Measurement and prediction of agrochemical movement in tropical sugar production

• Mauritius Sugar Industry Research Institute, Mauritius

Sugarcane and its manufacture are vital to the agricultural economies of Mauritius and coastal Queensland, Australia, but the huge yields depend on large inputs of fertilisers and pesticides. Both regions also derive considerable revenue from tourism and have expanding residential areas. When attractive scenery or pastimes are sullied by chemical contamination that may also endanger drinking water supplies, the sugar industry is often blamed.
The sugar industry needed better information about the fate of its chemical inputs so it could justify the agricultural practices it applied to growing crops in sensitive catchments and drainage basins. Growers wanted to ensure that their resources caused no damage to marine habitats including the Great Barrier Reef, or to town water supplies, or to the excellent fishing in local estuaries and on the coast itself. Apart from considerations of aesthetics, pleasure, health and responsibility to the natural environment, such damage would imply inefficient use of agricultural resources.

This project aimed to gather information about the fates of pesticides such as atrazine and chlorpyrifos and nutrients such as nitrates and phosphates, in relation to time of application, weather, soil types and hydrology. By project end the researchers from the Mauritius Sugar Industry Research Institute (MSIRI) and the Queensland Department of Natural Resources expected to produce recommendations to guide growers in management practices that would minimise movement of agrochemicals out of sugarcane crops.

Researchers chose which pesticides and nutrients to track in detail after they determined levels currently used in each sugarcane-growing catchment area of the respective countries. They measured the distributions of agrochemicals and major nutrients down the soil profiles of each experimental site, in the sediments carried in surface runoff, and along the beds of local streams. Runoff, leachates and groundwater were also analysed for pesticides.
Sample sites on farms in Mauritius and north Queensland were treated the same as sugar crops. However, there were experimental plots for researchers to test various techniques and timings. Researchers measured depth and volume of runoff, and the characteristics of leachate in relation to weather and crop management. In addition, bores were sunk to access groundwater for sampling and estimation of flow. Changes down the soil profiles at three or more sites and water quality in key local water bodies were monitored.
The new data added to established knowledge and complemented selected existing models. Incorporation of the data into these models gave researchers a broad picture of the soil and water processes involved in the transport of agrochemicals and their persistence over time - factors that determine their fate in sugarcane-growing areas.

A survey of chemical and nutrient inputs was completed. In the study of the hydrology of agrochemical movement it was planned that data should be collected from three full sugar-growing seasons. However, starting the project in January rather than July delayed the first season's data collection. As well, one season in each country was unusually dry; thus the data collected were not 'typical' and were therefore of limited value for modelling. There were no extremely wet seasons during the project and thus such events were not covered in the analysis. Data collection was undertaken for two more seasons in order to estimate deep drainage and further develop the model for pesticide and nutrient transport. This was included in an 18-month project extension. The scientists found that pesticides applied to sugar cane were most actively transported off-site when rain fell immediately after pesticide application. Mauritian soils were shown to vary considerably in their capacity to bind pesticides.
Reviewers commended efforts to establish the facility for chemical monitoring in Mauritius and the apparent commitment of the Mauritian team to good analytical practices. They also noted satisfactory progress in catchment-scale hydrology and modelling, and affirmed that although it would be difficult to develop a true decision support (DS) system from even 4-5 seasons' data the results should be valuable in developing industry guidelines for chemical use in cane cropping systems in both countries.
Training objectives in pesticide analysis and field hydrology were met. While two Mauritian project staff received short-term training in modelling, it was not the intention of the project to provide significant training in modelling to the overseas project partners.
The reviewers stressed that it was critical to the value of the project for technology transfer/ extension of the results to the Australian and Mauritian sugar industries to be given highest priority.