Seasonality | EPFL Graph Search

In time series data, seasonality is the presence of variations that occur at specific regular intervals less than a year, such as weekly, monthly, or quarterly. Seasonality may be caused by various factors, such as weather, vacation, and holidays and consists of periodic, repetitive, and generally regular and predictable patterns in the levels of a time series. Seasonal fluctuations in a time series can be contrasted with cyclical patterns. The latter occur when the data exhibits rises and falls that are not of a fixed period. Such non-seasonal fluctuations are usually due to economic conditions and are often related to the "business cycle"; their period usually extends beyond a single year, and the fluctuations are usually of at least two years. Organisations facing seasonal variations, such as ice-cream vendors, are often interested in knowing their performance relative to the normal seasonal variation. Seasonal variations in the labour market can be attributed to the entrance of

EFFECT OF WATER CONTACTS SEASONALITY ON SCHISTOSOMIASIS TRANSMISSION IN RURAL SAHELIAN AREA IN BURKINA FASO

Anna Faes

patterns due to economic and sociologic activities in African rural area are an important factor in the comprehension of the transmission of this disease. Through interviews and a participatory workshop, activities related to water-contact were identified, quantified and located for the study site of Tougou, situated in the north of Burkina Faso. Tougou, a field site for the 2iE Institute of Ouagadougou since 2004, has a sahelian climate; with a dry season of eight months and all precipitations occurring during the four other months of the year; and is affected by urinary schistosomiasis. The water-contact rates for the identified activities and the attractiveness of the different water points at the site, point out a seasonal variation in water-contact patterns. Using modelling to explore the effect of the seasonality of water-contacts in schistosomiasis transmission cycle, two different issues are observed depending on the considered system. In a local system, including one point of contact with water and one village, the seasonal variation of water-contacts leads to a decrease in the number of parasites per person. In opposite way, in a connected system of several villages having several water points where contacts occur, the seasonal variation of water-contacts lead to an increase in the number of parasites per person, due to the gathering of most water contacts at fewer number of water points during the dry period of the year. Table of content

2015

Geological heterogeneity in landslides

Boris Matti

Significant progress has been made these last decades in the development of hydrogeological numerical flow modelling for describing the hydrodynamic behaviour of landslides. However, these new sophisticated methods are still very seldom used in the problems of slope instability in particular because of the hydrogeological complexity which characterizes them; thin aquifers, discontinuous media, succession of saturated and unsaturated zones, low permeabilities, high hydraulic gradients, lithological heterogeneity, strong contrasts of permeabilities and heterogeneous infiltration. Predictive models of flow in the subsurface, which are often based on homogeneous porous media types of representation, are badly adapted to natural systems that are characterized by highly heterogeneous media such as landslides. These models are good and reliable on a landslide scale (regional scale), but their quality may be affected on a local scale by strong geological heterogeneities. Geological heterogeneities of the subsurface take part in determining the hydrodynamical and geomechanical behaviour of landslides. However, their spatial distribution is partially unknown. Thus, the principal objectives of this PhD thesis are: (i) To carry out an integrated multidisciplinary characterization study on the internal structure of landslides in flysch and Quaternary environments, in order to clarify the organisation of the geological heterogeneities and to identify the hydrodynamic implications. (ii) To propose a conceptual model representing the geological architecture and the hydrogeological functioning. (iii) To examine the effects of heterogeneity and anisotropy on flow systems. (iv) To better understand the influence of geological heterogeneities on the mechanical behaviour of large landslides by performing numerical sensitivity analyses, by means of different heterogeneity scenarios on the field parameters. (v) Finally, to test the incidences on slope stabilization techniques; evaluation of the efficiency of a drainage gallery work. The main test site of la Frasse landslide (VD, Switzerland) was chosen, and completed with additional landslide cases. The main results are the following: In most of the case studies, the landslide mass is composed of an old prehistoric stabilized mass, pinched between the active sliding mass and the bedrock, and playing an important hydrologic role. The stabilized mass and the bedrock form the substratum of the landslide. Landslides occurring in these types of media are defined by an organized heterogeneous environment with "fracture" flows and discontinuity porosity. The overall hydraulic conductivity is low, and locally high permeable zones exist. Regional groundwater circulations are limited and form local interconnected aquicludes organised in thin aquifers, and presenting saturated and unsaturated zones. The hydrogeological analyses showed that the system presents a bimodal permeability; (i) Low hydraulic conductivities characterizing the global matrix and defining the capacitive fraction, and (ii) high permeable features, with high hydraulic conductivities defining the conductive fraction, and favouring strong channelling effects. Besides, the observation shows that the aquifer system is generally very reactive with important magnitudes. Often, there is a straight correlation between water level variation and climatic conditions (rainy events). Landslides are characterized by two important inflows namely effective infiltration from the surface and lateral inflows from the neighbouring units. Water transfer between the stabilized mass and the active mass may be important and thus have to be considered. The existence of water transfer between the bedrock and the landslide mass (stabilized and active) is not well established. The bedrock and the landslide mass present a hydrological behavioural independence. Theoretical two- and three-dimensional flow models are used to investigate the effects of the spatial variability of the hydraulic conductivity on the underground flows. The role of the connectivity in generating flow channelling is examined thanks to the observation of close relations between the permeability and the hydraulic pressures. The sensitivity analysis shows clearly that the relation between local permeability and hydraulic pressures is not straight, and that the organization of the flows depends on the heterogeneity of the hydraulic properties and their spatial correlation. Strong channelling effects are observed in highly heterogeneous porous media. The development of flow channelling as a function of the variance of the natural log permeability values and the correlation lengths is demonstrated. The integrated multi-disciplinary geological characterization at the La Frasse test site combined with the hydrogeological and lithological data of several additional case studies led to the proposal of a global conceptual model. The following assumptions are considered to enable a subsequent quantification of flow components: The flow occurs under confined to leaky conditions, with leakage varying in space; The flow framework is controlled by a complex multi-layer system, isolated lenses or perched aquifer; The aquifer system is divided into interconnected hydrological zones presenting various degrees of saturation; Each hydrological zone may function individually from the others; Horizontally and vertically, the flow direction in the porous matrix is affected by prevailing structural patterns generating channeling effects; The flow is multidirectional, free and channelized, and is affected by temporal and spatial changes; The aquifer is under an unsteady flow regime due to seasonal variation of natural gradients; A conceptual model based on a simple reservoir approach is proposed. It allows the representation of most of the field observations and the main characteristics, namely the organized heterogeneity and the duality of the aquifers. The system is represented by various reservoirs more or less connected and saturated. Complex storage capacities and plug-flow effects may record past events and reactive sliding processes several months after the last important rainy event. The analysis shows that function of the capacity and the degree of saturation of the system, an important hydrological event is not necessarily associated to a reactivation. And, according to the degree of complexity of the system (saturation, connectivity...) a localized geological modification (variation of permeability, reservoir burst...) may produce a chain reaction, and generate failures in unexpected places. The conductive fraction favours the drainage of the system, whereas the capacitive fraction controls the distribution of the hydraulic heads. The role of the phreatic nappe, through the conductive fraction, is to drain and control the hydrologic equilibrium of the system. Therefore landslide remediation with the help of a deep drainage gallery is obviously the most valuable method for this type of landslide. It supports and enhances the natural effects of the conductive fraction in draining the system. Finally, in this context the efficiency of civil engineering works was evaluated according to the heterogeneity of the medium. This study describes transient hydrogeological and geomechanical models realized jointly in 2006 by the EPFL and GeoMod SA within the framework of the stabilization work of the La Frasse landslide. These models evaluate the impact of a deep drainage gallery with subvertical pipes towards the surface in terms of reduction of the deformation velocities and increase of the factor of safety of the landslide. Three variants consisting of different inter pipe spacings are tested. Considering the local heterogeneities, the results show that a mean spacing between the pipes of the order of 10 m is able to control the temporal head fluctuations between the wells within a range of some meters. Moreover, this solution induces a strong diminution of the predicted displacements during a specific crisis, from 101cm for the model without drainage to around 14 cm for the drained model, and a significant gain of security (from 1.05 to 1.30).

EPFL2008

Spatiotemporal modelling of extreme wildfires and severe thunderstorm environments

Jonathan Koh Boon Han

Environmental extreme events can have devastating impacts on society when they interact with vulnerable human and natural systems. Such events can result from natural causes, like phenomena related to the El Ni~no-Southern Oscillation or decadal/multi-decadal climate variations. These causes can follow an increase in human activity, e.g., through land-use changes or anthropogenic climate change, that can influence the frequency, intensity, spatial extent and timing of these events, and spur unprecedented extremes. To accurately understand and quantify the risks associated with these events, it is important to identify trends related to these causes, which may be measured or unmeasured. Fitting models for rare events is inherently difficult because of the paucity of data available. The most destructive extreme events are rarely isolated in space and time, so one must account for their spatial and temporal dependencies. This thesis deals with the parametric modelling of severe thunderstorms and wildfires using models motivated from limiting probabilistic results. The first part of this thesis explores influences on the magnitude and spatial extent of extremes of environments related to severe US thunderstorms. Our results show that the risk from severe thunderstorms in April and May is increasing in parts of the US where it was already high, and that the risk from storms in February increases during La Ni~na years. We also show that these extremes are more localized during spring/summer seasons than in the winter, and find that some of these seasonal differences are more pronounced during El Ni~no years.The second part of the thesis deals with predicting and explaining the spatial extent, frequency, intensity and timing of wildfires using meteorological and land-use covariates. Our first approach uses ideas from extreme-value theory in a machine learning context to give good prediction of the distributional tails of our data. The second approach uses a novel Bayesian hierarchical model designed specifically for extreme wildfires. We show that wildfire risk on the French Mediterranean basin is affected by significant random effects related to land-use and policy changes, and a seasonally-varying fire-weather index.