Designing Urban Pavement Sub-Bases to Support Trees

Design and Installation Methods

In Denmark, more than 800 trees were integrated into pavements at 30 different sites using a load-bearing matrix or different sand mixes. A number of matrix materials were used, and 3 mixing methods have been developed.

Sand mix

The Dutch experience has shown that uniform graded sand can be used to expand the rooting volume beneath pavements. The sand is mixed with up to 4% or 5% of humus in the form of compost or sphagnum. In Denmark, the sand mix has been used and installed as described by Couenberg (1994) (Figure 1). The intention of using sand mix is to achieve a compaction level of 70% to 80% proctor density (Couenberg 1994). Therefore, penetration resistance is measured to ensure that the degree of compaction does not exceed the critical limit for root growth. This quality check is not practiced in Denmark.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Use of sand mix at The Christians Havns Square, Copenhagen. The material is used in a 3-m wide, 57-m long pit, in which 7 linden trees (Tilia tomentosa) were planted. The soil in the planting pits is separated from the sand mix by a wire basket during the installation process.

Technical Construction of Planting Site

When topsoil and other materials containing humus are installed under pavements, there is a risk that the obstruction of aeration could cause anaerobic conditions. This risk is increased by insufficiencies in drainage (e.g., if the soil is compacted) (Harris 1992). Sufficient air movement in the soil or sand mix is achieved by installing a layer of stones or broken stones between the root growth area and the pavement above. The aeration layer receives air via pipes or the planting pit (Urban 1989) and should be covered with a suitable geotextile to prevent filling from the layers above (Figure 2). The same type of aeration layer is recommended when raising the grade around existing trees (Harris 1992). Although the effect of an aeration layer has not been documented in the Danish examples, it is considered to have a beneficial effect on the soil and the root growth (Smith 1995). The aeration layer may also be used to irrigate the trees (Urban 1989).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

Pavement construction with load-bearing matrix and aeration layer.

Applications in Denmark

More than 800 trees have been successfully planted since 1991 on more than 30 construction sites in Denmark using these alternative materials and methods.

Table 2 shows that stones or broken stones, in various size classes, are the most commonly used materials for load-bearing matrixes and are used with more than 75% of planted trees. Premixing before installation is the most frequently used installation method. Dry mixing during installation is only occasionally used. Each installation method has advantages and disadvantages (Table 3).

The major disadvantages of using the premix of stone and soil is that separation might occur during transportation and installation. A tackifier can help the soil adhere to the stones (Grabosky and Bassuk 1995). The calculation of mixing ratio is also problematic. In a stone and soil mix, a calculated void percentage (e.g., 25%) has often led to a recommended mixing ratio of 3:1. In one case, this ratio led to compaction of the soil in the voids, which caused subsequent ettling. The correct mixing ratio should have been 4:1 to secure 25% voids in a stone and soil mix. Dry mixing has been the most effective and practical method. However, the dry mixing method requires dry weather, dry stones, and dry soil.

Measurements of soil densities are included in the original Dutch method, in which sand mix was used (Couenberg 1994). Although the degree of compaction is intended to be optimized for plant growth, there is a risk of subsequent soil compaction of soil and settling of the pavement.

Conclusion

During the last 5 years, soil mixes, sand mixes, and load-bearing stone have been introduced for use in Denmark. Most landscape projects involving trees in paved areas use these new methods. No difficulties with load-bearing capacities, irregularities, or frost-heaving of pavements have been recorded. Compared to normal Danish conditions, the winter of 1995-1996 was long and cold, with continuous frost from the start of November to the middle of March. A top layer of at least 0.5 m (20 in.) was frozen, which influenced all layers of load-bearing mixes installed below pavements. None of the methods applied appear to have higher susceptibility to frost heave than does traditional pavement construction.

To reduce the visual effects of possible irregularities caused by heaves and settlements, landscape architects often changed the pavement on the load-bearing matrix materials so that it differed from the rest of the pavement. As a result, the possible height differences are less conspicuous than would be the case with uniform pavements. In addition, the change in pavement shows where the tree has its roots.

All examples presented in this article were established between 1991 and 1996. None of these trees have been measured for growth rates. Fast and accurate results were obtained by conducting growth experiments with 8 materials and 3 tree species beginning in spring 1994. These results illustrate the growth capacities of the different materials (Kristoffersen 1998).