Typically, the design process commences by drawing up a short list of species based on an understanding of different habitat stereotypes (see Tables 6.1-6.6). It is helpful to enter these names into a spreadsheet and reorganise them in terms of flowering season, i. e. spring, early summer; mid-late summer and autumn. This provides an indication of where the non-flowering times are likely to be in relation to likely public expectations of, and use, of the site. It also assists in the planning of colour combinations and the likely structure of the planting. If you find that half your short-listed species flower in spring to early summer, then you can compensate by adding additional later-flowering species.
At the end of this initial design phase, it is necessary to approximate the percentage of planting sites within the community to be created that are to be occupied by species a, b, c and so on. This allows you to refine the appearance and function of the planting mix. It is, of course, axiomatic that you can only do this if you have, or are prepared to develop, an in-depth understanding of the plant materials. You may, for example, have identified the 3 m tall yellow daisy, Coreopsis tripteris, as a late autumn component. If you decide to use this species as an occasional dramatic emergent, rather than a block that will obscure views through and across the planting, then you probably only need occupy 5% of the planting spaces with this species.
In a dry meadow planting you may have decided that the peak flowering display is to occur in late June and July, however you want to have some spring colour. As a result, you limit your spring flowering species, to blue Ajuga genevensis, yellow Primula veris, and acid yellow Euphorbia polychroma, and to no more than 10% of total planting spaces for each species. To create the required drama in midsummer you occupy 30% of your planting spaces with Salvia nemorosa, with a further 20% to Centaurea orientalis and 10% to Euphorbia seguieriana subsp. niciciana, and so on until all the planting spaces are allocated.
On a more subtle level, this approach allows you to plan the ecological structure of planting. For example, you may wish to have a distinctive spring flowering ground layer composed of species such as Ajuga and Primula, through which later flowering layers of plantings emerge. If this is to work, the shortest layers have to be extremely shade – tolerant (as in the examples given) or they will be eliminated by the later flowering, generally taller, species. If this is what you desire (it may assist in preventing an invasion of undesirable natives as well as providing some winter greenery), then the percentage of Ajuga and Primula will have to be increased from the previously identified 10%. Where shade-tolerant ground layer species spread rapidly by self-sowing, stolons or runners, lower percentages may be satisfactory.
In relatively open-plant communities, such as dry meadow and steppe, the lower layers in plantings may avoid being shaded because even the tallest species are relatively short and are widely spaced. Consequently, in these types of plantings, low-growing summer flowering species, such as Thymus and Sedum, may be used as a ground layer.
You can work out the approximate number of planting spaces in an area by deciding on a notional planting grid of 200-500 mm, giving between approximately 4 and 16 plants/m2. The grid spacing used will depend on the size and spread of plants, both at purchase and in the longer term, how closed a canopy is required, and how quickly. Closer spacings will generally produce a more weedresistant vegetation. For an average site of moderate productivity, 9-10 plants/m2 is a typical planting density. It is, however, important to re-state that where you design plantings that include plants of very different growth habits and rates, i. e. fast and slow, many of the slow-growing, shade-intolerant species will eventually be eliminated at these densities. Plant selection needs to ensure that most plants in a community are of similar growth rate and size. Despite adopting a rather confusing plant
On highly productive herbaceous plants, deep mulches of organic debris can be spread over the top of plantings to suppress weed colonisation without having an adverse effect on the desired species
With stress-tolerating species, crushed rock or gravel mulches are less effective but more visually and functionally appropriate
sociology based approach, Hansen and Stahl (1993) is a useful source of information on plant compatibility in planting.