Digital Theses Archive


Tesi etd-08012018-104046

Type of thesis
Weed management using a no-till system with stylosanthes guianensis cover crop in upland rice-based cropping systems in the Mid-West of Madagascar
Scientific disciplinary sector
Membro Prof. PE', MARIO ENRICO
  • cover crop
  • living mulch
  • Madagascar
  • Stylosanthes guianensis
  • upland rice
  • weed management
Exam session start date
Food production in developed countries relies strongly on intensive pesticide use, mineral fertilizers and even genetically modified crops, practices that directly or indirectly put at risk environmental and human health. In poor countries, agriculture is characterized by low external input use and an overall low productivity that is not able to satisfy the local food demand. Facing these situations, consumers, scientists, agronomists, farmers and politicians are becoming more and more aware of the need for more sustainable crop production models. Lately, it was realized that biodiversity can be used in a rational way to conceive crop production models that may help to reduce external inputs, and maintain or even improve crop production. Agroecological practices based on crop diversification such as crop rotation, intercropping and introduction of cover crops in agroecosystems became the focus of several studies in many countries around the world.<br>In the Mid-West of Madagascar, a few agroecological practices were tested and one of them appeared to be adapted to the local context: it is the use of Stylosanthes guianensis, managed as a living mulch cover crop, in upland rice-based cropping systems. Upland rice is indeed an important source of staple food for the fast growing population in this region. Rice was traditionally grown on lowlands were the soil is relatively fertile and weed infestations can be controlled by the flooding and transplanting technique. In the uplands, rice is direct-seeded and therefore loses the size advantage over weeds. Besides, upland rice cannot be submerged because of its topographic location, but receives water supply only by rainfall, losing the standing water that protects against aerobic weeds. Weed management in this system is done only by hand removal because farmers cannot afford herbicides which are anyway well known to be dangerous for environmental and population health. For these reasons, weed infestations in upland rice stands as a huge biotic constraint reducing the already low yield potential due to the poor soil fertility and lack of external inputs. Furthermore, farmers have limited time availability for weeding as a peak workload occurs shortly after the beginning of the rainy season. Many agricultural activities have to be achieved at the same time. Stylosanthes as a legume cover crop was proven to provide nitrogen to the cropping system thus enhancing soil fertility, and thanks to its ability to produce an important biomass it may provide the necessary mulch to cover the soil in order to suppress weed emergence and growth.<br>In order to understand if stylosanthes can be a useful tool in weed management, a few issues should be clarified. Which are the field and management characteristics that impact on weed abundance and composition on real farms? Does stylosanthes managed as a living mulch actually suppress weed infestation in upland cropping system, and which weed species are suppressed or promoted? Does it help to delay the weed-rice competition to save time for farmers? Does this system reduce or increase the rice yield? What are farmers’ perception about weeds and stylosanthes? <br>To answer to these questions a three years study was conducted in 2016, 2017 and 2018 in the Mid-West of Madagascar. A weed floristic recording was performed in 2016 and 2017 on a total of 91 farmer fields in order to describe the weed community composition in upland rice and define field and management characteristics that affected the weed community composition. Field experiments on farms were conducted in 2016 and 2017 to compare the conventional tilled system with the innovative no-till system with the stylosanthes cover crop in terms of weed community composition, weed infestation and rice yield. A split-plot experiment with tillage system in the main plot and fertilization type in the sub-plots was set up in the Ivory research station in the Mid-West in 2016, 2017 and 2018 to determine the treatment effects on weed community composition and abundance, and on rice yield components. Other field experiments on station were conducted in 2017 and 2018 to determine the weed-crop competition timing in the conventional system and in the innovative no-till system. Interviews with farmers were conducted in 2016 and 2017 to highlight farmers’ perception on weeds and the possible use of stylosanthes in weed management.<br>The main results indicated that the timing of the first weeding intervention, crop field age, the presence of maize intercropped with rice and soil management influenced the weed community. The no-till system with stylosanthes suppressed dominant weed species commonly infesting the conventional tilled system, such as Digitaria spp., Richardia scabra and Eleusine indica, consequently reduced the total weed biomass at harvest by 65%, and the total weed cover by 47% up to 93% depending on the sampling date. This innovative system, on the other hand, promoted other weed species namely Mitracarpus hirtus, Cyperus spp., Neojeffreya decurrens and Ageratum conyzoide. The weed suppression effect was mainly attributed to the stylosanthes mulch that limited weed emergence and growth early in the growing season. Mineral inputs combined with manure increased weed cover and weed biomass compared to manure alone. It increased particularly the abundance of Digitaria spp. In weed free conditions, rice yield was not reduced by the innovative system but was observed to be increased in some cases. It might be attributed to nitrogen supply by the legume cover crop and some improvement of the soil fertility such as soil texture or water regime reported in the literature. In case of high weed infestation, the yield loss due to weeds was reduced from 86% to 45% in average in the innovative system compared to the conventional system as a consequence of the weed suppression effect, from 39% to 26% if weed infestation is moderate, but may not reduce yield loss with weed infestation is very low. It indicated that the benefits from the innovative system with respect to yield is more expressed in high weed infestation condition. Mineral input was observed to increase rice yield in general, but decreased the yield when it promoted weed infestation at the same time. It was also observed that the suppression effect of weed emergence by the stylosanthes mulch delayed the critical period of weed interference in upland rice by 11 up to 18 days, and the presence of highly competitive weed species such as Richardia scabra may prompt the start of weed-crop competition at early date. The farmers’ general perception is in agreement with the results of the experiments on the fact that the innovative system decreases the total weed cover and poses no threat to rice yield, but may increase the yield instead.<br>The main limits of the innovative system are: the four to five years necessary for the system to reach its full potential, the investment in the roller-crimper necessary for its management, and the one of two year fallow period required by stylosanthes to accumulate enough biomass for the following rice cropping season. It is possible that farmers with limited crop fields and resources cannot afford the financial investment, the fallow period and the time required for the innovative system establishment. <br>Information provided in this research work should improve the knowledge about weeds in tropical countries, the interaction with cover crops and fertilization, and the implication of cover crops on the critical period of weed interference. The potential benefits of stylosanthes highlighted in this study can be used to address weed issues in upland rice systems in any context similar to the Mid-West of Madagascar.<br>