trans4num INSPIRE Hackathon 2024

Create experimental tools and educational content to demonstrate the benefits of Nature-Based Solutions (NBS) in agriculture, focusing on ecosystem services, biodiversity, and soil health.

This challenge aims to develop innovative demonstration tools that highlight the impact of Nature-Based Solutions (NBS) on ecosystem services and biodiversity in agriculture. The tools will focus on soil health, with live earthworms used as biological indicators in many of the demonstrations. The project will involve developing a mobile "NBS Help Tools" interactive kit, featuring elements such as soil compaction testers, erosion simulations, and composting process tools. The kit will be designed to visually and practically educate users on the importance of NBS in sustainable farming, with the goal of promoting broader adoption of these practices.

Introduction/Context

NBS can enhance agricultural resilience by improving soil health, water retention, and biodiversity. However, understanding and visualizing the benefits of NBS in a practical, educational context is essential to encourage widespread adoption. This challenge focuses on creating educational tools that simplify the complex interactions of NBS, helping to spread their use in agriculture.

Ambition of the challenge

The ambition is to create a comprehensive, mobile demonstration kit that can be used in educational and agricultural settings to showcase the real-world benefits of NBS. The tools will help farmers, students, and policymakers better understand how NBS interventions improve soil health and agricultural sustainability.

Next steps

The next steps involve refining and expanding the current NBS Help Tools kit, developing additional tools, and creating educational content for schools and agricultural training programs. Data on the effectiveness of these tools in improving understanding and adoption of NBS practices will be gathered to guide future improvements.

Mentors of the challenge

Zoltán Fűzfa & Katalin Miholicsné Orbán

Zoltán Fűzfa is the president and founder of Pistráng Kör Egyesület, where he plays a key role as an organizer and driving force. With a background as a biology-geography and Waldorf teacher, he is also an experienced canoe tour guide, combining his passion for education, nature, and community leadership.

Develop innovative microalgae-based bio-fertilizers to reduce greenhouse gas emissions and nutrient pollution, offering a sustainable alternative to conventional chemical fertilizers for climate-resilient agriculture.

The ALGAVERSE challenge focuses on leveraging microalgae biotechnology to create sustainable bio-fertilizers, addressing critical environmental issues in agriculture such as greenhouse gas emissions and water pollution caused by excessive chemical fertilizer use. By producing bio-fertilizers derived from microalgal biomass, ALGAVERSE offers an eco-friendly alternative that also captures CO2 and recycles nutrients from waste streams. The ultimate goal is to promote climate-resilient agricultural solutions that enhance soil health, improve food security, and reduce the agricultural sector's carbon footprint. Participants are tasked with developing scalable and cost-effective bio-fertilizers that can compete with traditional fertilizers, while also tackling food insecurity, energy crises, and climate change.

Introduction/Context

Modern agriculture is heavily dependent on chemical fertilizers, which contribute to significant environmental problems such as soil degradation, water pollution, and high levels of greenhouse gas emissions. The agricultural sector accounts for a large portion of methane, nitrous oxide, and carbon dioxide emissions. ALGAVERSE offers a transformative approach by developing microalgae-based bio-fertilizers that capture CO2 and recycle nutrients from waste streams. These bio-fertilizers present a sustainable solution that not only reduces emissions but also mitigates nutrient pollution in water bodies, making agriculture more resilient to climate change.

Ambition of the challenge

The ambition of ALGAVERSE is to revolutionize the agricultural industry by replacing conventional chemical fertilizers with microalgae-based alternatives. The challenge aims to create bio-fertilizers that are not only environmentally friendly but also cost-effective and capable of supporting sustainable food production. The overarching vision is to drastically reduce agricultural greenhouse gas emissions and nutrient pollution while promoting soil health. Participants are encouraged to develop innovative solutions that can be scaled globally, contributing to carbon removal on a gigaton scale and addressing key challenges like food insecurity and climate change.

Next steps

Participants will explore the development and scaling of microalgae bio-fertilizers, focusing on maximizing their CO2-capturing potential and nutrient recycling capabilities. The challenge will involve testing these fertilizers for their efficacy in various agricultural and horticultural applications, ensuring they meet the criteria for sustainability, cost-effectiveness, and scalability. Additionally, the solutions should be designed to reduce greenhouse gas emissions and nutrient pollution while improving soil health and crop yields. Collaboration with agricultural stakeholders will be key to implementing and scaling these innovative solutions across different regions and contexts

• Improvement of efficiency in the absorption and assimilation of nutrients.
• Tolerance to biotic or abiotic stresses, or improving any of their agronomic characteristics.
• They could complement, and in some cases substitute, chemical agro-products; that could improve plants metabolism and biochemical activities.

Mentor of the challenge

Nayab Raza

Nayab Raza, a PhD Scholar in Environmental Biology at the University of Manchester, UK, is a distinguished academic and entrepreneur with a passion for sustainable agriculture and water management. As the Founder and CEO of ALGAVERSE– Sustainable food for soil, she has spearheaded the development of innovative products such as BHAAN, a native freshwater microalgae-based biofertilizer. Her mission is to make innovative CO2 capturing Bio-Fertilizers derived from microalgal biomass that recover and recycle nutrients from waste-streams, with potential to provide solutions for food insecurity, energy crises and climate change. Furthermore, Nayab Raza is dedicated to promoting gender inclusion and climate-smart agriculture. She has trained over 500-600 farmers in Sindh, emphasizing the importance of sustainable farming practices and empowering women in agriculture. Her efforts serve as an inspiration for women in the world and beyond, encouraging them to actively engage in environmentally conscious agriculture.

Develop AI-powered solutions using ALIANCE technology to remove clouds from satellite imagery, improving crop monitoring for precision agriculture.

This challenge focuses on creating AI-driven solutions that enhance crop monitoring by removing cloud obstructions from satellite imagery. By utilizing ALIANCE, an advanced machine learning technology, satellite images from Sentinel-1 (radar) and Sentinel-2 (optical) can be integrated to create clear, accurate images for continuous crop surveillance. This approach is crucial for precision agriculture, optimizing resources such as water, fertilizers, and pesticides. The challenge also promotes the exploration of various data sources for precise monitoring, leading to better decision-making in agriculture under adverse weather conditions.

Introduction/Context

In precision agriculture, cloud-covered satellite images often hinder accurate crop monitoring, leading to inefficient resource use. The ALIANCE technology addresses this issue by removing clouds from satellite imagery, enabling clear and consistent crop health monitoring. This challenge seeks to advance the integration of radar and optical data for improved accuracy, ensuring continuous monitoring regardless of weather conditions.

Ambition of the challenge

The ambition is to enhance the accuracy of crop monitoring through AI-driven cloud removal from satellite imagery, leading to better-informed agricultural practices. This technology can also be applied to other sectors, such as land-use planning and crisis management, demonstrating its scalability and versatility.

Next steps

Participants will develop and optimize AI models for cloud removal, explore the integration of Sentinel-1 and Sentinel-2 data, and apply these solutions to precision agriculture. Future steps may include scaling the technology for broader applications and incorporating more data sources to improve monitoring accuracy further.

Mentors of the challenge

Petr Šimánek & Jiří Kvapil

Petr Šimánek is the Head of the AI Research Lab at Czech Technical University, Faculty of Information Technology, with a background in applied mathematics. He leads industrial research projects focused on machine learning for spatio-temporal analysis and prediction. His expertise includes AI for weather forecasting, merging physics with ML, and satellite data analysis.

Develop AI-driven solutions that improve local meteorological forecast accuracy using the ALIANCE platform, supporting agriculture and other sectors with precise, real-time weather data.

This challenge centers on creating innovative tools to enhance local meteorological forecasts by leveraging the ALIANCE technology platform, which integrates both local and global meteorological data. Using advanced machine learning techniques such as multilayer perceptrons, CatBoost, and LSTM networks ALIANCE generates high-resolution, near real-time forecasts. This technology significantly improves planning and management in agriculture by providing accurate weather data, but its applications extend to other sectors, including crisis management, construction, and logistics, making it a versatile forecasting solution.

Introduction/Context

Accurate weather forecasts are crucial for operational efficiency across numerous industries, especially agriculture. Traditional forecasting methods often lack the precision needed for local conditions, resulting in inefficiencies and increased risk. The ALIANCE platform, through its integration of advanced AI techniques and diverse data sources, offers a transformative approach to delivering high-resolution, real-time forecasts that optimize decision-making.

Ambition of the challenge

The challenge aims to revolutionize meteorological forecasting with the ALIANCE platform by significantly improving local forecast accuracy. The goal is to offer high-resolution forecasts that enhance agricultural planning and management, while also extending this capability to other critical sectors reliant on accurate weather data.

Next steps

Participants will work on refining the ALIANCE platform’s machine learning models to further improve forecast precision. The challenge includes developing super-resolution models for real-time forecasting and exploring applications beyond agriculture, such as crisis management and logistics, to broaden the platform’s impact.

Mentors of the challenge

Miroslav Čepek & Michal Kepka

Miroslav Čepek is a Research Fellow at the Faculty of Information Technology, CTU in Prague. With over a decade of expertise in data mining, pricing optimization, and big data analytics, he holds a PhD in Computer Science from Czech Technical University in Prague, specializing in genetic algorithms and knowledge discovery.

Leverage AI and geospatial data integration (GeoAI) to develop innovative solutions that enhance rural development strategies through improved decision-making and resource allocation.

This challenge invites participants to explore the intersection of AI, particularly large language models (LLMs), and geospatial analysis to drive advancements in rural development. Known as GeoAI, this approach enables the analysis of spatial data to generate valuable insights for addressing rural challenges. Participants can use existing LLM platforms as well as combine them with new models on provided infrastructure to craft AI-powered solutions that directly impact rural communities. The goal is to enhance decision-making, optimize resource allocation, and develop strategies that promote agricultural efficiency, land-use planning, and infrastructure development.

Introduction/Context

Rural development often faces challenges related to resource distribution, agricultural planning, and infrastructure needs. By combining AI and geospatial data, there is a unique opportunity to make more informed decisions that address these issues. GeoAI leverages the power of large language models and geographic information systems (GIS) to provide deeper insights into spatial data, offering scalable solutions for rural development initiatives.

Ambition of the challenge

The challenge aims to push the boundaries of AI in rural development by integrating LLMs with the models for geospatial analysis. Participants are tasked with creating solutions that not only improve decision-making processes but also drive sustainable growth in rural communities. The ambition is to develop tools that are practical, impactful, and scalable, allowing rural areas to benefit from advanced AI applications.

Next steps

Participants will explore how to effectively merge LLMs with GIS tools and other AI models that work with these tools to address real-world rural development needs. They will refine their AI models for specific tasks such as improving agricultural productivity, optimizing land use, and enhancing infrastructure planning. The challenge encourages the use of innovative AI techniques to solve complex geospatial problems, with a focus on delivering practical outcomes that can be applied to rural development projects.

Mentors of the challenge

Alexander Kovalenko & Karel Charvát

Alexander Kovalenko is an AI Research Scientist at the Faculty of Information Technology, CTU in Prague, focusing on applied and fundamental machine intelligence research. With a PhD in Physics from Czech Technical University and a Master’s in Aerospace Engineering, his expertise spans machine learning, algorithms, and deep learning. Alexander has also taught computational intelligence and data mining and contributed to charity through educational initiatives.

Develop web-scraping tools to automatically extract nutrient-related data from various sources to support models for regional nutrient balances and food-system level assessments.

This challenge seeks the development of a web-scraping solution to gather large-scale data from scattered sources, including scientific papers, reports, and websites, to support trans4num’s mission of upscaling Nature-Based Solutions (NBS) in agricultural and food systems. The challenge focuses on extracting critical data on nutrient contents in crops, feed, and food commodities to model regional nutrient balances and assess the contribution of these systems to population feeding. The goal is to create web-scraping code that gathers relevant data in a standardized format suitable for integration into NBS models.

Introduction/Context

As part of trans4num, nutrient balance and food-system level assessments require vast amounts of detailed data on nutrient characteristics. Unfortunately, much of this data exists in fragmented and unstructured formats across multiple online sources. This lack of accessibility limits the ability to conduct precise modeling and assessments. Web-scraping can automate the process of gathering this dispersed information, ensuring that it is collected in a structured, standardized form for use in models aimed at upscaling NBS and improving regional nutrient management.

Ambition of the challenge

The ambition is to develop an efficient, automated web-scraping tool capable of retrieving relevant data on nutrient content from multiple sources. This tool will enhance the accessibility of critical data needed for precise modeling of food systems and nutrient balances. By automating data collection, participants will significantly streamline the process of feeding relevant information into models for large-scale NBS implementation.

Next steps

Participants will program web-scraping algorithms that extract nutrient data from diverse sources, such as academic papers and websites. They will also ensure that the data is stored in a standardized format for easy integration into existing models for regional nutrient balances and food-system level assessments. This challenge will involve data science, coding, and developing techniques to handle a wide range of unstructured data sources.

Mentor of the challenge

Adrian Müller

Adrian Müller is a senior researcher at the Research Institute for Organic Agriculture FiBL in Switzerland. He works on food systems modelling, with a focus on organic and circular food systems and the role of livestock in them; further topics are climate change adaptation and mitigation in organic agriculture and agroecology, and climate policy in agriculture.

Regional nutrient balances can help to assess the effectiveness of NbS towards improved nutrient circularity. In this challenge, we would like to collect data and develop a modelling approach for calculating regional nutrient balances for the trans4num case study sites in Europe.

Regional nutrient balances provide crucial insights into current nutrient related imbalances and challenges and offer opportunities to assess the effectiveness of NbS in enhancing nutrient circularity on a regional scale. However, calculating these balances is often hindered by limited data availability. Existing datasets such as FAOSTAT or EUROSTAT contain inconsistencies and are usually on national or NUTS2 level and not on the desired spatial scale (such as landscape, river catchments, etc.) that would be relevant for NbS implementation.

In this challenge we seek to explore ways on how to calculate and visualize regional nutrient balances that are meaningful for the NbS case study sites. One way could be to examine to which extent downscaling from calculations with national or sub-national data to the regional level can be used for this purpose. Eventually, these outcomes should be evaluated by local experts and compared with existing locally-derived nutrient balance calculations. Furthermore, we aim to improve our nutrient balance calculations by gathering more accurate data for specific regions, which might only be accessible in local languages.

Therefore, we are seeking collaborators with local expertise, particularly from the countries involved in the trans4num NbS sites (Denmark, the Netherlands, Hungary, and the UK), for addressing this challenge.

Introduction/Context

In the context of trans4num, Nature-Based Solutions (NBS) aim to enhance nutrient circularity at a regional level. However, existing models for calculating regional nutrient balances often rely on coarse, generalized data, leading to high levels of uncertainty when applied to specific regions. This challenge addresses the need for more precise tools to assess how effectively NBS contribute to nutrient recycling and sustainability in localized contexts.

Ambition of the challenge

Ultimately, our ambition is to integrate the outcome of the challenge into the trans4num decision support tool for local stakeholders and decision-makers.

Next steps

Participants will explore ways to develop and refine nutrient balance models by integrating more localized data sources and improving the algorithms that calculate nutrient recycling efficiency. The challenge includes the development of new data collection methods. The outcomes of the challenge will help to inform strategies aimed at closing nutrient cycles and promoting sustainable practices in agriculture and resource management.

Mentors of the challenge

Hanna Frick & Adrian Müller

Hanna Frick holds a MSc.degree in environmental sciences and has worked on nitrate leaching losses from animal manure during her PhD. Currently, she is working as a scientist in the Department of Soil Sciences at FiBL, Switzerland. Her research focuses on nutrient management, manure and recycling fertilizers as well as reducing nutrient losses. Within trans4num: Work on decision-support tool for improved nutrient management on a regional scale using nature-based solutions (NbS).

Develop a flexible agent-based model framework to evaluate the adoption and impact of Nature-Based Solutions (NBS) across diverse local contexts in both Chinese and European settings.

This challenge aims to create a common agent-based modelling (ABM) framework capable of integrating diverse Nature-Based Solutions (NBS) case studies from different regions. NBS are inherently local but must be transferable across various societies and contexts to be scalable. Successful NBS adoption relies not only on stakeholder acceptance but also on the diffusion of innovations within farming communities. Participants will develop a joint Chinese-European ABM specification plan that can assess the social and technological impacts of NBS adoption. The challenge requires managing complexity and ensuring flexibility within the model by working with modular structures, limited data, and diverse dissemination pathways.

Introduction/Context

NBS are increasingly seen as vital for sustainable agricultural and environmental practices. However, the key challenge lies in their transferability between regions, contexts, and stakeholder groups. Current research highlights the importance of both local case studies and the wider diffusion of these solutions. In this context, agent-based modelling offers a unique opportunity to simulate the socio-economic impacts of NBS adoption, understanding the diffusion pathways and policy impacts in different environments. This challenge seeks to create a joint Chinese-European modelling framework that provides ex-ante impact assessments of NBS adoption.

Ambition of the challenge

The ambition is to develop a flexible, scalable, and transferable agent-based model that can evaluate the adoption and impact of NBS in different regions. By integrating various case studies from China and Europe, the model will provide a comprehensive framework for assessing the net effects of NBS on agricultural innovation, policy impacts, and diffusion pathways. This will enable policymakers and stakeholders to make more informed decisions regarding NBS implementation and scalability.

Next steps

Participants will collaborate to define the model specification, ensuring it addresses the complexities of working with limited data, modular structures, and varied agent types. The model will incorporate key elements such as innovative crop rotations, diffusion pathways, and policy impacts. Once developed, the model can be tested in both Chinese and European contexts, with a focus on evaluating the socio-economic benefits of NBS adoption and creating a toolkit for broader dissemination and scaling.

Mentor of the challenge

Anke Möhring

Anke Möhring is an agricultural economist with a strong background in policy evaluation and modelling. Her research focuses on the impact of agricultural and environmental policies on land use. She also holds a Master of Advanced Studies in Data Science. She is currently a senior researcher in agent-based modelling at the Research Institute of Organic Agriculture (FiBL) in Switzerland.