It is widely assumed that agricultural practices have a major impact on soil living organisms. However, the impact of agricultural practices on soil microbes is poorly known, notably for species richness, evenness, and taxonomic composition. The taxonomic diversity and composition of soil indigenous microbial community can be assessed now using pyrosequencing, a high throughput sequencing technology applied directly to soil DNA. Here, we studied the effect of agriculture management on soil bacterial and fungal diversity in a tropical grassland ecosystem of northeastern Laos using 454 pyrosequencing of 16S and 18S rRNA genes. We studied soil microbial diversity of agricultural soils 3 years after conversion from native grasslands. We compared five systems: one tillage, two no-tillage rotational, one no-tillage improved pasture, and one natural grassland. Our results show first that compared to the natural grassland, tillage decreases fungal richness and diversity by ?40 % and ?19 %, respectively and increases bacterial richness and diversity by +46 % and +13 %, respectively. This finding evidences an early impact of agricultural management on soil microbial diversity. Such an impact fits with the ecological concept of "intermediate perturbation"-the hump-backed model-leading to classify agricultural practices according to the level of environmental stress they generate. We found also that land use modified soil microbial taxonomic composition. Compared to the natural pasture, tillage decreased notably the relative abundance of Actinobacteria (by ?6 %), Acidobacteria (by ?3 %) and Delta-proteobacteria (by ?4 %) phyla, and by contrast increased the relative abundance of Firmicutes (by +6 %), Gamma-proteobacteria (by +11 %), and Chytridiomycota (+2 %) phyla. We conclude that soil microbial diversity can be modified and improved by selecting suitable agricultural practices. Moreover no-till systems represented intermediate situations between tillage and the natural pasture and appear therefore as a fair trade-off between the need for agriculture intensification and soil ecological integrity preservation. (Résumé d'auteur

Agricultural practices affect the physical and chemical characteristics of the soil, which in turn may influence soil microorganisms with consequences on soil biological functioning. However, there is little knowledge on the interactions between agricultural management, soil physicochemical properties, and soil microbial communities, notably in tropical ecosystems with few studies conducted in strongly weathered and acid soils. Here, we investigated the early effect of tillage and crop residues management on top soil physical, chemical, and microbial properties in an acid savannah grassland of northeastern Laos. We initiated a 3-year rotation of rice/corn/soybean under three no-till systems (NTs) distinguished by the cover crops associated prior to and with the main crops, and one conventional tillage-based system (CT). The effect of agricultural management was evaluated 2 years after land reclamation in reference to the surrounding natural pasture (PAS). Our results demonstrate that NTs improve soil physicochemical characteristics (aggregate stability, organic carbon, and cation exchange capacity) as well as microbial abundance (total biomass, bacterial and fungal densities). A significant discrimination of the genetic structure of soil bacterial community was also observed between NTs, CT, and PAS. Interestingly, bacterial abundance and diversity were differently influenced by soil environment changes: microbial density was affected by the quantity and diversity of crop residues, soil organic carbon, and exchangeable base contents, whereas soil bacterial genetic structure was mainly determined by exchangeable aluminum content, pH, cation exchange capacity, and C/N ratio. Altogether, our study represents one of the most complete environmental evaluations of agricultural practices in tropical agrosystems and leads to recommend no-till systems with high residue restitutions to improve the physical, chemical, and microbial properties of tropical acid soils and thus contribute to the sustainability of agriculture in these ecosystems.

Agricultural practices should modify the diversity of soil microbes. However, the precise relationships between soil properties and microbial diversity are poorly known. Here, we study the effect of agricultural management on soil microbial diversity and C turnover in tropical grassland of north-eastern Laos. Three years after native grassland conversion into agricultural land, we compared soils from five land use management systems: one till versus two no-till rotational cropping systems, one no-till improved pasture and the natural grassland. Soils were incubated in microcosms during 64 days at optimum temperature and humidity. Bacterial and fungal diversity were evaluated by metagenomic 454-pyrosequencing of 16S and 18SrRNA genes, respectively. Changes in soil respiration patterns were evaluated by monitoring 12C- and 13C-CO2 release after soil amendment with 13C-labelled wheat residues. Results show that residue mineralization increased with bacterial richness and diversity in the tilled treatment 7 days after soil amendment. Native soil organic C mineralization and priming effect increased with fungal richness and diversity in improved pasture and natural grassland. No-till cropping systems represented intermediate situations between tillage and pasture systems. Our findings evidence the potential of controlling soil microbial diversity by agricultural practices to improve soil biological properties. We suggest the promotion of no-till systems as a fair compromise between the need for agriculture intensification and soil ecological processes preservation. (Résumé d'auteur

In north-eastern Laos, the savannah grasslands of the Plain of Jars cover vast areas of potentially cultivable land. However, soil acidity, low inherent fertility, and the absence of alternatives to tillage represent significant constraints to the development of sustainable smallholder agriculture. Our objective was to evaluate the potential for conservation agriculture (CA) to enhance soil productivity and farming system profitability. A three-year rotation of rice/maize/soybean was tested under three fertilization levels and four agricultural systems: one conventional tillage-based (CT) system and three CA systems based on no-tillage with cover crops. After four cropping seasons, our results show that, compared with CT, CA systems led to similar-to-higher grain production, similar-to-higher profits, higher opportunity of livestock system intensification, and higher labour productivity regardless of fertilization levels. While CA represents a relevant alternative to current practices, our results suggest that its contribution to the emergence of a sustainable smallholder agriculture is conditioned by broader institutional transformations, including the enrolment of local manufacturers and traders for deploying no-till implements and seed market channels for cover crops, long-term public support to maintain active research and technical mentoring to farmers, and possibly the integration of ecosystem services in agricultural policy. (Résumé d'auteur