Previous Page Table of Contents Next Page


As the preceding discussion has already indicated, the cultivation of trees in an urban environment necessitates careful thought and forward planning at the time of establishment. A wide range of management skills are also involved, from the appropriate arboricultural treatment of individual trees to ensuring effective overall management of a given town or city's tree cover.

6.1. Tree establishment decisions

While some trees in urban environments have regenerated naturally and been allowed to grow, the majority are usually planted. A variety of factors should be considered when making the choice of species, provenance and even cultivar. Listed below are some of the major, interrelated points that ideally should play a part in the decision.

Arising from these considerations,

In most cases the decision involves the planting of a number of trees, in which case the potential complementarity of different species/provenances or cultivars must be considered.

For what?: The objective(s) of tree planting

Section four discussed a wide variety of potential benefits that trees can provide in city environments. Tree planting can fulfil a number of these at the same time, especially if an appropriate mix is chosen, but careful planning is required. Much of the skill of planning a planting scheme lies in the effective combination of different species.

An example of some of the attributes to be considered when selecting tree species for different objectives is given in box 6.1. The list is by no means complete, but serves as an indication of the variety of different issues that may need to be borne in mind or investigated.

Box 6.1.Tree planting for different objectives: key attributes to be considered  
objectivekey tree attributes to be considered
  • flowers: size, colour, flowering period
  • fruits: may be attractive, but not if they cause a mess when they fall and rot
  • foliage: shape, colour change, seasonality
  • bark: texture, colour
  • height
  • canopy size and shape
  • cultural significance
  • growth rate
  • canopy size, texture and shape
  • evergreen or deciduous
wildlife habitat
  • fruits edible for birds and animals
  • canopy size (for shelter)
  • particular species associations between different insects/birds/animals and trees
noise reduction
  • canopy size, texture and shape
  • height
reduction of air pollution
  • see section 4.2.
fruit production
  • consumer species/cultivar preference (domestic/market production)
  • nutritional value
  • time to production from establishment
  • seasonality of production (for poorer people, food during periods of deficiency?)
  • local species preference - probably dictated by seasonality (dry season fodder usually preferred), palatability and perceived fodder value
  • local consumer species preference
  • twig, leaf shedding (for gathered fuel)
  • consumer species preference
  • intended production (poles or roundwood)
  • growth rate.

Who makes the decision: different peoples's perspectives

Different people will have different perspectives on the choice of species to be planted. The identity of the person/people making this decision, particularly when he/they are different than those intended as the ultimate users, is therefore crucial. One determining factor is of course the ownership of the land on which the planting is to take place; this is probably of particular importance in developing countries. In developed countries, and especially in recent years, there has been a move towards greater interaction between the professionals involved in urban forestry (not only foresters, but also landscape architects, town planners, etc.) and urban residents. Thus whether planting occurs on public access or private land, some consultation regarding species choice is desirable between interested individuals. Certainly this approach is recognized as being integral to the concept of urban forestry (Johnston, 1992).

In developing countries, despite the spread of community and social forestry initiatives, decisions on species choice (as in other aspects of urban forestry) tend to take place in isolation. Thus an individual planting a tree on private land is unlikely to consult a professional on the subject. Similarly, local residents are rarely involved in deciding what species are to be planted on public access land. The decision seems to be taken in most cases by whichever body is responsible for implementing the planting, with little discussion with other professionals. Yet there may actually be major differences in opinion as to what constitutes a desirable species in a particular circumstance. A review of available literature revealed no study that investigated, by asking Third World urban dwellers directly, what attributes they value in urban trees. People belonging to different income and ethnic groups might well have different preferences, as suggested by a survey of trees growing in different localities of Bangalore (Gadgil and Parthasarathy, 1977). Unfortunately, the survey did not differentiate between private and public land, and it is not clear that plantings on the latter had been chosen by local residents. However, a clear species difference was recorded between certain types of residential areas. For example, “affluent localities with a strong element of indigenous culture” had a high proportion of indigenous ornamental species, while the “westernized upper class localities” had a high number of exotic ornamental trees. Fruit trees dominated the species assemblage in poorer areas, where few avenue trees existed.

Currently, the choice of species planted on public access land tends to be determined by professionals who see the establishment of trees as their main goal. This is illustrated by the following comment, made by an Indian forester on the subject of tree species suitable for street plantings,

“…It should not bear edible fruits (which would other-wise flood the area in later years with urchins darting from one side to the other). It should not be of fodder value (to keep the cowherd or goatherd from temptation to lop).” (Shyam Sunder, 1985:686)

Potential differences in opinion on species choice for public lands are further illustrated in box 6.2, which hypothesizes how certain attributes might be viewed by a technical forester versus the urban poor in a developing country.

The conflict situation illustrated in box 6.2 is not irreconcilable; it simply reflects a completely different perception of the function of the trees in question. Dialogue could do much to resolve this, as discussed in section five. Ideally, species choice should reflect the needs and desires of local people, as well as the characteristics of the particular site, purpose of planting, availability of seedlings of various species/varieties/cultivars, and future management requirements.

Box 6.2.Desirable attributes for a street tree: possible differences in perspective
attributetechnical foresterpoor urban dweller
edible fruitsnot favoured
  • attracts street urchins
  • fallen, rotting fruit creates a nuisance
  • a source of food
  • a possible source of income
leaves are a nutritious foddernot favoured
  • problems with tree establishment; damage by livestock necessitates extra maintenance costs
  • lopping by people when tree is mature causes unsightly damage
favoured by some
  • any source of fodder is valued to those keeping livestock in the city
shedding leavesvariable
  • shedding leaves are a nuisance on roads and pavements
  • leaf colour change and a short deciduous period give visual interest
  • not valued in trees used as shade (evergreen is preferable)
  • fallen leaves can be used as fuel or possibly livestock bedding
medicinal barknot favoured
  • harvesting will damage the tree
  • free medicine when needed
  • possibly a source of cash income, at least as a medicinal ingredient for faith healers
tendency for diebacknot favoured
  • dieback is unattractive
  • serious dieback may be a public hazard
  • dead twigs and branches represent a source of fuel
  • safety hazard unlikely as long as timely selective removal permitted
indigenousvariable, but not strongly favoured
  • exotics are often more familiar, easier to grow, and more ornamental
probably favoured
  • indigenous trees may be more familiar, especially to those with rural origins
  • indigenous trees may have cultural significance.

Where?: The site to be planted

As noted in section 3.2, the ownership of, or access to, the site to be planted may determine not only who decides what is planted, but also what they chose. Even if local people are involved in deciding what is planted on public access land as well as their own land, experience from community forestry suggests that they will choose to cultivate the most valuable trees only on land over which they have good control. Thus, for example, even if an overall priority were fruit production, one might find hardy, local varieties being chosen for cultivation on public access land, and grafted, less disease-resistant and more nutrient-demanding varieties being chosen for homegardens.

Site conditions
At the technical level, urban environments often pose a variety of particular problems for tree growth. It is important that as far as possible, the type and species/provenance/cultivar of tree chosen reflects a known adaptability to and tolerance of such conditions, some of which are listed in box 6.3.

The nature of urban site conditions is actually highly complex, and the response of trees to them is often poorly known even in countries where much research has been conducted on the subject, such as North America. For example, according to a recent study by Clark and Kjelgren (1990), the response of trees to water stress may be categorized into one of two main strategies (adaption or acclimatization); these entail a number of mechanisms for either maximising water uptake or limiting its loss. However, which category is most appropriate in an urban situation has not been determined, as species which respond well in urban environments are known to belong to both.

In developing countries, very little scientific study has been conducted at all on the response of different tree species or provenances to particular conditions in the urban environment However, surveys of existing urban trees can reveal a large amount of information, and give an indication of suitability to different urban sites (Jim, 1990a). The main drawback is that they are confined to an evaluation of species that have already been adopted. As a result of limited knowledge about species suitability to urban conditions is that, according to Jim (1990b:28),

“The tendency to be dictated by inertia and ‘easy’ species becomes inexorably entrenched. The lack of effective local arboricultural research traps the practitioners in a close circuit of conservatism.”

Box 6.3.       Site conditions common in urban environments, and desirable attributes for trees planted on such sites
site conditiondesirable attributes
soil compaction and poor drainage
  • robust rooting system
  • tolerance of some degree of anaerobic conditions
low soil nutrient levels
  • nitrogen fixation, or other microbial association (eg. mycorhizza)
  • deep rooting
  • low physiological demand for nutrients
low soil water availability
  • a variety of mechanisms to reduce water stress (either based on maximising water uptake or limiting water loss)
soil pollution - eg. saline conditions caused by salt application to roads (in temperate countries); high heavy metal levels, etc.
  • tolerance to the pollutants in question
air pollution - the level of particular pollutants varying in different locations
  • tolerance to the pollutants in question
  • certain general characteristics that render some trees better at removing pollutants from the atmosphere (see box 4.5)
proximity to buildings and other structures
  • rooting system: non-aggressive, deep rooting
  • tree height: short trees probably preferable; large trees may interfere with overhead lines, and provide excessive shade to buildings
exposure to salt spray (in coastal locations)
  • tolerance to salt spray
exposure to strong winds (more likely in coastal locations).
  • no tendency for branches to split and break in strong winds
  • deep rooting system

Landscape planning and design
The way in which planting schemes may be designed to maximise landscape enhancement is a topic beyond the scope of this document. However, the choice of species to be cultivated together is clearly an important one; matters to be born in mind, apart from visual complementarity, include relative growth rates, root competition, shade tolerance, etc. In Europe there has recently been a movement towards an ecological approach to landscape design (Tregay and Ertzgaard, 1979; Tregay and Moffat, 1980). This approach uses ecological principles to determine species choice and management, the result being dynamic, uneven aged plantations. For example, a woodland planted according to ecological principles in Warrington, UK had, after 20 years, an “open canopy of ‘climax’ trees and an emphasis on coppiced shrubs, light-demanding species, ‘edge’ thickets and meadow.” (Tregay and Moffat, 1980).

Future management requirements

In choosing a particular tree species or combination of species for planting, thought should be given to the need for future management. Factors that should be borne in mind include:

Such management issues are discussed further in section 6.2.

Availability of planting stock

The choice of tree to be planted depends ultimately, of course, on the planting stock that is readily available. Although a review of the available literature revealed no study on the source of stock for urban tree plantings in developed countries, it is clearly variable. Planting material may be produced in nurseries run by public bodies (municipal councils or departments such as forestry, horticulture, roads, or others); in private nurseries; or by individuals for their own use (or perhaps distribution among their neighbours).

Trees planted on public access land are often derived from nurseries run by public bodies. This is not always the case, however. For example, in the UK, a general tendency towards privatisation has led to the closure of many local authority nurseries. It appears that in many cities of developing countries, what is raised in public sector nurseries is of limited species diversity and completely determines what is planted, regardless of other considerations. For example, Malhotra and Vijayakumar (1987) mentioned seedling availability as the main factor determining the species planted as street trees in Calcutta, while Benavides Menza (1992) also noted a limited species availability in nurseries of Mexico City.

Private commercial nurseries exist in many Third World towns and cities. As they are often engaged in the production of ornamental and fruit tree species, they may add significantly to the total available choice of species.

In a few Third World cities, the establishment of tree nurseries by self-help groups or Non-Government Organizations (NGOs) has increased the availability of planting material to individuals, or for local community planting programmes. An example is the CONCERN Mozambique project (box 4.8).

It seems likely that for many urban homegardeners, their main source of planting material is that which they produce themselves, or perhaps exchange on an informal basis with their neighbours. In this matter there may be considerable, as yet poorly explored, local expertise.

It appears that, even in cities in which urban forestry is well developed, amenity plantings tend to be of limited diversity. For example, in Hong Kong, ten species make up 70% of all plantings (Webb, 1993). Profous (nd) found in a recent study of five cities of both developed and developing countries (notably Athens, Prague, Mexico City, Beijing and New York) that in each city five species alone accounted for 50 – 65% of all plantings. He also found that some of the most common urban trees, such as black locust and Japanese pagoda trees, have worldwide distribution. It would be of interest to investigate the genetic origin of this planting stock; it is quite possible that many commonly planted amenity trees have a very narrow genetic base, having all been propagated from a few original introductions. Current amenity plantings, according to Profous's study, are at least of wider species diversity.

Much of the reason for the lack of diversity in planting material can be traced back to the limited knowledge about the suitability of different species, as mentioned above. Apart from a limited knowledge base, factors limiting species availability probably include difficulties in seed supply, and unfamiliarity or problems with propagation techniques for “different” species.

Many writers have noted the paucity of indigenous species in urban amenity plantings, particularly in developing countries (Jim, 1991; Benavides Menza, 1992). Low tolerance to urban conditions is often used as an argument for planting exotic species; many indigenous species are said to perform poorly in cities, both in comparison with their growth in rural localities and with the growth of commonly planted exotics (Douglas, 1983; Webb, 1991). However, in many countries the preponderance of exotics is also a reflection of lingering colonial influences and more recent fashions, as mentioned in section five.

6.2. Arboricultural practices

This section largely concerns trees planted for environmental benefits, rather than trees which are primarily cultivated for production. There is a vast amount of general literature, not pertaining to urban environments in particular, on the latter subject. (This is particularly so in horticulture, on the management of fruit and spice producing trees; and in forestry, on silvicultural techniques for timber and fuelwood production, which is beyond the scope of this paper to review). It is in urban tree planting for amenity and landscape purposes that specific arboricultural techniques have been developed.

Many of the general principles of arboriculture are applicable throughout the world, although specific management requirements will be dictated by such factors as the species and climatic regime in question. A poor choice of species can be ameliorated to a certain extent by appropriate arboricultural treatment; for example, a tree which has grown too tall for its surroundings can be pruned. However, this is undesirable for tree health, aesthetically unappealing, and wasteful in terms of maintenance costs. It would have been far better to select in the first place a tree that only reaches a low height on maturity. Important arboricultural principles of general application include site preparation, tree establishment and early maintenance, tree surgery and protection, and the removal of tree waste. Each are discussed briefly here, drawing attention to recent changes in thinking and practice. The extent to which these principles are currently followed in Third World cities is generally uncertain, although mention is made if any information on this subject was available.

Site preparation

As noted above, urban soils are often poorly suited to tree growth. Common problems are low levels of available nutrients, and high compaction due to the impact of human and vehicular traffic. This is compounded by low soil organic matter levels. The ‘traditional’ response to this has been radical site amelioration prior to planting. Current thinking entails a more modified approach, depending on site conditions. In general, alleviation of soil compaction is seen as being more important than other soil treatments. This may be achieved in a number of ways, the most important being subsoil ripping of the planting site. Current arboricultural practice also places far more emphasis on choosing a species to fit the given site rather than modifying the site to fit the desired species.

Tree establishment and early maintenance

Planting stock
It is particularly crucial in urban settings to plant nursery stock of good form and quality, with a healthy root-to-shoot ratio. Saplings without this are unlikely to survive planting in compacted urban soils, but even if they do, they may become a hazard in later life, being more prone than trees with a well-developed root system to being blown over, or to other damage. Similarly, saplings which have damaged stems will grow into trees with an unbalanced branch system; this may not only look unsightly, but could be dangerous.

The size of seedlings or saplings planted in urban situations is often considerably larger than those used in normal plantation forestry. However, the use of seedlings of no more than 60 cm height is now considered by many professionals to be sounder practice. They are often so much more vigourous than larger saplings that they catch up in size with the latter after a short period (Hill, pers. comm.).

Planting techniques
Following site preparation, it is of course important to ensure that planting pits are dug to an adequate size, and preferable that they are prepared well in advance of, rather than at the moment of, planting. It may also be necessary to take into account the existence of underground utility services (water pipes, etc.) when planning where to plant. In professionally conducted urban plantings, particularly in the case of street trees, a variety of techniques may be employed to ensure good tree establishment and to guard against future problems (Webb, 1991).

Watering and mulching
Watering is often considered essential for the establishment of urban trees, but it may be very difficult to provide in some circumstances. One example is provided by tree growing in self-help housing areas where even drinking water is in short supply. Apart from taking the obvious precaution of timing planting to coincide with the beginning of the rains, and ensuring that the pit is at least well watered at the time of planting, various options may be considered. These include the use of drought-tolerant species, regular weeding, the application of mulch, irrigation (possibly using wastewater), and the incorporation at the time of planting of a pipe to facilitate water penetration.

Weed control may be effected either by manual weeding or by the use of appropriate herbicides, both before planting and on a regular basis afterwards. While the use of herbicides in the case of UK plantings is often recommended (Davies, 1987), this may be less appropriate in developing countries for a variety of reasons, both social (employment generation may be desirable) and technical (herbicides may act differently in warmer, wetter climates).

In general, staking is no longer recommended in amenity plantings, as it inhibits the development of sturdy stems and can cause problems when finally removed. Forgotten stakes and ties can also badly damage growing trees. It is now considered preferable to plant seedling/saplings which are small enough to require no support (Patch, 1987).

Tree surgery and protection

Since the mid-1980s, arboricultural theory and practice in the Western world has been strongly influenced, if not “revolutionized”, by the work of the American arboriculturalist, Alex Shigo (Denne, pers. comm.). While his ideas are not universally accepted, or even necessarily new, 1 his practical recommendations are now quite widely implemented in North America, Europe, Australia and cities elsewhere with a strong urban forestry programme such as Hong Kong. Possibly the most celebrated of Shigo's theories is that of CODIT, Compartmentalization of Decay In Trees. CODIT is a model that describes a tree's defense system - its response to wounding. It is used to explain why traditional pruning regimes using flush cuts are bad for tree health, and that ‘Natural Target Pruning’ (NTP), cutting off the branch at an angle (leaving the branch collar, if one exists, intact), is better practice. CODIT theory is also used as an argument against treating wounds with a seal to guard against pathogens. In place of sealants, a variety of gel formulations with systemic fungicidal properties are now more commonly recommended (Clifford and Gendle, 1987).

It is uncertain to what extent the ideas of Shigo are put into practice in developing countries, although it is unlikely that they are widely known or followed (Denne, pers. comm.).2 In general, knowledge and skill in tree surgery varies widely by country. For example, Hill (1992) reports that in Quito, Ecuador, poor pruning practices were attributed to excessive demand for services and poor tools, rather than lack of experience and skill. In South China, however, both Jim (1991:155) and Johnston (pers. comm.) have noted that despite quite high standards of tree selection and planting design, pruning is conducted with very scant regard to tree health. Johnston added that this was certainly not merely a reflection of a lack of suitable equipment.

Timely and efficient tree surgery is one of the most important means of preventing the spread of disease in trees once it occurs, but good arboricultural practice also implies minimising the possibility of pest and disease attacks. Attention to this in the early years of tree establishment, including an appropriate choice of species/provenance/cultivar is clearly important, as is the prompt and complete removal of any infected tree material (which should be duly destroyed).

1 Johnston (pers. comm.) notes that some of Shigo's “revolutionary” practices are actually old practices that have been forgotten. He quotes La Suer (1933), who in The care and repair of ornamental trees (the standard British textbook of the time) recommended leaving the ‘branch ring’ intact in exactly the same manner as Shigo now refers to the ‘branch collar’.

2 Denne also reports, from her experience of teaching overseas students at the University College of North Wales, Bangor, U.K. that in many developing countries urban forestry seems to be given scant attention in forestry curricula. Furthermore, that which is taught appears to be somewhat out-dated.

The removal of tree waste

Key issues in the removal of tree waste are public safety, utilization of removed material as appropriate, and (as mentioned above) limiting the spread of pests and diseases. As indicated in section 4.1, tree “waste” may actually serve a variety of useful purposes, from fuelwood to timber. In North America and Europe, many urban authorities make use of tree prunings by shredding it for compost or mulch, and such recycling is becoming an increasingly important feature of urban forest management.

Inevitably urban trees grow old and must be removed, or parts of them removed, before they cause any damage. Technically, there are often complications in this due to the close proximity of buildings and other urban infrastructure. However, perhaps the most important issue for urban foresters is advance planning - ensuring that removal operations are timed to avoid any public hazard, rather than responding to one.

6.3. Urban tree inventories

The systematic planning of urban forest management is difficult, if not impossible, without sound information on the number of trees within the town/city; their species composition, age, and condition. For example, an urban tree survey of Hong Kong conducted in 1985 found that two thirds of the street trees in Hong Kong and Kowloon had arboricultural problems (Webb, 1991 quoting Jim, 1985). Had the survey not taken place, many of these problems could probably have gone un-noticed for some time to come, with potentially dangerous consequences. Miller (1988) provides a list of questions that the urban forester should ask before deciding that an inventory might be necessary; this is reproduced in box 6.5. Although it relates to an urban forestry service that is well-developed and responsible for arboricultural management on request, it provides useful guidelines for developing countries in which urban forestry is receiving increasing attention.

Urban tree inventories are certainly an increasingly important urban forestry tool in countries where the profession is well-developed. Management Information Systems for ease of information processing and recall have also been devised (Gerhold et al. 1988), as have a variety of methods for putting a value on individual trees. The most commonly used of these, at least in Britain, is the Helliwell method, under which a tree is rated according to six criteria, given a score (1–4) on each, and then all scores multiplied together (Arboricultural Association, 1990). The advantage of this method is its simplicity and apparent consistency. Its disadvantage is its subjectivity and lack of a sound mathematical basis (Price, pers. comm). Alternative valuation methods have been proposed (for example, Matthews et al. 1990), but are not currently in common usage.

Box 6.4Urban Tree Inventories:
Questions to consider when evaluating their necessity
1. How much work is currently needed?
2. Are there a large number of planting spaces available, or are the streets up to full stocking?
3. Is pruning done on a scheduled basis, or by request?
4. How are tree stump removal decisions reached?
5. Are priorities set based on tree needs or other considerations?
6. How is work scheduled and assigned?
7. How accurately are existing needs reflected in current priorities?
8. Do estimates of the volume of work correspond with actual needs of the tree population, or are there discrepancies?
9. Is the public satisfied with the level of service, or are there numerous complaints?
10. Is service performed routinely or by request; if by request, how long does it take to respond?
Source: Miller R.W. (1988) ‘Urban Forestry Planning and Managing Urban Greenspaces’

Computer-based recording of detailed information on urban trees is a recent development of the last decade. The extent to which this level of information management is relevant to tree cultivation in Third World towns and cities is open to question, but a reliable estimate, at least, of current stocking rates, age and species composition of trees on public and private land is necessary for future planning. In most developing countries, such information does not appear to have been collected.

Previous Page Top of Page Next Page