Jonathan Coney

MRes Student in Climate and Atmospheric Science 2019-20

Index | About | Computer Project | Files


Hello, world.

I am an MRes student in the School of Earth and Environment at the University of Leeds in Yorkshire. My MRes project is called Goldmine or bust? Crowdsourced meteorological data for atmospheric science, and is supervised by Ben Pickering, Dr David Dufton and Dr Maryna Lukach.

This is a small website to host some information about my MRes project, which involves investigating the effectiveness of Netatmo smart home weather stations for meteorological applications.

There are over 5000 of these stations in the UK alone, and the map to the right shows the location of all of these. A large proportion of these stations are located in urban areas - noticeably around the south east of England. Also plotted for comparison are stations used by the Met Office for temperature observations - in urban areas the number of Netatmo observations far exceeds the number of observations from Met Office stations. Conversely, the opposite is true in more rural areas, such as the Scottish highlands. Netatmo data was used to show the existence of an urban heat island (UHI) in London, and data from Netatmo weather stations is being used as part of the Birmingham Urban Observatory.

The project consists of two parts: calibrating the Netatmo outdoor temperature sensor in a climate chamber; and comparing quality control techniques of Netatmo temperature data.

Unfortunately the chamber testing was cut short due to the COVID-19 outbreak, but I managed to successfully gather temperature and pressure data for each of the 7 sensors I had my disposal. I gathered some humidity data too, however only from 3 sensors and this data was not fantastic in quality. I also plan to quantify the lag time for the temperature sensor and investigate whether the lag is uniform across the temperature range I measured (from -10 to +40 ℃), and from sensor to sensor. Pictured below right is one of the Netatmo outdoor temperature sensors in the Atmospheric Measurement & Observation Facility (AMOF) climate chamber at the National Centre for Atmospheric Science (NCAS) in Leeds. It was about to undergo a temperature calibration test.

Having written code to retrieve data from the Netatmo API for all the Netatmo weather stations in the UK, and save these data as a netCDF file (more information on the Computer Project pages), the next stages involve using UK Met Office data to act as a truth for the Netatmo stations to measure the effectiveness of Netatmo weather stations for recording temperature changes in the UK, and comparison of various quality control methods. The two methods I am focusing on are outlined in these papers: by Nipen (2020) and Meier (2017). If time allows, I will develop my own quality control method, hopefully incorporating machine learning.

As the resolution of models improves, we need more observations to assist with this. As the map to the right shows, for every 1 Met Office observation station, there are about 15 Netatmo weather stations in the UK. Crowdsourcing offers a potential source of these additional observations, particularly around urban areas. However, if we are to trust the data from these stations (for example, how can we ensure these stations have not been left indoors or in direct sunlight?).

Netatmo weather data is being used as part of MET Norway's forecasting models, and as long as a sufficient quality control method is employed, data from home weather stations can complement existing traditional observation networks by providing a dense spatial network of observations, especially in urban areas where microclimates can exist. Future applications include drone deliveries - where microscale observations are required to ensure it is safe for deliveries to take place.

Thank you to my supervisors your time, help and patience, and to Barbara Brooks from the AMOF for your guidance and instruction in operating the chamber safely and effectively. Thank you also to the Birmingham Urban Observatory team from the University of Birmingham, who kindly lent this project 6 Netatmo temperature sensors to calibrate, before they are to be deployed in schools and homes around Birmingham.

Netatmo UK Map
Map of UK Netatmo Smart Home Weather stations, as well as Met Office stations for comparison
Netatmo sensor in Climate chamber
Netatmo temperature sensor in AMOF Climate chamber