The practice of inoculating forage legumes with rhizobia strains is widespread. It is assumed that the inoculated strain determines the performance of the symbiosis and nitrogen fixation rates. However, native-naturalized strains can be competitive, and actual nodule occupancy is often scarcely investigated. In consequence, failures in establishment, and low productivity attributed to poor performance of the inoculant may merely reflect the absence of the inoculated strain in the nodules. This study lays out a strategy followed for selecting a Rhizobium leguminosarum sv. trifolii strain for white clover (Trifolium repens) with competitive nodule occupancy. First, the competitiveness of native-naturalized rhizobia strains selected for their efficiency to fix N₂ in clover and tagged with gusA was evaluated in controlled conditions with different soils. Second, three of these experimental strains with superior nodule occupancy plus the currently recommended commercial inoculant, an introduced strain, were tested in the field in 2 years and at two sites. Plant establishment, herbage productivity, fixation of atmospheric N₂ (¹⁵N natural abundance), and nodule occupancy (ERIC-PCR genomic fingerprinting) were measured. In both years and sites, nodule occupancy of the native-naturalized experimental strains was either higher or similar to that of the commercial inoculant in both primary and secondary roots. The difference was even greater in stolon roots nodules, where nodule occupancy of the native-naturalized experimental strains was at least five times greater. The amount of N fixed per unit plant mass was consistently higher with native-naturalized experimental strains, although the proportion of N derived from atmospheric fixation was similar for all strains. Plant establishment and herbage production, as well as clover contribution in oversown native grasslands, were either similar or higher in white clover inoculated with the native-naturalized experimental strains. These results support the use of our implemented strategy for developing a competitive inoculant from native-naturalized strains.

Water management impacts both methane (CH₄) and nitrous oxide (N₂O) emissions from rice paddy fields. Although controlled irrigation is one of the most important tools for reducing CH₄emission in rice production systems it can also increase N₂O emissions and reduce crop yields. Over three years, CH₄ and N₂O emissions were measured in a rice field in Uruguay under two different irrigation management systems, using static closed chambers: conventional water management (continuous flooding after 30 days of emergence, CF30); and an alternative system (controlled deficit irrigation allowing for wetting and drying, AWDI). AWDI showed mean cumulative CH₄ emission values of 98.4 kg CH₄ ha⁻¹, 55 % lower compared to CF30, while no differences in nitrous oxide emissions were observed between treatments ( p > 0.05). No yield differences between irrigation systems were observed in two of the rice seasons ( p > 0.05) while AWDI promoted yield reduction in one of the seasons ( p< 0.05). When rice yield and greenhouse gases (GHG) emissions were considered together, the AWDI irrigation system allowed for lower yield-scaled total global warming potential (GWP). Higher irrigation water productivity was achieved under AWDI in two of the three rice seasons. These findings suggest that AWDI could be an option for reducing GHG emissions and increasing irrigation water productivity. However, AWDI may compromise grain yield in certain years, reflecting the importance of the need for fine tuning of this irrigation strategy and an assessment of the overall tradeoff between relationships in order to promote its adoption by farmers.