Level 2 Detail of experimental conditions



The study monitored runoff, soil erosion, nutrient loss and crop yield over a range of tillage practices on cropping lands. The cropping model system PERFECT was tested against the observed data.

When and Where?

From 1998 – 2000 a tudy was undertaken on the Kairi research station (Atherton tablelands) approximately 90km from Cairns, North Queensland.

The soil type is predominantly of red ferrosols, The soil profile class, Tolga, is a deep red, structured, uniform, clay soil with a well drained profile. The clay percentage range between 50% (surface) and 75% (1m) and pH (water) between 6.6 (surface) and 5.9 (1m). Bulk density ranges between 1.1 g/cm3 (0 – 0.15m) and 1.3 g/cm3 (0.50 – 1.20m). The soil is formed from basalt flows that lie on level to gently undulating planis and rises.

The experiment incorporated twelve plots (20 X 5m) divided into four treatments, each with three replications, using an incomplete randomised block design. The experiment area covered approximately 0.5 ha with a slope of 6% The tillage practices included conventional tillage (CT), reduced tillage (RT), grass pasture, and bare. CT involved the use of a disc plough during the dry seasons followed by offset discs prior to planting rains in November and December. RT was consisted of two passes of the chisel plough trailing a dead rod. A single pass was made during the mid dry season and the other prior to planting rains, for weed control.

The cropping sequence remained under the conventional and reduced tillage treatments were peanuts (1997 – 1998) and maize (1999-2000, 2000-2001). The grass plots were converted from reduced tillage in 1997-1998 and seeded with Braccharia decumbens. Following establishment (1 year) grass plots were mowed approximately twice each year. Grass cuttings were not removed from any plots. Bare plots were initially disc ploughed early in the season, followed by an offset disc and rotary hoe. Plots remained bare by the use of herbicides to remove any emergent weeds. Ripping in all cropped and bare plots was performed to 0.55-0.65m during July (ever 2-3 years). Herbicides were also used in the cropped plots to control weeds.

Sowing took place for late November to late December. Maize and peanuts were sown at a row spacing of 0.90 m. At sowing fertilisers were banded at approximately 60mm beside the seed. Fertilisers were applied to maize at concentration of 80 kg/ha (N), 25-30 kg/ha (P) and 25 kg/ha (K) and peanuts at concentrations of 40 kg/ha (P) and 40kg/ha (K).

Crop yield parameters (plant density, vegetative yield and grain yield) were recorded randomly by selecting two 5m rows within each plot.  Vegetative yield was measured from 2.5m of each 5m row. Cobs were de-husked and shelled by hand and grain moisture was determined. Vegetative yield was dried and weighed. Peanuts were hand harvested from a randomly selected 2m row for all plots. Nuts were cleaned of soil and weighed in their shell. Kernels were then sorted by quality indices using the local grading standard of the peanut marketing board. Grass plots were sampled annually using a randomly located quadrat. Grass was then cut to ground level using hand shears. Cover measurements (combined canopy and contact) were recorded visually at intervals throughout the wet seas. Gravimetric moisture content was measured with the use of a hand auger. Soil samples were collected at a depth of 1m.

At the bottom of each plot was a trough that lead to a 4-L tipping bucket with a dual tipping mechanism. Campbell data loggers were used to count each time a tip occurred. Two pluviometers were connected to the dataloggers to record rainfall. Runoff water and sediment were recorded after a significant rain event. A splitter device collected a subsample (approximately 22mL) from each tipping bucket. Runoff was collected 6 hours of runoff from the splitter and refrigerated. Bedload sediment was collected from the bottom of the trough and oven dried at 40°C. Subsamples of runoff water, suspended sediment, and bedload were also taken for subsequent chemical analysis for total N and P.

Suspended load was calculated by filtering runoff via a glass fibre filter. Total nitrogen and phosphorus were measured via the Bran+Luebbe (1990) technique. The plant-available water capacity of the Ferrosol was determined based on soil survey results from Malcolm et al. (1998) and previous site sampling undertaken at wet and dry seasons. The leaf area index (LAI) crop model within PERFECT was used to simulate crop growth. Crop parameters adjusted included root depth, daily root growth and the proportion of growing season at a specified LAI and growing degree-days.

Data from runoff loggers were analysed using internal Department of Natural Resources and Water processes, while the EI30 was calculated using the Wishchmeier and Smith (1978) calculation. Enrichment ratios were calculated using the Hashim et al. (1998) calculation, assuming a topsoil depth of 0.10m. Data were analysed for significant differences using ANOVA, while linear regression model identified the level of agreement among parameters measured and modelled values

Project administration

Site identifier code: na

Principal investigator: Lex Cogle

Principal data manager: -na

Principal organizations: Queensland Department of Natural Resources and Water

The Department of Primary Industries

Data custodian: Department of Environment and Resource Management (DERM)

Key co-operators: -na

Data access policy: Research has been published.

Planned pathway for data: completed study

Data warehousing: for ongoing studies -na

Planned data upload frequency: for ongoing studies - na

Key references and sources of this data synthesis

  1. Cogle A.L, Keating M.A, Langford P.A, Gunton J, Webb I.S. 2011. Runoff, soil loss, and nutrient transport from cropping systems on Red Ferrosols in tropical northern Australia. Soil Research. 49:87-97.


Cropping, North Queensland, runoff, soil erosion, nutrients, crop yield, tillage practice


Level 1, level 3, level 4 and level 5