A multi-purpose rainfall simulator for field infiltration and erosion studies.
The rainfall simulator (RFS) design described in this paper was developed primarily for the project `Post-mining Landscape Parameters for Erosion and Water Quality Control', which was jointly funded by the Australian Coal Association Research Programme and BHP Australia Coal Ltd, Callide Coalfields Limited, Capricorn Coal Management Pty Ltd, Curragh Queensland Mining Limited, MIM Holdings Limited, and Pacific Coal Pty Ltd.
In the initial 3 years of the project, field rainfall simulator studies of soil and soil erodibility were carried out on 14 open-cut coal mines and one mining lease, with a geographic spread of sites of approximately 1000 km. In a subsequent 3-year extension of the project, field studies of effects of vegetation and surface roughness were carried out at 6 of the mine sites.
User requirements for the rainfall simulator
Two aspects of the rainfall simulator design were accepted from the outset. The first was that the simulator would use nozzles spraying downwards, to minimise problems with wind blowing `rainfall' off the plots prepared to receive it. The second was that the simulator would use the Veej et 80100 nozzles widely used in previous USDA-ARS and QDPI research (Meyer and McCune 1958; Barnett 1977; Loch and Donnollan 1983a, 1983b). There is some variation in reported kinetic energies produced by the nozzles. Perhaps the most accurate assessment comes from Duncan (1972). He used photographic methods to measure both drop size and velocity, and found that with a fall height of 2.4 m, the smaller drops had velocities greater than the terminal velocities of similar-sized drops in still air. Kinetic energy calculated for the Veejet 80100 nozzle on the basis of measured drop size and velocity was 29.49 J/[m.sup.2].mm, similar to the kinetic energy reported for natural rain at intensities [is greater than] 40 mm/h by Rosewell (1986). From distrometer measurements that actually counted only 1.96% of the drops, D. M. Silburn and J. L. Foley (pers. comm.) estimated a kinetic energy for the 80100 nozzle of 24 J/[m.sup.2].mm. However, this can be expected to be a slight underestimate due to the extremely low proportion of drops counted, and the fact that those drops would have been mainly at the edge of the spray pattern where the drop size distribution is slightly finer (Duncan 1972). Despite these apparent uncertainties, it can be concluded that the 80100 nozzles give reasonable simulation of kinetic energies of intense natural rain.
Other user requirements, based on technical and logistical considerations, were: (i) flexibility in use, with ability to vary plot size and hence, the erosion processes studied, and also ability to alter the rainfall intensity applied; (ii) suitability for use on steep slopes (up to 40%), with the ability to minimise variation in nozzle pressures as a result of head differences in the manifold to which nozzles were attached being important; (iii) high portability, not only from plot to plot, but also for transport from one site to another; (iv) economy in water use (as supplies of good quality water were limited in some of the areas visited); (v) reliability and ease of repair; and (vi) suitability for operation by a field work team of 2 or 3 persons. From these requirements, it was concluded that the simulator should be of a modular design, with a robust structure that was also lightweight and simple in design.
The modular design allows the length of each manifold to which nozzles are attached to be kept short, thus reducing the magnitude of differences in head and of resulting variations in operating pressure at the nozzles. It also enables plot sizes to be varied, as the number of modules joined together to rain on a plot can be varied greatly (within the limitations of resources).
Development from previous rainfall simulator designs
Rainfall simulation has been widely used for erosion/infiltration research in the Darling Downs region of southern Queensland. The initial designs used were a rainulator (McKay and Loch 1978) used for erosion studies (e.g. Loch and Donnollan 1983a, 1983b; Loch and Thomas 1987; Loch and Donnollan 1988), and a rotating disk rainfall simulator (Morin et al. 1967)used for studies of infiltration (e.g. Connolly et al. 1991; Foley et al. 1991).
Subsequently, a laboratory rainfall simulator was constructed for studies of aggregate breakdown under rainfall wetting (Loch 1989; Loch and Foley 1992). This simulator was based on the design of Bubenzer and Meyer (1965), and used 2 Veejet 80100 nozzles mounted on a manifold that oscillated so that the spray fans swept across the plot. The oscillation of the manifold was driven by an automotive windscreen wiper motor. An electronic control system was built to allow the rate of oscillation of the manifold (and hence, the rainfall intensity) to be varied.
Satisfaction with the performance of the laboratory rainfall simulator led to the construction of a similar unit for field infiltration studies. Using 3 Veejet 80100 flat fan nozzles, the simulator allowed a plot 2.0 by 1.8 m to be studied' or 2 adjacent plots each 0.9 by 2.0 m. The oscillating manifold was supported by an aluminum A-frame structure, and the machine was transported on a trailer fully assembled. It has been used widely throughout southern and central Queensland for studies of soil infiltration properties and nutrient movements (Bridge and Bell 1994; Costantini et al. 1995; Loch et al. 1995; Bell et al. 1997, 1998), and for field days and demonstrations to farmer groups.
The field rainfall simulator using the oscillating manifold formed the basis for the rainfall simulator that is described in the following sections of this paper. Considerable design and development were necessary to ensure that the rainfall simulator met the user requirements outlined in the previous section.
Description of design
The RFS frame is constructed as base modules to which additional modules can be connected to increase the length of rainfall simulator plot, depending on experimental requirements (Fig. 1). The base modules can be of either 2-nozzle or 3-nozzle configurations. This corresponds to module lengths of 2 or 3 m, with nozzles set at a 1-m spacing along the manifold, with upslope and downslope nozzles being placed 0.5 m from the ends of each module. (This ensures that the 1-m nozzle spacing is maintained when modules are joined together.) The base module consists of a triangular prism with the sloping sides of equal lengths (see Fig. 2). Additional horizontal members of lighter material are added to enable attachment of the droplet catch trays and the nozzle boom. The module dimensions could be scaled up or down to suit special applications
The frame is constructed from aluminum tubing of 38 mm outside diameter (O.D.) and a 3-mm wall thickness. The bases of the upright tubes are flattened and bent parallel to the ground surface to form feet. Holes drilled in the flattened section enable the unit to be pegged to the ground, if required. Operation on sloping ground may require the module to be guyed to a peg or solid object. Modules are connected by nylon joiners that fit inside the main horizontal frame members. The joiners are machined from solid nylon stock and are about 200 mm in length. Locking pins ensure that the multiple units are securely interconnected.
The nozzle boom consists of a 2-m or 3-m length of 38-mm-O.D. (1.6-mm wall) stainless steel tubing. The boom rotates in 2 plastic bushes (graphite impregnated) that also prevent lateral movement of the boom. Male threaded (1/2" BSPT) unions are welded onto the boom at 1-m spacings. Check valves are threaded onto the bushes to prevent nozzle flow or dripping when the unit is not in use, and Veejet nozzles are fitted to the check valves. The water supply inlet is via a 1 1/4" BSPT tee fitting attached to the boom opposite to the row of nozzles. A tapping from this fitting provides a connection for a pressure gauge, so that the operating pressure of each module can be monitored.
The operational sequence of this RFS relies on a continuous water flow through the raindrop nozzles. Flow not required (as rainfall on the test surface) is recycled via droplet catch trays on either side of the oscillating nozzles (see Figs 1 and 2). During operation, the nozzles oscillate through ~107 [degrees]. Of this travel, the middle 67 [degree s]pies raindrops onto the trial surface, and 20 [degrees] is used at either end of the travel for overlap into the catch trays. Adjusting the frequency of nozzle oscillation alters rainfall intensity from the RFS. Calibration and performance details of the RFS are given in a subsequent section.
The heavy-duty, geared 12 V DC electric motor used initially provided the oscillating action via a simple Pitman arm and lever system. The current system uses a 1 Amp stepper motor with a 25: 1 reduction gearbox, operated at 36 V. Advantages of the stepper motor are that it is able to drive at the same speed in both directions, and that it provides repeatable speeds irrespective of operating conditions (load, temperature). A simplified drive using a small roller chain is now used to connect the motor to the manifold. A sprocket is attached to the stepper motor and a drive pin (which fits inside a chain link) is welded to the nozzle boom.
Water is supplied to all RFS modules from a single electric, 1500 W water pump (Onga Hi-Flo centrifugal model 142) housed in a steel frame with a 120-L reservoir. The single pump-reservoir unit is located at the base of the slope and recycles return flow from the RFS modules. Its reservoir is topped up, via 75-mm-diameter lay-flat hose and a float valve, from 10000- or 5000-L storage tanks at the head of the slope.
The pressure pipe near the pump is fitted with a non-return valve to prevent back-flow through the pump from the supply line and RFS units when the pump is turned off. A mesh filter is fitted at this point to filter water prior to being pumped to nozzles. A 75-mm camlock connects the pump to the main supply line, which consists of 2-m sections of 75-mm aluminum irrigation pipe extending the length of the linked RFS units. Outlets at 2-m spacings are fitted with 38-mm brass ball valves and 32-mm plastic cam-locks to allow individual pressure control to each module. Pressure gauges (0-100 kPa) are mounted on each RFS unit at nozzle level.
Catch trays are used to collect water from the nozzles that is not sprayed onto the plot. The trays have provision for slight adjustment up or down to ensure that the sprays overlap the plot sides sufficiently to ensure even coverage, but wasteful, excessive application is prevented. This also ensures that areas reasonably close to the plots remain dry, assisting the movement of personnel engaged in the trials.
On early versions, catch trays extending the full length of the module were used to collect excess flow from all nozzles on that module. To reduce weight, smaller catch trays fabricated from galvanized tinplate are now used to collect excess flow from individual nozzles. Water collected in the catch trays is gravity fed via flexible hose into a common manifold (100-mm-diam. PVC drainage pipe), and returned to the pump storage tank under gravity.
The control system is composed of a controller circuit board, a stepper motor power supply and driver module, and a manifold position feedback module.
The controller circuit includes a power supply, a micro-controller (Motorola M68hc 11), thumbwheel switches, and transistor switching circuits. The micro-controller runs in single chip mode using only internal random access memory (RAM) and electrically erasable, programmable read-only memory (EEPROM) for data and program storage. The thumbwheel switches set the time period between consecutive sweeps of the spray manifold in 0.1-s increments. The transistor switching circuit is required to provide the correct voltage and current levels to the stepper motor driver module.
The stepper motor power supply provides 36 V DC to the motor driver module. The stepper motor driver module is a commercial unit, with inputs for setting motor direction, step resolution (half or full steps), and motor current setting adjustment.
The manifold position feedback module comprises 2 Hall Effect integrated circuits (ICs) and interface cable mounted on the motor/gearbox assembly. An aluminum disk with 2 small magnets embedded into one surface is attached to the drive sprocket. The magnets are positioned to provide feedback signals indicating the start and end of the manifold movement cycle. The magnet polarities are reversed to eliminate any potential for false triggering resulting in position error.
When powered on, the controller reads the signal from the Hall Effect ICs to determine the manifold's position and if, necessary, moves the manifold to the start position. Five seconds after the unit is switched on, the controller starts to oscillate with a period set by the thumbwheel switches. This period can be changed at any time while the unit is operating, either by changing the thumbwheel setting or via the serial interface to allow multiple RFS units to be controlled simultaneously. This ability to alter rain intensity via computer allows plots to be subjected to rain intensity profiles.
Plot edgings and runoff collection
Plots are enclosed by metal plate of 3 mm thickness hammered into the soil, usually to a depth of 75 mm. Where a single module is used, the plot edging is constructed in a single piece, occupying the top and sides of the plot, with a separate gutter for collection of runoff at the downslope outlet. For larger plots, the edging is composed of overlapping sheets of steel plate, 1200 mm long and 150 mm wide.
The use of tipping buckets for runoff monitoring and their logging through time by computer are described by Loch et al. (1998).
Despite the use of nozzles spraying downwards, in windy conditions there is still a requirement for some form of windbreak to reduce movement of `rain' from the target plot. The form of windbreak used varies with the size of simulator used. Where several modules are joined together, a shade-cloth wind break (30% open cloth) is erected to run the full length of the RFS. The wind break uses the A-frame of the RFS modules for support in the centre and is pegged to the ground along the windward side. Where only a single module is used, it has been more effective to have the windbreak material sewn and joined to obtain complete enclosure of the RFS.
A dual-axle fully enclosed trailer was constructed and fitted out to transport the RFS units and associated equipment between trial sites. The 2.0-t trailer measured 1.8 m wide, 2.0 m high, and 3.0 m long. The RFS units were stacked along either side of the trailer with an access walkway through the middle and full width opening doors to the rear. A separate compartment housed a 4 kVA petrol generator, which supplied power for all electrically operated field equipment.
Rainfall simulator performance
Spatial uniformity of intensity
Two modules were joined together to give a 5-m-long RFS covering a plot 4 m long and 1.5 m wide. The RFS was calibrated by measuring volumes of rainfall intercepted by rows of plastic jars located directly under and midway between each nozzle so that each row was 0.5 m apart. Each row consisted of 5 jars, with the centre jar being exactly in the centre of the plot. The outer jars were located so that their outer edge was 0.75 m from the centre of the plot, with the other jars being centred 0.37 m from the centre of the plot. Additional jars were placed on the centre line of the plot, midway between each row. Jar diameters were 10.96 cm. Plastic mesh was placed in the jars to prevent splash losses.
Results obtained after some initial adjustment of the RFS are shown in Tables 1 and 2, which indicate coefficients of variation (CV) of 8-10% for the body of the plot (with the end rows excluded). When the end rows were included, CVs for the data in Tables 1 and 2 were in the range 12.2-13.4%.
Table 1. Spatial distribution of measured intensities (mm/h) for a cycle time of 3.5, simulator run for 20 min Rows in bold were directly under nozzles Distance Measured rainfall intensities (mm/h) downslope (m) for distances across plot (m) of: 0.05 0.37 0.75 1.12 1.43 0 32 40 40 38 37 0.25 48 0.50 38 46 48 48 45 0.75 49 1.00 44 56 57 54 54 1.25 52 1.50 38 45 48 45 44 1.75 46 2.00 49 59 56 56 52 2.25 51 2.50 40 45 47 45 41 2.75 51 3.00 45 51 53 51 43 3.25 51 3.50 43 48 51 48 43 3.75 48 4.00 37 38 48 41 35 Table 2. Spatial distribution of measured intensities (mm/h) for a cycle time of 2.0, simulator run for 9.5 min Rows in bold were directly under nozzles Distance Measured rainfall intensities (mm/h) downslope (m) for distances across plot (m) of: 0.05 0.37 0.75 1.12 1.45 0 60 70 70 67 64 0.25 74 0.50 74 84 87 80 80 0.75 94 1.00 77 94 97 94 87 1.25 100 1.50 84 94 94 87 87 1.75 87 2.00 94 98 94 94 87 2.25 105 2.50 85 90 94 90 84 2.75 87 3.00 80 94 100 94 87 3.25 87 3.50 74 77 82 80 72 3.75 82 4.00 72 74 78 78 67
Earlier calibrations by D. N. Orange and B. J. Bridge (unpublished data) for a single module using a windscreen wiper motor drive system gave similar results. For a plot area 2 m long and 2 m wide, the CV ranged from 11.5% to 14.3% for measurements at intensities ranging from 21.5 to 184 mm/h. The greatest variation was noted at the corners of the plots, and if those 4 points were deleted from the data, the CV was reduced to 9.4-12.5%. For a central area 1.5 m long and wide, the CV was only 6.3%, illustrating that increasing plot area also increases spatial variability due to inclusion of edge effects. Effects of intensity on the CV were small, with a slight increase in CV as intensity decreased.
This degree of variability is low relative to other simulator types. For example, Marston (1980) reports intensity data for a rotating disc simulator (Morin et al. 1967) that show a CV of 24.6% over a plot of only 1 [m.sup.2] area. Lascelles et al. (2000) report uniformity coefficients largely in the range 0.7-0.85 (roughly equivalent to CVs of 15-30%) for a simulator using 4 full-cone nozzles covering a plot area of 7 [m.sup.2].
Because the simulator is based on flat fan nozzles that sweep to and fro across the plot, there is consistent variation in intensity in 2 directions.
The major component of spatial variability is up and down the plot, with peaks under the nozzles and troughs between them. This is minimised by overlap of the nozzles, but still causes consistent variation as shown in Fig. 3.
The minor component of spatial variability is across the plot, with intensity reaching a peak in the centre of the plot, and falling away towards the edges of the plots. This variation can be minimised by adjustment of the catch troughs on the simulator (Fig. 4) to reduce any interception of the sprays until the fan clears the edge of the plot.
For erosion studies, the low variation in intensity across the plots is desirable, as it does not create continuous downslope lines of high rainfall erosivity with increased potential for rill formation. Equally, the repeating pattern of intensity variation down the plot should have a minimal effect on erosion processes associated with downslope accumulation of flow and flow tractive force.
The other main points of intensity variation are at the upper and lower ends of the plot. Intensity falls away at these points because some of the overlapping of nozzles is lost. On longer plots, this has less effect on calculated coefficients of variation than on shorter plots (where its effect on such statistics is greater).
Control of intensity
The measured relationship between the setting of the control boxes and intensity is shown in Fig. 5. In practice, for a range of reasons, this relationship could be expected to vary slightly, and it should at no time be used as a substitute for adequate monitoring of rainfall intensities during field experimentation.
Use of the simulator for erosion and infiltration studies
The size of the simulator and the potential to consider a range of plot sizes have enabled studies to consider inter-rill erosion on short plots ([is less than] 3 m long), rill plus inter-rill erosion on plots 12 m long, and rill erosion by application of overland flows to the 12-m-long plot. The approach used will naturally depend on the type of soil erosion model for which parameters are to be obtained. Rill and inter-rill processes are considered separately when determining parameters for process-based erosion models, whereas a 12-m-long plot can give reasonable estimates of the Modified Universal Soil Loss Equation (MUSLE) (Onstad and Foster 1975) parameters in situations where the plot length and dominant erosion processes are not unduly different to those found under field conditions. The derivation and application of a range of erosion model parameters from differing plot sizes are illustrated by Loch (2000b), and Loch et al. (2000).
For infiltration studies, a range of plot sizes have been used, from 18 [m.sup.2] to 1.5 [m.sup.2]. Desirable levels of replication for infiltration studies vary considerably, depending on the spatial variability of the treatment or situation being studied. As well, there are options to use either large plots that cover the range of site variation, or small plots to assess the various components of site variability.
Intermittency of rain
Like many simulators, this machine applies rain in pulses as the nozzles sweep across the plot. Although noting that natural rain seldom occurs in a steady or uniform fashion, it can also be pointed out that the pulsed rainfall application does not appear to affect or distort the parameters obtained. For example, Connolly et al. (1991) and Connolly and Silburn (1995) found that infiltration parameters derived from rainfall simulation studies (using a simulator that applied rain intermittently) gave good prediction of observed runoff hydrographs from field catchments. Kinnell (1993) found no difference between continuous and intermittent rain in the time-averaged sediment concentrations carried by inter-rill flows.
Field studies using 6 modules to cover a plot 12 m long and 1.5 m wide were found to need typically a field crew of 3 staff, and 1 or 2 laboratory technicians. Depending on the range of measurements made and the proximity of plots, an experienced field crew could run 2-3 plots per day. For example, typical `runs' included raining on a 12-m-long plot for 30 min, followed by inter-rill erosion measurement on a shorter plot, and then overland flow studies on the 12-m-long plot. These measurements are described by Loch (2000b).
Automation of runoff measurement and simplification of data recording described by Loch et al. (1998) gave some streamlining of operations, usually allowing some of the field crew to prepare a new plot for study while the simulator was still running on an existing plot.
Since its construction in 1993, the 6-module simulator has been used in studies on open cut coal mines in southern and central Queensland (Loch 2000a; Sheridan et al. 2000), and also in studies on mines in New South Wales (Loch 2000b) and Western Australia. It has been used in New South Wales and Queensland for studies of pesticide movement in runoff from irrigated cotton (Silburn et al. 1996, 1998; Silburn and Connolly 1998), for studies of erosion and nutrient movement from freshly planted pine plantations, and for studies of movement of nutrients from dairy effluent applied to pastures in southern and northern Queensland. Smaller numbers of modules have been used in laboratory studies (Sheridan et al. 2000), and in studies of erosion from forest roads (Costantini et al. 1999).
The mobility of the simulator from plot to plot and its ease of transport over large distances have clearly been an advantage. As well, its reliability has been a major factor in enabling efficient use of rainfall simulation in a wide range of environments and locations.
Barnett AP (1977) A decade of K-factor evaluation in the Southeast. In `Soil erosion: prediction and control', pp. 97-104. (Soil Conservation Society of America: Ankeny, IA)
Bell M J, Bridge B J, Harch GR, Orange DN (1997) Physical rehabilitation of degraded Krasnozems using ley pastures. Australian Journal of Soil Research 35, 1093-1113.
Bell M J, Moody PW, Connolly RD, Bridge BJ (1998) The role of active fractions of soil organic matter in physical and chemical fertility of Ferrosols. Australian Journal of Soil Research 36, 809-819.
Bridge B J, Bell MJ (1994) Effect of cropping on the physical fertility of krasnozems. Australian Journal of Soil Research 32, 1253-1273.
Bubenzer GD, Meyer LD (1965) Simulation of rainfall and soils for laboratory research. Transactions ASAE 8, 73 and 75.
Connolly RD, Freebairn DM, Bell MJ (1998) Change in soil infiltration associated with leys in southeastern Queensland. Australian Journal of Soil Research 36, 1057-1072.
Connolly RD, Freebairn DM, Bridge BJ (1997) Change in infiltration characteristics associated with cultivation history of soils in south-east Queensland. Australian Journal of Soil Research 35, 1341-1358.
Connolly RD, Silburn DM (1995) Distributed parameter model (ANSWERS) applied to a range of catchment scales using rainfall simulator data. II. Application to spatially uniform catchments. Journal of Hydrology 172, 105-125.
Connolly RD, Silburn DM, Ciesiolka CAA, Foley JL (1991) Modelling hydrology of agricultural catchments using parameters derived from rainfall simulator data. Soil and Tillage Research 20, 33-44.
Costantini A, Loch R J, Connoly, RD, and Garthe, R (1999). Sediment generation from forest roads: beds and eroded sediment size distributions, and runoff management strategies. Australian Journal of Soil Research 37, 947-964.
Costantini A, Loch R J, Glanville SF, Orange DN (1995) Evaluation of the potential to dispose of sewage sludge. I. Soil hydraulic and overland flow properties of Pinus plantations in Queensland. Australian Journal of Soil Research 33, 1041-1052.
Duncan MJ (1972) The performance of a rainfall simulator and an investigation of plot hydrology. MAgrSc thesis, Lincoln College, University of Canterbury, New Zealand.
Foley JL, Loch RJ, Glanville SF, Connolly RD (1991) Effects of tillage, stubble, and rainfall energy on infiltration. Soil and Tillage Research 20, 45-55.
Kinnell PIA (1993) Sediment transport by shallow flows impacted by pulsed artificial rainfall. Australian Journal of Soil Research 31, 199-207.
Lascelles B, Favis-Mortlock D, Parsons A, Guerra A (2000) Spatial and temporal variation in two rainfall simulators: implications for spatially explicit soil erosion modelling. Earth Surface Processes and Landforms 25, 709-721.
Loch RJ (1989) Aggregate breakdown under rain: its measurement and interpretation. PhD Thesis, University of New England, Armidale, NSW.
Loch RJ (2000a). Effects of vegetation cover on runoff and erosion under simulated rain and overland flow on a rehabilitated site on the Meandu Mine, Tarong. Australian Journal of Soil Research 38, 299-312.
Loch RJ (2000b) Using rainfall simulation to guide planning and management of rehabilitated areas: I. Experimental methods and results from a study at the North Parkes mine. Land Degradation and Development 11,221-240.
Loch RJ, Bourke JJ, Glanville SF, Zeller L (1998) Software and equipment for increased efficiency of field rainfall simulation and associated laboratory analyses. Soil & Tillage Research 45, 341-348.
Loch RJ, Connolly RD, Littleboy M (2000) Using rainfall simulation to guide planning and management of rehabilitated areas: II. Computer simulations using parameters from rainfall simulation. Land Degradation and Development 11, 241-255.
Loch RJ, Costantini A, Barry GA, Best EK (1995) Evaluation of the potential to dispose of sewage sludge. II. Potential for off-site movements of solids and solutes. Australian Journal of Soil Research 33, 1053-1062.
Loch RJ, Donnollan TE (1983a) Field rainfall simulator studies on two clay soils of the Darling Downs, Queensland. I. The effects of plot length and tillage orientation on erosion processes and runoff and erosion rates. Australian Journal of Soil Research 21, 33-46.
Loch RJ, Donnollan TE (1983b) Field rainfall simulator studies on two clay soils of the Darling Downs, Queensland. II. Aggregate breakdown, sediment properties and soil erodibility. Australian Journal of Soil Research 21, 47-58.
Loch RJ, Donnollan TE (1988) Effects of the amount of stubble mulch and overland flow on erosion of a cracking clay soil under simulated rain. Australian Journal of Soil Research 26, 661-672.
Loch RJ, Foley JL (1992) Effects of plot size on size distributions of water-stable material at the soil surface under simulated rain. Australian Journal of Soil Research 30, 113-118.
Loch RJ, Thomas EC (1987) Resistance to rill erosion: observations on the efficiency of rill erosion on a tilled clay soil under simulated rain and run-on water. Catena Supplement 8, 71-83.
Marston D (1980) Rainfall simulation for the assessment of the effects of crop management on soil erosion. MNatResour thesis, University of New England, Armidale.
McKay ME, Loch RJ (1978) A modified Meyer rainfall simulator. In `Proceedings Conference on Agricultural Engineering, Toowoomba'. pp. 78-81. (Institution of Engineers: Australia.)
Meyer LD, McCune DL (1958) Rainfall simulator for runoff plots. Agricultural Engineering 39, 644-8.
Morin J, Goldberg D, Seginer I (1967) A rainfall simulator with a rotating disc. Transactions ASAE 10, 74-77, 79.
Onstad CA, Foster GR (1975) Erosion modeling on a watershed. Transactions ASAE 18, 288-292.
Rosewell CJ (1986) Rainfall kinetic energy in eastern Australia. Journal of Climate and Applied Meteorology 25, 1695-1701.
Sheridan G, So HB, Loch R J, Pocknee C, Walker CM (2000) Using laboratory scale rill and interill erodibility measurements for the prediction of hillslope scale erosion on rehabilitated coal mine soils and overburdens. Australian Journal of Soil Research 38, 285-299.
Silburn DM, Connolly RD (1998) Some science behind best practices for managing pesticides in runoff-recent experience in the cotton industry. In `Proceedings of the National Symposium on Pesticide Management in Catchments'. 4-5 Feb. 1998, University of Southern Queensland, Toowoomba. pp 107-117. (Condamine-Balonne Water Committee: Dalby, Qld)
Silburn DM, Hargreaves P, Budd N, Glanville SG (1996) Endosulfan on cotton plants--persistence and washoff during rain. In `INTERSECT 96--International Symposium on Environmental Chemistry and Toxicology Sydney, 14-18 July 1996. Abstract No. 058. (RACI: North Melbourne)
Silburn DM, Waters DK, Connolly RD, Simpson BW, Kennedy IR (1998) Techniques for stabilising soil erosion on cotton farms. In `Minimising the impact of pesticides on the riverine environment: key findings from research with the cotton industry'. (Eds NJ Schofield, VE Edge) Occasional Paper 23/98, pp. 99-105. (Land and Water Resources Research and Development Corporation: Canberra)
Manuscript received 15 June 2000, accepted 9 October 2000
R. J Loch(AB), B. G. Robotham(AC), L. Zeller(A), N. Masterman(AD), D. N. Orange(AE), B. J. Bridge(HI), G. Sheridan(FG), and J.J. Bourke(A)
(A) Queensland Department of Primary Industries, PO Box 102, Toowoomba, Qld 4350, Australia.
(B) Current address: Landloch Pty Ltd, 67 Bridge St, Toowoomba, Qld 4350, Australia.
(C) Current address: Bureau of Sugar Experiment Stations, Private Bag 4, Bundaberg DC, Qld 4670, Australia.
(D) Current address: PO Box 74, Auburn, SA 5451, Australia.
(E) Current address: Queensland Department of Natural Resources, Leslie Research Institute, Toowoomba, Qld 4350, Australia
(F) Department of Agriculture, University of Queensland, St Lucia, Qld 4072, Australia; Current address: Victorian Dept Natural Resources and Environment, Arthur Rylah Research Institute, 123 Brown St, Heidelberg, Vic. 3084, Australia.
(G) CSIRO Division of Soils, Leslie Research Institute, Toowoomba, Qld 4350, Australia.
(H) Current address: 24 Amos Crs, Toowoomba, Qld 4350, Australia.
|Printer friendly Cite/link Email Feedback|
|Author:||Loch, R. J.; Robotham, B. G.; Zeller, L.; Masterman, N.; Orange, D. N.; Bridge, B. J.; Sheridan, G.;|
|Publication:||Australian Journal of Soil Research|
|Article Type:||Statistical Data Included|
|Date:||May 1, 2001|
|Previous Article:||A new procedure to determine soil water availability.|
|Next Article:||Soil factors affecting the availability of potassium to plants for Western Australian soils: a glasshouse study.|