At the edge of the forests in Mumbai, crossing that boundary can affect the way an animal is managed, the way a community is treated and the extent to which its knowledge is taken seriously. The boundary of Sanjay Gandhi National Park (SGNP) appears simple on a plan; forest on one side, and the city on the other. The line serves to define jurisdiction, development rights and responsibilities. It is an important planning tool. However, the meaning of that line changes if you take some time walking along the edge of the park.

Leopards travel through the park, residential areas, institutional grounds and infrastructure corridors. Their movements respond to vegetation cover, prey availability, traffic patterns and daily human routines^[1]. This is not a surprising observation, and landscape architects are also already well aware of the mismatch between ecological processes and administrative boundaries.

What is more interesting is what happens institutionally when a leopard crosses the park boundary. Within SGNP, a leopard is classified as protected wildlife. Outside, the same animal may be refered to as an intrusion, a threat or a management problem. The animal itself has not changed, but the institutional meaning of its presence has.

The same shift can be found beyond wildlife management. Around forest edges, planning categories help determine what belongs, what requires protection, whose knowledge is considered relevant and who carries the burden of living with environmental risk. Besides the basic organisation of space, boundaries also organise authority.

For landscape architects, this creates a practical question. If legal boundaries are necessary and fixed, how do we act responsibly in landscapes which are defined by relationships that extend beyond these boundaries?

What the boundary cannot show

SGNP covers approximately 104 square kilometres within Mumbai and Thane^[2]. A recent camera-trap survey identified 54 adult leopards and four cubs across SGNP and neighbouring Aarey Milk Colony^[3]. These animals live alongside some of the highest urban population densities in the world.

The landscape surrounding the park is part of this ecology. Domestic dogs form an important part of the leopard’s diet, and dog populations are influenced by food waste and feeding practices^[4]. Roads affect movement patterns, and vegetation offers cover. Areas that are busy during the day may become movement routes at night.

The responsibility for these conditions rarely sits with a single organisation as waste collection, road management, public safety, landscaping and wildlife response are often handled separately. Together, however, they determine how people and animals encounter one another. The park boundary remains indispensable as it protects land from development and establishes legal responsibilities. The problem emerges, however, when the boundary is mistaken for a complete picture of the landscape. Many of the relationships that influence what happens on site originate elsewhere.

What classification does

Landscape architects inherit classifications long before they begin designing. The land may already be defined as habitat, development land, protected open space, reserve, informal settlement or infrastructure corridor. These categories are useful because they make decisions possible. At the same time, they inevitably privilege certain ways of understanding a place while sidelining others.

The leopard illustrates this particularly clearly. Within a protected area, its presence confirms the value of conservation. Outside the boundary, attention may shift towards capturing the animal, monitor or remove it. Such responses may sometimes be necessary, especially where public safety is concerned. The point is that classification can make some causes easier to see than others.

An encounter between people and wildlife may be linked to waste management, changes in vegetation, altered movement routes or transitions in urban development. Yet these underlying conditions can disappear from view once the issue is framed solely as a wildlife problem. A similar dynamic applies to the way people are seen in landscapes.

Ananya Roy’s work on urban informality shows how planning systems distinguish between authorised and unauthorised forms of occupation^[5]. These distinctions have profound consequences for security, permanence and access to resources. Around Mumbai’s forest edge, public agencies, developers, institutions and local communities often have very different levels of influence over how land is categorised and managed.

By the time a project reaches a landscape architect, many of these decisions already appear settled. They arrive embedded within policy documents, land ownership plans and project briefs. The classifications themselves are visible while the negotiations and conflicts that produced them remain out of sight.

Knowledge that predates the brief

The forest edge is not experienced solely through planning documents and wildlife surveys. Long before contemporary conservation frameworks were established, Indigenous Warli communities had developed cultural practices for living alongside large carnivores. Their relationship with big cats is reflected through Waghoba, a deity associated with protection, danger and reciprocity. Researchers documented 150 Waghoba shrines across the study area and found that worship, oral histories and the annual Waghbaras festival continue to influence understandings of coexistence^[6].

The significance of this example is not that traditional knowledge offers a simple solution to human-wildlife conflict because it does not. Forest-edge communities experience real risks, including attacks, livestock loss and unequal exposure to danger. Nor do all residents share the same views. What the example demonstrates is that important knowledge about a landscape often exists before a project brief is written. In practice, consultation frequently begins after a site has been chosen and a problem has already been defined; communities are then asked to comment on proposed responses.

The sequence matters. If the challenge has already been identified as “preventing leopard entry”, fencing may quickly emerge as the preferred intervention. A different conversation might reveal concerns about unmanaged waste, concentrations of stray dogs, vegetation around footpaths, poor lighting or confusion over reporting wildlife sightings. The perceived problem may remain the same, but the range of potential responses becomes wider. The task is not to replace technical expertise with local knowledge. The more promising approach is to recognise that different forms of knowledge identify different aspects of the landscape.

Working differently within fixed boundaries

Landscape architects are not in the position to alter legal boundaries, but they can decide whether those boundaries define the extent of the project. One practical approach is to distinguish between the project site and the area of enquiry. The project site remains the place where construction, funding and permissions are concentrated. The area of enquiry includes the systems that influence the project regardless of ownership or jurisdiction. This may include drainage networks, animal movement routes, waste infrastructure, informal paths, transport corridors or patterns of daily use depending on the context.

Around Aarey, for example, understanding leopard encounters may require looking at reported sightings alongside vegetation cover, waste collection points and dog concentrations. None of these factors fit neatly within a single property boundary. This shift has implications for representation as well as analysis.

The drawings can show the legal site while also identifying wider areas of influence. They can distinguish between statutory records, field observations, community knowledge, inferred movement routes and ecological data. These sources should remain visible as different forms of evidence rather than being merged into a single account.

Consultation also changes when viewed through this lens. Instead of asking communities to respond to predetermined proposals, engagement can help define the problem itself. The disagreements may become valuable information rather than an obstacle to consensus. One group may see vegetation as welcome shade; another may associate it with reduced visibility and safety after dark.

Neither view is inherently wrong, but the role of design is not to remove such tensions from the drawing. It is to understand where they exist and how an intervention might redistribute them.

For example, a wildlife corridor should not be evaluated merely on the basis of habitat connectivity since questions relating to access, safety and routine movement are also important. A fence which reduces risk at one site may shift that risk to another location, and a buffer zone might improve ecological protection while at the same time affecting the livelihoods, access routes or cultural practices. Relevant design cannot eliminate these trade-offs. But it can make them visible.

The red line

Landscape architecture often describes sites as shared landscapes. The phrase is appealing, but it can obscure significant differences in power. Not everyone has an equal ability to shape land use, influence policy or absorb the consequences of environmental change. Some groups are better positioned than others to determine where risk, restriction and responsibility fall. For that reason, the challenge is not simply to acknowledge that ecological processes cross boundaries because we already know they do.

The more difficult task is learning how to work within legal and administrative boundaries while remaining attentive to the wider relationships they leave outside the frame. That may mean expanding the survey area beyond the project site. It may mean drawing systems rather than parcels, engaging communities before solutions are identified, or assessing who benefits and who bears costs when an intervention is implemented.

The red line remains necessary as it defines where decisions can be authorised and where work can be built.
But it should not determine the limits of observation, consultation or responsibility. The land inside a commission and the landscape affected by it are rarely identical. Recognising that distinction does not make boundaries disappear, but it does alter how we work with them.

References

Notes

This article is based on Aditi Sunder Nair’s MLA Landscape Architecture thesis at the Bartlett School of Architecture, UCL. The research explores Mumbai’s forest edge, focusing on how planning boundaries intersect with ecology, culture and everyday human-wildlife coexistence around Sanjay Gandhi National Park and Aarey Milk Colony.