TRANSPORT OF MICROORGANISMS THROUGH SOIL JAMAL ABU-ASHOUR 1, DOUGLAS M. JOY 1, HUNG LEE2., HUGH R. WHITELEY 1 and SAMUEL ZELIN1 I School of Engineering and 2 Department of Environmental Biology, University of Guelph, Guelph, Ontario, Canada N1G 2W1 (Received April 21, 1992; accepted August 4, 1993)
Abstract. Microorganisms migrating into and through soil from sources on the land surface may cause a serious threat to both ground and surface waters. It has been estimated that microorganisms can migrate significant distances in the field. Results from various studies suggested that preferential flow through macropores, worm holes, cracks, and fractures is the main reason for such observations. However, a quantitative representation of this phenomenon has not been provided. Microorganisms migrate through soil by advection and dispersion, while being subjected to effects of filtration, adsorption, desorption, growth, decay, sedimentation and chemotaxis. Both laboratory and field investigations have contributed important information on bacterial movement in soils. Qualitative comparisons are generally transferable from laboratory to field situations. Quantitative agreement is much more difficult to establish. Available mathematical modelling of microbial transport is limited in practical application because of the simplifying assumptions used in its development.
1. Introduction Environmental and public health problems associated with the spreading of sewage on land have been observed since the dawn of the 20th century. Instances of land application of sewage are increasing because this disposal process removes some of the pollutants from the applied sewage, constitues a possible aquifer recharge source, and increases crop yields by supplying essential nutrients and by improving soil properties (Lance et al., 1982; Tim et al., 1988). However, disadvantages of land application may include degradation of quality of surface and groundwater through chemical and microbial contamination, and accumulation of heavy metals in soil. Spreading agricultural wastes may constitute a source of pathogens to the groundwater, surface water and soil. The application of these wastes to agricultural lands can cause environmental problems even when the application procedures are within the current guidelines. Problems have been demonstrated in Ontario by Dean and Foran (1990a, b, 1991), Fleming etal. (1990) and Palmateer etal. (1989) where application of liquid manure to agricultural fields have resulted in rapid movement of a tracer bacterium, nalidixic acid-resistant Escherichia coli, through the soil and under drain systems leading to contamination of surface receiving waters. Microbial contamination of water and soil due to land application of liquid manure and other liquid wastes is difficult to treat, because once applied, manure * Corresponding author. Telephone: (519) 824-4120, Exentension 3828. Fax: (519) 837-0442; Email:
[email protected] Water, Air and Soil Pollution 75: 141-158, 1994. (~) 1994 Kluwer Academic Publishers. Printed in the Netherlands.
142
JAMALABU-ASHOUR'~ETAL.
becomes a potential non-point source of pollution, less susceptible to correction than a point source (Crane et al., 1983; Khaleel et al., 1980). Pathogenic bacteria and viruses known to cause disease have been detected in groundwater. Contaminated groundwater causes almost half of the outbreaks of water-borne diseases each year in the United States (Craun, i979, 1984). The most important pathogenic bacteria and viruses that might be transported to groundwater include Salmonella sp., Shigella sp., Escherichia coli and Vibrio sp., and hepatitis virus, Norwalk virus, echovirus, poliovirus and coxsackievirus (Corapcioglu and Haridas, 1984; Craun, 1984; Gerba and Keswick, 1981). Land application systems are designed and installed with the assumption that the soil can act as a living filter with the potential for self purification through biological processes that reduce microbial concentrations (Tim et al., 1988). However, both field and laboratory observations have shown that microorganisms can migrate significant distances through soil in both vertical and horizontal directions (Chen, 1988; Keswick et al., 1982; Stewart and Reneau, 1981; Viraraghaven, 1978). Bacterial migration up to 830 m and viral migration up to 408 m have been reported (Gerba et al., 1975; Keswick and Gerba, 1980). The ability of microorganisms to migrate through soil increases the probability of water contamination. The chance of contamination will increase further if microorganisms have the ability to survive for long periods of time. In laboratory studies reported by Gerba et al. (1975), E. coli survived up to 4.5 months in groundwater maintained in darkness. Under the same conditions, Gerba and Keswick (1981) found that a pathogenic strain ofE. coli survived for 4 months and a saprophytic strain of E. coli survived 5.5 months. This survival occurred despite a reduction of 99.9998% of E. coli and 99.9995% of faecal streptococci in 20 days. In another study, Chandler et al. (1981) assessed the persistence of indicator bacteria (faecal coliforrns and faecal streptococci) on land to which pig manure had been applied. They found the time required for a 90% reduction in number of indicator bacteria in the top 30 mm of soil ranged from 7 to 20 days. The dieoff rate constant for E. coli and faecal streptococci in groundwater ranged from 0.16 to 0.36 day -1 and from 0.03 to 0.23 day -1, respectively (Gerba and Bitton, 1984). Survival of viruses was found to vary widely. The experimental data of Gerba and Keswick (1981) in the study above indicated a reduction of 99.9% in viral concentration in 20 days. The die-off rate constant ranged from 0.046 to 0.77 day -1 and from 0.39 to 1.42 day -1 for poliovirus and coliphage f2, respectively (Gerba and Bitton, 1984). The potential problems associated with land application of sewage or liquid agricultural wastes are receiving increasing attention from the public and regulatory agencies. Regulations, recommendations and disposal guidelines have been established to reduce the risks involved in such practices. Morrison and Martin (1977), as reported in Crane et al. (1983), have made recommendations to reduce health risks resulting from the application of manure or slurries to agricultural land. These authors recommended that no direct contact should occur between
TRANSPORT OF MICROORGANISMS THROUGH SOIL
143
the applied wastes and a crop. Further, subsequent application of liquid waste to the same land should be made only after an adequate period of time to maximize the die-off of pathogenic bacteria and to avoid any potential for buildup. They also recommended that populated areas not be irrigated with liquid wastes. Application of liquid agricultural wastes is not recommended on frozen ground. In a field study conducted in Ontario, Canada, Culley and Phillips (1982) found that manure applications in winter resulted in significantly higher faecal coliform and faecal streptococcus counts in the surface runoff, and faecal streptococcus counts in subsurface discharge when compared with applications during other seasons. In this review we will focus on factors that affect microbial transport through soil. Knowledge of microbial transport mechanisms is needed to contribute to better land application practices that will minimize health and environmental problems.
2. Microbial Transport Through Soil Bacteria and viruses have been shown to travel through porous media with the distance travelled being dependent on the type of porous medium. In studies by Gerba et al. (1975), coliforms travelled from 0.6 m in fine sandy loam to 830 m in sand-gravel; bacteriophage T4 travelled up to 1.6 km in a carbonate rock terrain area. Stewart and Reneau (198 l) detected migration of coliforms from septic tank drainfields in both vertical and horizontal directions to monitoring wells of 152and 305-cm depth located within 30 m of the drainfields. The extent of migration in both directions varied depending on the position of the monitoring well relative to the drainfield. They attributed these differences to variations in water flow. The movement of microorganisms through soil can be very fast. smith et al. (1985) compared the movement of a streptomycin-resistant E. coli K12 strain and C1- tracer through soils of different texture. With the Huntington silt soil contained in 0.28-m undisturbed columns, about 90% of the E. coli applied initially moved through the column in 17 min, while about 70% of the applied C1- moved through within the same time. The authors suggested that such rapid movement resulted from the presence of continuous macropores. In a study by McCoy and Hagedorn (1979), they found E. coli strains were transported in the subsurface at an apparent maximum speed of 17 cm min -a. In another study in Humberside, U.K., bacteriophages were injected into an aquifer by boreholes at 366 and 122 m from a pumping well (Skilton and Wheeler, 1988). The results showed that bacteriophages moved rapidly, reaching a maximum speed of 2.8 cm s -1. Many studies of bacterial movement through soil have been conducted in the field and the results generally show a rapid movement and high concentration of bacteria reaching receiving waters. The explanation normally provided is that the observed phenomena are due to preferential flow of microorganisms through macropores, cracks, fractures, worm holes and channels formed by plant roots, or animals in the soil. Preferential flow through macropores has been observed in both laboratory and field studies (Chandler et al., 1981; Thomas and Phillips, 1979; van Elsas et
144
JAMAL ABU-ASHOUR ET AL.
al., 1991). The influence of macropores on transport of dissolved and suspended matter through soft was reviewed by White (1985). He suggested the flow along preferential paths was the cause of rapid movement of dissolved and suspended matter through soil. Studies of chemical transport in preferential towpaths may give guidance on transport of microorganisms. Rice et al. (1988) studied movement of solutes and herbicides under irrigated fields. They found that preferential flow resulted in solute and herbicide movement velocities of 1.5 to 2.5 times greater than those expected based on water balance considerations. The results from other studies showed also the flow of water and chemicals through macropores is more rapid than that through a soil matrix (Beven and Germann, 1982). Everts and Kanwar (1988) used a hydrograph separation technique to quantify the preferential and matrix flow components to a tile line. They found that preferential flow contributed less than 2% of the total water flow. However, flow of bromide and nitrate account for up to 25% of these tracer chemicals found in the tile line. This large contribution occurred because these chemicals moved through preferential paths at the applied concentrations. Many factors have been observed to affect the survival and movement of microorganisms in soil. These are mainly related to interactions between soil, water, microorganisms and the surrounding environment (Crane et al., 1983; Gerba and Bitton, 1984; Tim et al., 1988). These factors are summarized in Tables I and II. The effect of these factors on movement of microorganisms in soil have been a subject of study by many researchers, van Elsas et al. (1991) studied the influence of soil properties on the vertical movement of a genetically-marked Pseudomonas fluorescens bacterial strain through 50-cm long soil columns of loamy sand. They added bacterial cells at the top of soil columns of 5.3 and 13% moisture content, and measured the concentration of cells translocated to various depths. They found the ratio of cell concentration in the dryer soil to that in the wetter one ranged from 68% at 20 cm depth to 98% at the soil surface. These researchers also studied the effect of soil bulk density on bacterial transport. They observed a trend towards a higher degree of transport to lower soil layers at the lowest bulk density (1 g cm -3) compared to higher bulk densities (1.15 and 1.3 g cm-3). Studies by Huysman and Verstraete (1993a) also demonstrated the strong influence of soil bulk density on bacterial transport. In their work an increase in the bulk density from 1.27 to 1.37 g cm -3 resulted in up to a 60% decrease in the migration of bacteria in laboratory columns. 9 Soil texture can affect bacterial movement through soils. Smith et al. (1985) compared the movement of streptomycin-resistant E. coli through both undisturbed and repacked soils of different texture. In undisturbed soils, 22% of the applied E. coli passed through a 0.28- m column of Maury silt soil while 44 and 79% of the microorganisms passed through similar columns containing the Crider silt and Bruno silt loam soils, respectively. When columns were repacked with the same soils, at least 93 % of the applied E. coli were retained in the soil core over the same
TRANSPORT OF MICROORGANISMS THROUGH SOIL
145
TABLE I Factors affecting the survival of enteric bacteria and viruses in soil Factor 1.
Microorganisms and their physiological state.
2.
Physical and chemical nature of receiving water. pH
-
Soil water content Organic matter content Texture and particle size distribution -
Temperature
-
Availability of nutrients Adsorption properties
Comments
- Shorter survival time in acidic soils (pH 3-5) than in alkaline soils. - longer survival time in wet soils and during times of high rainfall. - Increased survival and possible growth when sufficient amount of organic matter is present. - Finer soils especially clay minerals and humic substances increase water retention by soil which increases survival time. - Longer survival at lower temperature. Increases survival times. - Microorganisms appear to survive better in -
sorbed state. 3. -
4. -
-
Atmospheric conditions Sunlight Water (vapor and precipitation) Temperature Biological interactions Competition from indigenous microflora Antibiotics Toxic substances
5. -
- Shorter survival time at the soil surface. - Same as in (2) above. - Same as in (2) above.
- In sterile soil, survival is increased. - Many microorganisms cannot survive in the presence of antibiotics Same as antibiotics -
Application method Technique Frequency of application Organism density in waste material
Sources: Crane et al. (1983); Gerba and Bitton (1984).
146
JAMAL ABU-ASHOUR ET AL.
TABLE II Factors affecting movement of microorganisms in soil Soil physical characteristics - Texture - Particle size distribution - Clay type and content - Organic matter type and content pH - Pore size distribution - Bulk density -
2.
Soil environment and chemical factors - Temperature Soil water content Soil water flux
-
-
3.
Chemical and Microbial factors Ionic strength of soil solution pH of infiltrating water Nature of organic matter in waste effluent solution (concentration and size) - Type of microorganism Density and dimensions of the microorganism Presence of larger organisms -
-
-
-
-
4.
Application method Soil drying between applications Time of application (winter, spring)
-
-
Bitton et al. (1979); Culley and Phillips (1982); Crane et al. (1983); Gerba and Bitton (1984); Opperman et al. (1987); Peterson and Ward (1989); van Elsas et al. (1991).
elution volume. The Size and morphology of microorganisms may affect their transport through soil. Gerba and Bitton (1984) reported in a study that when E. coli and coliphage f2 were injected together into an aquifer, the larger E. coli were detected 150 m down gradient in an observation well ahead o f the smaller coliphage. The reason for this rapid m o v e m e n t o f E. coli is not known. On the other hand, Kott (1988) who studied the m o v e m e n t of different types o f bacteria in sand columns, showed that bacterial size and morphology did not affect the filtration efficiency of the sand. Fontes et al. (1991) studied the effects of ionic strength of artificial groundwater, cell size, mineral grain size and the presence o f heterogeneities within the porous media on bacterial transport. They found the grain size was the most important factor while cell size and ionic strength were about equally important, but of lesser
TRANSPORT OF MICROORGANISMS THROUGH SOIL
147
importance than the grain size in controlling bacterial transport. Gannon et al. (1991a) studied the relationship between cell size, cell surface hydrophobicities and surface charges of 19 bacterial strains and their transport through Kendaia loam soil. They found transport was related to cell size, with bacteria shorter than 1 #m being transported in higher percentages through soil than longer ones. There was no correlation between the transport or retention of these strains by soil and their hydrophobicities or net surface charges. They also found no relation between the presence of flagella and the extent of bacterial transport through soil. Scholl et al. (1990) studied the influence of soil mineral composition, ionic strength and pH on bacterial attachment to aquifer materials. Their results indicated that interactions between mineral grains in the aquifer and bacterial cells influenced adhesion of cells to the mineral grains, and hence play an important role in determining the movement of bacteria through saturated porous media. Gannon et al. (199 lc) studied the influence of NaC1 in the carrying solution, cell density and flow velocity on transport of Pseudomonas sp. strain KL2 through 0.3-m columns of aquifer sand under saturated conditions. When 108 bacterial cells in 0.01 M NaC1 were applied to the column at a flux of 10-4m s -1, only 1.5% of the applied bacteria passed through the column within 2 h of application. However, when distilled water was used as the carrying solution, 60% of the applied bacteria passed through a similar column under the same flow conditions. Their results clearly indicated that movement of bacteria added to sandy aquifers may be enhanced or reduced by modifying the chemical composition of the carrying solution. The presence of plants and large living organisms may affect the persistence and movement of microorganisms in soils. For example, bacteria in soil are subject to competition and predation from other bacteria such as streptomycetes, myxobacter and Bdellovibrio, and larger soil organisms such as protozoa and nematodes (Peterson and Ward, 1989; Ramadan et al., 1990; Tim et al., 1988). In some instances, the presence of such organisms may enhance mixing of microorganisms within soil. Opperman et al. (1987) studied the effect of the earth worm Eisenia Foetida (Savigny) on movement of cattle slurry through soil. The slurry was obtained from drainage ditches beneath the cattle shed floor. Ten worms were introduced to 17.5cm sand columns. Their activity was found to mix the slurry with the sandy soil to a depth of 17.5 cm as indicated by the movement of coliform bacteria through the soil.
3. Transport Mechanisms Several laboratory and field studies showed that average velocity of microorganisms moving through soil was greater than that of a chemical tracer such as chloride or bromide (Harvey et al., 1989) or the flow of ambient groundwater without any tracer (Wood and Ehrlich, 1978). In a field study, Harvey et al. (1989) injected a mixed bacterial population, collected from ground water and stained with a fluorochrome dye DAPI, together with bromide into a sandy aquifer and followed
148
JAMAL ABU-ASHOUR ET AL.
their movement through the aquifer sediment. They discovered that the stained bacteria travelled faster than the bromide tracer, the peak bacterial concentration was reached 1 to 2 h before that of bromide. This suggests the existence of one or more mechanisms that accelerated microbial movement, and these mechanisms may differ from those involved in transport of the chemical tracer. Several studies investigated mechanisms of microbial transport and attempts have been made to quantify the contributions of each mechanism. Microbial transport through soil depends on a complex set of physical and chemical conditions (Harvey, 1989) which are presently not well understood. There are disagreements in the literature as to the relative importance of various mechanisms to microbial transport through soil. Gray and Williams (1971), as reported in Wollum and Cassel (1978), summarized four modes of microbial transport through soil. They are: (1) movement in water films due to motility of the microorganisms; (2) hyphae elongation, a mode whereby microorganisms move from one water film to another; (3) microbial growth which may contribute to microbial transport; and (4) microbial dispersion through soil by water movement. The latter mode is independent of microbial motility or growth. In addition, microorganisms can be carded by the water flow. Tim et al. (1988) suggested that transport and attenuation mechanisms which affect microbial movement through soil can be divided into physical, geochemical and biological processes. We have adopted these sectional headings in the discussion that follows. 3.1.
PHYSICAL PROCESSES
The principle physical pro~esses for microbial movement through porous media are convection or advection and hydrodynamic dispersion. In advection, microorganisms are carried with bulk water flow (Yates and Yates, 1990) and their movement is governed by the velocity of water (McCoy and Hagedorn, 1979). In a simple model, advection is equal to the average velocity of groundwater as determined from the product of hydraulic conductivity and hydraulic gradient all divided by porosity (Corapcioglu and Haridas, 1984). Hydrodynamic dispersion is the spreading of microorganisms as they move along the water path as a result of both microscopic and macroscopic effects (Tim et al., 1988). It is measured by determining the concentration vs. time dependence of a tracer with respect to a sampling point in the flow path. It may be described by the general transport equation (Equation (1)) in vectorial form (Matthess et aL, 1988; Pekdeger and Matthess, 1983): OC (D ) Vw Ot = d i v RddgradC --Rdd g r a d C - A C
where
(1)
TRANSPORTOFMICROORGANISMSTHROUGHSO~
149
D = coefficient of hydrodynamic dispersion gradC = concentration gradient v~o = average groundwater velocity Rd = retardation factor A = elimination constant. Two distinct processes are operative in dispersion: molecular diffusion and mechanical mixing (Yates and Yates, 1990). In addition, the mobility of microorganisms may cause some dispersion (Tim et al. 1988). Diffusion is defined as the spreading of microorganisms due to a concentration gradient. It is considered to be of negligible importance in bacterial transport compared to mechanical mixing (Yates and Yates, 1990). However, diffusion is an important transport mechanism when small particles (
