Apr 24, 2016

The effects on soil biology of agricultural chemicals

By Alan Broughton
February 2013.

In the last few years the soil science community has showed a new interest in soil biology. This is possibly because of the growing influence of the organic and biological farming sectors, coupled with the government’s Carbon Farming Initiative and the increased costs of fertilisers. However it is of concern that many of the scientists advising farmers on methods to enhance the benefits provided by soil biological components treat agricultural chemicals as essential for farming and therefore untouchable. 

There is a large body of research that shows agricultural chemicals do have varied and significant effects on soil biology and must be included in any discussion about capturing the services of soil biology for the benefit of agriculture, farm sustainability and soil carbon sequestration. The large number of successful organic and biological farmers demolishes the myth that agricultural chemicals are essential. 

The term “agricultural chemicals” refers to inorganic fertilisers and pesticides in the broad sense that includes herbicides, insecticides, fungicides, miticides, nematicides and anthelmintics (worming drenches).

Inorganic fertilisers

A review of the literature by the CSIRO in 2006 concluded that there were few adverse effects on soil biology by inorganic fertilisers, and often positive effects because of the increased growth which produced greater amounts of organic matter in the soil that supported soil biology. This reflects conventional wisdom of many soil scientists and government departments but does not reflect reality over the long-term. While the total microbial biomass increases in the short term there are major structural changes in the composition of this biomass.  

Nitrogenous fertilisers

The long-term use and overuse of nitrogenous fertilisers acidifies soil and burns up organic matter. Increased use of inorganic fertiliser, particularly nitrogen, use does not lead to an increase in soil organic matter over the long-term, instead causing a reduction. The nitrates stimulate microbial communities that feed on nitrogen; these communities use up readily available soil carbon to balance their diet then start decomposing more stable forms such as humus. As they do so the soil loses its ability to hold nitrogen and the nitrogen that is surplus to the plants’ requirements leaches, causing water pollution, or evaporates as nitrous oxide which is 300 times more potent as a greenhouse gas than carbon dioxide. At the same time the soil hardens and becomes more acidic, providing worsening conditions for beneficial microorganisms. Water holding capacity is also reduced. In the long-term, after 10 or so years of regular applications of anhydrous ammonia or urea, soil acidification has increased in many cases and microbial biomass decreased.

Nitrogenous fertilisers negatively affect the most important nutrient releasing organisms in the soil, mycorrhizae and Rhizobium. As inorganic nitrogen fertiliser rates increase, phosphorus and other nutrient release by mycorrhizal fungi, and nitrogen fixation by Rhizobium, goes down.

The effects on mycorrhiza of soluble nitrogen and phosphorus

Mycorrhizal fungi associated with plant roots release unavailable phosphorus into a form that the plant can use. At the same time other nutrients, particularly trace elements, and water become more available. Mycorrhizae are important not just for nutrient and water uptake, but most importantly for carbon sequestration. Glomalin is a long-lasting high carbon glue-like substance produced only by mycorrhizae, making up 27% of the soil’s organic carbon. The substance was discovered only in 1996. Mycorrhizal colonisation is greatly affected by phosphatic and nitrate fertilisers, fungicides, some herbicides, and cultivation of the soil. 

Most of the soluble nitrate fertilisers reduce root colonisation by mycorrhizae and inhibit spore germination. Mycorrhizae are also known to release nitrogen from soils, and when nitrogenous fertilisers are used this activity is suppressed. 

Several studies have shown a decrease in root colonisation by mycorrhizae by the addition of inorganic phosphorus. 

While in very low phosphorus soils the addition of small amounts of superphosphate can increase mycorrhizal activity, as applications increase beyond these levels there is a decline in activity. This is probably because of less root exudate and lower levels of soluble carbohydrates in the plant roots as the plants become adequately supplied with phosphorus. Many studies have found a direct inverse correlation between the level of available phosphorus and the amount of mycorrhizal infection. 

Experiments with faba beans showed that not only does superphosphate reduce the mycorrhizal inoculation but also reduces Rhizobium nodulation. A side benefit of good mycorrhizal association is increased drought tolerance - superphosphate thus indirectly reduces the ability of plants to extract water from soil.

There are much lower mycorrhizal spore counts and infection rates in soils with high available phosphorus and nitrogen, whether applied as fertiliser or farmyard manure. Spore counts can be four times higher in an unfertilised field and root colonisation percentage 20 times high. Nitrogen application has been found to be just as suppressive of mycorrhizae as phosphorus application. 

A CSIRO research team found ammonium phosphorus applications (MAP and DAP) reduced mycorrhizal colonisation of wheat crops by between 33% and 75%. A side effect was significantly lower zinc levels in the wheat grain.

Potassium chloride

The organic movement has claimed that potassium chloride is deadly on soil biology, but there seems to be no evidence to support this opinion when normal application rates are used. Chloride is an essential nutrient. In many cases the amount of chloride in the fertiliser is far less than that supplied in irrigation water. However if levels build up in soils that are too poorly drained to allow leaching of the excess, the effect is similar to salinity. Many scientific articles show that potassium sulphate produces better yields than potassium chloride, either because of the benefit of the sulphate as a nutrient, or the toxic effects of the chloride.

The effects of salinity on soil biology are well documented. There is a steady decline in soil microbial biomass as the salinity of soil increases. Salinity reduces microbial biomass and activity, enzyme activity and mineralisation from decomposition. The only benefit of potassium chloride over potassium sulphate is its price.

Heavy metal contamination

There are often contaminants of lead, mercury and cadmium in fertiliser which could have a greater negative effect that the fertiliser itself. Microorganisms are far more sensitive to heavy metal stress than plants growing on the same soil. Excess zinc and cadmium are the most damaging to Rhizobium. Zinc levels are known to build up with continued use of poultry litter as a fertiliser, and cadmium is a common contaminant of phosphorus fertilisers. 

Of the four most common heavy metal pollutants cadmium is the most toxic, followed by copper, zinc and lead. The nitrogen fixation process by Rhizobium is very sensitive to the metals, at levels as low as 30 ppm for zinc, 15 ppm for copper, 2 ppm nickel and 2 ppm cadmium. Decomposition of organic matter, enzyme activity and microbial biomass are all also negatively affected by heavy metals. Azotobacter, a non-symbiotic nitrogen fixer, is very sensitive. The effect on mycorrhizae varies according to species, and more tolerate ones can provide a protection against toxicity in the plant. 


It is commonly reported that microbial biomass remains the same after herbicide application, allowing some researchers to conclude that no harm is being done. However, more specific research shows there can be significant changes in the biological community, particularly in some of the most valuable functional groups including nitrogen fixers and mycorrhizae. 

Herbicides and nitrogen fixation

Herbicides have a severe impact on Rhizobium. Tests of various concentrations of many common herbicides including diquat, paraquat, 2,4-D, trifluralin, glyphosate and amitrole show nodulation decreases to zero as the concentrations rise. While at recommended application rates the effect on nodulation is slight, the accumulated effects of regular usage and build up of residues in soil must be taken into account.

Trials on the effects of three common herbicides, 2,4-D amine, Roundup and atrazine, on nitrogen fixation by Rhizobium bacteria found that all three caused a marked decline in total nodule numbers (to 10% of the control). Dry weight of nodules (to 50% of the control), plant shoot length and plant shoot dry weight also declined. Nodule formation on side roots was non-existent. The authors thought that the effect was due to either herbicide injury to the plant, or to the Rhizobium, or to both. Application rates were at Minimum Inhibitory Concentration (applications at and above these levels killed the plants). 

A large range of herbicides negatively affects Azotobacter, while total bacteria populations are often stimulated by the herbicides.

Herbicide effects on mycorrhizae and other fungi

Four herbicides recommended for use on wheat crops in the UK were tested for their effects on mycorrhizae at field application rates. The brand names were Avenge 630 (difenzoquat methyl sulphate), Ceridor (bifenox and mecoprop), Dicurane (chlorotoluron) and Harrier (mecoprop, ioxynil and clopyralid). Avenge completely prevented spore germination, Ceridor and Harrier prevented spore germination at low application rates but not at high rates, while Dicurane showed no effect on spore germination. 

Many herbicides have been found to negatively affect the root colonisation of Eucalyptus grandis by arbuscular mycorrhizae, including glyphosate which also depresses spore production. 

Ectomycorrhizae (those associated with many forest trees) are also affected by herbicides. Glyphosate has a severe effect at field rates, followed by 2,4-D, hexazinone and triclopyr. 

The herbicides paraquat, linuron, simazine and MCPA are all toxic to several species of soil fungi, reducing spore germination and fungal growth. Simazine shows the least effect of the four. The anti-pathogenic soil fungus Trichoderma viride (highly valued for controlling several important root diseases including Armillaria, Rhizoctonia and Sclerotium) is particularly sensitive to paraquat, and paraquat completely stops spore germination of Rhizopus stolonifer (black bread mould, an important decomposer in the soil). On the other hand the root disease Fusarium culmorum is very resistant to paraquat. 

In a trial of 21 herbicides, only one (oryzalin) did not inhibit the spore germination and/or growth of the insect-controlling fungus Beauveria bassiana. Twelve of the herbicides were significantly inhibitory at low doses. Some (diuron, pronamide, simazine and terbacil) inhibited spore germination but not growth of the fungus.

It was noticed in Victorian pine plantations that the fungal disease Phytophthora cinnamomi ceased to be a problem following the substitution of herbicides. Simazine and propazine stimulated the root disease while glyphosate and chlorthal dimethyl exhibited a fungicidal effect on the pathogen. On the other hand the root diseases Pythium and Fusarium are greatly stimulated by glyphosate.

Glyphosate effects on soil biology

There have been many studies done on glyphosate, the active component of Roundup, with various conclusions as to its effect on mycorrhizae, Rhizobium, earthworms and various other soil biology components. It is very toxic to reducing microbes (those that convert insoluble oxides of manganese and iron to plant available forms) and stimulates oxidising microbes (those that reduce the availability of nutrients to plants). Glyphosate is also toxic to many disease-controlling organisms and directly stimulatory to root diseases including Fusarium, Phytophthora, Pythium, Gaeumannomyces graminis (take-all disease of cereals) and Rhizoctonia. Glyphosate works by immobilising metallic trace elements (copper, iron, nickel, cobalt, manganese); Rhizobium cannot function because the nickel that the process requires is made unavailable. 

Glyphosate does not readily biodegrade but does get immobilised by cations, and accumulates in the soil and plant tissue. Phosphorus can reverse the immobilisation and allow glyphosate to re-enter plants many years after application. There can be significantly increased bacterial populations following glyphosate application as they feed on the carbon, nitrogen and phosphorus compounds in the chemical. 

Research in Western Australian wheatfields showed a strong correlation between take-all disease (Gaeumannomyces graminis) and use of the herbicides Roundup (glyphosate) and Sprayseed (diquat and paraquat). Treflan (trifluralin) on the other hand was found to have no effect. The authors believed the herbicide effects were due to their suppression of natural biological controllers of take-all rather than any physiological change in the wheat plant. 

The Institute for Responsible Technology provides a huge reference list for effects of glyphosate. Figures show that at only one fortieth of the recommended application rate the transport of iron, manganese and zinc through the plant is reduced by at least 90%. More than 40 plant diseases are promoted. Glyphosate accumulates in soil and perennial plant (such as lucerne) tissue and can be reactivated by phosphorus application.

Herbicide effects on algae

Many herbicides kill soil algae, which are excellent producers of sugars that microbes convert to humus. Many cause total suppression, while some herbicides have little effect on some species of algae. When tested, five herbicides, chlortoluron, terbutryne, metabenzthiazuron, chloridazon and dinosebacetate, all caused total suppression of soil algal growth at recommended application rates. 

Cyanobacteria (blue-green algae) growth is inhibited by low concentrations of the herbicides diuron, atrazine and paraquat, while glyphosate causes inhibition at higher levels. Trifluralin is very suppressive. Green algae are tolerant of diquat but cyanobacteria and diatoms are severely affected at rates below the levels that are commonly detectable by testing laboratories.

Herbicide effects on earthworms and nematodes

Many herbicides have little effect on earthworms. However glyphosate, while not killing earthworms, reduces their weight and prevents the formation of cocoons and the hatching of eggs, and 2,4-D is fatal. 

Pendimethalin (usually used as a pre-emergent herbicide to prevent the establishment of annual ryegrass and wireweed) was found by several studies to reduce nematodes and other soil invertebrates at low rates of application. 


The effects of insecticides and nematicides are more significant than those of herbicides. The organophosphates (eg. chlorpyrifos, malathion, dimethoate, diazinon) reduce bacterial numbers while stimulating fungi, affecting various soil enzymes, reducing collembolan numbers, and interfering with earthworm reproduction. The carbamate insecticide carbaryl significantly affects earthworm enzymes and reduces phosphatase activity (conversion of phosphorus from unavailable to available) in soils. The organochloride insecticides (such as DDT and lindane), and arsenic compounds, have shown more severe negative effects.

Some insecticides are very toxic to earthworms while others have little effect. Most toxic are the organo-phosphates acephate, azinphos methyl, chlorpyrifos, ethoprofos, ethyl-parathion and phorate. The carbamates carbaryl and carbofuran are extremely toxic at field application levels. There is very strong avoidance of endosulfan by earthworms and of chlorpyrifos by collembolans, and both insecticides reduce the reproduction of earthworms and collembolans. 

There is little toxic effect of a range of insecticides on beneficial nematodes, except for methomyl (Lannate), and the miticides tebufenpyrad and fenpyroximate which are harmful. 

Chlorpyrifos, a common organophosphate insecticide, affects cyanobacteria by inhibiting photosynthesis, the effect varying depending on the species and the dose. 


Fungicides cause great damage, especially to soil fungi. Copper sprays are among the worst, causing long-term decline in earthworm populations and significant reductions in microbial biomass. Copper accumulates and does not break down, so effects are very long lasting. Benomyl causes long-term reductions in mycorrhizae, reduced fungal levels compared to bacteria, avoidance by earthworms, and reduction in nematode and collembolan numbers. Chlorothalonil and azoxystrobin both suppress control agents for Fusarium wilt. Captan negatively affects fungal length and density. Matalaxyl and mefenoxam are both toxic to nitrogen fixers. 

The following fungicides are toxic to earthworms: thiabenzadole seed dressing, chlorthalonil (Bravo), and the carbamates benomyl and carbenzadim. 

Research has been done on the effects of fungicides on mycorrhizae. The systemic fungicide calixin severely reduces mycorrhizal infection of maize (78% reduction) while triforine has no effect. The effect of fungicides on mycorrhizae depend on many variables – the type of fungicide, the species of mycorrhiza, the genetics of the host plant, the soil ecosystem and the soil physical and chemical properties. 

The seed dressing fungicide Funaben T (a combination of thiram and carbendazim) has been found to cause a severe 95% suppression of the activity of the lucerne nitrogen fixer Sinorhizobium melioti, and a significant decrease in Azotobacter, while actinomycetes are stimulated. 

Copper fungicides are commonly used in orchards in Australia, up to 15 times per year for avocados, causing a rapid build up in soils and a profound impact on soil biology. The effect includes the elimination of earthworms, accumulation of undecomposed litter, reduced mycorrhizal associations, reduced nematodes, mites and overall biological activity, and an increase in pest and disease attack. Depending on the soil type, as little as 4 ppm of available copper can negatively affect earthworms. Some avocado orchard soils in Northern NSW contain over 200 ppm copper.


More research is needed, especially on the effects of long-term use of agricultural chemicals in the field. There can be a considerable difference in effect in short-term pot experiments compared to long-term field tests, and between acute toxicity and chronic toxicity. Another issue to be considered is the difference between sites, in soil texture, soil fertility, organic matter level, pH, crop grown, climate and existing soil biological composition, resulting in different effects in different circumstances. Many of the pesticides registered for use in Australia have never been tested for their effect on soil biology. Combinations of chemicals are not tested. 

It is not legitimate to claim that any effects of agricultural chemicals on soil biology are transitory, or that if some species of microbial life are taken out others will replace them. While it is reasonably accurate to say that microbial biomass remains constant in the short-term, this is not the case over the long-term, and it is not the case that biological function is unaffected. The evidence shows the important functions of nitrogen fixation by Rhizobium and Azotobacter, and phosphorus and trace element release by mycorrhizae, are severely compromised by inorganic fertilisers and pesticides. These are not replaceable, except by increased use of more inorganic fertilisers, which is not sustainable. Neither is the disease causing effect of glyphosate. 

A study of 75 paired comparisons of organic and non-organic farms has found that on average organic farms hold 3.5 tonnes per hectare more carbon than the non-organic. This is partly due to the application of organic matter but must largely be attributed to the non-use of agricultural chemicals.

There is little point in extolling the virtues of soil biology without considering the role played in biological suppression by agricultural chemicals. Grazing management, biological supplements, integrated pest management and reduced cultivation play a significant part in enhancing soil biology but are only part of the solution. 

A well functioning soil biological community removes the necessity for inorganic fertilisers and pesticides. 

A more detailed version of this article will go up on the OAA website (www.organicagriculture.asn.au), including an extensive reference list.