Soil quality assessment in rice production systems

In the state of Rio Grande do Sul, Brazil, rice production is one of the most important regional activities. Farmers are concerned that the land use practices for rice production in the Camaquã region may not be sustainable because of detrimental effects on soil quality. The study presented in this thesis aimed (a) to describe and understand how rice farmers assess soil quality; (b) to propose a minimum data set (MDS) to assess soil quality; (c) to establish which soil quality indicator(s) can be used to guide management leading to sustained crop production and (d) to reconcile local and scientific knowledge. To accomplish these objectives the research was based on two methodological procedures to assess soil quality: qualitative (local knowledge) and quantitative (scientific knowledge). The qualitative study led to the understanding of soil quality from a rice farmers’ perspective. Farmers from Camaquã were found to have detailed knowledge and a holistic view of the quality of the soil they are cultivating. Eleven indicators were mentioned as good soil quality indicators. Four out of these 11 soil quality indicators were considered “significant” by the farmers (soil colour, earthworms, soil organic matter and soil friability), while three indicators were found to be “useful” by the farmers in their decision-making: spontaneous vegetation, rice plant development and soil colour. In order to assess soil quality following a scientific approach, the three main management systems for irrigated rice in Rio Grande do Sul were chosen: conventional (dry seedbed preparation and sowing, high tillage intensity), semi-direct (dry seedbed preparation and sowing, low tillage intensity), and pre-germinated (seedbed preparation and sowing on inundated fields, high tillage intensity). Twenty-one rice fields covering these management systems and two different soil great groups were selected for investigation. In each field, five replicate plots were randomly laid out within an area of 3 ha for sampling. From each plot, 29 soil properties were analysed to establish a MDS using statistical tools in a novel way. The MDS consists of eight significant soil quality indicators: available water, bulk density, mean weight diameter, organic matter, Zn, Cu, Mn and earthworm number. In order to define the usefulness of this scientific approach, and to compare this with the local knowledge, a soil quality index (SQI) was determined. This study demonstrated that soil quality was best assessed when using the entire indicators set of 29 indicators. However, the MDS and farmers’ indicators sets performed almost equally well as the entire indicators set in showing the same trends in differences between management systems, soil textural classes and soil functions. The semi-direct management system resulted in the highest overall SQI, followed by the pre-germinated and conventional systems. A further study was undertaken to assess the effects of the different rice management systems on the physical and chemical soil quality. The results also indicate that the semi-direct management system is more sustainable, whereas the pre-germinated and conventional systems appear to contribute to soil degradation. Finally, a review of the state of knowledge on earthworm diversity in Rio Grande do Sul revealed 36 species and 20 genera, belonging to a total of 7 families in the state. Nine species were found in the rice fields of Camaquã (all new records for the region), two species were reported for the first time for the Rio Grande do Sul state and a new native genus and species of Criodrilidae (Guarani camaqua) was described. It is concluded that statistical procedures identified soil quality indicators which can be applied to monitor soil quality of rice fields and to support management decisions. Using an integrated approach may therefore increase the understanding of the complex nature of soil.