Weed management in rainfed rice cropping systems in Madagascar primarily relies on manual weeding. This labour-intensive practice often coincides with peak workloads on the smallholder farms, leading to delayed weeding, one of the major causes of rice yield losses. The aim of this study was to determine the critical period of weed interference (CPWI) in the specific context of low-input rainfed rice-based cropping systems under conventional tillage (CT). We further assessed whether no-tillage with a living mulch of Stylosanthes guianensis (NTLM) could postpone the optimal timing of weeding as cover crops and mulch are often proposed as a promising strategy to reduce weed infestation. Field experiments were conducted during the 2016/17 and 2017/18 cropping seasons. Two controls (weedy or weed-free during the full crop cycle) and six treatments combining two different weeding regimes, increasing weedy period or increasing weed-free period, to three durations (20, 40, and 60 days after sowing) were compared. Rice grain yield was about 2.69 +/- 0.32 t ha-1 and 0.04 +/- 0.04 t ha-1 in weed-free and weedy controls, respectively. Weed interference throughout the entire rice growing season caused yield losses of up to 99% in CT. In CT, the CPWI started 10 to 18 days after sowing in both years. By contrast, the NTLM system delayed the onset of the CPWI by about 18 days, reducing yield losses to 51% (rice yields of 3.3 +/- 0.36 t ha-1 and 1.55 +/- 0.51 t ha-1 in weed-free and weedy controls, respectively). Our results highlight the importance of early weeding in rice fields to avoid significant grain yield losses. No-tillage with a living mulch is a promising option, providing additional time to farmers during peak labour periods at the start of the rice growing season. However, further research is needed to confirm these findings and to identify the conditions and support mechanisms that would enable farmers to adopt this type of no-tillage cropping system.

Weed management in dryland rice systems in Madagascar primarily relies on manual weeding. This is a labour-intensive process at a time when the workload on the smallholder farms is particularly high and often leads to delayed weeding, one of the major causes of yield losses in rice. The aim of this study was to determine the critical period of weed interference (CPWI) in the specific context of low-input dryland rice-based cropping systems under conventional tillage (CT). We further assess whether no-tillage with a living mulch of Stylosanthes guianensis (NT) could postpone the optimal timing of weeding. Field experiments were conducted during the 2016/17 and 2017/18 cropping seasons. Different weeding regimes were compared: weedy and weed-free controls, and treatments with an increasing weedy period and increasing weed-free period, respectively. Results showed that the CPWI started 10 to 18 days after sowing (DAS) in CT. Rice grain yield was about 2.69 ± 0.32 t ha-1 and 0.04 ± 0.04 t ha-1 in weed-free plots and weedy controls, respectively. Weed interference throughout the entire rice growing season caused yield losses of up to 99% in CT. The NT system delayed the CPWI by about 18 days, and reduced the impact of weeds on rice yield, resulting in only a 51% loss (rice yields of 3.3 ± 0.36 t ha-1 and 1.55 ± 0.51 t ha-1 in weed-free and weedy controls, respectively). Our results indicate the need for early weeding in rice fields to avoid grain yield losses. No-tillage with a living mulch is a promising option providing additional time to farmers during peak labour at the start of the rice growing season. Yet, farmers will need support to establish and implement these cropping systems, which require specific equipment and additional financial investment.

Weeds are a major threat in tropical regions where climate conditions favor their growth and development. This is particularly true in low-input rice-based cropping systems in the Malagasy highlands, where weed management is mainly done by manual removal. Crop rotation is often promoted as an efficient way to control weed infestations, while the role of fertilization is more controversial. In this study, we compared rice monoculture to three rainfed rice-based two-year rotations: rice followed by groundnut, rice followed by sorghum-cowpea mixture, and rice followed by a velvet-bean crotalaria mixture. Each rotation was tested with two levels of fertilization (5 t DM ha(-1) organic manure, sole or in combination with mineral fertilizer - 400 kg ha(-1) NPK + 200 kg ha(-1) urea). We assessed the effect of rotation and fertilization on weed composition, diversity, biomass and rice yield. Additionally, the farmers' perception of weed harmfulness and the relation between their assessment of weed harmfulness and rice production was tested. Our results showed that weed biomass significantly decreased rice yield but only under the low fertilization level. The rotation of rice with the velvet beancrotalaria mixture was efficient in reducing weed biomass, modified weed community composition and allowed to achieve the highest rice yield. A significant negative relationship was found between weed community harmfulness index and weed species richness. Yet, the lowest rice yield was observed under rice monoculture despite a higher species richness over years and under high fertilization level. The lack of significant correlation between the harmfulness index and the actual rice yield is probably because our index is partly based on farmer's perception, and only on major weeds. More studies on tropical weed harmfulness are needed to support the design of ecologically intensified cropping systems.

In Madagascar, upland rainfed rice is traditionally grown on tilled soil without transplanting. Weed infestation by annual species is a serious issue in such a system. Stylosanthes guianensis, a legume cover crop, can be grown with rice as a living mulch but its effect on the weed community and rice yield was not yet well studied. Therefore, a field experiment was conducted in Mid-West Madagascar from 2016 to 2018. A no-till system with stylosanthes (NT) managed as a living mulch was compared to the conventional tilled (CT) system. Two fertilisation regimes were included: organic manure (F1) and organic manure combined with mineral fertiliser (F2). Weed floristic surveys were performed, and weed harmfulness evaluated by comparing weedy and weed-free plots. The living mulch harmfulness was evaluated using weed-stylosanthes-free plots. Results showed that NT suppressed dominant grass weeds Digitaria spp. and Eleusine indica leading to a decrease in total weed biomass. Rice yield was significantly higher in NT than in CT; in weed-free condition, it was 2.4 versus 2 t ha(-1) in NT and CT, respectively, and 1.5 versus 0.6 t ha(-1) in NT and CT, respectively, in weedy condition. The living mulch reduced rice yield by 14% but it remained significantly higher in NT than in CT. The addition of mineral fertiliser to the organic manure had a positive effect on rice yield but only in weed-free conditions. This study showed that a no-till system with a living mulch can help to control the most competitive weeds without undermining rice production.

Description of the subject. Weed pressure is a main biotic constraint in tropical agriculture. Cover crop mixtures have increased in popularity to limit weed growth through competition for resources, but the relationship between cover crop diversity and weed suppression is still under debate. Objectives. This study aimed to assess the impact of increasing cover crop diversity (one to four species) on weed control during two growing seasons (tropical summer and winter) in Reunion Island. Method. Weed control was expressed regarding ground cover by weeds and weed aboveground dry mass in the mixtures during four months of growth and its response to cover crop traits was tested using structural equation models. Results. While cover crops reduced weed ground cover and dry mass by 60% and 68% on average in summer and winter, respectively, a higher number of cover crop species within a mixture did not increase mean weed control. Nonetheless, weed control was influenced by the mixture composition and improved when including Guizotia abyssinica. Additionally, cover crop traits explaining weed control differed between growing seasons. In summer, weed control was mainly explained by the final cover crop aboveground biomass and leaf area (depletion strategy). In contrast, weed control was mainly explained by the cover crop rate of increase in ground cover (obstruction strategy) in winter. Conclusions. Using traits to characterize cover crop mixture enables us to identify mixtures of species and traits adapted to different growing conditions. Our study suggests that particular attention on species identity rather than diversity should be paid in mixture to improve weed control in tropical conditions.

Weeds cause more damage to crops than all other pests, particularly in tropical areas, but despite years of work and many studies, they are still insufficiently understood to develop functional approaches to tropical weed communities at a global level. Amatrop Dataverse contains 30 datasets from weed surveys carried between 1988 and 2020 in tropical cropping systems mainly in Africa and Islands in the Indian Ocean. All the datasets were homogenized and standardized to facilitate the concatenation of files for global statistical analyses. An additional dataset contains the reference weed flora (1,789 taxa) contained in all the datasets, with updated botanical nomenclature. Amatrop Dataverse currently contains records of 8,718 field surveys and of the occurrence of 109,120 weed species in multi-site and single-site surveys, weighted and unweighted floristical records from farmers' fields or experimental studies. This new dataverse already contains considerable knowledge on tropical weeds for a better understanding of weed behaviour in cropping systems and to improve weed management by farmers.

During the year 2014, 619 floristic observations were made in cotton fields in Côte d'Ivoire. 80 % of cotton plots, wich average size was about 5.4 ha, were treated with a pre-emergence herbicide. Additional weeding occured when necessary (three times on average). This study identified the dominant weeds in cotton crop from sowing to harvest. Rottboellia cochinchinensis (Lour.) Clayton, Euphorbia heterophylla L., Ageratum conyzoides L. and Commelina benghalensis L. were the most common species. The floristic composition evolved along the cropping season: at the beginning, Euphorbia heterophylla and Commelina benghalensis were the most frequent species, while Ageratum conyzoides grew especially late in the cycle. The species were also distributed according to the different climate areas. Weed infestation was correlated with control practices.

Supporting organic farming aims to find alternative solutions to synthetic pesticides and antibiotics, using local plants, to protect crops. Moreover, in the One Health approach (OHA), a pesticidal plant should not be harmful to humans, meaning it cannot be toxic if the crop is consumed or should have a limited and conscious use if it is used for medical care. Knowledge on plant use presented in the scientific literature was compiled in a knowledge base (KB). The challenge is to develop a KB exploration method that informs experts (including farmers) about protection systems properties that respect OHA. In this paper, we present a method that extracts the Duquenne-Guigues basis of implications from knowledge structured using Relational Concept Analysis (RCA). We evaluate the impact of three data representations on the implications and their readability. The experimentation is conducted on 562 plant species used to protect 15 crops against 29 pest species of the Noctuidae family. Results show that consistently splitting data into several tables fosters less redundant and more focused implications.

The objective of this study was to investigate the effect of general factors such as crop, regional climate or country on the weed flora and weed abundance in rainfed crops in tropical regions. For this purpose, we used 24 weed survey datasets from 4 climates, 11 crops and 11 countries in the tropics. These datasets are available in public access on the CIRAD Amatrop dataverse. This represents a set of 6 069 weed surveys, covering a total of 1 388 taxa. The data were analysed using different complementary approaches, including infestation diagrams, species correlation and co-occurrence matrices, corrected ecological profiles, and PCAIV. The results show that in the tropics there is a background of fairly common and ubiquitous species that are found in varying abundance in almost all situations. Ageratum conyzoides is the only Major general species, while Rottboellia cochinchinensis, Digitaria horizontalis, Bidens pilosa, Cyperus rotundus, Commelina benghalensis, Euphorbia heterophylla, Eleusine indica, Euphorbia hirta and Tridax procumbens are General species, present in more than 15% of the records. The climate factor, depending on rainfall and altitude, which influences temperature, selects certain indicator species for different situations. The crop factor does not have a particular selective effect and the country highlights local flora. Crops are themselves dependent on climate. It is nevertheless possible to know which species are most likely to be found in a plot of a crop, under a given climate and in a given country. However, it seems that it is the combination of the edaphic and climatic nature of the plot that most accurately determines the floristic composition of the weed flora, although we were unable to analyse this due to the lack of complete information on these factors for all the datasets used. Finally, out of the 1 388 taxa taken into account, only about 300 species constitute the core of the weeds of tropical rainfed crops, because of their frequency, their abundance or their character as indicators of particular agro-ecological conditions.

Anticipating the establishment of pest reservoirs, and therefore pest infestation in sugarcane agrosystems, is a challenge for the implementation of integrated pest management (IPM) programs. The objective of this work was to develop a decision support system that suggests host plant species and pest natural enemies located in a production area. A knowledge base system (KBBI) was developed and coupled to DECIPESTS, a decision support system for PEST management in sugarcane. According to an observed damage, KBBI suggests the wild and cultivated plants that host the potential pests previously identified by DECIPESTS. The comparison with a local floristic inventory enables to determine pest reservoirs. Applied to a case study in Senegal, the system showed, for instance, that Eldana saccharina can be hosted by nine wild plant species located in the irrigation canals and two neighboring cultivated crops of socioeconomic importance in the area. This latter result indicates that the management of Eldana saccharina has to be tackled jointly by local farmers to be successful. While DECIPESTS uses a tactical approach to identify possible causes of pest infestation, its combination with KBBI makes it a strategic tool to enhance IPM strategy at a local scale.

Description of the subject. Cover plants have a high potential to deliver ecosystem services such as weed, pest or erosion control, as well as nitrogen fixation. Due to the large diversity of cover plants available, the choice of a plant to deliver these services is crucial because the spatial and temporal combinations with the main crop will determine the performances of the cropping system. Objectives. In this study, we aim to assess the growth and trait response of 16 cover plant species depending on four sowing dates in Reunion island. Method. Comparison of species was based on various traits such as aboveground biomass, ground coverage, height or radiation use efficiency. Results. Some species were performing well in both warm and cool season (e.g. Penisetum glaucum, Guizotia abyssinica), while others were more specific to warm (e.g. Crotalaria juncea, Canavalia ensiformis) or cool (e.g. Avena sativa, Pisum sativum) season. Among species and sowing dates, ground coverage and plant height were positively correlated with the ratio between leaf and aerial biomass. Conclusions. Considering the species maximum height and radiation use efficiency, we discussed a trait-based selection of species adapted to crop mixture or fallow with sugarcane in Reunion island. Even based on few traits, this approach allowed us to identify cover crop to test in sugarcane based cropping systems.

Description of the subject. Cover plants have a high potential to deliver ecosystem services such as weed, pest or erosion control, as well as nitrogen fixation. Due to the large diversity of cover plants available, the choice of a plant to deliver these services is crucial because the spatial and temporal combinations with the main crop will determine the performances of the cropping system. Objectives. In this study, we aim to assess the growth and trait response of 16 cover plant species depending on four sowing dates in Reunion island. Method. Comparison of species was based on various traits such as aboveground biomass, ground coverage, height or radiation use efficiency. Results. Some species were performing well in both warm and cool season (e.g. Penisetum glaucum, Guizotia abyssinica), while others were more specific to warm (e.g. Crotalaria juncea, Canavalia ensiformis) or cool (e.g. Avena sativa, Pisum sativum) season. Among species and sowing dates, ground coverage and plant height were positively correlated with the ratio between leaf and aerial biomass. Conclusions. Considering the species maximum height and radiation use efficiency, we discussed a trait-based selection of species adapted to crop mixture or fallow with sugarcane in Reunion island. Even based on few traits, this approach allowed us to identify cover crop to test in sugarcane based cropping systems.

Le contexte tropical Objectif : Utiliser des approches relatives a la science des donnees pour analyser les contraintes d'enherbement dans les cultures tropicales a une echelle globale. Une masse de donnees heterogenes complexes a assembler, stocker, partager et analyser