Articles | Volume 14, issue 8
https://doi.org/10.5194/amt-14-5637-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/amt-14-5637-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Machine learning calibration of low-cost NO2 and PM10 sensors: non-linear algorithms and their impact on site transferability
Grantham Institute – Climate Change and the Environment, Imperial College London, London SW7 2AZ, UK
Department of Physics, Imperial College London, London SW7 2AZ, UK
Data Science Institute, Imperial College London, London SW7 2AZ, UK
Climatic Research Unit, School of Environmental Sciences, University of East Anglia, Norwich NR4 7TJ, UK
Lev Konstantinovskiy
AirPublic Ltd, London, UK
Hannah Gardiner
AirPublic Ltd, London, UK
John Cant
AirPublic Ltd, London, UK
Related authors
Xiao Lu, Yiming Liu, Jiayin Su, Xiang Weng, Tabish Ansari, Yuqiang Zhang, Guowen He, Yuqi Zhu, Haolin Wang, Ganquan Zeng, Jingyu Li, Cheng He, Shuai Li, Teerachai Amnuaylojaroen, Tim Butler, Qi Fan, Shaojia Fan, Grant L. Forster, Meng Gao, Jianlin Hu, Yugo Kanaya, Mohd Talib Latif, Keding Lu, Philippe Nédélec, Peer Nowack, Bastien Sauvage, Xiaobin Xu, Lin Zhang, Ke Li, Ja-Ho Koo, and Tatsuya Nagashima
Atmos. Chem. Phys., 25, 7991–8028, https://doi.org/10.5194/acp-25-7991-2025, https://doi.org/10.5194/acp-25-7991-2025, 2025
Short summary
Short summary
This study analyzes summertime ozone trends in East and Southeast Asia derived from a comprehensive observational database spanning from 1995 to 2019, incorporating aircraft observations, ozonesonde data, and measurements from 2500 surface sites. Multiple models are applied to attribute to changes in anthropogenic emissions and climate. The results highlight that increases in anthropogenic emissions are the primary driver of ozone increases both in the free troposphere and at the surface.
Kevin Debeire, Lisa Bock, Peer Nowack, Jakob Runge, and Veronika Eyring
Earth Syst. Dynam., 16, 607–630, https://doi.org/10.5194/esd-16-607-2025, https://doi.org/10.5194/esd-16-607-2025, 2025
Short summary
Short summary
Projecting future precipitation is essential for preparing for climate change, but current climate models still have large uncertainties, especially over land. This study presents a new method to improve precipitation projections by identifying which models best capture key climate patterns. By giving more weight to models that better represent these patterns, our approach leads to more reliable future precipitation projections over land.
Peer Nowack and Duncan Watson-Parris
Atmos. Chem. Phys., 25, 2365–2384, https://doi.org/10.5194/acp-25-2365-2025, https://doi.org/10.5194/acp-25-2365-2025, 2025
Short summary
Short summary
In our article, we review uncertainties in global climate change projections and current methods using Earth observations as constraints, which is crucial for climate risk assessments and for informing society. We then discuss how machine learning can advance the field, discussing recent work that provides potentially stronger and more robust links between observed data and future climate projections. We further discuss the challenges of applying machine learning to climate science.
Jingyu Wang, Gabriel Chiodo, Timofei Sukhodolov, Blanca Ayarzagüena, William T. Ball, Mohamadou Diallo, Birgit Hassler, James Keeble, Peer Nowack, Clara Orbe, and Sandro Vattioni
EGUsphere, https://doi.org/10.5194/egusphere-2025-340, https://doi.org/10.5194/egusphere-2025-340, 2025
Short summary
Short summary
We analyzed the ozone response under elevated CO2 using the data from CMIP6 DECK experiments. We then looked at the relations between ozone response and temperature and circulation changes to identify drivers of the ozone change. The climate feedback of ozone is investigated by doing offline calculations and comparing models with and without interactive chemistry. We find that ozone-climate interactions are important for Earth System Models, thus should be considered in future model development.
Philipp Breul, Paulo Ceppi, and Peer Nowack
EGUsphere, https://doi.org/10.5194/egusphere-2025-221, https://doi.org/10.5194/egusphere-2025-221, 2025
Short summary
Short summary
We explore how Pacific low-level clouds influence projections of regional climate change by adjusting a climate model to enhance low cloud response to surface temperatures. We find significant changes in projected warming patterns and circulation changes, under increased CO2 conditions. Our findings are supported by similar relationships across state-of-the-art climate models. These results highlight the importance of accurately representing clouds for predicting regional climate change impacts.
Sarah Wilson Kemsley, Paulo Ceppi, Hendrik Andersen, Jan Cermak, Philip Stier, and Peer Nowack
Atmos. Chem. Phys., 24, 8295–8316, https://doi.org/10.5194/acp-24-8295-2024, https://doi.org/10.5194/acp-24-8295-2024, 2024
Short summary
Short summary
Aiming to inform parameter selection for future observational constraint analyses, we incorporate five candidate meteorological drivers specifically targeting high clouds into a cloud controlling factor framework within a range of spatial domain sizes. We find a discrepancy between optimal domain size for predicting locally and globally aggregated cloud radiative anomalies and identify upper-tropospheric static stability as an important high-cloud controlling factor.
Hendrik Andersen, Jan Cermak, Alyson Douglas, Timothy A. Myers, Peer Nowack, Philip Stier, Casey J. Wall, and Sarah Wilson Kemsley
Atmos. Chem. Phys., 23, 10775–10794, https://doi.org/10.5194/acp-23-10775-2023, https://doi.org/10.5194/acp-23-10775-2023, 2023
Short summary
Short summary
This study uses an observation-based cloud-controlling factor framework to study near-global sensitivities of cloud radiative effects to a large number of meteorological and aerosol controls. We present near-global sensitivity patterns to selected thermodynamic, dynamic, and aerosol factors and discuss the physical mechanisms underlying the derived sensitivities. Our study hopes to guide future analyses aimed at constraining cloud feedbacks and aerosol–cloud interactions.
Xiang Weng, Grant L. Forster, and Peer Nowack
Atmos. Chem. Phys., 22, 8385–8402, https://doi.org/10.5194/acp-22-8385-2022, https://doi.org/10.5194/acp-22-8385-2022, 2022
Short summary
Short summary
We use machine learning to quantify the meteorological drivers behind surface ozone variations in China between 2015 and 2019. Our novel approaches show improved performance when compared to previous analysis methods. We highlight that nonlinearity in driver relationships and the impacts of large-scale meteorological phenomena are key to understanding ozone pollution. Moreover, we find that almost half of the observed ozone trend between 2015 and 2019 might have been driven by meteorology.
Carl Thomas, Apostolos Voulgarakis, Gerald Lim, Joanna Haigh, and Peer Nowack
Weather Clim. Dynam., 2, 581–608, https://doi.org/10.5194/wcd-2-581-2021, https://doi.org/10.5194/wcd-2-581-2021, 2021
Short summary
Short summary
Atmospheric blocking events are complex large-scale weather patterns which block the path of the jet stream. They are associated with heat waves in summer and cold snaps in winter. Blocking is poorly understood, and the effect of climate change is not clear. Here, we present a new method to study blocking using unsupervised machine learning. We show that this method performs better than previous methods used. These results show the potential for unsupervised learning in atmospheric science.
Alexander Kuhn-Régnier, Apostolos Voulgarakis, Peer Nowack, Matthias Forkel, I. Colin Prentice, and Sandy P. Harrison
Biogeosciences, 18, 3861–3879, https://doi.org/10.5194/bg-18-3861-2021, https://doi.org/10.5194/bg-18-3861-2021, 2021
Short summary
Short summary
Along with current climate, vegetation, and human influences, long-term accumulation of biomass affects fires. Here, we find that including the influence of antecedent vegetation and moisture improves our ability to predict global burnt area. Additionally, the length of the preceding period which needs to be considered for accurate predictions varies across regions.
James Keeble, Birgit Hassler, Antara Banerjee, Ramiro Checa-Garcia, Gabriel Chiodo, Sean Davis, Veronika Eyring, Paul T. Griffiths, Olaf Morgenstern, Peer Nowack, Guang Zeng, Jiankai Zhang, Greg Bodeker, Susannah Burrows, Philip Cameron-Smith, David Cugnet, Christopher Danek, Makoto Deushi, Larry W. Horowitz, Anne Kubin, Lijuan Li, Gerrit Lohmann, Martine Michou, Michael J. Mills, Pierre Nabat, Dirk Olivié, Sungsu Park, Øyvind Seland, Jens Stoll, Karl-Hermann Wieners, and Tongwen Wu
Atmos. Chem. Phys., 21, 5015–5061, https://doi.org/10.5194/acp-21-5015-2021, https://doi.org/10.5194/acp-21-5015-2021, 2021
Short summary
Short summary
Stratospheric ozone and water vapour are key components of the Earth system; changes to both have important impacts on global and regional climate. We evaluate changes to these species from 1850 to 2100 in the new generation of CMIP6 models. There is good agreement between the multi-model mean and observations, although there is substantial variation between the individual models. The future evolution of both ozone and water vapour is strongly dependent on the assumed future emissions scenario.
Abdul Malik, Peer J. Nowack, Joanna D. Haigh, Long Cao, Luqman Atique, and Yves Plancherel
Atmos. Chem. Phys., 20, 15461–15485, https://doi.org/10.5194/acp-20-15461-2020, https://doi.org/10.5194/acp-20-15461-2020, 2020
Short summary
Short summary
Solar geoengineering has been introduced to mitigate human-caused global warming by reflecting sunlight back into space. This research investigates the impact of solar geoengineering on the tropical Pacific climate. We find that solar geoengineering can compensate some of the greenhouse-induced changes in the tropical Pacific but not all. In particular, solar geoengineering will result in significant changes in rainfall, sea surface temperatures, and increased frequency of extreme ENSO events.
Xiao Lu, Yiming Liu, Jiayin Su, Xiang Weng, Tabish Ansari, Yuqiang Zhang, Guowen He, Yuqi Zhu, Haolin Wang, Ganquan Zeng, Jingyu Li, Cheng He, Shuai Li, Teerachai Amnuaylojaroen, Tim Butler, Qi Fan, Shaojia Fan, Grant L. Forster, Meng Gao, Jianlin Hu, Yugo Kanaya, Mohd Talib Latif, Keding Lu, Philippe Nédélec, Peer Nowack, Bastien Sauvage, Xiaobin Xu, Lin Zhang, Ke Li, Ja-Ho Koo, and Tatsuya Nagashima
Atmos. Chem. Phys., 25, 7991–8028, https://doi.org/10.5194/acp-25-7991-2025, https://doi.org/10.5194/acp-25-7991-2025, 2025
Short summary
Short summary
This study analyzes summertime ozone trends in East and Southeast Asia derived from a comprehensive observational database spanning from 1995 to 2019, incorporating aircraft observations, ozonesonde data, and measurements from 2500 surface sites. Multiple models are applied to attribute to changes in anthropogenic emissions and climate. The results highlight that increases in anthropogenic emissions are the primary driver of ozone increases both in the free troposphere and at the surface.
Kevin Debeire, Lisa Bock, Peer Nowack, Jakob Runge, and Veronika Eyring
Earth Syst. Dynam., 16, 607–630, https://doi.org/10.5194/esd-16-607-2025, https://doi.org/10.5194/esd-16-607-2025, 2025
Short summary
Short summary
Projecting future precipitation is essential for preparing for climate change, but current climate models still have large uncertainties, especially over land. This study presents a new method to improve precipitation projections by identifying which models best capture key climate patterns. By giving more weight to models that better represent these patterns, our approach leads to more reliable future precipitation projections over land.
Peer Nowack and Duncan Watson-Parris
Atmos. Chem. Phys., 25, 2365–2384, https://doi.org/10.5194/acp-25-2365-2025, https://doi.org/10.5194/acp-25-2365-2025, 2025
Short summary
Short summary
In our article, we review uncertainties in global climate change projections and current methods using Earth observations as constraints, which is crucial for climate risk assessments and for informing society. We then discuss how machine learning can advance the field, discussing recent work that provides potentially stronger and more robust links between observed data and future climate projections. We further discuss the challenges of applying machine learning to climate science.
Jingyu Wang, Gabriel Chiodo, Timofei Sukhodolov, Blanca Ayarzagüena, William T. Ball, Mohamadou Diallo, Birgit Hassler, James Keeble, Peer Nowack, Clara Orbe, and Sandro Vattioni
EGUsphere, https://doi.org/10.5194/egusphere-2025-340, https://doi.org/10.5194/egusphere-2025-340, 2025
Short summary
Short summary
We analyzed the ozone response under elevated CO2 using the data from CMIP6 DECK experiments. We then looked at the relations between ozone response and temperature and circulation changes to identify drivers of the ozone change. The climate feedback of ozone is investigated by doing offline calculations and comparing models with and without interactive chemistry. We find that ozone-climate interactions are important for Earth System Models, thus should be considered in future model development.
Philipp Breul, Paulo Ceppi, and Peer Nowack
EGUsphere, https://doi.org/10.5194/egusphere-2025-221, https://doi.org/10.5194/egusphere-2025-221, 2025
Short summary
Short summary
We explore how Pacific low-level clouds influence projections of regional climate change by adjusting a climate model to enhance low cloud response to surface temperatures. We find significant changes in projected warming patterns and circulation changes, under increased CO2 conditions. Our findings are supported by similar relationships across state-of-the-art climate models. These results highlight the importance of accurately representing clouds for predicting regional climate change impacts.
Sarah Wilson Kemsley, Paulo Ceppi, Hendrik Andersen, Jan Cermak, Philip Stier, and Peer Nowack
Atmos. Chem. Phys., 24, 8295–8316, https://doi.org/10.5194/acp-24-8295-2024, https://doi.org/10.5194/acp-24-8295-2024, 2024
Short summary
Short summary
Aiming to inform parameter selection for future observational constraint analyses, we incorporate five candidate meteorological drivers specifically targeting high clouds into a cloud controlling factor framework within a range of spatial domain sizes. We find a discrepancy between optimal domain size for predicting locally and globally aggregated cloud radiative anomalies and identify upper-tropospheric static stability as an important high-cloud controlling factor.
Hendrik Andersen, Jan Cermak, Alyson Douglas, Timothy A. Myers, Peer Nowack, Philip Stier, Casey J. Wall, and Sarah Wilson Kemsley
Atmos. Chem. Phys., 23, 10775–10794, https://doi.org/10.5194/acp-23-10775-2023, https://doi.org/10.5194/acp-23-10775-2023, 2023
Short summary
Short summary
This study uses an observation-based cloud-controlling factor framework to study near-global sensitivities of cloud radiative effects to a large number of meteorological and aerosol controls. We present near-global sensitivity patterns to selected thermodynamic, dynamic, and aerosol factors and discuss the physical mechanisms underlying the derived sensitivities. Our study hopes to guide future analyses aimed at constraining cloud feedbacks and aerosol–cloud interactions.
Xiang Weng, Grant L. Forster, and Peer Nowack
Atmos. Chem. Phys., 22, 8385–8402, https://doi.org/10.5194/acp-22-8385-2022, https://doi.org/10.5194/acp-22-8385-2022, 2022
Short summary
Short summary
We use machine learning to quantify the meteorological drivers behind surface ozone variations in China between 2015 and 2019. Our novel approaches show improved performance when compared to previous analysis methods. We highlight that nonlinearity in driver relationships and the impacts of large-scale meteorological phenomena are key to understanding ozone pollution. Moreover, we find that almost half of the observed ozone trend between 2015 and 2019 might have been driven by meteorology.
Carl Thomas, Apostolos Voulgarakis, Gerald Lim, Joanna Haigh, and Peer Nowack
Weather Clim. Dynam., 2, 581–608, https://doi.org/10.5194/wcd-2-581-2021, https://doi.org/10.5194/wcd-2-581-2021, 2021
Short summary
Short summary
Atmospheric blocking events are complex large-scale weather patterns which block the path of the jet stream. They are associated with heat waves in summer and cold snaps in winter. Blocking is poorly understood, and the effect of climate change is not clear. Here, we present a new method to study blocking using unsupervised machine learning. We show that this method performs better than previous methods used. These results show the potential for unsupervised learning in atmospheric science.
Alexander Kuhn-Régnier, Apostolos Voulgarakis, Peer Nowack, Matthias Forkel, I. Colin Prentice, and Sandy P. Harrison
Biogeosciences, 18, 3861–3879, https://doi.org/10.5194/bg-18-3861-2021, https://doi.org/10.5194/bg-18-3861-2021, 2021
Short summary
Short summary
Along with current climate, vegetation, and human influences, long-term accumulation of biomass affects fires. Here, we find that including the influence of antecedent vegetation and moisture improves our ability to predict global burnt area. Additionally, the length of the preceding period which needs to be considered for accurate predictions varies across regions.
James Keeble, Birgit Hassler, Antara Banerjee, Ramiro Checa-Garcia, Gabriel Chiodo, Sean Davis, Veronika Eyring, Paul T. Griffiths, Olaf Morgenstern, Peer Nowack, Guang Zeng, Jiankai Zhang, Greg Bodeker, Susannah Burrows, Philip Cameron-Smith, David Cugnet, Christopher Danek, Makoto Deushi, Larry W. Horowitz, Anne Kubin, Lijuan Li, Gerrit Lohmann, Martine Michou, Michael J. Mills, Pierre Nabat, Dirk Olivié, Sungsu Park, Øyvind Seland, Jens Stoll, Karl-Hermann Wieners, and Tongwen Wu
Atmos. Chem. Phys., 21, 5015–5061, https://doi.org/10.5194/acp-21-5015-2021, https://doi.org/10.5194/acp-21-5015-2021, 2021
Short summary
Short summary
Stratospheric ozone and water vapour are key components of the Earth system; changes to both have important impacts on global and regional climate. We evaluate changes to these species from 1850 to 2100 in the new generation of CMIP6 models. There is good agreement between the multi-model mean and observations, although there is substantial variation between the individual models. The future evolution of both ozone and water vapour is strongly dependent on the assumed future emissions scenario.
Abdul Malik, Peer J. Nowack, Joanna D. Haigh, Long Cao, Luqman Atique, and Yves Plancherel
Atmos. Chem. Phys., 20, 15461–15485, https://doi.org/10.5194/acp-20-15461-2020, https://doi.org/10.5194/acp-20-15461-2020, 2020
Short summary
Short summary
Solar geoengineering has been introduced to mitigate human-caused global warming by reflecting sunlight back into space. This research investigates the impact of solar geoengineering on the tropical Pacific climate. We find that solar geoengineering can compensate some of the greenhouse-induced changes in the tropical Pacific but not all. In particular, solar geoengineering will result in significant changes in rainfall, sea surface temperatures, and increased frequency of extreme ENSO events.
Cited articles
Breiman, L.: Random forests, Mach. Learn., 45, 5–32, https://doi.org/10.1201/9780429469275-8, 2001. a, b
Breiman, L. and Friedman, J. H.: Predicting multivariate responses in multiple linear regression, J. Roy. Stat. Soc.-B, 59, 3–54, https://doi.org/10.1111/1467-9868.00054, 1997. a
Casey, J. G. and Hannigan, M. P.: Testing the performance of field calibration techniques for low-cost gas sensors in new deployment locations: across a county line and across Colorado, Atmos. Meas. Tech., 11, 6351–6378, https://doi.org/10.5194/amt-11-6351-2018, 2018. a, b
Casey, J. G., Collier-Oxandale, A., and Hannigan, M.: Performance of artificial neural networks and linear models to quantify 4 trace gas species in an oil and gas production region with low-cost sensors, Sensor.
Actuat. B-Chem., 283, 504–514, https://doi.org/10.1016/j.snb.2018.12.049, 2019. a
Castell, N., Dauge, F. R., Schneider, P., Vogt, M., Lerner, U., Fishbain, B., Broday, D., and Bartonova, A.: Can commercial low-cost sensor platforms contribute to air quality monitoring and exposure estimates?, Environ. Int., 99, 293–302, https://doi.org/10.1016/j.envint.2016.12.007, 2017. a, b, c, d
Cross, E. S., Williams, L. R., Lewis, D. K., Magoon, G. R., Onasch, T. B., Kaminsky, M. L., Worsnop, D. R., and Jayne, J. T.: Use of electrochemical sensors for measurement of air pollution: correcting interference response and validating measurements, Atmos. Meas. Tech., 10, 3575–3588, https://doi.org/10.5194/amt-10-3575-2017, 2017. a
De Vito, S., Esposito, E., Salvato, M., Popoola, O., Formisano, F., Jones, R., and Di Francia, G.: Calibrating chemical multisensory devices for real world applications: An in-depth comparison of quantitative machine learning approaches, Sensor. Actuat. B-Chem., 255, 1191–1210, https://doi.org/10.1016/j.snb.2017.07.155, 2018. a, b, c
De Vito, S., Esposito, E., Formisano, F., Massera, E., Auria, P. D., and Di Francia, G.: Adaptive Machine learning for Backup Air Quality Multisensor Systems continuous calibration, 2019 IEEE International Symposium on Olfaction and Electronic Nose (ISOEN), 26–29 May 2019, Fukuoka, Japan, 1–4, https://doi.org/10.1109/isoen.2019.8823250, 2019. a
Dormann, C. F., Elith, J., Bacher, S., Buchmann, C., Carl, G., Carré, G., Marquéz, J. R., Gruber, B., Lafourcade, B., Leitão, P. J., Münkemüller, T., Mcclean, C., Osborne, P. E., Reineking, B., Schröder, B., Skidmore, A. K., Zurell, D., and Lautenbach, S.: Collinearity: A review of methods to deal with it and a simulation study evaluating their performance, Ecography, 36, 27–46, https://doi.org/10.1111/j.1600-0587.2012.07348.x, 2013. a
Eilenberg, S. R., Subramanian, R., Malings, C., Hauryliuk, A., Presto, A. A., and Robinson, A. L.: Using a network of lower-cost monitors to identify the influence of modifiable factors driving spatial patterns in fine particulate matter concentrations in an urban environment, J. Expo. Sci. Env. Epid., 30, 949–961, https://doi.org/10.1038/s41370-020-0255-x, 2020. a
Esposito, E., De Vito, S., Salvato, M., Bright, V., Jones, R. L., and Popoola, O.: Dynamic neural network architectures for on field stochastic calibration of indicative low cost air quality sensing systems, Sensor.
Actuat. B-Chem., 231, 701–713, https://doi.org/10.1016/j.snb.2016.03.038, 2016. a
European Environment Agency: Air quality in Europe – 2019 report, available at: http://www.eea.europa.eu/publications/air-quality-in-europe-2012 (last access: 1 November 2020), 2019. a
Green, D. C., Fuller, G. W., and Baker, T.: Development and validation of the volatile correction model for PM10 – An empirical method for adjusting TEOM measurements for their loss of volatile particulate matter, Atmos. Environ., 43, 2132–2141, https://doi.org/10.1016/j.atmosenv.2009.01.024, 2009. a
Hagan, D. H., Isaacman-VanWertz, G., Franklin, J. P., Wallace, L. M. M., Kocar, B. D., Heald, C. L., and Kroll, J. H.: Calibration and assessment of electrochemical air quality sensors by co-location with regulatory-grade instruments, Atmos. Meas. Tech., 11, 315–328, https://doi.org/10.5194/amt-11-315-2018, 2018. a, b, c, d, e
Hagler, G. S., Williams, R., Papapostolou, V., and Polidori, A.: Air Quality Sensors and Data Adjustment Algorithms: When Is It No Longer a Measurement?, Environ. Sci. Technol., 52, 5530–5531, https://doi.org/10.1021/acs.est.8b01826, 2018. a
Hoerl, A. E. and Kennard, R. W.: Ridge Regression: Biased Estimation for Nonorthogonal Problems, Technometrics, 12, 55–67, https://doi.org/10.1080/00401706.2000.10485983, 1970. a
James, G., Witten, D., Hastie, T., and Tibshirani, R.: An Introduction to Statistical Learning, Springer Science+Business Media, New York, https://doi.org/10.1007/978-1-4614-7138-7, 2013. a
Jiao, W., Hagler, G., Williams, R., Sharpe, R., Brown, R., Garver, D., Judge, R., Caudill, M., Rickard, J., Davis, M., Weinstock, L., Zimmer-Dauphinee, S., and Buckley, K.: Community Air Sensor Network (CAIRSENSE) project: evaluation of low-cost sensor performance in a suburban environment in the southeastern United States, Atmos. Meas. Tech., 9, 5281–5292, https://doi.org/10.5194/amt-9-5281-2016, 2016. a
Keller, C. A. and Evans, M. J.: Application of random forest regression to the calculation of gas-phase chemistry within the GEOS-Chem chemistry model v10, Geosci. Model Dev., 12, 1209–1225, https://doi.org/10.5194/gmd-12-1209-2019, 2019. a
Lewis, A. C., Lee, J. D., Edwards, P. M., Shaw, M. D., Evans, M. J., Moller, S. J., Smith, K. R., Buckley, J. W., Ellis, M., Gillot, S. R., and White, A.: Evaluating the performance of low cost chemical sensors for air pollution research, Faraday Discuss., 189, 85–103, https://doi.org/10.1039/c5fd00201j, 2016. a, b, c, d
Lewis, A. C., von Schneidermesser, E., and Peltier, R. E.: Low-cost sensors for the measurement of atmospheric composition: overview of topic and future applications, Tech. rep., World Meteorological Organization, available at: https://www.ccacoalition.org/en/resources/low-cost-sensors-measurement-atmospheric-composition-overview-topic-and-future (last access: 1 November 2020), 2018. a, b, c, d
Liu, H. Y., Schneider, P., Haugen, R., and Vogt, M.: Performance assessment of a low-cost PM2.5 sensor for a near four-month period in Oslo, Norway, Atmosphere, 10, 41, https://doi.org/10.3390/atmos10020041, 2019. a
Malings, C., Tanzer, R., Hauryliuk, A., Kumar, S. P. N., Zimmerman, N., Kara, L. B., Presto, A. A., and R. Subramanian: Development of a general calibration model and long-term performance evaluation of low-cost sensors for air pollutant gas monitoring, Atmos. Meas. Tech., 12, 903–920, https://doi.org/10.5194/amt-12-903-2019, 2019. a, b, c, d, e, f, g, h, i
Malings, C., Tanzer, R., Hauryliuk, A., Saha, P. K., Robinson, A. L., Presto, A. A., and Subramanian, R.: Fine particle mass monitoring with low-cost sensors: Corrections and long-term performance evaluation, Aerosol Sci. Tech., 54, 160–174, https://doi.org/10.1080/02786826.2019.1623863, 2020. a
Mansfield, L., Nowack, P., Kasoar, M., Everitt, R., Collins, W. J., and Voulgarakis, A.: Can we predict climate change from short-term simulations using machine learning?, npj Climate and Atmospheric Science, 3, 44, https://doi.org/10.1038/s41612-020-00148-5, 2020. a
Masson, N., Piedrahita, R., and Hannigan, M.: Quantification method for electrolytic sensors in long-term monitoring of ambient air quality, Sensors, 15, 27283–27302, https://doi.org/10.3390/s151027283, 2015. a, b, c
Mead, M. I., Popoola, O. A., Stewart, G. B., Landshoff, P., Calleja, M., Hayes, M., Baldovi, J. J., McLeod, M. W., Hodgson, T. F., Dicks, J., Lewis, A., Cohen, J., Baron, R., Saffell, J. R., and Jones, R. L.: The use of electrochemical sensors for monitoring urban air quality in low-cost, high-density networks, Atmos. Environ., 70, 186–203, https://doi.org/10.1016/j.atmosenv.2012.11.060, 2013. a, b, c, d, e, f
Moltchanov, S., Levy, I., Etzion, Y., Lerner, U., Broday, D. M., and Fishbain, B.: On the feasibility of measuring urban air pollution by wireless distributed sensor networks, Sci. Total Environ., 502, 537–547, https://doi.org/10.1016/j.scitotenv.2014.09.059, 2015. a
Munir, S., Mayfield, M., Coca, D., Jubb, S. A., and Osammor, O.: Analysing the performance of low-cost air quality sensors, their drivers, relative benefits and calibration in cities – a case study in Sheffield, Environ. Monit. Assess., 191, 94, https://doi.org/10.1007/s10661-019-7231-8, 2019. a, b
Nowack, P. and Konstantinovskiy, L.: Code in support of Nowack et al. (2021) in Atmospheric Measurement Techniques (Version 2), Zenodo [code], https://doi.org/10.5281/zenodo.5215849, 2021. a
Nowack, P., Braesicke, P., Haigh, J., Abraham, N. L., Pyle, J., and Voulgarakis, A.: Using machine learning to build temperature-based ozone parameterizations for climate sensitivity simulations, Environ. Res. Lett., 13, 104016, https://doi.org/10.1088/1748-9326/aae2be, 2018. a, b, c, d
Nowack, P., Ong, Q. Y. E., Braesicke, P., Haigh, J. D., Luke, A., Pyle, J., and Voulgarakis, A.: Machine learning parameterizations for ozone: climate model transferability, in: Conference Proceedings of the 9th International Workshop on Climate Informatics, 2–4 October 2019, Paris, France, 263–268, 2019. a, b, c
Nowack, P., Runge, J., Eyring, V., and Haigh, J. D.: Causal networks for climate model evaluation and constrained projections, Nat. Commun., 11, 1415, https://doi.org/10.1038/s41467-020-15195-y, 2020. a
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E.: Scikit-learn: Machine learning in Python, J. Mach. Learn. Res., 12, 2825–2830, 2011. a, b, c, d, e
Popoola, O. A., Stewart, G. B., Mead, M. I., and Jones, R. L.: Development of a baseline-temperature correction methodology for electrochemical sensors and its implications for long-term stability, Atmos. Environ., 147, 330–343, https://doi.org/10.1016/j.atmosenv.2016.10.024, 2016. a
Rai, A. C. and Kumar, P.: Summary of air quality sensors and recommendations for application, Ref. Ares, p. 65, available at: https://www.iscapeproject.eu/wp-content/uploads/2017/09/iSCAPE_D1.5_Summary-of-air-quality-sensors-and-recommendations-for-application.pdf (last access: 1 November 2020), 2018. a, b, c, d, e
Runge, J., Heitzig, J., Petoukhov, V., and Kurths, J.: Escaping the curse of dimensionality in estimating multivariate transfer entropy, Phys. Rev. Lett., 108, 258701, https://doi.org/10.1103/PhysRevLett.108.258701, 2012. a
Runge, J., Nowack, P., Kretschmer, M., Flaxman, S., and Sejdinovic, D.: Detecting and quantifying causal associations in large nonlinear time series datasets, Science Advances, 5, eaau4996, https://doi.org/10.1126/sciadv.aau4996, 2019. a
Sadighi, K., Coffey, E., Polidori, A., Feenstra, B., Lv, Q., Henze, D. K., and Hannigan, M.: Intra-urban spatial variability of surface ozone in Riverside, CA: viability and validation of low-cost sensors, Atmos. Meas. Tech., 11, 1777–1792, https://doi.org/10.5194/amt-11-1777-2018, 2018. a
Sayahi, T., Garff, A., Quah, T., Lê, K., Becnel, T., Powell, K. M., Gaillardon, P. E., Butterfield, A. E., and Kelly, K. E.: Long-term calibration models to estimate ozone concentrations with a metal oxide sensor, Environ. Pollut., 267, 115363, https://doi.org/10.1016/j.envpol.2020.115363, 2020. a
Sherwen, T., Chance, R. J., Tinel, L., Ellis, D., Evans, M. J., and Carpenter, L. J.: A machine-learning-based global sea-surface iodide distribution, Earth Syst. Sci. Data, 11, 1239–1262, https://doi.org/10.5194/essd-11-1239-2019, 2019. a
Spinelle, L., Gerboles, M., Villani, M. G., Aleixandre, M., and Bonavitacola, F.: Field calibration of a cluster of low-cost available sensors for air quality monitoring. Part A: Ozone and nitrogen dioxide, Sensor. Actuat. B-Chem., 215, 249–257, https://doi.org/10.1016/j.snb.2015.03.031, 2015. a, b, c
Spinelle, L., Gerboles, M., Villani, M. G., Aleixandre, M., and Bonavitacola, F.: Field calibration of a cluster of low-cost commercially available sensors for air quality monitoring. Part B: NO, CO and CO2, Sensor. Actuat. B-Chem., 238, 706–715, https://doi.org/10.1016/j.snb.2016.07.036, 2017. a
Tanzer, R., Malings, C., Hauryliuk, A., Subramanian, R., and Presto, A. A.: Demonstration of a low-cost multi-pollutant network to quantify intra-urban spatial variations in air pollutant source impacts and to evaluate environmental justice, Int. J. Environ. Res. Pub. He., 16, 2523, https://doi.org/10.3390/ijerph16142523, 2019. a
Vikram, S., Collier-Oxandale, A., Ostertag, M. H., Menarini, M., Chermak, C., Dasgupta, S., Rosing, T., Hannigan, M., and Griswold, W. G.: Evaluating and improving the reliability of gas-phase sensor system calibrations across new locations for ambient measurements and personal exposure monitoring, Atmos. Meas. Tech., 12, 4211–4239, https://doi.org/10.5194/amt-12-4211-2019, 2019.
a
Zimmerman, N., Presto, A. A., Kumar, S. P. N., Gu, J., Hauryliuk, A., Robinson, E. S., Robinson, A. L., and R. Subramanian: A machine learning calibration model using random forests to improve sensor performance for lower-cost air quality monitoring, Atmos. Meas. Tech., 11, 291–313, https://doi.org/10.5194/amt-11-291-2018, 2018. a, b, c, d, e, f, g, h, i, j
Short summary
Machine learning (ML) calibration techniques could be an effective way to improve the performance of low-cost air pollution sensors. Here we provide novel insights from case studies within the urban area of London, UK, where we compared the performance of three ML techniques to calibrate low-cost measurements of NO2 and PM10. In particular, we highlight the key issue of the method-dependent robustness in maintaining calibration skill after transferring sensors to different measurement sites.
Machine learning (ML) calibration techniques could be an effective way to improve the...