MOLONARI stands for "monitoring local des échanges nappe-rivière", which translates in English to local monitoring of stream-aquifer exchanges (LOMOS). This project aims to develop a comprehensive environmental monitoring solution for water and heat exchanges in riverbed environments.
This repository provides detailed descriptions and scripts on how to build, prepare and deploy a 1D MOLONARI device, from hardware deployment to scientific analysis. A MOLONARI1D device provides temperature and pressure sensors in the river bed, at depths ranging from 0 to ~40cm. In the long run, these measures give us informations on water and heat exchanges between the river and the aquifer, and on the aquifer's properties (porosity, permeability, etc.)
MOLONARI 1D v1.0 is composed of two devices, measuring depth distributed temperatures and the associated hydraulic head difference over the riverbed. MOLONARI 1D v2.0 integrates all the sensor in one device only
Key attention points - The MOLONARI1D device is designed with the following key concepts in mind :
- low-tech production
- open-source and replicability
- scalability
Here is an overview of the main features developed so far in the MOLONARI1D project.
Arduino-based underwater sensors network
- Sensor Nodes: Waterproof packages with temperature/pressure sensors
- Relay Stations: Data aggregation and LoRaWAN communication
- Communication Protocols: Custom LoRa (local) + LoRaWAN (wide-area)
- Power Management: 8-12 months autonomous operation
Bayesian inference for hydrological parameter estimation
- Bayesian inversion, MCMC based: Infers physical properties of 1D riverbed columns - multichain MCMC also available
- Data Integration: Direct coupling with sensor data streams from monitoring systems
- Uncertainty Quantification: Provides parameter estimates with confidence intervals
- Research Extensions: Modular architecture supporting experimental features
PyQt5-based GUI for operational monitoring
- Device Registration: Laboratory and sampling point hierarchy management
- Data Pipeline: Quality control workflows from raw sensor data to analysis-ready datasets
- Analysis Integration: Direct launching of pyheatmy inference workflows
MOLONARI1D implements a multi-tier monitoring network architecture designed for scalable, autonomous environmental monitoring.
Data Flow Pipeline:
Underwater Sensors → Relay → Gateway → Server → Database → Analysis Tools
(Arduino) (LoRa) (LoRaWAN) (Internet) (SQL) (Python ML/GUI)
- Field Sensors: Battery-powered Arduino devices collect temperature/pressure every 15min
- Local Communication: Custom LoRa protocol transmits sensor data to relay daily
- Wide Area Network: LoRaWAN gateway forwards data to internet-connected server
- Quality Control: Server database processes and validates incoming sensor data - backend of Molonaviz
- Analysis Interface: GUI manages devices and launches scientific computing - frontend of Molonaviz
- Scientific Inference: pyheatmy performs Bayesian MCMC inversion for flux estimation
MOLONARI1D/
├── hardware/ # Arduino-based monitoring devices
│ ├── sensors/ # Temperature, pressure, and demo sensors
│ ├── relay/ # Data aggregation stations
│ ├── shared/ # Common libraries and protocols
│ ├── tests/ # Hardware validation tests
│ └── docs/ # Hardware documentation
├── pyheatmy/ # Scientific computing engine
├── Molonaviz/ # Device management GUI
├── data/ # Sample datasets
└── dataAnalysis/ # Analysis tools and notebooks
- 🔬 Research Users: Operate monitoring systems and analyze flux data
- 💻 Software Developers: Extend Python ecosystem and analysis tools
- 🔧 Hardware Developers: Build and deploy sensor networks
- 📡 Protocol Engineers: Develop communication systems
- 🏭 Fablabs: Manufacture and deploy monitoring hardware
Current TRL: 5-6 - Laboratory validation complete, field deployment in progress
The MOLONARI1D system has successfully demonstrated complete functionality in laboratory underwater testing. See TRL_ASSESSMENT.md for detailed Technology Readiness Level evaluation, including:
- ✅ Complete system validated underwater in laboratory (TRL 5)
- 🔄 Transitioning to field deployment and operational validation (TRL 6)
- Evidence-based assessment of hardware, firmware, and system integration
- Recommendations for advancing to higher TRL levels
- 📖 Documentation: See component-specific README files
- 🧪 Examples: Check
data/anddataAnalysis/directories - 🐛 Issues: Open GitHub issues for support and bug reports
- 💬 Community: Participate in collaborative development
This version of MOLONARI1D is a development version. Some features are not yet implemented or are incomplete. Some bugs may also appear. We therefore do not guarantee any reliability on the resulting values of the calculations, however the data format will remain constant during their full implementation. Think of this code as a template that will remain persistent when the features are reliable in their results.
🚀 Enhanced Development Workflows: Comprehensive repository reorganization with team-oriented structure, consolidated shared libraries, and automated CI/CD validation for all components.
📚 Comprehensive Documentation: Added 15+ README files for Arduino sketches, complete API documentation (MOLONARI1D_API.md), and integration guides between hardware and software components.
🔧 Improved Hardware Integration: Better integration between the original Device/ structure and new organized hardware/ directory, with clear migration paths and backward compatibility.
✅ Quality Assurance: Fixed CI issues, validated all test suites, and ensured robust development workflows for multi-team collaboration.
MOLONARI1D is a teaching and research project held at Mines Paris - PSL since 2021, under the supervision of Nicolas Flipo, Aurélien Baudin, Agnès Rivière, Thomas Romary, and Fabien Ors.
- Nicolas Flipo supervises the development of the MOLONARI 1D ecosystem. He manages the project, the infrastructure design and deployment, the coordination of teaching, production and research, as well as the github repository and the developpers' teams. With Fabien Ors and Pierre Guillou, he is in charge of the source code development either for hardware programming, scientific computing, and IT
- Aurelien Baudin, Agnès Rivière, and Nicolas Flipo supervise the monitoring device development and their in situ deployment
- Nicolas Flipo and Thomas Romary supervise the development of pyheatmy
- Fabien Ors contributed to the 2021 and 2022 editions by conceptualising molonaviz, and since 2024 securizing it
- Since 2024, Pierre Guillou secures the software deployment
The MOLONARI1D project is integrated into the french research project EQUIPEX+ TERRA FORMA led by CNRS. Students contributions are listed in contributors.md
© Contributors to the MOLONARI1D softwares.
All rights reserved. This software and the accompanying materials are made available under the terms of the Eclipse Public License (EPL) v2.0 which accompanies this distribution, and is available at http://www.eclipse.org/legal/epl-v20.html.
pyheatmy and molonaviz were first released in 2021 under the MIT license.