On farm assessment of rice yield variability and productivity gaps between organic and conventional cropping systems under Mediterranean climate

Summary

Organic rice production is characterised by high yield variability and substantial productivity gaps with respect to conventional systems. Variability may be accentuated in areas of erratic climate, such as in the Mediterranean region of La Camargue in southern France. While management recommendations for organic cropping systems are not readily available, innovative farmers develop strategies to achieve high, and less variable, yields. The objectives of this study were to identify the main factors affecting yield variability and the farmer management strategies used to sustain crop productivity while reducing input use. Participatory monitoring of farmer fields for yields, yield components, soil condition, weeds and management practices from 1992 to 2009 resulted in a database of more than 380 entries. These data included continuous, discrete and nominal variables. They were explored using classification and regression trees to describe management strategies under conventional and organic systems and to identify and categorise the main variables associated with rice yield variability. Rice yields varied between 0.5 and 10 t ha−1 under conventional and between 0 and 9 t ha−1 under organic management. Weed competition was the main factor affecting yield for both conventional and organic systems. The gap between current average yields and the estimated yield potential of 10 t ha−1 was on average 2.7 t ha−1 under conventional and 5 t ha−1 under organic management. The latter can be largely attributed to the effect of weed competition. The productivity gap between conventional and organic management fluctuated between c. 1 t ha−1 under conducive conditions and c. 4 t ha−1 under limiting conditions (e.g., severe salinity problems). Strategies to attain high yields under conventional and organic management differed: Under conventional management, a low initial plant stand associated with early sowing was compensated by high tillering rates induced through N fertilisation, while weeds were controlled by herbicides. Under organic management, late sowing allowed a higher initial plant density due to higher temperatures during emergence. This higher density ensured greater competition with weeds and sufficient number of panicles per unit area at harvest. If organic rice production is to be further promoted in Mediterranean regions, such innovations should be supported by technical means such as short cycle varieties adapted to late sowing under high latitudes. Other alternatives to outcompete and/or control weeds without the need for herbicides, notably through irrigation water management, crop rotation or use of cover crops must also be explored. These results indicate that farmer innovations may show possible pathways towards the ecological intensification of current agriculture.