Abstract

Key Words

Soil microbiology, Pseudomonas spp., grapevine, cellulolytic

Introduction

Soil health and microbial diversity have become vital issues for the grape growing industry. This study examines the effect on soil microbial biodiversity of vineyard floor management practices similar to those found in many organic vineyards (i.e. low input systems with continuous plant cover) as compared with conventional systems which maintain weed free under-vine and/or inter-rows by herbicide application.

Weed free vineyards have been described as essentially monocultures with high input and output characterised by a lack of stability due to their inherent ecological simplicity (Tedders 1983). Vineyards with few ecological niches for predators are susceptible to pests, pathogens and weed competition. On the other hand, natural ecosystems are characterised by high biological diversity so that mass multiplication of any one organism is impossible due to competition, predators and parasites. This enables disturbances and balance shifts to be self-regulated internally. Continuous plant cover also results in increases in the soil organic matter, leading to improved vineyard soil structure, nutrient storage capacity, water infiltration, water holding capacity and microbial density (Gulick et al 1994, Bugg and van Horn 1997, Pinamonti et al 1996). Soil health has been described as implying “ecosystem sustainability, diversity, functional connectedness, and resilience in response to a disturbance or stress” (van Bruggen and Semenov 1999) with soils with low microbial biodiversity being less sustainable and resilient.

The effect of herbicides on vineyard soil microbial populations is likely to involve not only the direct effect of the herbicide but also the indirect effect of reduction in rhizosphere exudates and organic material due to the lack of vegetation after repeated herbicide applications. Glyphosate, a widely used broad spectrum, non-selective, post-emergence herbicide in Australian viticulture, has been shown to reduce soil bacterial populations in peach orchard and soybean field soils (Araújo et al 2003) and in wheat soils (Mekwatanakarn and Sivasithamparam 1987) and to reduce bacterial, actinomycete and fungal populations in forest soil (Gorlach-Lira et al 1997) and vineyard soil (Encheva and Rankov 1990). On the other hand, some studies report increased populations of soil actinomycetes and fungi with glyphosate (Araújo et al 2003), increased soil microbial biomass (Haney et al, 2002) or no long-term change in microbial populations (Busse et al 2001). Another herbicide widely used in Australian viticulture consists of a mixture of diquat and paraquat. Both glyphosate and paraquat were reported to cause activation in soil urease and invertase soil enzymes but suppression of phosphatase enzymes (Sannino and Gianfreda 2001) whilst diquat and paraquat increased fungal populations, decreased the population of some fungal antagonists to take-all and increased the take-all incidence in wheat soil (Mekwatanakarn and Sivasithamparam 1987). In a cropping field study, paraquat increased populations of bacteria, fungi and actinomycetes at normal application rates but at higher application rates were toxic to fungi (Camper et al, 1973). Few studies have demonstrated long-term impacts on soil microbial populations of herbicide applications. It appears that in viticulture, only Encheva and Rankov (1990) have studied the long-term effect of glyphosate and none have reported on the long-term effect of diquat and paraquat.

This study investigated the effects of herbicide application versus site adapted permanent swards in two vineyards: one in a warm climate and the other in a cooler climate. Soil fungi, cellulolytic bacteria and Pseudomonas spp. populations were monitored for 2.5 years and population differences in winter 2004 are reported here.

Methods

Vineyard field trial

There were three floor management treatments: a complete spray out with herbicide both in the undervine and inter-row regions (C), slash only with no herbicides applied (S) and under-vine herbicide spray only (U). These approaches represent a broad range of floor management techniques being utilised in Australian viticulture today. The treatments were applied at two field sites located in New South Wales vineyards (Vitis vinifera cv.Chardonnay) in regions representing warm climate and cooler climate viticulture.

The warm climate site was located at the Charles Sturt University vineyard at Wagga Wagga. The region has an average annual rainfall of 522 mm with most of the rain received through winter, although rainfall was below average throughout the trial period (Table 1). Summer temperatures can often reach 40^°C. Canopy management at this site was a simple single cordon wire and one foliage wire. A cover crop of cocksfoot and subterranean clover mix was sown in June 2002 but severe rainfall deficiencies during winter and spring resulted in poor establishment, so that the non-herbicided plots contained a diversity of plant species with a mixture of grasses such as Cynodon, Stellaria, Erogrostis, Digataria and Panicum, and other herbaceous species. Twelve bulked soil cores (10 cm deep, 60mm diameter) were used to describe the soil. The soil was a brown sandy loam and NaHCO₃-extractable P was 36 mg/kg soil in the inter-row (Colwell 1963). Soil pH (1:5, 0.01 M of CaCl₂) was 7.1 and oxidisable organic carbon (dichromate oxidation method) was 0.91%. Vine rows were 3 m apart and each panel contained 3 grapevines spaced 2 m apart. The experimental plots were 3 panels long and 4 mid-rows wide with a buffer row between each plot. Four replicates of the 3 treatments were arranged in a randomised block design.

The cooler climate site was located in Tumbarumba. This region has an annual rainfall of 968 mm although rainfall was below average throughout the trial period (Table 1). Frost is often experienced in the early part of the growing season. Excessive vine vigour can be a problem and canopy management relies on split canopy systems. A subterranean clover pasture dominated the non-herbicided inter-row area (23% to 25% of plant species). The soil was a brown clay-loam and NaHCO₃-extractable P was 20 mg/kg soil in the inter-row (Colwell 1963). Soil pH (1:5, 0.01 M of CaCl₂) was 5.9 and oxidisable organic carbon (dichromate oxidation method) was 1.7 %. Vine rows were 3m apart and each panel contained 8 grapevines spaced 1m apart. The experimental plots were 2 panels long and 4 mid-rows wide with a buffer row between each plot.

Table 1. Rainfall for Wagga Wagga and Tumbarumba, 2002 to 2004.

	Long term average rainfall (mm)	Rainfall in 2002 (mm)	Rainfall in 2003 (mm)	Rainfall from 1st January to 1st July, 2004. (mm)
Wagga Wagga	533	373	438	130
Tumbarumba	968	752	872	271

Herbicide applications

At the Wagga Wagga trial site, a diquat and paraquat mixture (2.78 L/ha) was applied to the appropriate treatment plots in April, September, October and November of 2002 and in June, November and December of 2003. A mixture of carfentrazone-ethyl and glyphosate (1.85 L/ha) was applied in March and October of 2003.

At the Tumbarumba trial site, glyphosate (1.85 L/ha) was applied to the appropriate treatment plots in July 2002. A diquat and paraquat mixture (2.78 L/ha) was applied in October and November of 2002; and in January, June, November and December of 2003; and in January of 2004. A mixture of carfentrazone-ethyl and glyphosate herbicide was applied in September and October of 2003.

Dilution plating

Six soil cores (10 cm deep, 60 mm diameter) were collected from the inter-row region at a position 50 cm from grapevine trunk and bulked. The bulked soil samples were vortex mixed with phosphate buffered saline (pH 7.2) for 10 s, sonicated at 260W/cm² for 15 s and orbitally shaken at 290 rpm for 30 min on ice. Culturable bacteria and actinomycetes were then isolated by dilution plating (with three plate replicates) on Pseudomonas Agar CCF (Oxoid) (PS) which is selective for pseudomonads, and Cellulose Bacterial Agar (CBA) (Tuitert et al 1998). Fungi were isolated by dilution plating on Dichloran Rose Bengal Chloramphenicol (DRBC) agar (Pitt and Hocking 1985). All plates were placed in plastic bags loosely wrapped so that there was ample air supply and incubated at 25°C in darkness for 1 week before colonies were counted.

Statistical analysis

Data were subjected to a 2-way analysis of variance (ANOVA) test using Genstat for Windows, 7^th Edition. Least significant differences (l.s.d.) were calculated at P < 0.05.

Results and Discussion

Table 2. Cellulolytic bacteria, Pseudomonas spp. and fungi isolated from non-herbicide and herbicide treated inter-row soil at Wagga Wagga and Tumbarumba field sites, July 2004.

	Cellulolytic bacteria (105 cfu1 / cm3 dry soil)	Pseudomonas spp. (104 cfu / cm3 dry soil)	Fungi (103 cfu / cm3 dry soil)	% Soil water content in July 2003	% Soil water content in July 2004
Wagga Wagga, nil herbicide2	349*	28*	170*	12.1	10.9
Wagga Wagga, herbicide3	168	7	117	11.7	11.0
l.s.d (P<0.05)	171	14	45	1.0	1.2
Tumbarumba, nil herbicide2	489*	110*	253*	13.7*	18.6
Tumbarumba, herbicide3	191	3	154	16.7	18.4
l.s.d (P<0.05)	228	92	63	1.9	1.4

¹ colony forming units
² Treatments S and U (slash only and undervine spray-out)
³ Treatment C (complete spray-out)
* Treatments without herbicide are significantly different from treatments with herbicide (P < 0.05)

In July 2004, the populations of soil cellulolytic bacteria, Pseudomonas spp. and fungi were significantly lower (P < 0.05) in the inter-rows that had been treated with herbicide, indicating a significant effect of the herbicide treatment on soil microbial biodiversity. The herbicide treatment decreased the pseudomonad population by 75% in Wagga Wagga and by 97% in Tumbarumba. It also decreased the population of cellulolytic bacteria by 52% in Wagga Wagga and by 61% in Tumbarumba. The fungal population was decreased by 31% in Wagga Wagga and by 39% in Tumbarumba.

The major herbicides used in this trial were glyphosate, diquat, paraquat and carfentrazone-ethyl. These herbicides effectively decreased the vegetation in the inter-row to low levels so it was not possible to determine whether the changes in soil microbial population were due to the direct effect of the herbicides on the soil microbes alone or in combination with the indirect effect of the elimination of inter-row plant growth and therefore reduction in rhizosphere exudates and organic material after repeated herbicide applications. Interestingly, in Tumbarumba the decrease in microbial population occurred despite the fact that the soil moisture level had been significantly higher in the bare herbicided soil in 2002 (data not supplied) and in winter 2003.

The decline in the population of soil cellulolytic bacteria after herbicide treatment was in agreement with Voets et al (1974) who found that cellulolytic bacteria were permanently reduced in an apple orchard kept free from direct vegetative cover by the long term application of atrazine herbicide. Cellulolytic bacteria and fungi have a key role in the decomposition and transformation of organic matter in soil. As many as 90 % of soil fungi are capable of utilizing cellulose (Sims 1990)

Dry conditions in 2002 and 2003 caused low soil moisture levels at the trial sites of this study (Table 2). Pseudomonads, being less resilient to dessication than spore-forming bacteria or bacteria capable of forming a non-multiplying resting state (Labeda et al 1976) did not survive well in the herbicide treated inter-row in the absence of plant roots. The Pseudomonas genus contains species that are beneficial (Bavaresco and Fogher 1996) and other species that are pathogenic to grapevines (Waschkies et al 1994). Pseudomonads are generally R-strategists (adapted to grow and multiply in high nutrient environments) so they are usually much more numerous on plant roots rather than in bulk soil (Smit at al 2001). Marilley and Aragno (1999) found that plant roots were selective towards members of the Pseudomonas genus to the detriment of Gram positive bacteria and those belonging to Acidiobacterium division. Różycki et al (1999) reported that most of the bacteria isolated from roots of pine and oak trees belonged to the genera Pseudomonas and Bacillus and that the majority of these bacteria had nitrogenase activity (i.e. they were diazotrophs, or nitrogen fixers). Similar studies of grapevine roots have not been made although Bavaresco and Fogher (1996) reported that the inoculation of grapevine roots with fluorescent pseudomonads improved the establishment of arbuscular mycorrhizal fungi. Barka et al (2000) also showed that inoculation of grapevine plantlets with Pseudomonas sp. (strain PsJN) induced a significant growth promotion which made them more hardy and vigorous and induced resistance to Botrytis cinerea.

Loss of microbial biodiversity can affect the functional stability of the soil microbial community.

Griffiths et al (2001) reported that soils of reduced biodiversity due to low plant species diversity were less ‘resistant’ (i.e. able to withstand the immediate effects of) and ‘resilient’ (i.e. able to recover from) to perturbations such as copper sulfate application and heat stress. Soil contains significant populations of microorganisms with the ability to attack or suppress plant pathogenic fungi. Fungistasis has been shown to be strongest in resistant soils with high organic matter content and microbial activity, with soil bacteria such as cellulolytic bacteria and pseudomonads, but also fungi such as Trichoderma spp. implicated as major causes of such fungistasis (Sturz et al 1997, Whipps 2001).

Further work will investigate the effect on inter-row microbial populations of the above commonly used viticultural herbicides. The effect of the herbicides on the grapevine rhizosphere and undervine soil microbial populations will also be examined.

Acknowledgments

This project is funded by the GWRDC which is a partnership between and jointly funded by the Commonwealth and the Australian winegrape and wine industry. NWGIC staff Rhonda Smith, Andrew Smith and Sylvie Sicard have provided valuable assistance in the field and laboratory. Thanks must also go to Charles Sturt University winemaker Greg Gallagher and Stuart Barclay of Tumbarumba Wine Estates who have provided the experimental sites at Wagga Wagga and Tumbarumba respectively.

References

Araújo ASF, Monteiro RTR, Abarkeli RB (2003) Effect of glyphosate on the microbial activity of two Brazilian soils. Chemosphere 52, 799-804.

Barka EA, Belarbi A, Hachet C, Nowak J, Audan J (2000) Enhancement of in vitro growth and resistance to gray mould of Vitis vinifera co-cultured with plant growth-promoting rhizobacteria. FEMS Microbiology Letters 186, 91-95.

Bavaresco L , Fogher C (1996) Lime induced chlorosis of grapevine as affected by rootstock and root infection with arbuscular mycorrhiza and Pseudomonas fluorescens. Vitis 35, 119-123.

Bugg RL, van Horn M (1997) Ecological soil management and soil fauna: best practices in California vineyards. ‘Proceedings Australian Society Viticulture and Oenology, Viticulture Seminar, Viticultural Best Practice, Mildura’. pp 23-34.

Busse MD, Ratcliffe AW, Shestak CJ, Powers RF (2001) Glyphosate toxicity and the effects of long-term vegetation control on soil microbial communities. Soil Biology and Biochemistry 33, 1777-1789.

Camper ND, Moherek EA, Huffman J. (1973) Changes in microbial populations in paraquat-treated soils. Weed Research 13, 231-233.

Colwell JD (1963) The estimation of P requirements of wheat in southern NSW by soil analysis. Australian Journal of Experimental Agriculture and Animal Husbandry 3, 190-197.

Encheva H, Rankov V (1990) Pochvoznanie-i-Agrokhimiya 25, 66-73.

Gorlach-Lira K, Stefaniak O, Slizak W, Owedyk I (1997) The response of forest soil microflora to the herbicide formulations Fusilade and Roundup. Microbiological Research 152, 319-329.

Griffiths BS, Bonkowski M, Roy J, Ritz K. (2001) Functional stability, substrate utilisation and biological indicators of soils following environmental impacts. Applied Soil Ecology 16, 49-61.

Gulick SH, Grimes DW, Munk DS, Goldhamer DA. (1994) Cover-cropped-enhanced water infiltration of a slowly permeable fine sandy loam. Soil Science Society of America Journal 58, 1539-1546.

Haney RL, Senseman S, Hons FM (2002) Effect of Roundup Ultra on microbial activity and biomass from selected soils. Journal of Environmental Quality 31, 730-735.

Labeda DP, Kang-Chien L, Casida LE (1976) Colonization of soil by Arthrobacter and Pseudomonas under varying conditions of water and nutrient availability as studied by plate counts and transmission electron microscopy. Applied and Environmental Microbiology 31, 551-561.

Marilley L, Aragno M. (1999) Phylogenetic diversity of bacterial communities differing in degree of proximity of Lolium perenne and Trifolium repens roots. Applied Soil Ecology 13, 127-136.

Mekwatanakarn P, Sivasithamparam K (1987) Effect of certain herbicides on soil microbial populations and their influence on saprophytic growth in soil and pathogenicity of take-all fungus. Biology and Fertility of Soils 5, 175-180.

Pinamonti F, Stefanini M, Dalpiaz A (1996) Soil management effects on nutritional status and grapevine performance. Wein-Wiss. 51, 76-82.

Pitt JI, Hocking AD (1985) Fungi and food spoilage. CSIRO Division of Food Processing: North Ryde, NSW, Australia.

Różycki H, Dahm H, Strzelczyk E, Li CY (1999) Diazotrophic bacteria in root-free soil and in the root zone of pine (Pinus sylvestris L.) and oak (Quercus robur L.) Applied soil Ecology 12, 239-250.

Sannino F, Gianfreda L (2001) Pesticide influence on soil enzymatic activities. Chemosphere 45, 417-425.

Sims GK. (1990) Biological degradation of soil. Advances in Soil Science 11, 289-330.

Smit E, Leeflang P, Gommans S, van den Broek J, van Mil S, Wernars K (2001) Diversity and seasonal fluctuations of the dominant members of the bacterial soil community in a wheat field as determined by cultivation and molecular methods. Applied and Environmental Microbiology 67, 2284-2291.

Sturz AV, Carter MR, Johnston HW (1997) A review of plant disease, pathogen interactions and microbial antagonism under conservation tillage in temperate humid agriculture. Soil Tillage Research 41, 169-189

Tedders WL (1983) Insect management in deciduous orchard ecosystems: habitat manipulations. Environmental Management. 7, 29-34.

Tuitert G, Szczech M, Bollen GJ, 1998, Suppression of Rhizoctonia solani in Potting Mixtures Amended with Compost Made from Organic Household Waste. Phytopathology 88, 764-772

van Bruggen AHC, Semenov AM (1999) A new approach to the search for indicators of root disease suppression. Australasian Plant Pathology. 28, 4-10.

Voets JP, Meerschman P, Verstraete W (1974) Soil microbiological and biochemical effects of long-term atrazine applications Soil Biology and Biochemistry. 6, 149-152
Waschkies C, Schropp A, Marschner H (1994) Relations between grapevine replant disease and root colonisation of grapevine (Vitis sp.) by fluorescent pseudomonads and endomycorrhizal fungi. Plant and Soil 162, 219-227.

Whipps JM (2001) Microbial interactions and biocontrol in the rhizosphere. Journal of Experimental Botany 52, 487-511.