Optimization of drip irrigation using a sensor network in Sahel Region (Burkina Faso) - Analysis of soil water and plant dynamics in the framework of project Info4Dourou 2.0

The research project Info4Dourou 2.0 's main goal is to improve agriculture in semi-arid regions by developing a support system that optimizes agricultural production and water consumption based on continuous soil humidity measurements using a sensor network. In this context, the main purpose of the study was to give more scientific support to the agronomic model on which the sensor network is based. In particular the research aims at understanding the soil water and plant dynamics in order to give recommendations on the system design. The main feature of the system is to indicate when irrigation must be triggered and in which soil moisture conditions. The thesis developed a numerical model based on the software HYDRUS 2D to acquire a precise knowledge on the soil water dynamics and was calibrated and coupled with field experiments on two vegetable crops of eggplant and cabbage in Burkina Faso using drip kit irrigation systems. A great focus was given on the plant response to different irrigation schedules and to water stress, considering aerial biomass development, root distribution and final yields. Water stress was linked to continuous measurements of the soil matrix potential at different depth and locations. The field experiments showed that a daily irrigation frequency resulted to better canopy development during the first part of the growth but that great water savings are possible by optimizing the schedule. Water stress was difficult to track precisely by following daily sap flow behaviour which suggested that plant stress occurs before transpiration reduction, especially when biomass is building up. The adaptability of the root distribution was demonstrated and was clearly correlated with the wetted zone which depended on the irrigation schedule. From the numerical model, it was shown that using one sensor at a depth of about 10 cm was most appropriate in order to pilot irrigation during the whole crop growth. The analysis showed that most of the water savings could be achieved at relatively high threshold values and that low thresholds mainly resulted to transpiration reduction which was not advisable. A threshold of -20 kPa was proposed for the beginning of the growth stage which then decreases to a value of -50 kPa at full canopy development. These thresholds are believed to be adequate for most vegetable crops since they allow keeping the soil matrix potential in the root zone at values that are tolerated by most plants. An analysis of the influence of the soil texture also showed that similar values seem to perform well for most soils, if assuming an adaptation of the root distribution to the wetted zone. The study focused mainly on water stress but more research could be done on the impact of the irrigation system on nutrients availability for example.

The influence of cattle activity on tree regeneration in wood-pastures

Charlotte Vandenberghe

Wood-pastures are semi-natural, highly biodiverse systems maintained by traditional extensive agriculture. They are used for grazing and timber. The shifting mosaics of grassland, shrub thickets and woodland patches in these systems are driven by large herbivores. Current changes in agricultural practices are leading to either intensification or abandonment and they result in a segregation of grassland and forest. This presents a complex conservation problem for these endangered ecosystems. Large herbivores have been identified as an important factor preventing the establishment of trees and the regeneration of woodland. However, successful management is severely hindered by the lack of ecological knowledge about the effects of large herbivores on woody species. This research focussed on the quantification of the impact of cattle activity on the early stages of tree establishment for four tree species (Abies alba, Picea abies, Acer pseudoplatanus and Fagus sylvatica) common in the wood-pastures of the Swiss Jura Mountains. Cattle can have both direct and indirect effects on tree establishment, and the interaction between cattle activity and both competition and facilitation by surrounding vegetation was studied. Experimental field studies (involving seed sowing and sapling planting) were carried out. Both cattle activity and competition from neighbours reduced first-year tree seedling recruitment in wood-pastures. Seedling establishment decreased with increasing cattle grazing intensity. The effects of gaps in the vegetation on seedling emergence were complex, as a result of biotic (i.e. competition, facilitation and predation from rodents, invertebrates and large herbivores) and abiotic (e.g. moisture, temperature and light) interactions, leading to different results in different years. Although overall the chances for tree seeds to make it to seedlings were low, seedling establishment was relatively high for Picea, intermediate for Abies and Acer and low for Fagus. The probability of a sapling of being browsed increased with sapling size and decreased with the height of surrounding vegetation and in the presence of nurse shrubs. It was similar for the four species and under low and high grazing intensities. Saplings had species- and size-specific responses in survival and growth to cattle browsing. Under both low and high grazing intensity, small coniferous saplings (36 %) had a lower survival rate than small deciduous saplings (53 %), which we hypothesize is due to the larger biomass losses stemming from a more vulnerable plant architecture. Under low grazing intensities, large Picea (80 %) and Fagus (67 %) saplings had a higher survival than large Acer (50 %) and Abies (33 %) saplings. This was likely due to differences in sapling tolerances to loss of biomass. Simulated browsing damage resulted in relatively smaller growth losses when small saplings were growing slowly in the presence of neighbours. Shade increased this effect. At least in the short term, the degree of small sapling tolerance was not related to plant performance as saplings that compensated almost for biomass losses still had lower survival and growth rates than less tolerant saplings. Picea was the only species not negatively affected by strong irradiance as found in an open pasture. For all tree species, the facilitative effect of nurse shrubs on small saplings was highest at intermediate levels of grazing intensity. Shrubs were more heavily damaged at high intensity than low intensity. Consequently, escaping browsing, sapling survival and growth was significantly increased by shrubs under low grazing intensity but not under high grazing intensity. Moreover, the positive effects of shrubs tended to be higher for coniferous species, and in particular Picea which was more sensitive to both competition from neighbours and browsing. In conclusion, the resistance of trees to cattle activity varies among the early stages of tree establishment and is affected by tree species, grazing intensities and environmental conditions. Both competitive and facilitative interactions found between young trees and surrounding vegetation illustrate the complex nature of plant-plant and plant-animal interactions during the tree establishment phase in wood-pastures. The insight provided into the interactions between cattle grazing and tree establishment in wood-pastures will contribute to improve ecological theory and models, in addition to informing the management and conservation of wood-pasture systems.

EPFL2006

A model-based optimization framework for the inference of regulatory interactions using time-course DNA microarray expression data

Vassily Hatzimanikatis, Eleftherios Papoutsakis

Background: Proteins are the primary regulatory agents of transcription even though mRNA expression data alone, from systems like DNA microarrays, are widely used. In addition, the regulation process in genetic systems is inherently non-linear in nature, and most studies employ a time-course analysis of mRNA expression. These considerations should be taken into account in the development of methods for the inference of regulatory interactions in genetic networks. Results: We use an S-system based model for the transcription and translation process. We propose an optimizationbased regulatory network inference approach that uses time-varying data from DNA microarray analysis. Currently, this seems to be the only model-based method that can be used for the analysis of time-course "relative" expressions (expression ratios). We perform an analysis of the dynamic behavior of the system when the number of experimental samples available is varied, when there are different levels of noise in the data and when there are genes that are not considered by the experimenter. Our studies show that the principal factor affecting the ability of a method to infer interactions correctly is the similarity in the time profiles of some or all the genes. The less similar the profiles are to each other the easier it is to infer the interactions. We propose a heuristic method for resolving networks and show that it displays reasonable performance on a synthetic network. Finally, we validate our approach using real experimental data for a chosen subset of genes involved in the sporulation cascade of Bacillus anthracis. We show that the method captures most of the important known interactions between the chosen genes. Conclusion: The performance of any inference method for regulatory interactions between genes depends on the noise in the data, the existence of unknown genes affecting the network genes, and the similarity in the time profiles of some or all genes. Though subject to these issues, the inference method proposed in this paper would be useful because of its ability to infer important interactions, the fact that it can be used with time-course DNA microarray data and because it is based on a non-linear model of the process that explicitly accounts for the regulatory role of proteins.

2007

Recyclage des eaux usées en irrigation

Yeli Mariam Sou

The present study takes place in Ouagadougou (Burkina Faso), a typical Sub-Saharan city where water shortage and food crisis necessitate wastewater reuse in urban agriculture. Different issues linked to this practice were studied in two components separated in two experimental sites. The first part treats about agronomic and sanitary aspects and take place in the 2iE (International Institute for Water and Environmental Engineering) experimental site of Ouagadougou. The second component was emphasised on industrial effluents impacts on irrigated soil quality and takes place in Kossodo, an industrial suburb of Ouagadougou. The 2iE site is supply with treated wastewater from domestic origin. The effluent was used to realise the following objectives: (1) managed wastewater fertilizers to fit the plant needs; (2) assess different crops sanitary quality and the effectiveness of methods used for this assessment. Three treatments were tested: (i) irrigation with the treated wastewater (ii) irrigation with the treated wastewater and fertilizers management, adjusted to plant needs and (iii) conventional treatment i.e. fresh water irrigation and total doses of commercial fertilizers. Each treatment was tested on three different crops: lettuce (Lactuca sativa), carrot (Daucus carota) and eggplant (Solanum melongena). The first objective was achieved by comparing crops yields (on basis of wet and dry matter) obtained with the different treatments. The three crops, irrigated with only wastewater, led to yield reduction between the first and the second year. At the contrary, the adjusted treatment showed higher yield for lettuce, with comparable values to conventional treatment yield. However, the effects on carrot and eggplant did not demonstrate better results; indicating that further investigations must be completed for crops with longer growing periods. Most likely, the fertilizers management should better correspond to the crop period demand. The sanitary assessment (second objective) was performed by measuring the heavy metals concentration (Cd, Cu, Pb, Ni, Cr) and the faecal levels (faecal coliform, Escherichia coli and four pathogens among which Salmonella typhimirium and Vibrio cholerae). Chromocult Coliform Agar medium was used for simultaneous detection of faecal coliform and Escherichia coli whereas pathogens were analysed by Polymerase Chain Reaction (PCR). From methodological point of view, faecal coliform were not sufficiently discriminatory in the case of this study since they were naturally present in the soil as well as in the fresh water. At the contrary, Escherichia coli were better linked to wastewater contamination. However, some "false positive" results were detected meaning that each presumptive colony of Escherichia coli must be confirmed. Nevertheless, microbial quality assessment indicated no Escherichia coli on edible parts of carrot or eggplant irrigated with wastewater. Carrots were expected to have the poorest microbiological quality. Therefore carrot root bacterial quality must be further examined. Results on eggplant are not surprising as these vegetables are somewhat far from the ground and they receive significant solar radiation which acts like a disinfectant. Lettuce from wastewater treatment plots, however, exhibited presence of Escherichia coli, meaning that lettuce is likely to be contaminated by pathogens. However, no pathogens were detected neither on crops irrigated with fresh water nor on those irrigated with wastewater. Heavy metals results demonstrated that the crop's consumption did not involve higher risk for consumers when the vegetables were irrigated with domestic wastewater, in comparison with fresh water irrigation. Kossodo treatment plant receives mainly industrial wastewater, which are characterized by high levels of sodium and bicarbonates. The treated wastewater supplies a market garden in which space has been defined for the study of this second component. The main objective was to assess the impact of treated wastewater on soil quality during two years of irrigation. Early in the second year, the presence of black spots, suspected to be black alkali (dissolved organic matter), appeared at the surface of plots irrigated with treated wastewater. These indications led to put forward the hypothesis of alkalinisation/sodisation on soil receiving wastewater. The methodological approach used was based on in situ measurement of infiltration at the end of the second year of irrigation, supplemented by shrinkage curve modelling and chemical analysis. The hypothesis of an alkalinization/sodisation and dissolved organic matter (black alkali) presence in plots irrigated with treated wastewater were validated by the chemical analysis. Shrinkage curve modelling showed a reduction of the structural porosity on wastewater irrigated plots. The reduction was more marked in the subsurface horizons than in surface, which is probably caused by dispersed clay leached in depth. These structural changes were confirmed in situ by an infiltration capacity greatly reduced on plots irrigated with treated wastewater, compared to control plots (irrigated with fresh water) where the infiltration capacity is higher and comparable to that of non-irrigated soil. The study confirmed the fertilizing value of domestic wastewater while emphasizing the need to concentrate future investigations on sustainable effluent/fertilizers management. The sanitary assessment with different crops regarding their respective distances to the soil was very useful since it showed the strong interaction of soil microbial community on the crop microbial quality. It also revealed the importance to determine a faecal indicator for the crop, specifically related to wastewater contamination. Soil quality assessment reiterated the necessity to verify wastewater quality before its reuse in agriculture. In the case of this study, the sodic and alkaline properties of the industrial effluents make them definitively unfit for irrigation.

EPFL2009