First part of a series from Growing Today, July 1983
by R.L. Hathaway
National Plant Materials Centre,
Provision of adequate shelter from wind is essential for the successful production of most horticultural crops in New Zealand.
Good shelter not only results in improved yield and quality of crops; it often determines whether or not the crop can be grown at all.
A large proportion of horticultural crops are grown in coastal areas, which are often more windy than inland areas. Most areas in N.Z. are affected by the prevailing westerlies, which lower temperatures and increase evaporation and transpiration losses. These factors result in reduced growth rates, lower yields, and lower sugar levels in fruit crops.
Occasional very strong winds from the easterly quarter can result in serious physical damage to crops and trees.
Briefly, the benefits of shelter are:
- Minimising physical damage to leaves, blossom, fruit, branches etc.
- Improvement of microclimate, with higher temperatures, higher humidity, and reduced evaporation and transpiration.
- Improvement in pollination by bees which are encouraged to work more often.
- Increasing the range of crops that can be grown.
The greatest sheltering effect for a given input of resources will only be achieved if an integrated system is planned and implemented. Consider both present and future requirements of the crop, as well as management and shelterbelt replacement. The cumulative effect of shelterbelts on wind reduction is often underestimated but is an important factor in reducing wind velocities.
Shelter system design will depend to a large extent on the crop to be grown. The first step is to decide how much shelter is required by that crop to ensure a good crop performance. In other words, what level of exposure can the crop withstand. This will depend on several factors:
- Susceptibility of the crop to wind damage. This can include mechanical damage, such as branch breakage, fruit rub, desiccation of foliage etc. For example, kiwifruit is much more susceptible to wind damage than grapes or vegetable crops and requires a much greater degree of shelter for economic production.
- Frost susceptibility of the crop. Highly sheltered areas are more prone to cold air ponding; this can result in more severe frost damage than would occur in areas with adequate cold air drainage.
- The period during the year when wind protection is required. Some crops need protection in early spring (e.g. stone fruits) or late autumn and early winter (e.g. kiwifruit and tamarillos). This affects the selection of shelter species to be included in the system, particularly with regard to deciduous versus evergreen species.
- Whether the crop is wind or insect pollinated. Wind pollinated crops require a certain level of air movement for good pollination. However, with insect pollinated crops a higher level of shelter is desirable to encourage bees to work. Overseas work has shown that fruit set in pip fruit crops can be considerably greater in the most sheltered area of the orchard (e.g. Smith and Lewis, 1972).
- The self sheltering ability of the crop.
This factor applies mainly to larger trees, the effect being quite small initially, but increases as the trees approach maturity.
- The degree of shelter required will depend to a large extent on the exposure and windiness of the site. For example, in more windy areas of New Zealand it may be necessary to reduce mean wind velocities by 50% or more to obtain an acceptable level for the production of many horticultural crops, while in less windy areas a reduction in mean velocity of 25% may be all that is required.
Shelterbelt performance is directly related to its height, length, and permeability.
HEIGHT: The distance downwind over which a shelterbelt is effective is proportional to its height.
Some reduction in wind speed, leeward of the shelterbelt, can be detected up to a distance of 20 times the height of the windbreak.
LENGTH: Where the wind strikes the shelterbelt at right angles a continuous length of at least 24 times the height of the shelterbelt is regarded as the minimum. There is an increase in wind speed around the ends of and through gaps in the belt.
PERMEABILITY: The relationship between permeability, extent of sheltered area, and reduction in wind velocity has been the subject of many investigations and is still debated. It is generally accepted that Fig 1 approximates the wind profiles of shelterbelts of different densities, although other factors such as the flexibility of the barrier, windward surface roughness and atmospheric turbulence can affect profiles.
Shelterbelts of medium density result in the greatest sheltered area, although wind reduction very close to the shelterbelt is not as great as with dense or very dense belts. Very dense belts result in turbulence in the lee of the belt which can sometimes cause greater damage than unhindered wind flows. This turbulence is caused by wind shear at the top of the barrier, where the air flow is deflected sharply upwards, and the velocities of the underlying layers are rapidly reduced.
The upper layer of air quickly descends to mix violently with the lower layers. To decrease turbulence, permeable shelterbelts are required; it has been found that medium dense belts with a permeability of 40-50% are generally the most satisfactory. In addition, vertical changes in permeability should be gradual.
From Fig 1, it can be seen that a moderately dense shelterbelt (40-50% permeability) will give a wind reduction at ground level of 50% at a distance of 8H from the shelterbelt, where H is the height of the shelterbelt. The height of the shelterbelt should be taken as the height above the top of the crop. Thus to achieve a wind reduction of at least 50% over a planting of 5m tall fruit trees, with shelterbelts 15m tall, the shelterbelts should be spaced no further than 80m apart.
Alternatively shelterbelts 10m tall spaced 40m apart would provide the same degree of shelter.
If a wind reduction of 25% is required, 10m tall shelterbelts placed 65m apart, or 15m tall shelterbelts placed 130m apart would be satisfactory. If a wind reduction greater than 60% is required, dense belts placed no further apart than 25m for 10m tall belts, or 50m for 15m tall belts are required. For taller fruit trees it is obvious that shelterbelts must be either taller or more closely spaced.
For kiwifruit shelter, the current Ministry of Agriculture and Fisheries recommendation is 35m between north-south shelterbelts (i.e. giving protection from easterlies and westerlies), and 80-120m between north-south shelterbelts. Close spacing between north-south shelterbelts is essential in some districts to obtain adequate protection during the establishment period, a distance of 50-60m between shelterbelts 10m tall should be satisfactory in less windy districts.
Having determined the optimum height, spacing and permeability for shelterbelts in the system, consideration must then be given as to how this can be achieved.
Main shelterbelts for protecting horticultural crops should be oriented in a north-south direction to minimise shading. Where prevailing winds are from the west or damaging easterlies occur, as is the case in most horticultural districts, this arrangement can often be achieved. Long narrow blocks provide the most protection, with the long dimension at right angles to the wind.
Where protection is required from winds which strike the north-south shelterbelts at an angle, shorter blocks with more frequent east-west belts are required to prevent funnelling of the wind down the length of the block. It may also be necessary to place the north-south belts closer together than is required where the wind strikes at right angles to the shelterbelt.
Gaps for access through shelterbelts should be avoided as far as possible in north-south shelterbelts, or in those exposed to the most damaging winds. The positioning of access points is more critical where the wind does not strike the shelterbelt at right angles. In this situation, access points in east-west belts should be positioned on the most sheltered side of each block.
In many situations, the orientation of property boundaries, topographic features, the presence of tile drains, electric power or telephone lines or irrigation races prevent the adoption of the ideal layout and a compromise must be made. Tile drains are a difficulty encountered on many properties, and roots of many species, not only poplars and willows, will cause blockages.
Undulating land presents further problems, and the placement of shelterbelts for optimum effect on this type of country is much more difficult. On gently rolling land, wind flow follows the contour fairly closely, thus shelterbelts should be placed on ridges and upper slopes to take advantage of the extra height obtained.
On steeper land, turbulence, and eddying can be much more severe than on flat country, and no general shelterbelt pattern can be recommended. Intersecting shelterbelts across valleys are often necessary to limit funnelling. Advantage should be taken of any naturally sheltered areas.
If drainage is present or planned, a section of sealed pipe is necessary at any position where a shelterbelt crosses a tile drain. Where drains run parallel to shelterbelts, these should be at least 8m away from the trees. Lateral root growth can be controlled by deep ripping alternate sides of the shelterbelt every second year.
Double-row shelterbelts comprising a fast growing species for tall shelter, such as radiata pine, poplar, willow, or eucalypt, in combination with a slower growing species, generally evergreen, to provide bottom shelter, and to reinforce the sheltering effect of the fast growing species in time, are the most satisfactory for main north-south shelterbelts. In addition, the second row provides a safety factor should any deaths occur in the taller species. The slower growing species should be chosen carefully to be fully compatible with the faster growing species.
It needs to be either quite shade tolerant, or else planted sufficiently far away from the fast growing species not to be suppressed (1½-2m).
Side trimming on the inside of the fast growing species may also be necessary. The slower growing species should always be planted on the northern side of east-west shelterbelts, and on the windward side of north-south shelterbelts, to minimise over-topping.
The use of dense evergreen species such as radiata pine or Leyland cypress in east-west shelterbelts may be undesirable because of the heavy shade they cast during winter when the sun is at a low angle. In this situation single rows of deciduous species are preferred.
Internal north-south shelterbelts consisting of single rows of deciduous species can sometimes be satisfactory in areas where wind velocities are not high. However, it is likely that they will become open at the base in time, and may need underplanting. In general, at least every second north-south shelterbelt should be a double row.
In some districts boundary shelter may require three or more rows to break the full force of the wind. Where unproductive land is available on boundaries, it may be possible to establish small woodlots of tall growing species, particularly on southern and western boundaries.
Temporary shelterbelts of fast growing poplar or willow species may assist in establishing the crop trees earlier than would otherwise be possible. These can be planted in an intermediate position between permanent shelterbelts, and removed at a later stage when the permanent shelterbelts are tall enough to shelter the total area. A further row of crop trees can then be planted in their place.
Deciduous species can sometimes be included in the same row as evergreen species to obtain the required permeability, or to allow for cold air drainage on frosty sites. It is however difficult to avoid competition effects between the two species unless the evergreen is particularly shade tolerant (e.g. Cryptomeria japonica).
(Density can be controlled by side trimming and pruning. In areas subject to severe frosts, it may be necessary to cut holes in the bottom of downslope shelterbelts which have become too dense, to allow cold air to drain away.
In many of the more exposed districts of New Zealand now being considered or undergoing horticultural development, tall shelterbelts of satisfactory density throughout can only be achieved by planting two rows of trees. Single-row windbreaks of species such as Populus Flevo, Salix matsudana, or Casuarina species may appear to be moderately dense under light wind conditions, but when subject to strong winds, large gaps can appear which increase the permeability from 40% to nearer 80%.
Satisfactory density can be achieved by single rows of evergreen species such as Leyland cypress which are regularly trimmed, although these can often become too dense.
It can be seen that in planning a shelter system, many compromises must be made, especially where live trees are used. Because of the variability in site conditions, and differing requirements of many horticultural crops, there can be no single design to suit all situations. Artificial shelter can minimise many of the compromises which must be accepted with live shelterbelts, but at the expense of very high cost, including the possibility or replacement after a relatively short period.
Species selection, irrigation requirements, and the use of artificial shelter are also important aspects to be considered in detail at the planning stage.
Caborn, J.M. 1965: Shelterbelts and windbreaks. Faber & Faber Ltd, London. 288p
Smith, B.D. and LEWIS, T. 1972: The effects of windbreaks on the blossom-visiting fauna of apple orchards and on yield.
Ann. Appl. Biol. 72: 229-238.