Student project: Outdoor nitrogen monitoring

Comparison and improvement of calibration models for a low-cost nitrogen measurement station.

What you will do

High emissions, concentrations and depositions of nitrogen compounds pose a great risk to the balance of ecosystems. The necessity and urgency to reduce the impact of anthropogenic nitrogen have been recognized in the Netherlands and the other member states of the European Union. The excess of nitrogen compounds can result in a surplus of nutrients for vegetation. Plants that need nutrient-poor soil and water disappear and are replaced by species that do thrive on the surplus, leading to a loss in biodiversity. Ammonia (NH3) and nitrogen oxides (NO¬x) are the two major compounds that carry nitrogen through the nitrogen cycle. NH3 is mainly emitted by agricultural activity, NO¬x by traffic and industry. To reduce emissions various measures are taken across the board. However, their effectiveness in practice remains difficult to establish. OnePlanet Research Center has developed a low-cost measurement station to monitor concentrations of nitrogen compounds with high temporal resolution. These stations support the assessment of mitigation measures effectiveness.

Raw measurements of the built-in sensors  are reported in nanoamperes. These values need to be calibrated and converted to a concentrations measurement to provide researcher and decision makers with meaningful insights. Such calibration must account and compensate for factors that would potentially result in biased measurements, such as humidity. There are various methods to calibrate the stations, all with their own benefits and drawbacks. Ideally, the stations are deployed at their intended locations immediately, and calibration is performed while the boxes are deployed in the field (field calibration).

Main tasks:

• Compile a collection of feasible field-calibration approaches based on domain knowledge and literature review.
• Collaborate, brainstorm, and document with team on pros and cons of different field-calibration approaches for different nitrogen compounds.
• Setup code pipeline to evaluate field calibration using machine learning techniques.
• Simulate field calibration using existing datasets.
• Evaluate performance and differences in performance between calibration approaches.

What we do for you

• We have a challenging problem where you have freedom to explore and deliver solutions.
• We have a diverse team of experts in data science, machine learning, hardware, software, sensors, environmental science, and agriculture who can coach you and provide you with advice in the development of the assignment.
• You will join the Digital Twin team of OnePlanet, which employs state-of-the-art knowledge on machine learning and the frameworks necessary to perform these big data tasks on a large scale.
• You will be able to exchange views and knowledge with the OnePlanet and Imec community of experts and scientists, widening your professional network.
• We can help you improve your coding skills up to industry standards.
• You will have access to our cloud infrastructure to solve this problem allowing you to process large amounts of data within a reasonable time. 

Who you are

• Knowledge of Python (or R).
• Knowledge of supervised and unsupervised machine learning algorithms, testing and validation methods.
• Knowledge of time series analysis.
• Knowledge of the working principles of (chemical) sensors is a plus.
• Knowledge of Git is a plus.

Interested

Does this position sound like an interesting next step in your career at imec? Don’t hesitate to submit your application by clicking on ‘APPLY NOW’.