Landscape Maintenance Without Cultivation

Paper given to Horticulture 100, 1994, © Muriel Robinson 2008

Summary

The goal of inexpensive, chemical-free weed control is unattainable with present knowledge.

The use of soil acting herbicides is being questioned increasingly on environmental grounds. Greater use is likely to be made of glyphosate and other non persistent herbicides that have little or no effect on the environment. However repeated use of glyphosate (or other herbicide) over a number of years is likely to result in a build-up of resistant biotypes.

Weed smothering plants are often aggressive and can become as difficult to control as the weeds they were meant to suppress.

Mulches are effective if used properly but are expensive and often have unexpected side effects. Digging out weeds that will eventually grow through a mulch will quickly impair its weed suppressing properties as soil and weed seeds will be brought up into the germination zone. Better understanding of the properties of mulches will enable them to be used more effectively and economically. Wood chippers are likely to become more popular as a means of providing a relatively cheap organic mulch in situ from tree and shrub prunings.

In the future greater emphasis will be placed on an integrated approach using combinations of methods, such as mulching plus the judicious use of those herbicides with minimal effect on the environment or herbicides plus physical methods.

Introduction

It is surprising how some people still hanker after cultivation -good clean cultivation - as if this was the ideal way to manage amenity plantings. Yet plants and soil coexisted for millions of years without cultivation and many modern problems, such as water run-off, erosion and soil compaction only became widespread when man began to cultivate the soil.

The main problems caused by soil cultivation are root injury, the encouragement of further weed seed germination and damage to soil structure. Most plants tend to root prolifically in the surface soil layers and damage to surface roots, even by shallow, careful hoeing, is much more severe than most people realise. The roots damaged by cultivation are the most effective roots - those growing in the fertile upper soil layers. Moreover, cultivation not only damages plants directly by reducing the plants ability to absorb mineral elements and water but also indirectly, as the plant uses up energy reserves in compartmentalising wounds caused by cultivation.

Cultivation encourages weed seed germination both by creating a favourable seed bed and also by bringing up viable, but dormant, weed seeds from lower depths into the germination zone.

After cultivation, soil structure looks good to the human eye. This is because hoeing and cultivation bury the surface soil layer, which has been degraded by passing traffic or rainfall, and brings up non compacted soil. Many experiments with fruit crops have shown that, where the soil is not cultivated, a crust at the surface does not continue to deteriorate annually. After the first year, structure-forming processes, such as frost action and cycles of wetting and drying, act on the degraded surface. The structure of the soil below this compacted surface layer is better than that where land is cultivated because of the absence of soil disturbance and more continuous pores.

There is abundant evidence therefore that 'non-cultivation' is good management practice, particularly in view of the high cost of manual labour. The major imponderables at present are how best to control weeds and how to balance environmental concern with practicality and cost. No one solution to the ubiquitous weed problem will satisfy everybody. Similarly no single method can solve the problem because of the versatility and adaptability of weeds.

Potential 'non cultivation' methods of weed control include the use of mulches, thermal techniques, biological agents, ground cover plants and herbicides.

Mulches

Weeds can be successfully controlled with plastic mulches and with organic materials provided they are applied at a suitable depth. The depth or thickness necessary to control weeds will depend on the type of material used. A depth of about 100 mm when settled is required for straw (Bushnell and Weaver 1930), 75 mm for bark (Campbell-Lloyd 1986) and about 50 - 75 mm for sawdust (Rowe-Dutton 1976). A deep mulch (100 mm) of hardwood bark chips, sawdust or crushed corn cobs gave more effective control than a shallow mulch (50 mm) of the same materials. Provided a 75 mm layer of bark mulch is properly applied, 95% weed control should be achieved over a three year period (Campbell-Lloyd 1986).

Apart from the smothering effect of mulches, chemicals in the materials used can also affect weeds. Phenols and tannins in coniferous bark and sawdust improved the degree of weed control and reduced cultivation costs compared with chopped bark and straw and a mixture of chopped shrub waste and limestone gravel (Kolb et al 1985). Similarly better weed control was obtained in roses with sawdust than with rockwool (Paterson et al 1979). Although organic mulches can be effective against annual weeds they usually have little effect on established perennials.

Perennial weeds can emerge through deep layers of organic mulch and, if present when the mulch is applied are likely to thrive because of the absence of competition from annual weeds. Opaque plastic materials give excellent control of annual weeds and are more effective than organic mulches against perennials. However, some of the more aggressive perennial species, or those with sharply pointed shoots such as couch grass (Elymus repens) can penetrate thin polythene (38 micron) film (Davison 1983).

Despite the general efficacy of mulches against weeds, problems can arise with both opaque synthetic and organic mulches. If plastic mulches are torn, accidentally or to assist water penetration, weed growth through the holes will be vigorous (Rowe-Dutton 1976). Even with undamaged plastic vigorous weed growth can occur at the gap around the stem or stake.

Although transparent plastic film results in higher soil temperature than opaque film and crop growth may be enhanced initially (Agulhon 1975), the vigorous weed growth that occurs beneath clear plastic is a severe limitation. Consequently transparent films are little used for mulching perennial plants. Some organic mulches, such as coir, spent mushroom compost, composted bark and sewage sludge, break down quickly and provide a good substrate for weed growth (Insley 1981). Finely pulverised grades of bark, in particular, tend to be colonised rapidly (Campbell-Lloyd 1986). Moreover, the weed problem may be increased by weed seeds introduced in certain mulches, such as fresh manure or hay cut when seeding.

Although mulching is generally regarded as a good horticultural practice, undesirable side effects can also occur, so that mulches must be used judiciously. The value of mulching, like many other horticultural practices, depends very much on local circumstances, such as soil, climate and the type and growth stage of the plants to be mulched. As far as possible, clear specifications should be produced both with the regard to the mulching material and also the way it is used.

It is not surprising therefore any advice given on the use of mulches must contain many qualifiers. For example a peat mulch can be particularly beneficial to rhododendrons and many other ericaceous species (Harig and Witt 1984), but is some circumstances it can be detrimental by absorbing and holding moisture and releasing it again to the air by evaporation. 

Organic mulching material is bulky and costly to transport. Consequently organic mulches are likely to be of greatest use in situations where near perfect weed control is required, unless material is available cheaply as a local waste product (Rowe-Dutton 1957). Bark has become an important mulching material in the last decade but a wide range of other materials are also used. These include organic substances such as pine needles (Paterson et al 1979), leaves, sawdust, straw, grass clippings and corn cobs (Gartner 1978); mineral material such as sand, gravel, stones and granite chippings (Wright 1985), rock wool (Paterson et al 1979) and manufactured materials such as plastic, paper, glasswool, metal foil, cellophane and urethane foam (Creagur and Katchur 1975).

Proprietary mulches have also been developed e.g. strawdust consisting of resin impregnated granules of wheat straw. This material is long lasting, sterile and contains a slow release nitrogen fertiliser (Tuefel 1983).

Unlike herbicides, whose only function is weed control, mulches affect plant growth in a number of other ways, such as the prevention of loss of water from the soil by evaporation, reduction of temperature fluctuations, light reflection and prevention of erosion.

Soil moisture

By providing a protective barrier at the soil surface a mulch reduces water evaporation (Scholl and Schwemmer 1982). Suppression of weed growth also reduces moisture loss through transpiration. This enables the important surface soil layer to remain moister for a longer period than an unmulched soil surface.

Water penetration into the soil may be improved by certain mulches such as straw. This is due to a number of factors including the protection provided by the mulch against rain impact at the soil surface, reduced soil compaction (Ingle 1981), and probably to increased activity by soil fauna.

Where moisture conservation is a major consideration, organic mulches such as bark, straw or wood chips will be preferable to plastic as they tend to retain higher moisture levels (Ashworth and Harrison 1983). However there are large differences between organic mulches in their moisture conservation properties, straw being more effective than manure with hay and wood shavings intermediate (Harris and Yao 1923). The moisture holding properties of any individual material is affected by its physical condition. Finely ground bark with particle sizes up to 25 mm, retains more moisture than coarsely ground bark (75 mm) with a medium grind (50 mm) intermediate (Gartner 1978).

Impervious plastic and other solid synthetic mulches may prevent water penetration and cannot be laid over large areas without some means of enabling water to reach the soil, such as perforating the film. In addition the high temperatures that develop under clear plastic will also tend to reduce soil moisture levels (Ashworth and Harrison 1983). Plastic formulations are now available that allow moisture and nutrients to pass downwards, but are still effective in preventing weed growth.

Higher soil moisture does not always benefit plant growth. On low lying, poorly drained sites or on soils that do not dry out quickly, excessive moisture under mulches during wet springs, may result in root asphyxiation (Agulhon 1975). Impeded drainage and nitrogen deficiency were suggested as the likely causes of the better response of the moisture-loving alder than sycamore and lime to mulches of bark, sewage sludge and press board sheets (Insley 1981). 

Soil temperature

Each mulch type creates its own unique soil temperature regime and big differences have been recorded between organic and plastic mulches. Under organic mulches soil temperatures are lower during the day and slightly higher at night than those in bare soil (Paterson et al 1979). Temperature fluctuations are therefore reduced (Ashworth and Harrison 1983) but organic mulches have less effect on seasonal temperature variation (Bredell and Barnard 1974).

The effect on soil temperature will depend on the depth and thickness of the mulch material. A deep mulch (100 mm) of hardwood chips, sawdust, jointer curls or corn cobs had a better insulation value than a shallow 50 mm mulch (Gartner 1978).

Dark coloured mulches such as black polythene, absorb the sun's rays and may increase soil temperature considerably especially during sunny periods. In a trial in Wisconsin , plastic mulches resulted in significantly increased soil temperatures at 100 mm depth compared with organic mulches (Ashworth and Harrison 1987). Higher soil temperatures were also recorded under black plastic than under bark or hay and the differences were greatest early in the season (Litzow and Pellet 1983).

Effect of mulches on soil nutrients and physical properties

Different mulches will affect soil fertility and structure in a variety of ways. In a comparison of nine mulches, including straw and synthetic materials, the soil under bark showed the highest pH, organic matter content and potassium levels (Ashworth and Harrison 1983). Bark mulch also gave a greater increase in potassium, calcium and magnesium than sawdust or corn cobs (Gartner 1978). Spent mushroom compost is a rich source of nutrients, an application of 20 tonnes per acre will provide 36 kg nitrogen, 34 kg phosphorus and 158 kg potassium per acre (Maher 1990).

Plants of Spiraea japonica mulched with conifer sawdust were smaller in size than unmulched plants after six years, but growth of both unmulched and mulched plants was greatly improved with applications of ammonium sulphate (Kolb 1984, Karbe 1984). The growth of rhododendrons was temporarily improved by a mulch of fermented bark while their growth was adversely affected by unfermented bark throughout a three year study period (Harig and Witt 1984). Better growth of trees and shrubs was recorded with a combination of bark mulch plus nitrogen fertiliser than with fertiliser or mulch alone (Whitcomb 1978).

Apart from the effect of mulching material on nitrification, plant growth may also be affected positively or negatively by chemicals in the mulching material. The growth of seedlings of Sitka spruce and Lodgepole pine was improved by the nutrients washed out of a mulch of bracken fern applied to nutrient deficient acid heath soil and removed before the tree seeds were sown (Holmes and Faulkner 1953).

Plant growth and crop yield 

In view of the complex effect of mulches on many different soil functions, it is not surprising that many different types of mulch affect plant species in different ways and that no single type of mulch performs best in all situations or for all species (Ashworth and Harrison 1983). Good response in terms of increased plant vigour has been obtained with a wide variety of different plant materials including pine bark (Insley 1981), hay, black plastic, calcined clay (Litzow and Pellet 1983), heavy duty green plastic (Ashworth and Harrison 1983), bark and sawdust (Kolb et al 1983), and plastic and gravel mulch (Werken 1981).

Mulches of black plastic (38 or 125 microns thick) have given consistently good results on woody ornamentals without adverse effect on any plant species (Davison 1982). In these trials polythene-mulched plants made more growth than those kept weed-free with herbicides or with hand weeding. However, Whitham (1982) obtained a strong correlation between growth of Eucalyptus and weed control with most mulching and herbicide treatments tested. In this work, mixtures of simazine and aminotriazole gave greater growth than pine bark, grass hay, sawdust, black plastic, scoria, newspaper and hoeing.

In Britain, a mulch or a combination of mulch plus herbicides often gives higher yields of apples, blackcurrants, raspberries and nursery stock than the use of herbicides alone (e.g. Davison and Baily 1980). However, in Ireland there has been a general tendency for fruit crops on herbicide-treated, unmulched plots to outyield mulched ones (Robinson and O'Kennedy 1978).

The reduction of yield on mulched plots may be due in some cases to lower air temperatures and increased frost injury. Alternatively the effect of an organic mulch in reducing soil temperatures during the growing season or in maintaining excessively moist conditions in the soil may also be responsible.

More extensive root systems have been recorded under mulches (Agulhon 1975, Allmaras and Nelson 1971). The total weight of a vine root system in the 0 to 0.60 m soil layer for a mulched plant was 150% higher than for a non-mulched, cultivated plant after one year. Initially plastic film tends to encourage shallow rooting, 92% of the roots of a mulched vine rootstock occurring in the 0 - 30m layer and none in the 0.45 and 0.60 m layer, compared with only 19% of the roots of the cultivated vines in the 0 - 0.30 m layer. The absence of exploitation of the deeper soil layers by mulched plants is normally shortlived.

Further studies with vines in France showed that after three years, the depth of the root system was similar both for mulched and cultivated plants, but that the superiority of mulched plants was due to their more vigorous root growth in the 0 - 0.15 m layer (Agulhon 1975).

Not only do plants respond differently to different mulches but different plant species may respond differently to the same mulch. Mulching with bark or sawdust significantly increased the growth of Hypericum, Potentilla and Cotoneaster but not of Spiraea, Arundinaria or Geranium (Kolb et al 1985).

Plant establishment

The rapid development of an extensive shallow root system on vines mulched with plastic enabled plants to recover more rapidly after planting (Agulhon 1975). This was attributed to the greater development of surface roots and the better balanced root system of young plants. The establishment of ornamental perennials and shrubs was also improved by organic mulching materials especially bark and sawdust (Kolb et al 1983).

Appearance of mulches

The appearance of a mulch is more important in landscape situations than in field grown food crops or nursery stock. Bark, both chips from softwood and coarse ground and screened hardwood, make decorative mulches (Gartner 1978). Of nine different mulches tested in Wisconsin , bark was considered to be the most satisfactory since it blended with the surrounding soil (Ashworth and Harrison 1983). In prestige situations where cost is not a limiting factor, the moisture-conserving and weed-suppressing advantage of plastic may be combined with the pleasant appearance of natural mulches by covering the plastic mulch with a thin layer of bark or granite chips (Wright 1982).

Such a surface covering will protect the plastic from degradation by ultra violet light.

Economics

The use of mulches can greatly reduce the labour required for weed control. In Britain , a hand hoeing/forking bare soil regime between shrubs and trees required about 33 man hours per 100 sq m per year compared with 5 man hours for a mulch layer (Wright 1982). In Germany , cultivation costs were reduced on average by 23% to 40% with bark and sawdust mulches, by 10 to 22% by straw and by 5 to 8% by chopped shrub waste and limestone gravel compared with unmulched controls (Kolb et al 1985). Reducing the thickness of the mulch from 30 mm to 19 mm reduced the cost of the mulching treatment but reduced its effectiveness.

Organic mulches such as bark would be used much more widely if they were not so expensive. In 1990 bark cost around £1.50 -£2.50/sq. metre and polythene 12p/sq. metre compared with 1p/sq. metre for glyphosate (Bisgrove 1990).

The economics of mulching does not depend only on the initial cost but also on the longevity of the material used. Newspaper and a thin mulch of straw had virtually disappeared after six months, but chopped bark, sawdust and black polythene lasted well and were not substantially different after six months (Ingle 1981).

The transport of bulky organic mulches adds considerably to their costs (Insley 1981). About 500 cu. m of mulching material is required per hectare to give a minimum mulch depth of 50 mm. Heavy duty plastic can be handled more easily, but light synthetic mulches are often difficult to lay even in a light wind (Ashworth and Harrison 1983).

Despite the high price of many organic materials, the cost of mulching was recouped in one year in Germany due to reduced maintenance cost and improved plant establishment (Kolb et al 1983). However a comparison of weed control methods in Sweden showed that chemical weed control was least expensive, followed by a combination of chemical treatments and mulching and by mulching alone (Christensson 1982).

Thermal Techniques

The development of new machinery, including flame guns and microwaves, during the 1980s has increased interest in the possible use of high temperatures at the soil surface to control weeds. Modern flame guns using liquified petroleum gas are being used by some local authorities on hard surfaces to suppress weeds either by direct heat or by infra red whereby a heated surface radiates energy onto the weeds. Flame guns work by applying sufficient heat to the weed cells and one of their major disadvantages is that it is difficult to assess the extent of the damage after one pass of the flame (Royal Horticultural Society 1992).

Good results have been obtained in Denmark with 'flame cultivation' on bare ground, but the selective use of high temperatures among growing plants is very difficult (Vester 1988). The results of this work showed the importance of a level soil surface and vigorous 'crop' growth at the time of treatment. In general, flame control is more expensive than chemical or cultural control and so its usage is likely to be restricted to situations where cultivation (e.g. hard surfaces) or herbicides cannot be used.

With the development of mobile microwave units, field use of this technology is theoretically possible (Moosmann and Koch 1988). Considerable control of weed seeds has been achieved down to a depth of 5 cm with 111 Kilojoules of energy but 140 Kj were needed to penetrate to 15 cm. Mobile microwaves will be very expensive and the main use of this technology is more likely to be in the area of compost sterilisation.

Biological Control

The prospects for biological control of weeds in amenity plantings using insects, nematodes and pathogens are not encouraging. Although striking successes against weeds have been achieved in a number of countries, the weed species involved in most cases have been introduced range weeds, e.g. Opuntia spp controlled in Australia with the moth Cactoblastis cactorum from Argentina . In this type of control the introduced agent is self-sustaining once established. Despite research programmes in many countries, developments have been slow, largely because of the risk that an introduced agent may damage desirable plants. In any case, control of one or two weed species only would be of limited value in most amenity plantings which are normally affected by a dozen or more species.

Endemic mycoherbicides

A mycoherbicide contains the spores of a naturally occurring pathogen, specific to a particular weed species, but which is not normally present in the environment in sufficient quantity to kill the weed. When bulked up and applied in high concentration at a vulnerable growth stage of the weed, mycoherbicides can be effective.

Two mycoherbicides, 'Devine' (Phytophthora palmivora) and 'Collego' (Colletotrichum gloeosporiodes Penz. f.sp. aeschynomenae) are registered in the US (Cullen and Hassan 1988). In Canada a related mycoherbicide, 'Biomal' (C. gloeosporiodes Penz f. sp. malvae) controls Malva pulsilla, a prairie weed that can not be controlled effectively by chemical means (Del Serrone, 1989).

Work is in also in progress with mycoherbicides for the control of some of the most common weeds of north west Europe , including Senecio vulgaris, Stellaria media and Chenopodium album ( Sunderland 1990). Puccinia lagenophorae shows some promise for the control of Senecio vulgaris (Paul and Ayres 1987). Further research is likely to produce other fungi that could be developed as mycoherbicides but which are ineffective in nature because of poor overwintering capacity.

Although some progress along these lines is expected, there is no practical application for mycoherbicides in amenity horticulture at present, and the possibilities for the future in this area are not very bright.

Ground Cover Plants

Once they are established, ground cover plants, such as Erica carnea and Cotoneaster dammeri, can suppress weeds effectively. Until these plants form a cover over the soil surface, they require as much attention to weed control as other plants.

Particularly vigorous plants, such as Lamium galeobdolon, are sometimes recommended as weed suppressors because they establish quickly from small rhizome fragments and their vigorous growth competes well with many weed species. Such species are, however, very aggressive and can become as difficult to control as the weeds they were meant to suppress.

Chemical Weed Control

Herbicides are widely used in amenity horticulture in some areas and a large number of commercial products, based on some 17 active ingredients, are approved for use in the UK (Clay and Stephens 1992).

Chemical weed control is particularly successful in plantings of trees and shrubs; the range of herbicides available enables most weeds to be controlled effectively, plant vigour benefits from the elimination of cultivation and the cost of maintenance can be reduced substantially.

The two main problems facing chemical weed control are:-

1. public pressure against the use of pesticides with particular concern about the possible contamination of drinking water. 

2. the risk of a further build up of resistant biotypes.

Public fears about the use of herbicides is widespread and has led to some Local Authorities banning the use of certain approved chemicals. These fears need to be put into perspective. With very few exceptions, herbicides have a remarkable record of safe use. Public concern has been fuelled to some extent by the conspicuous protective clothing that operatives are obliged to wear.

There is a degree of ambivalence on the part of the public towards herbicides. Along with the pressures against their use, there is also strong public demand for satisfactory weed management standards and criticism if these standards slip. Land managers can do much to reduce the use of herbicides by formulating clear objectives about the type of landscape needed (intensively managed landscape or area of rough land) and by deciding clearly in advance about appropriate weed management standards.

Clay and Stephens (1992) suggest that public fears could be reduced if less toxic herbicides were available and spray operators did not have to wear obvious protective clothing. These requirements might also be reduced by the use of spinning disc sprayers (CDA) which diminish sprayer contamination as all the droplets reach their target.

The contamination of ground water with traces of atrazine and simazine has understandably heightened public concern about the use of these herbicides. These fears persist, even though the Advisory Committee on pesticides has confirmed that the levels detected in the UK do not endanger the health of consumers or the environment. It seems likely that the residues in ground water result from the use of triazines for total weed control on uncropped ground (Clay and Stephens 1992). In these situations high doses are often used on inert, inorganic material where there would be little or no adsorption of the chemicals. In contrast, where low doses of simazine are applied to non cultivated soil containing average amounts of organic matter (2 or 3%) and planted up with shrubs and trees, there is evidence that simazine is retained largely in the top 100 mm (Burschel 1961). Bioassay tests in Ireland using lettuce as a test crop failed to detect any simazine except in the top 50 mm even though the area planted up with shrubs had received 48 applications of simazine at 1.67 kg a.i./ha during the last 24 years.

As a result of the detection of triazines in ground water, the Government announced restrictions in May 1992 on the use of simazine and atrazine on non-cropped ground. The restriction on the use of high doses of triazine herbicides on non cropped ground has been widely welcomed because this was a likely cause of ground water pollution. Since the use of simazine at low doses on crops is unlikely to have been responsible, its use on crops such as blackcurrants, apples and nursery stock is still permissable. It is also permissable to use simazine on private amenity plantings of trees and shrubs but not in public places. 

It seems anomalous that simazine can continue to be used on shrubs in nurseries, in private gardens and in fruit and vegetable crops but not at low doses in shrubs in parks and other amenity situations. This will cause considerable difficulties in the maintenance of amenity plantings as simazine was the least expensive of all commonly used herbicides.

It is likely that simazine will continue to be used on woody plants where it is legal to do so and where resistant biotypes are not a problem. In other cases where a residual weed killer is considered necessary, simazine will be replaced by other herbicides, such as napropamide, diuron, lenacil and dichlobenil depending on the tolerance of the plants being grown.

Much can be done to control weeds in amenity plantings with a rotation of approved residual, contact and translocated herbicides. Glyphosate, for example, is rapidly biodegraded in soil to harmless compounds and is unlikely to contaminate ground water. Some woody ornamental species show a degree of tolerance of glyphosate at certain times of the year (Skroch 1987) and more testing of this and other herbicides will benefit the landscape industry.

Although glyphosate (isopropylamine salt of N-phosphonomethyl glycine) is a valuable herbicide in amenity plantings, present advertisements advising land managers that the use of simazine should be replaced by applications of glyphosate in the spring and autumn are unsound technically. This practice if repeated for several years is likely to lead to the build-up of glyphosate resistant biotypes. Glyphosate inhibits the EPSP-synthease enzyme which is a key enzyme in the pathway responsible for the synthesis for some aromatic amino acids. In the USA Convolvulus arvensis resistant to glyphosate has already developed, possibly because the weed produces an excess amount of EPSP-synthease or because an alteration of the EPSP-synthease gene results in a lower affinity to bind with glyphosate (Hass and Streibig 1990).

The results in USA confirm that unless land managers rotate herbicides and systems of weed control it is only a matter of time before severe problems will appear with glyphosate resistant biotypes. When this occurs, the value and efficacy of this useful herbicides will be greatly reduced.

Apart from the development of glyphosate resistant biotypes, it is also certain that, if a programme of spring and autumn glyphosate application is used for several years as the method of land management, weeds that emerge after the spring application and seed early will become prevalent and cause problems.