Diarrheas are responsible for 1.57 million deaths. The highest health burden occurs in Sub-Saharan Africa where 85 million people, living mainly in rural areas, are using surface water for domestic uses, and where bacteria are the 2nd leading cause of diarrhea. Moreover, climate change is expected to impact water resources both in quantity and quality and to potentially boost the presence, dissemination and transmission of pathogens. MAMIWATA aims at using satellite and environmental data to estimate health hazard of bacterial origin. However, important knowledge gaps remain 1) on Escherichia coli (E. coli) dynamics in surface water, their links with hydro-meteorological determinants and their role as fecal indicator bacteria (FIB) in West Africa, and 2) on the role of environment in the transmission of diarrheal disease. MAMIWATA project propose to study the environmental determinants playing a major role on bacteriological health hazard of lakes and ponds in tropical rural areas in West Africa (Burkina Faso, Niger) and to assess the potential of satellite data to monitor this health hazard of bacterial origin. Estimating E. coli and other water borne pathogens by remote sensing is challenging because it requires the investigation of several environmental determinants that play a role in the spatio-temporal dynamics of pathogens. MAMIWATA project propose to overcome these issues by combining unique collections of radiometric data, environmental determinants, E. coli measurements, potential pathogen data using loop-mediated isothermal amplification (LAMP) specific to pathogens, and epidemiological data over 4 sites in Burkina Faso and Niger. MAMIWATA will 1) study the fecal contamination of surface waters and establish whether E. coli is a good FIB in tropical semi-arid waters characterized by an abundance of very fine particles; 2) investigate the “key” environmental determinants for E. coli dynamics and the links with epidemiological data (diarrheal cases); 3) design environmental monitoring methods using satellite data to monitor the bacteriological health hazard in a context of climate and land use changes.

A huge trend in recent earth observation missions is to target high temporal and spatial resolutions (\emph{e.g.} SENTINEL-2 mission by ESA). Data resulting from these missions can then be used for fine-grained studies in many applications. In this project we will focus on three key environmental issues: agricultural practices and their impact, forest preservation and air quality monitoring. Based on identified key requirements for these application settings, MATS project will feature a complete rethinking of the literature in machine learning for time series, with a focus on large-scale methods that could operate even when little supervised information is available. In more details, MATS will introduce new paradigms in large-scale time series classification, spatio-temporal modelling and weakly supervised approaches for time series. Proposed methods will cover a wide range of machine learning problems including domain adaptation, clustering, metric learning and (semi-)supervised classification, for which dedicated methodology is lacking when time series data is at stake. Methods developed in the project will be made available to the scientific community as well as to practitioners through an open-source toolbox in order to help dissemination to a wide range of application areas. Moreover, the application settings considered in the project will be used to showcase benefits offered by methodologies developed in MATS in terms of time series analysis.

The TELKANTE LAB project between UMR LETG (CNRS) and ALKANTE aims to develop technical and software solutions for integrating remote sensing data into spatial data infrastructures (SDIs). Indeed, the complementary nature of the LETG and ALKANTE teams opens promising expectations towards scientific and technological advances to facilitate the use of spatial information for a wide range of applications (agriculture, urban planning, ecology, climatology, etc.), benefiting from the launch of numerous satellites and easier access to remote sensing images. On the one hand, ALKANTE, a 55-employee SME based in Brittany, has long-standing experience in the implementation of SDIs via the development of the PRODIGE interministerial software, which promotes the use of geographic information by private or institutional users (e.g. Ministries of Ecological Transition and Health). However, this open source software solution, although widely used by local authorities, does not currently integrate information from remote sensing. For its part, LETG is recognized for its expertise in processing remote sensing data (optical and radar imagery, 2D and 3D) and producing indicators for monitoring socio-environmental dynamics (e.g. urban dynamics, monitoring agricultural practices, climate change, etc.), but is struggling to make its results visible beyond the academic world. The TELKANTE LAB project is therefore structured around three research axes covering the entire technological chain, from data identification to the production and visualization of geographic information by and for users. Axis 1 - Integration of remote sensing data into SDIs: LETG will be responsible for identifying remote sensing data and their derived products to be integrated into GDIs. ALKANTE will ensure data access and interoperability. In particular, it will anticipate the need to process « Big Earth Observation Data ». Axis 2 - Geographic information processing: LETG will develop indicators for monitoring socio-environmental dynamics based on remote sensing, drawing on the latest scientific advances in image processing (time series analysis, artificial intelligence). ALKANTE will be responsible for ensuring the implementation of the processings in the SDIs. Axis 3 - User experience: This axis aims to develop modules enabling users (citizens, government representatives, companies) to play a central role in SDIs, as contributors ("citizen sensor") and interpreters of data according to their own experiences, intentions and motivations ("experience sharing"). The project is part of a dynamic sectoral context supported by the France 2030 plan to develop industrial competitiveness and forward-looking technologies to meet public sector needs for spatial data and associated services. In this context, the innovative nature of TELKANTE LAB concerns the development of solutions aimed at improving the ability to identify, process and interpret geographic information. To achieve this, the 54-month project will focus on territorial contexts that are priorities for the LETG and at the heart of major contemporary socio-environmental issues: (1) urban dynamics in a context of climate change, and (2) agricultural dynamics in the Amazon. These two application cases will enable us to develop a wide range of new solutions in the form of service offers and products aimed at turning the initial idea (TRL3 - integration of remote sensing data into IDGs) into prototypes (TRL5 to 7 - demonstrator) and then into a product accessible online and available on the market (TRL9).

The population of Moorea, an island located in the Society archipelago of French Polynesia (FP), doubled between 1983 and 2002. This significant demographic growth generated a strong tension on the coastline and the lagoon, which has resulted in an intensification of fishing activities, physical degradation (embankments, extraction of coral materials for the construction of infrastructures, etc.) and an increase in the discharge of pollutants (wastewater, pesticides, etc.). Because of this anthropic pressure, the municipality of Moorea-Maiao established in 2004 a Marine Spatial Management Plan (MSMP) mainly composed of Marine Protected Areas (MPA) and Regulated Fishing Zones (RFZ). Sixteen years after its formalization, the MSMP presents mitigated outcomes. Although the results remain encouraging from an ecological perspective, it is quite different from a societal point of view. Actually, the zonings and their regulations do not win the support of the lagoon fishermen. This failure can be explained by the cognitive and cultural inadequacy of the graphic representation of MPAs and PRZs. Lagoon fishing is indeed much more than a simple lagoon usage whose representation by zonings with fixed boundaries would be sufficient in itself. It is an ancestral practice linked to a singular socio-cultural context which, contrary to Western mental representations, is characterized by a triple continuum between nature and culture, land and sea, and humans and non-humans. Exploratory and descriptive, the HITI project aims to produce operational cartographic materials (i.e., usable by a maximum of people) that better reflect the experience of Moorea's lagoon fishermen through the consideration of this cultural principle. Called "non-Aristotelian," these maps, which transcend the Aristotelian principle of non-contradiction, contain three basic elements: (1) physical markers that shape the collective mental representation of the lagoon; (2) spatial affectivities (i.e., the affective relationships that fishermen have with some parts of the lagoon); and (3) geosymbols (i.e., places collectively recognized by the community). To produce these maps, three experiments involving 60 fishermen will be conducted. First, spatial mental representations will be captured through sketch maps that will be analyzed individually and then collectively. Spatial affectivities will be assessed subsequently using narratives that will be spatialized in the form of individual graphs. The third experimentation will then consist in an interpretation of these supports in order to complete the speeches collected previously (hermeneutics). Generated from a Geographic Information System based on the theory of fuzzy subsets, the non-Aristotelian cartographies of the lagoon fishing practices in Moorea will compile in the end all the information obtained from these three experiments. These will provide policy makers tangible interpretation keys of the island's fishermen experience that could be used to improve the MSMP. They will also serve as a tool to valorize and save the Polynesian intangible cultural heritage that is lagoon fishing in FP.

MULTISCALE is a research project that aims at providing a complete and integrated framework for multiscale image analysis and learning with hierarchical representations of complex remote sensing images. While hierarchical representations of RS images has led to an effective and efficient scheme to deal with panchromatic or at most multiband data, their application to complex data is still to be explored. In addition, despite their ability to encode structural and multiscale information, their so far exploitation have not reached beyond a mere superposition of monoscale analysis. In this context, the MULTISCALE project defines new methods for the construction of hierarchical image representations from multivariate, multi-source, multi-resolution and multi-temporal data, and provides some dedicated image analysis and machine learning tools to perform multiscale analysis. The new methodology will be implemented in various toolboxes used by the community to favor the dissemination of the results. Success of the project will be assessed by benchmarking the proposed framework on two remote sensing applications. Substantial breakthroughs over classical methods are expected, both in terms of efficiency and effectiveness.