BRGM unveils its 2021 prizes

BRGM Geosciences for a Sustainable Earth FRENCH GEOLOGICAL SURVEY 2021 INNOVATION PRIZE The DROP Project: "Developing Research for Operational Predictivity". Mineral Resources and the Circular Economy Hello. I'm Guillaume Bertrand and I'm the Head of the Geology and Mineral Resource Economy Unit in the BRGM geo-resources department. I participated in the DROP Project: Developing Research for Operational Predictivity. Presenting the Project DROP is the fruit of a collaborative effort between heads in the geo-resources and IT departments at BRGM. The project sought to develop a tool for predictive mapping for mineral-resource exploration. Predictive mapping seeks to create maps where a given phenomenon is likely to occur. In this case, it's a concentration of a mineral that we're looking for. It's a means of predicting areas with high potential or low potential for discovering a deposit, in order to better target explorations. DROP's Three Innovations In reality, DROP is not one but three innovations. The first is the development of an original methodology for predictive mapping called CBA, or "Cell-Based Association". What's unique about this method is it takes into account the environment and the broader mineralization context to reduce the biases of conventional techniques, such as inaccurate locating from mapping data. We quickly realized that this method was an original way to associate data for processing. Going further and optimizing calculations of values of potential, we added an artificial-intelligence approach, random forests, which allowed for predicting based on a large number of decision trees. That was the second innovation. It forced us to break open a bolt in science because we had to identify true negatives, or "non-deposits". Without getting too into technical details, we selected a rather elegant and original way to determine non-deposits that was not purely rooted in geology, but rather on a statistical approach. That was the third innovation. Rolling Out the Project We assessed the performance of the DROP approach at a competition organized within a conference on predictive mapping for mineral resources at BRGM. A number of teams processed the same dataset, similar to what a mining-exploration company would have to make a predictive map. The DROP approach performed better based on numbers of deposits found on the least amount of explored area, which is the goal of an exploration. Conclusions and Possibilities DROP is the fruit of three innovations. It's now a digital tool that was developed into a prototype. Our goal is to continue its development to create an operational predictive mapping tool that can be used by BRGM as well as offered to third parties. DROP by no means should be limited to use in mineral resources, because the approach is versatile and can apply to other geoscience fields like water, geothermals, natural disasters, and much more.

BRGM Geosciences for a Sustainable Earth FRENCH GEOLOGICAL SURVEY 2021 Scientific Publication Prize Combining narrative scenarios, local knowledge, and land-use-change modeling to provide an overall view of soil erosion My name is Cécile Hérivaux. I work at BRGM in Montpellier developing economic approaches to protecting soil and water resources. Hi, I'm Fabrice Vinatier. I work at the UMR LISAH doing digital mapping of landscapes to study the impact of changing land use on those same resources. I'm Mohammed Sabir. I work at the ENFI on approaches to conserving soil. Context and goals of the project Today, we hear a lot of talk about soil erosion. It can be partly caused by water erosion, moving soil particles from one place to another. Playing a part are different factors like climate change, land use, and farming practices. Mediterranean catchment areas are greatly threatened by erosion, which leads to diminished soil fertility and silt washing into dams. The research project We absolutely have to investigate the uses of these areas to limit erosion and damage. That's what the ALMIRA project we participated in looks at. Our goal Typical approaches confront narrative scenarios, stories, which are difficult to use to simulate biophysical events like water erosion of soils. We wanted to create scenarios that are plausible, spatial, and of use for simulating water-erosion events. We devised an innovative approach that combines prediction, modeling changes in land use, and working with stakeholders throughout the process. The study area Approach Application to the Tleta Catchment Area in Morocco It's a largely rural basin that's rapidly becoming urbanized and thus facing major erosion problems. The dam lake at the basin outlet is used to produce drinking water for Tangiers, but it lost 1/3 of its storage capacity in 45 years due to erosion. We wanted to understand how land use in the basin could develop over time The approach First, we looked at satellite images and statistical data in the region, then launched a series of interviews with farmers in the basin as well as institutional stakeholders. From there, we formulated hypotheses about future changes and expressed the scenarios in story form written in news articles. The stories were then translated into spatial and quantifiable hypotheses working with local stakeholders, then turned into a model of changing land use to generate scenario maps through 2040. We had to simulate the growth of small villages and add new crops in the catchment area based on the hypotheses that we formulated. Results We came up with three different scenarios for how land use would change, leaving various margins for rainfed agriculture and the urban area. The scenarios were then submitted to debate in workshops with local stakeholders. That allowed us to estimate the positive impact of rainfed-agriculture development on agricultural production in general and stemming soil erosion, with changing impacts depending on the scenario and whether farmers are located in the plains or the hillier regions. This work is important for getting stakeholders involved in thinking about solutions for sustainable management of soils across the entire basin.

BRGM Geosciences for a Sustainable Earth FRENCH GEOLOGICAL SURVEY 2021 BRGM THESIS PRIZE Hydrological Signatures of Karst Basins Managing Groundwater Hello. I'm Martin Le Mesnil and I carried out my thesis work at LISAH, or the Soil- Agrosystem-Hydrosystem Interaction Lab, and at BRGM, the French Geological Survey in Montpellier. My thesis is called "Hydrological Signatures of Karst Basins". It looks at the influence of karst groundwater on river flooding. It aims to improve understanding of flood phenomena and make it easier to model to better manage flood risks in karst catchment areas. River catchment areas take in all the rain that feeds the river. What is Karst? Let's explain what karst is. Karst develops in regions made up of soluble rock. This rock can split up with tectonic movements and dissolve as water infiltrates it. This creates and grows channels for water exchange between the surface and groundwater. It can create open structures that humans can enter called karst terrain. Quantifying and Representing To carry out my thesis, I looked into the influence of karsts on flooding. The influence is talked about but hard to quantify, because karst basins are complex and groundwater activity is hard to monitor. It can vary basin to basin and depending on the conditions of a flood. The primary challenge of my thesis was producing quantifiable, representative results. For that, I had to roll out multiple approaches at multiple sites. I chose a number of precise indicators called hydrological signatures that would allow for quantifiable results. Methods Used I worked at 120 monitoring stations spread over three regions of France in catchment areas with karst components: Cévennes, Jura, and Normandy. For each of the 120 stations, I centralized the available data on river flows and climate. I also quantified the major flows of the water cycle: precipitation, streaming, evaporation, and infiltration. This water-balance method applied to 120 study sites over the course of a year helped to locate the regions where rivers lose water via infiltration toward aquifers or gain water from resurging groundwater. On the scale of a flood, this balance method was coupled with analyses of flow-rate curves and simulation of losses and gains between two monitoring stations. That's how I proved the influence of karst basins on reducing flooding phenomena, meaning they reduce maximum flow rates and extend the length of the flood. To better understand the processes in reality associated with these flows, we installed monitors at two catchment areas to monitor water mineralization from electrical conductance. Mineralization was proportional to the quantity of ions in the water, and in karst areas to the time the water spent in the karst groundwater. Through this analysis, I traced back groundwater or surface-water origin of the flooding water based on karst type and the season. Finally, hydrological models helping to simulate river flows as a function of the precipitation received were used at every monitoring site. I then analyzed variations in parameters based on karst presence. This approach showed that the influence of karst can be taken into account during modeling by adapting parameters based on how saturated the karst is or dominant groundwater flows. Conclusions All of this work helped create a typology of roles karst plays in flooding based on the major types of karst one finds. It's a significant result in light of the fact that so few large-scale studies exist. It allows for better accounting for karst in modeling as performed by organizations tasked with anticipating floods such as Schapi, which runs monitoring site Vigicrues.

BRGM Geosciences for a Sustainable Earth FRENCH GEOLOGICAL SURVEY 2021 SPECIAL JURY PRIZE Critical mineral supply policies: a study of cobalt flows and the EU cobalt value chain by Raphaël Danino Perraud Mineral Resources and the Circular Economy Hi. I'm Raphaël Danino Perraud. My thesis was on supply policies concerning strategic minerals, specifically an analysis of value chains and flows of cobalt in Europe. I carried out my work at BRGM and the Laboratoire d'économie in Orléans. My thesis looked at three points that satisfied the expectations of my supervisors. The first was analyzing flows of materials, so an MFA, for the European Union from 2008 to 2017. Second was to study the societal metabolism of cobalt across the entire value chain. Third was carrying out a geopolitical analysis according to these different methodologies. What is cobalt? Why cobalt? Cobalt is a substance primarily used in lithium-ion batteries that we use more and more in our everyday lives. There's a lot of data on cobalt, so it was relatively easy to carry out a flow analysis. What's a value chain? A value chain is all the steps of use and production of a product or a substance. For example, we know cobalt is a by-product of copper and nickel. So, it comes from stratiform deposits of copper or deposits of nickel-cobalt sulfide. It can be used in batteries, superalloys, so in electronics, smartphones, computers, or planes and other turbines. What is a material flow analysis? A material flow analysis is a method that tries to analyze the incoming and outgoing flows of a substance or a product in a given geographic space. In my case, it was the European Union, during a given time, so from 2008 to 2017. What were the challenges? The first challenge I encountered was going into databases and understanding how they worked. It's not intuitive at the outset. Then the challenge was gathering data and translating it into cobalt-containing products. That was kind of fun, really, because I needed to analyze cobalt contained in batteries and alloys from Japan, the US, and Korea. I don't know if you know how many grams of cobalt are in a Samsung or Panasonic battery, or in an A380 plane, but I certainly didn't. And I still don't know, nobody does. But there's a lot of data explaining more or less how much cobalt is in what kind of product. We take averages, which makes it easier. For other types of products, chemical products, there's no data and it requires modeling. What were the results? As a conclusion, I was able to produce a material flow analysis of cobalt in the EU, a dynamic flow analysis, which had never been done, or only partially. That was the scientific objective. I was able to make a bibliographical review of cobalt contained in different products at different stages of production and use. That had never been done before, or only partially. That completed the work. It allowed me the opportunity to take a new look at the geopolitics of raw materials by broadening the scope to the geopolitics of supply chains of raw materials, taking a more complete, systemic view on these topics. Conclusion I think what I'm most proud of was creating a dialogue between different disciplines such as economics, geography, geology, and different concepts like material flow analysis, value chains, geopolitics, and between people from different institutions who never thought they would have to interact but who were able to talk, discuss to make for a quality thesis. Quality would have suffered had I focused on only one or two disciplines or one or two concepts. That's why I want to thank BRGM and everyone who helped me along the way to make the work as high quality as it is.