Search

This paper gives a perspective of the debates around the « margins » of France. It shows that the theme allows to reconsider the geography of France and to de/reconstruct space because the margins do not fit into a definitive and normative geography. So the approach of place by marginal populations as a social stage of encounters and affirmation of difference is linked with a very high mobility of these resource places which, according to the social contingencies, sketch an ephemeral geography. It offers to understand the logics of the dominant, of integrated populations and urban poles and, on the contrary, it initiates centers of contestation and innovation. The margin is therefore an eminently political geographical object. It defines itself in space but does not find borders and limits, while it is hard mobilize the usual tools of the discipline: transition, areola, gradient, relative, hidden, fractals; marginal places seem to be kept at a distance. Margins require a plurality of approaches and perceptions of the territory as methodology. The theme so allows to study the frictions between the shape of the world and the way of living in it. It opens on a study of emotions, feelings, and sometimes resentment.

Deep artificial neural networks are feed-forward architectures capable of very impressive performances in diverse domains. Indeed stacking multiple layers allows a hierarchical composition of local functions, providing efficient compact mappings. Compared to the brain, however, such architectures are closer to a single pipeline and require huge amounts of data, while concrete cases for either human or machine learning systems are often restricted to not-so-big data sets. Furthermore, interpretability of the obtained results is a key issue: since deep learning applications are increasingly present in society, it is important that the underlying processes be accessible and understandable to every one. In order to target these challenges, in this contribution we analyze how considering prototypes in a rather generalized sense (with respect to the state of the art) allows to reasonably work with small data sets while providing an interpretable view of the obtained results. Some mathematical interpretation of this proposal is discussed. Sensitivity to hyperparameters is a key issue for reproducible deep learning results, and is carefully considered in our methodology. Performances and limitations of the proposed setup are explored in details, under different hyperparameter sets, in an analogous way as biological experiments are conducted. We obtain a rather simple architecture, easy to explain, and which allows, combined with a standard method, to target both performances and interpretability.

Computer science offers a large set of tools for prototyping, writing, running, testing, validating, sharing and reproducing results; however, computational science lags behind. In the best case, authors may provide their source code as a compressed archive and they may feel confident their research is reproducible. But this is not exactly true. James Buckheit and David Donoho proposed more than two decades ago that an article about computational results is advertising, not scholarship. The actual scholarship is the full software environment, code, and data that produced the result. This implies new workflows, in particular in peer-reviews. Existing journals have been slow to adapt: source codes are rarely requested and are hardly ever actually executed to check that they produce the results advertised in the article. ReScience is a peer-reviewed journal that targets computational research and encourages the explicit replication of already published research, promoting new and open-source implementations in order to ensure that the original research can be replicated from its description. To achieve this goal, the whole publishing chain is radically different from other traditional scientific journals. ReScience resides on GitHub where each new implementation of a computational study is made available together with comments, explanations, and software tests.

The emergence of the modern concept of the sustainable city raises afresh the longstanding issue of the place and role of vegetation in urban and peri-urban areas in Europe. The awareness of biodiversity and the exploration of the services provided by ecosystems both lead to the development of ecological networks based on green spaces in and around the city. The establishment of these networks converges with the control of urban growth and urban sprawl, with the ‘green belts’. Drawing on the development of public policy governing the place of vegetation in Berlin, London and Paris, this article seeks to show the correspondences that have developed in the discussions of urban policy carried on in the major industrialized countries, and also the conflicting goals which these policies are meant to implement.

The rapid invasion and spread of Aedes albopictus (Skuse, 1894) within new continents and climatic ranges has created favorable conditions for the emergence of tropical arboviral diseases in the invaded areas. We used mosquito abundance data from 2014 collected across ten sites in northern Italy to calibrate a population model for Aedes albopictus and estimate the potential of imported human cases of chikungunya or dengue to generate the condition for their autochthonous transmission in the absence of control interventions. The model captured intra-year seasonality and heterogeneity across sites in mosquito abundance, based on local temperature patterns and the estimated site-specific mosquito habitat suitability. A robust negative correlation was found between the latter and local late spring precipitations, indicating a possible washout effect on larval breeding sites. The model predicts a significant risk of chikungunya outbreaks in most sites if a case is imported between the beginning of summer and up to mid-November, with an average outbreak probability between 4.9% and 25%, depending on the site. A lower risk is predicted for dengue, with an average probability between 4.2% and 10.8% for cases imported between mid-July and mid-September. This study shows the importance of an integrated entomological and medical surveillance for the evaluation of arboviral disease risk, which is a precondition for designing cost-effective vector control programs.

This article analyzes the changes which have occurred in land occupation and vegetation cover over the past few decades in a number of villages scattered along the bioclimatic gradient in the South-West of Niger, from the Tillabéry-Filingué line in the North to Gaya (on the Niger-Benin border) in the South. Diachronic analyses were performed based on satellite photographs (CORONA March and October 1965) or aerial photographs (IGN mission in March 1975) which were compared to current Google Earth photographs. These analyses were then confronted to interviews conducted in the various villages as well as to surveys of the state of the vegetation. If researchers emphasize the resumption of rainfall since the mid-1990s, they also show that the South-West of Niger paradoxically presents an increased socio-environmental vulnerability. It seems possible to generalize this observation at the scale of the two regions. However, the research undertaken here leads to a more nuanced conclusion : when data are disaggregated, the situation looks very diversified, revealing, independently from the bioclimatic gradient and from the meridian differentiation linked with the fossile valleys (dallols), situations where the vulnerability of both populations and ecosystems appears more or less pronounced.

Scientists have to appropriate the sustainable development concept. Multipartenarial research programs involving southern and northern researchers and institutions are already applying this concept. The authors selected two cases of landscape evolution dealing with coastal resources and geographical uses in the Petite Côte region and in the Sine-Saloum delta (Senegal). The cases studied were built-up growth of the principal city on the region and the mangrove dynamics of the Saloum delta.

This interdiciplinary research focuses on explaining the process of bush growth progression in the Causse Méjean. The paper presents the method by which heterogeneous data in time and space scales have been matched and hierarchized. Data concerning vegetation, environment conditions, history of property and land use, social practices and representations have been treated by the P.C.A. (to describe vegetation) and by Canonical Analysis (to check its correlations with explanatory factors). This research opens some ways to model the interactions between biophysical and social systems.