NeroSense

💎 Idea

Water monitoring today is fragmented, rigid, and often inefficient. Existing solutions typically rely on fixed infrastructures or isolated data sources, limiting their ability to adapt to different environments, data availability, and monitoring objectives. As a result, decision-makers lack flexible tools to efficiently track, validate, and respond to dynamic water quality issues.

NeroSense is a scalable, modular water intelligence system designed to transform how water environments are monitored and analyzed. The platform is built around a remote sensing–first approach, where large-scale environmental observations are used to guide more targeted, high-resolution in-situ measurements. NeroSense enables users to:

• Define the water quality indicators they want to monitor (e.g. phytoplankton, turbidity, pollution risk)
• Use a set of integrated remote sensing data sources available within the platform (e.g. satellite data), while also having the ability to connect additional custom sources such as drone imagery or external APIs
• Integrate in-situ hardware systems (e.g. autonomous boats, IoT sensors), which act as a complementary layer to validate and enrich remote sensing insights

Based on the selected configuration, NeroSense performs intelligent analysis and strategic planning, transforming raw environmental data into targeted measurement actions. At the core of this process is a data-driven prioritization model, which identifies where in-situ measurements will bring the highest value. This strategy is dynamically generated based on three key components:

1. Confidence of remote sensing data evaluates data reliability based on factors such as cloud coverage, land interference, and sensor limitations, ensuring measurements focus on areas where uncertainty is highest.
2. Environmental risk and intensity identifies areas with elevated indicator values, highlighting potential hotspots and regions of concern.
3. Temporal dynamics (change over time) detects trends, anomalies, and rapid changes in environmental conditions, prioritizing areas where the system observes significant development or instability.

These components are combined into a unified spatial scoring model, generating priority maps that highlight the most informative locations for in-situ sampling. This approach enables NeroSense to move beyond static monitoring by continuously adapting its measurement strategy based on real-time environmental conditions. Each new in-situ measurement feeds back into the system, refining the model and improving the accuracy of future predictions and strategies.

As a result, NeroSense supports earlier detection of critical events, reduces unnecessary sampling efforts, and provides actionable insights for environmental agencies, water utilities, and researchers responsible for managing water resources.

For the purpose of this hackathon, we demonstrate NeroSense through a focused use case: monitoring phytoplankton dynamics in the Iskar Reservoir, a key water resource for Sofia.

To support this, we develop a mobile autonomous hardware unit (surface robot) equipped with sensors measuring key environmental parameters, which are used to validate and enrich remote sensing data related to phytoplankton monitoring. Through this use case, we demonstrate how such hardware can be designed, connected to the platform, and guided by remote sensing analysis to perform targeted measurements - showcasing how NeroSense operates as an integrated, adaptive system.

🛰️ EU Space Technologies

NeroSense is designed as a data-agnostic platform capable of integrating remote sensing inputs from multiple Earth observation programs, including Copernicus, Landsat, commercial satellite providers, and user-defined data sources.
This flexibility ensures that the system can adapt to different environments, resolutions, and monitoring requirements.

For the purpose of this hackathon and our focused use case - monitoring phytoplankton dynamics in the Iskar Reservoir - we specifically leverage data from the Copernicus Programme, as it provides the optimal balance between spatial resolution, spectral capabilities, revisit frequency, and open accessibility.

In this context, we utilize:

• Sentinel-2 serves as the primary data source for this use case due to its high spatial resolution (10–20 m) and suitable spectral bands in the visible and near-infrared range. These characteristics make it particularly effective for monitoring inland water bodies and deriving key indicators such as chlorophyll-a proxies, turbidity, and suspended matter, enabling detailed, localized analysis of phytoplankton dynamics.
• Sentinel-3 acts as a complementary source focused on spectral accuracy and consistency. While offering lower spatial resolution, Sentinel-3 provides enhanced spectral sensitivity for water color analysis, making it valuable for validating chlorophyll-related signals and identifying broader environmental trends.
• Landsat 8/9 provides long-term historical observations, enabling temporal analysis and trend detection. Although it has a lower revisit frequency compared to Sentinel-2, Landsat data is valuable for retrospective studies and cross-validation, supporting a more robust understanding of water quality evolution over time.
• MODIS offers high temporal resolution with near-daily global coverage, making it useful for tracking rapid changes and large-scale dynamics. While its spatial resolution is coarse, MODIS is effective for identifying temporal patterns, seasonal behavior, and anomaly detection, particularly when combined with higher-resolution sources.

These datasets are openly accessible, frequently updated, and scientifically validated, making them highly reliable for environmental monitoring applications. Within NeroSense, users can seamlessly access these integrated EU space data sources, while also having the flexibility to extend the system with additional remote sensing inputs tailored to their specific use case.

EU space technologies provide the foundation for:

• Wide-area environmental observation 
• Detection of anomalies and patterns 
• Guiding targeted in-situ measurements

By combining these capabilities with adaptive in-situ sensing, NeroSense enhances both the accuracy and efficiency of environmental monitoring.

💧 EU Space for Water

NeroSense addresses the challenge of “Tracking and preventing water pollution” by transforming how Earth observation data is used in real-world monitoring workflows.

While EU space technologies already enable large-scale observation of water bodies, their full potential is often limited by static analysis and lack of integration with in-situ systems. NeroSense builds on this foundation by turning satellite data into actionable, adaptive measurement strategies.

Water pollution processes, such as nutrient-driven phytoplankton blooms, are spatially uneven and time-sensitive. Traditional monitoring approaches rely on fixed sampling locations or periodic measurements, which can miss localized or rapidly developing events.

NeroSense enhances the value of EU space data by:

• Converting satellite observations into dynamic priority maps, identifying where measurements are most needed
• Reducing uncertainty in Earth observation data through targeted in-situ validation
• Adapting measurement strategies in real time, based on evolving environmental conditions
• Enabling earlier detection of critical events, such as rapid phytoplankton growth or pollution spikes

By combining EU space data with adaptive sensing and intelligent analysis, NeroSense moves beyond passive monitoring and enables a closed-loop environmental intelligence system, where observation, validation, and decision-making are continuously connected.

This results in more efficient use of resources, improved data reliability, and faster response to emerging water quality risks, contributing to more effective protection and management of vital water resources.

🚀 Team

• Viktoria Todorova - Data Engineer & Analytics Specialist: brings data-focused thinking and technical versatility to the team. With experience in Python, SQL, and data processing, she is driven to turn raw data into actionable insights. Passionate about learning and innovation, she focuses on building efficient solutions and continuously improving her skills in real-world problem solving.
• Yoan Trendafilov - Embedded Systems & Sensor Integration Engineer: combines expertise in embedded systems, sensor integration, and data acquisition with real-world experience in environmental and atmospheric data processing. He works at the intersection of hardware and data, building systems that collect, process, and automate complex datasets. His strength lies in turning physical measurements into usable, data-driven solutions.
• Nikoleta Evtimova - Automation & Robotics Engineer: combines expertise in automation, Python, and robotics with a strong drive for solving complex challenges. She thrives in dynamic environments, building efficient systems and adapting quickly to new problems. With a hands-on approach and high motivation, she brings speed, precision, and execution power to the team.
• Nikola Lozanov - Full Stack Engineer: a Senior Full Stack Developer with deep expertise in backend systems, databases, and scalable web applications. With years of experience optimizing performance and building robust APIs, he focuses on creating efficient, high-quality solutions. He brings strong engineering discipline, speed, and technical leadership to the team.
• Martina Dimitrova - Remote Sensing & Environmental Scientist: combines a background in hygrobiology with advanced studies in remote sensing, focusing on analyzing water ecosystems through satellite data. She brings deep understanding of aquatic environments and translates environmental processes into measurable, data-driven insights. Her expertise connects scientific knowledge with real-world monitoring and decision-making.
• Elena Deleva - Water Systems & Environmental Engineer: specializing in water supply, wastewater systems, and environmental infrastructure. With hands-on experience in designing treatment systems and sustainable engineering solutions, she bridges technical design with real-world environmental challenges. She brings critical domain expertise and practical knowledge to support data-driven water management solutions.
• Nikolay Vasilev - AI & Data Scientist: a data scientist and backend developer specializing in AI, machine learning, and data-driven systems. With experience in NLP, computer vision, and time series analysis, he builds scalable solutions for real-world problems. Passionate about research and space, he focuses on applying AI to Earth observation and space technologies.
• Bojan Baidanoff - Business Strategy Specialist & GIS Consultant: specializes in GIS, data visualization, graphic and business design. He is responsible for the market research and pricing. His responsibility is the ethical use of AI-driven business development within the team, providing data and strategic partnerships.
  
  

               GitHub-NeroSense

•