human use of the site remains a structuring factor in its ecological evolution.

4.2 Adapting management strategies to building use

The adaptive management of the site has been essential in responding to both ecological needs and evolving user interactions. The prairie restoration, for example, has demonstrated the effectiveness of traditional hay meadow management techniques, with initial results showing improvements in soil organic matter and the recruitment of native plant species from donor meadows. Following this experiment, hay mowing has been synchronized with traditional fenaison periods (mid-June), and the meadow restoration has incorporated winter soil preparation and direct overseeding techniques.

While the woodland is relatively stable, improvements are still needed. Advances in urban soil ecology—particularly in Technosols made with recycled urban waste (Fabbri et al., 2021; Minixhofer et al., 2022)―and the increasing availability of indigenous plants offer new opportunities to refine its management. Building on the use of the woodland and early experiences, the focus has shifted to selective tree thinning, enriching the understory with native ground cover species―now safeguarded by fences from the trampling of schoolchildren―and enhancing habitat complexity through the addition of organic matter (Jusselme et al., 2019). This approach could be further complemented by the addition of earthworms to support the green roof’s ecosystem services, as demonstrated in other experiments (Jusselme et al., 2019).

Physical separation to balance ecological conservation and human use as proposed by Goddard et al. (2010) was implemented on the rooftop. The woodland area was fenced off as a sanctuarized space, while pedagogical gardens―extended to the school’s balconies and designed with planters―were given clear visual boundaries. This approach successfully reconciled ecological preservation with educational activities, demonstrating its effectiveness in fostering biodiversity while accommodating human needs. The educational gardens, expanded to rooftop edges and balconies, have served as both pedagogical tools and biodiverse features that further diversified the rooftop ecosystem.

These interventions highlight the dynamic nature of biodiverse architecture, where ecological processes unfold over extended timescales. The continued monitoring and adaptive management of the site will allow further refinements in these strategies, ensuring that both ecological functions and user needs are met.

Overall, these adapting management strategies can be considered as part of a restoration ecology method for disturbed site called the “minimum intervention approach” (Prach and Hobbs, 2008). In this context, spontaneous succession and passive restoration are advocated but combined with technical intervention measures to manipulate them toward a target (Prach and Hobbs, 2008; Prach and Moral, 2015).

4.3 Limitations of the study

While the study provides unique longitudinal insight into the co-evolution of architecture, biodiversity, and human practices, its scope remains constrained by practical and methodological limits. The ecological data, though extensive, are irregular and partly qualitative due to resource constraints and the participatory nature of the monitoring. As a single-case study, results cannot be generalized but instead serve as a model for integrating ecological monitoring into architectural practice. Future research should formalize long-term interdisciplinary protocols to better link ecological data with adaptive design decisions across different building typologies.

4.3.1 Sampling bias and methodological constraints

As this study was primarily practice-based and conducted voluntarily by the original design—research team, it inevitably bears certain methodological limitations. The frequency and timing of surveys varied between years and were often coordinated with the school calendar and climatic conditions rather than fixed sampling schedules. Consequently, temporal gaps may have influenced detection probabilities, particularly for short-lived or migratory species. Seasonal and interannual variability, as well as the opportunistic character of certain observations, may also have produced moderate sampling bias in favor of conspicuous taxa. These limitations do not invalidate the overall findings―species persistence and habitat evolution were repeatedly confirmed through multiple observation cycles―but they underline the need for standardized long-term and systematic protocols in future research.

4.3.2 Contextual and comparative limits

This study focused on a single, plot-scale case to observe, over a long period, the internal dynamics of one building conceived as an ecosystem. This in-depth, practice-based approach aimed primarily to inform adaptive management strategies rather than to achieve statistical generalization. While this focus necessarily narrows external comparability, it provides a rare opportunity to document the intertwined evolution of architectural habitats, ecological colonization, and user practices within a real project. Comparable situations―combining multiple habitat types, user interactions, long-term monitoring, and low-maintenance intensive roofs in a temperate climate―remain scarce. The absence of a formal control group thus reflects both the specificity of the case and the exploratory nature of this methodological framework. Future studies could build upon this foundation by replicating similar longitudinal observations across diverse contexts to progressively strengthen comparative knowledge.

4.3.3 Architectural and analytical scope

Architectural parameters―façade morphology, substrate depth, exposure, and material texture—were analyzed qualitatively through their observed ecological performance rather than through a quantitative correlation model. While this limits the generalizability of results, it reflects the interdisciplinary and iterative mode of architectural inquiry. Future research could build on these findings by formalizing analytical models linking bioreceptivity parameters (Sanmartín et al., 2021) and design factors (Jim, 2015) such as material properties and morphology, with microclimatic and biodiversity data.

4.3.4 Material and life-cycle considerations

The long-term performance of bioreceptive materials also raises important life-cycle questions. While low-maintenance strategies and spontaneous colonization reduce operational impacts, the embodied footprint of specialized concretes or substrates remains to be assessed. Future research could include life-cycle assessment (LCA) to evaluate the balance between material impacts and the ecological benefits generated over time.

4.4 Lessons learned and future directions

The School of Sciences and Biodiversity serves as a rare example of long-term ecological integration within an urban educational facility. This case study underscores several key takeaways for biodiverse architectural projects.

1) Architecture as an evolving ecosystem

◦ The project demonstrates that buildings can function as dynamic ecological systems, requiring long-term monitoring and adaptive management rather than static maintenance regimes.

◦ The biodiversity observed at the site is comparable to that of an urban park, but achieving this required continuous ecological interventions and habitat structuring.

2) The importance of adaptive management

◦ The post-occupancy phase is not the endpoint but rather the beginning of an ongoing adaptative management and ecological process.

◦ Strategies such as shifting from direct seeding to transplanting nursery-grown plants, refining meadow management techniques, and enhancing soil conditions have been essential in adapting to unforeseen ecological and user challenges.

◦ Similar to protected natural reserves, a 10-year management plan is necessary for the continued coexistence of biodiversity and human activity.

◦ A “minimum intervention approach” combining spontaneous succession and technical reclamation is adequate.

3) Balancing ecological goals with human use

◦ The project’s success depends not only on ecological processes but also on the behavior and engagement of users.

◦ The post-occupancy audits highlighted the need for structured spatial planning, including restricting access to sensitive areas, allocating specific gardening zones to reduce pressure on the ecological areas.

◦ The ongoing dialogue between architecture, ecology, and pedagogy is essential in ensuring that natural dynamics are integrated into daily school activities without compromising biodiversity goals.

4) The need for a new architectural approach

◦ Future biodiverse buildings should integrate ecologists, landscape architects, and biodiversity specialists from the early design phases.

◦ The structural capacity of the building, volumetry, and load-bearing elements should be considered to support diverse vegetation types, including rooftop woodlands and climbing plant structures.

◦ Architectural interventions should be designed to anticipate long-term ecological dynamics, creating environments that mature and evolve rather than degrade over time.

◦ As a future direction, the school’s meadows were leveraged to enhance another project under construction―a mixed-use building featuring a vegetated roof. The two projects were synchronized: the meadow in Boulogne’s School was mowed while preserving its ecosystem, and the cuttings were immediately harvested, transported, and spread on the same day onto the future meadow areas of the rooftop on the second site.

This study provides a valuable reference for future biodiversity-integrated projects, reinforcing the need for multidisciplinary collaboration, adaptive design strategies, and long-term ecological management. The next phase of this project will focus on the implementation of the 2023 action plan―some of the actions have already been taken―further refining the balance between human engagement and ecological resilience.

This project aligns with key principles of ecological design outlined by Shu-Yang et al. (2004), such as supporting human needs while supporting ecosystem integrity and indigenous biodiversity and fostering environmental literacy to build social support for sustainable development. However, there is still room for improvement in reducing reliance on high-carbon materials like concrete. Future developments could further enhance sustainability by incorporating more renewable resources, recycling, and efficient material and energy use while incorporating. Life Cycle Assessments, both on materials (Zabalza Bribián et al., 2011) and greening systems (Ottelé et al., 2011), must be widely adopted in green building projects to prevent architectural greenwashing (Perini and Novi, 2016; Riley, 2017). However, existing cost-benefit frameworks are not yet suited to projects of this type and scale, as they typically focus on smaller systems (Perini and Rosasco, 2013), and biodiversity along with its positive effects remains difficult to quantify. The trade-off between the environmental cost of using concrete and the benefits of supporting biodiversity in a highly constrained, dense urban district remains an open discussion.

5 Conclusion: towards an ecosystemic architecture

The School of Sciences and Biodiversity in Boulogne-Billancourt serves as a benchmark for biodiverse architecture, demonstrating how buildings can support biodiversity rather than simply hosting greenery. The project highlights both the potential and challenges of embedding biodiversity in urban design, requiring a shift from viewing architecture as fixed and immutable to adaptive and evolving.

Key findings emphasize the need for long-term ecological management, transdisciplinary collaboration, and site-specific renaturation strategies. Unlike traditional green infrastructure, this project shows that architecture must accommodate ecological flux, requiring continued observation and adjustments. The integration of biodiversity also calls for a new design culture, where ecologists and architects work together to create functioning ecosystems within built environments.

Beyond its technical innovations, this study contributes to broader discussions on urban renaturation. In cities, restoring lost ecosystems is impossible, but recreating ecological functions is achievable. This case study offers a reference point for future research, informing methodologies for biodiverse and ecosystemic architecture and shaping new construction approaches that actively contribute to ecological resilience.

At a broader scale, the adaptive management framework developed here aligns with emerging biodiversity policies such as the Loi pour la reconquête de la biodiversité (2016) in France and the European Nature Restoration Law (adopted in 2024), both of which call for the integration of ecosystems into the urban fabric and the built environment. By operationalizing these ambitions within a concrete architectural project, this research illustrates how design practice can directly support policy goals for ecological transition.

Ultimately, this research challenges conventional architectural paradigms, advocating for buildings to participate actively to generate urban ecosystems. As more projects integrate biodiversity, this work lays the foundation for new ecological design strategies, ensuring that cities evolve as multi-species environments rather than solely human-centered spaces.

References

[1] Alexandri, E., Jones, P., 2008. Temperature decreases in an urban canyon due to green walls and green roofs in diverse climates. Build. Environ. 43, 480—493.

[2] Auer, T., Radi, M., Brkovi, M., 2019. Green facades and living walls: a review establishing the classification of construction types and mapping the benefits. Sustainability 11 (17), 4579.

[3] Barcelo, L., Kline, J., Walenta, G., Gartner, E., 2014. Cement and carbon emissions. Mater. Struct. 47 (6), 1055—1065.

[4] Barra, M., Johan, H., 2021. Ecology of Green Roofs. Summary of the GROOVES (Green Roofs Verified Ecosystem Services) Study. Agence Régionale De La Biodiversité En Ile-de-France. Institut Paris Région.

[5] Barriuso, F., Urbano, B., 2021. Green roofs and walls design intended to mitigate climate change in urban areas across all continents. Sustainability 13 (4), 1—14.

[6] Beatley, T., 2011. Biophilic urban design and planning. In: Beatley, T. (Ed.), Biophilic Cities: Integrating Nature into Urban Design and Planning. Island Press, pp. 83—129.

[7] Blanco, E., Cruz, E., Lequette, C., Raskin, K., Clergeau, P., 2021. Biomimicry in French urban projects: trends and perspectives from the practice. Biomimetics 6, 27.

[8] Bustami, R.A., Belusko, M., Ward, J., Beecham, S., 2018. Vertical greenery systems: a systematic review of research trends. Build. Environ. 146, 226—237.

[9] Canepa, M., Mosca, F., Barath, S., Changenet, A., Hauck, T.E., Ludwig, F., Roccotiello, E., Pianta, M., Selvan, S.U., Vogler, V., Perini, K., 2022. Ecolopes, beyond greening. A multi-species approach for urban design. AGATHÓN | International Journal of Architecture, Art and Design 11, 238—245 (online).

[10] Catalano, C., Marcenò, C., Laudicina, V.A., Guarino, R., 2016. Thirty years unmanaged green roofs: ecological research and design implications. Landsc. Urban Plann. 149, 11—19.

[11] Charoenkit, S., Yiemwattana, S., 2016. Living walls and their contribution to improved thermal comfort and carbon emission reduction: a review. Build. Environ. 105, 82—94.

[12] Chayaamor-Heil, N., Vitalis, L., 2021. Biology and architecture: an ongoing hybridization of scientific knowledge and design practice by six architectural offices in France. Front. Archit. Res. 10—2, 240—262.

[13] Connop, S., Nash, C., 2018. Blandscaping that erases local ecological diversity. In: The Nature of Cities, 9 January 2018.

[14] Cruz, M., Beckett, R., 2016. Bioreceptive design: a novel approach to biodigital materiality. Architectural Research Quarterly 20 (1), 51—64. Cambridge University Press.

[15] De Flores, M., Loïs, G., 2017. Zoom sur le Lierre grimpant. Insectes 187, 13—16.

[16] Dunnett, N., Kingsbury, N., 2004. Planting Green Roofs and Living Walls. Timber Press.

[17] Fabbri, D., Pizzol, R., Calza, P., Malandrino, M., Gaggero, E., Padoan, E., Ajmone-Marsan, F., 2021. Constructed technosols: a strategy toward a circular economy. Appl. Sci. 11, 3432.

[18] Fortel, L., Henry, M., Guilbaud, L., Mouret, H., Vaissiere, B.E., 2016. Use of human-made nesting structures by wild bees in an urban environment. J. Insect Conserv. 20 (2), 239—253.

[19] Francis, R.A., Lorimer, J., 2011. Urban reconciliation ecology: the potential of living roofs and walls. J. Environ. Manag. 92 (6), 1429—1437.

[20] Francis, R.A., 2011. Wall ecology: a frontier for urban biodiversity and ecological engineering. Prog. Phys. Geogr. 35 (1), 43—63.

[21] Fuller, R.A., Irvine, K.N., Devine-Wright, P., Warren, P.H., Gaston, K.J., 2007. Psychological benefits of greenspace increase with biodiversity. Biol. Lett. 3 (4), 390—394.

[22] Goddard, M.A., Dougill, A.J., Benton, T.G., 2010. Scaling up from gardens: biodiversity conservation in urban environments. Trends Ecol. Evol. 25 (2), 90—98.

[23] Guillitte, O., 1995. Bioreceptivity: a new concept for building ecology studies. Sci. Total Environ. 167 (1—3), 215—220.

[24] Gunnell, K., Grant, G., Williams, C., 2012. Landscape and Urban Design for Bats and Biodiversity. Bat Conservation Trust.

[25] Gunnell, K., Murphy, B., Williams, C., 2013. Designing for Biodiversity. A Technical Guide for New and Existing Building, second ed. RIBA Publishing, Londres.

[26] Hitti, Y., Chapelat, J., Wu, B.-S., Lefsrud, M., 2021. Design and testing of bioreceptive porous concrete: a new substrate for soilless plan growth. ACS Agricultural Science & Technology 1 (4), 285—293.

[27] Hostetler, M.E., Holling, C.S., 2004. Detecting the scales at which birds respond to structure in urban landscapes. Urban Ecosyst. 4, 25—54.

[28] Inventaire National du Patrimoine Naturel, 2018. Information System on Nature and Landscapes of île-De-france — Naturalist Data Outside île-De-france Recorded in Cettia-idf. UMS PatriNat (OFB-CNRS-MNHN), Paris.

[29] IPBES, 2019. In: Brondizio, E.S., Settele, J., Díaz, S., Ngo, H.T. (Eds.), Global Assessment Report on Biodiversity and Ecosystem Services of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services. IPBES Secretariat.

[30] Jim, C.Y., Chen, W.Y., 2011. Bioreceptivity of buildings for spontaneous arboreal flora in compact city environment. Urban For. Urban Green. 10 (1), 19—28.

[31] Jim, C.Y., 2015. Greenwall classification and critical designmanagement assessments. Ecol. Eng. 77, 348—362.

[32] Jusselme, M.D., Pruvost, C., Motard, E., Giusti-Miller, S., Frechault, S., Alphonse, V., Balland-Bolou-Bi, C., Dajoz, I., Mora, P., 2019. Increasing the ability of a green roof to provide ecosystem services by adding organic matter and earthworms. Appl. Soil Ecol. 143, 61—69.

[33] Kondratyeva, A., Knapp, S., Durka, W., Kühn, I., Vallet, J., Machon, N., Martin, G., Motard, E., Grandcolas, P., Pavoine, S., 2020. Urbanization effects on biodiversity revealed by a two-scale analysis of species functional uniqueness vs. redundancy. Front. Ecol. Evol. 8, 73.

[34] Lewandowski, D., Robain, H., Clergeau, P., Le Roy, R., 2023. Bioreceptivity of living walls: interactions between building materials and substrates, and effect on plant growth. Urban For. Urban Green. 83, 127912.

[35] Lundholm, J.T., 2006. Green roofs and facades: a habitat template approach. Urban Habitats 4 (1), 87—101.

[36] Madre, F., Clergeau, P., Machon, N., Vergnes, A., 2015. Building biodiversity: vegetated façades as habitats for spider and beetle assemblages. Glob. Ecol. Conserv. 3, 222—233.

[37] Manso, S., Mestres, G., Ginebra, M.P., De Belie, N., Segura, I., Aguado, A., 2014. Development of a low pH cementitious material to enlarge bioreceptivity. Constr. Build. Mater. 54, 485—495.

[38] Manso, S., Calvo-Torras, M.Á., De Belie, N., Segura, I., Aguado, A., 2015. Evaluation of natural colonisation of cementitious materials: effect of bioreceptivity and environmental conditions. Sci. Total Environ. 512, 444—453.

[39] Mayrand, F., Clergeau, P., 2018. Green roofs and green walls for biodiversity conservation: a contribution to urban connectivity? Sustainability 10 (4), 985.

[40] Mayrand, F., Clergeau, P., Vergnes, A., Madre, F., 2018. Vertical greening systems as habitat for biodiversity. In: Pérez, G., Perini, K. (Eds.), Nature Based Strategies for Urban and Building Sustainability, vol 2018, pp. 227—237.

[41] McKinney, M.L., 2002. Urbanization, biodiversity and conservation. Bioscience 52 (10), 883—890.

[42] McKinney, M.L., 2006. Urbanization as a major cause of biotic homogenization. Biol. Conserv. 127 (3), 247—260.

[43] Medl, A., Stangl, R., Florineth, F., 2017. Vertical greening systems — a review on recent technologies and research advancement. Build. Environ. 125, 227—239.

[44] Mennill, D.J., Gamboa, S.M., Bolger, E.M., Bygrove, M.I., Carlini, M., Cesca, K.A., Drew, H.L., Emerick, N.A., Gaisinsky, A., Miller, R.M., Nguyen, L., O’Leary, L.D., Regonda, S., Robinson, D.N., Tessier, E., Doucet, S.M., 2025. Avian biodiversity in the urban green spaces of paris: higher bird species richness in larger parks and park centers. Avian Conserv. Ecol. 20 (2), 1.

[45] Miller, A.Z., Sanmartín, P., Pereira-Pardo, L., Dionísio, A., Saiz-Jimenez, C., Macedo, M.F., Prieto, B., 2012. Bioreceptivity of building stones: a review. Sci. Total Environ. 426, 1—12.

[46] Minixhofer, P., Scharf, B., Hafner, S., Weiss, O., Henöckl, C., Greiner, M., Room, T., Stangl, R., 2022. Towards the circular soil concept: optimization of engineered soils for green infrastructure application. Sustainability 14 (2), 905.

[47] Muratet, A., Barra, M., Hardion, L., Chiron, F., 2024. Origins and drivers of roof plant assemblages: designing green roofs for biodiversity conservation. Urban For. Urban Green. 94, 128247.

[48] Mustafa, K.F., Prieto, A., Ottelé, M., 2021. The role of geometry on a self-sustaining bio-receptive concrete panel for facade application. Sustainability 13 (13), 7453.

[49] Ottelé, M., Perini, K., Fraaij, A.L.A., Haas, E.M., Raiteri, R., 2011. Comparative life cycle analysis for green façades and living wall systems. Energy Build. 43 (12), 3419—3429.

[50] Pech, P., 2013. Les Milieux Rupicoles. Les Enjeux De La Conservation Des Sols Rocheux (Rupicolous Environments). Synthèses Collection, Editions Quae, Versailles, p. 159. ISSN 1777-4624.

[51] Perini, K., Rosasco, P., 2013. Cost-benefit analysis for green façades and living wall systems. Build. Environ. 70, 110—121.

[52] Perini, K., Novi, F., 2016. Is greening the building envelope a sustainable design practice? Journal of Civil Engineering and Architecture 10, 661—666.

[53] Prach, K., Hobbs, R.J., 2008. Spontaneous succession versus technical reclamation in the restoration of disturbed sites. Restor. Ecol. 16 (3), 363—366.

[54] Prach, K., Moral, R.D., 2015. Passive restoration is often quite effective: response to Zahawi et al. (2014). Restor. Ecol. 23 (4), 344—346.

[55] Raymond, C.M., Frantzeskaki, N., Kabisch, N., Berry, P., Breil, M., Nita, M.R., Geneletti, D., Calfapietra, C., 2017. A framework for assessing and implementing the co-benefits of nature-based solutions in urban areas. Environ. Sci. Pol. 77, 15—24.

[56] Riley, B., 2017. The state of the art of living walls: lessons learned. Build. Environ. 114, 219—232.

[57] Riley, B., de Larrard, F., Malécot, V., Dubois-Brugger, I., Lequay, H., Lecomte, G., 2019. Living concrete: democratizing living walls. Sci. Total Environ. 673, 281—295.

[58] Sanmartín, P., Miller, A.Z., Prieto, B., Viles, H.A., 2021. Revisiting and reanalysing the concept of bioreceptivity 25 years on. Sci. Total Environ. 770, 145314.

[59] Schneider, A., Landis, M., Bousselot, J., 2021. Observations on the survival capacity of 118 plant taxa on a green roof in a semi-arid climate: 12 year update. Journal of Living Architecture 8 (1), 19—40.

[60] Shu-Yang, F., Freedman, B., Cote, R., 2004. Principles and practice of ecological design. Environ. Rev. 12 (2), 97—112.

[61] Steele, J., 2005. Ecological Architecture: a Critical History. Thames & Hudson, London, pp. 272—p.

[62] Vale, B., Vale, R.J.D., 1991. Green Architecture: Design for an energy-conscious Future. Little, Brown.

[63] Veeger, M., Ottelé, M., Prieto, A., 2021. Making bioreceptive concrete: formulation and testing of bioreceptive concrete mixtures. J. Build. Eng. 44, 102545.

[64] Werner, P., Kelcey, J.G., 2017. Urban green and biodiversity. In: Tan, P., Jim, C. (Eds.), Greening Cities. Springer, pp. 131—154, 2017.

[65] Wilson, E.O., 2017. Biophilia and the conservation ethic. In: Mysterud, I. (Ed.), Evolutionary Perspectives on Environmental Problems. Routledge, Washington DC, pp. 263—272.

[66] Zabalza Bribián, I., Valero Capilla, A., Aranda Usón, A., 2011. Life cycle assessment of building materials: comparative analysis of energy and environmental impacts and evaluation of the eco-efficiency improvement potential. Build. Environ. 46 (5), 1133—1140.

Rights & permissions

2095-2635/2026 The Authors. Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd.