Why Planting Plans Don’t Always Work

James Hitchmough

Much debate around sustainable planting focuses, across the world, on the issue of the native v the exotic, but in actual fact this is often a red herring. Far more fundamental is the question of ‘how do we design for the change that must occur in all planting?’

Change in vegetation is driven by two factors. Firstly the external; the response of plants to the environment in which they are placed and how they cope with this. Secondly, the internal factors; the response of plants to their neighbours. For any plant, the outcome of these interactions lie across a sliding scale from death to dominance, and everything in-between.

These processes have huge impact on the aesthetic perceptions of landscape users. If one does not adequately consider these factors as part of the design process, most plantings other than the simplest imaginable – trees planted at wide spaces in grass or in a hard surface – end up becoming something you didn’t want. Do we have the resources to design it again as part of an eternal groundhog day?

Why planting plans don’t work

The difficulty in conceptualizing and responding to change in designed vegetation is that it is continuous and fine grained. We don’t really notice what’s happening, and even if we do its difficult for our brains to process, until we have arrived at the consequences of these accumulations, by which time it is often too late. We do not help develop our capacity to imagine change when the outcome of most planting design thought is a conventional planting plan. These are convenient devices, but tend to fix our thinking in some future spatial realm, through entrenching the static, inviolate nature of the territories of our chosen species.

Learning from nature

Is it possible to design planting to build in capacity to respond to change in a more nuanced and clever way? Whilst planted blocks of species A and B are easy to conceptualise for both design and maintenance, they are fundamentally flawed as models for dealing with longer term change and human perception of the acceptability of that change. The species age at different rates, short lived, long lived, so bits of the composition fall apart at different times. Once species A comes to the end of its existence it is generally replaced by urban weed species of dubious desirability. If this is the inevitable end point then why bother with the preceding design phase? This descent into wretchedness is not something you typically see in nature where plants exist as multi-species communities.

Are designed communities the way to deal more elegantly with the change in landscape architectural vegetation? Change is just as relentless in a naturally occurring plant community, but you don’t really notice it. This is because every square metre is occupied by a variety of different species competing for light and space. The long-lived species tend to be slow growing, the short lived species fast. The fast initially dominate. If there are not too many of these per unit area, the long-lived species gradually take over as the fast species wane. The long lived species are themselves rarely immortal, but if they can produce seedlings to replace themselves, they can appear so.

Building the planting machine

I like the idea of planting as a sort of self-assembling, quasi-immortal machine. I have spent much of my professional life thinking about how these things work in semi-natural communities, and tinkering through research and practice with how to apply these ideas to landscape architectural practice. Just imagine a planting that is initially in sun, but contains trees so gradually becomes more shaded. If the designed plant communities contain a diversity of both shade and sun tolerant species, the former will gradually dominate as the latter decline, often quite elegantly. There is no need to guess where the shade might be at year 5, 10 or 20 years, so as to tether plants to a fixed point in space in a planting plan. You will in any case get this wrong. Or you can build in the same resilience to changing gradients of moisture stress, as in a drainage swale. This is neat; self-fixing vegetation “skins” without overly heavy reliance on human agency in the form of maintenance and often does not turn up.

The problems of predicting the future

The challenge however is how does one know what the machine will do after you push the start button? What it will become? The answer to this is that one cannot fully know this. It is possible to know the general direction of travel, but there will always be two or three possible destinations.   This is unsettling, although we are perfectly happy day in and day out designing vegetation of which we have little or no idea of the possible destinations.

In designing these types of ecological machines, I use my experience of observing plant species as both individuals and aggregations in nature and in designed plantings, to try to predict the probably outcomes. I then use research methodologies to see if my guesses, informed yes, but guesses non-the-less, are anywhere near the mark. We construct plant communities with species in differing proportions to their neighbours, and subject them different management regimes, and record what happens over the next 5-10 years.

I mainly work with herbaceous vegetation, which is relatively small, does things quickly (well relatively so), and fits into the size of spaces we have for research. The processes of competition etc., which we are trying to understand are however largely scale independent, so the ideas can be just as well applied to a woodland. Colleagues at SLU Alnarp in Sweden have been working in parallel on designed woodlands for the last 30 years.


The magic of spreadsheets

But what about the planting plan? Communities, by their very nature, are more likely to involve species arranged, at least initially in random patterns, so planting plans are only useful for the few largest species, who by virtue of their superior size and hence light capturing capacity, will persist in the longer term, like the trees in the park. Whilst doodling is good, ultimately you really need spreadsheets to iterate the proportions and densities of each species in the planting community. David Thompson of LDA Design once referred to my doing this at the London Olympic Park my “magic”. Much as I like the idea of being a magician, the actual process is anything but.

The problem with designed plant communities is that however clever they might be negotiating change, they do not appear humanized. Unease of disorder is widespread in the urban world, although if the flowering display of a community is dramatic enough, for long enough, this acts as a powerful salve for many. You really do need to read my colleague Anna Jorgensen’s blog on urban wilderness…..

james   James Hitchmough

designed ecology   Designed Ecology

2 thoughts on “Why Planting Plans Don’t Always Work”

  1. I think it is possible to design with blocks (in large-scale landscapes) which don’t fall apart but gradually cross boundaries over time (as in tall herb perennials). Or if the designer is skillful enough, subordinate species which may tolerate drought better or different soil nutrients etc., can come to the fore when conditions do not suit the tall, nutrient demanding forbs or can occupy spaces withing the larger blocks.

    I think change is more relentless (and perceived to be) in highly designed ‘artiface-type’ plantings. Or with designed annual plantings which can look very scrappy after the first year (weeds come in quickly into the gaps). Although multi-species annuals are a good foil until the perennials start to spread and assume visual dominance. Strongly growing clonals (rhizomes & runners) are best here perhaps.
    Getting the densities of species right is the hardest thing (in nature this is often achieved with seed dormancy or animal transported seed).
    When woody species are included in the design, as you point out the shade tolerant forbs will eventually take over, so the skill will be in selecting those that can lie low until their time in the shade arrives (i.e. act as subordinates) and they should also be able to tolerate light during this time.
    Resilience therefore, must be built into the scheme for this reason and if a major (of minor) disturbances occur, then things really happen!
    You are the only designer who comes close to these ‘magic’ formulae.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: