Feature Scoping Study: Developing a compact, low-cost DELTA-Mi with flow sensing for low-time resolution measurement of atmospheric gases and aerosols

Tang Y.S1., Peters D2., McGurk C2., Ul-Haq I2. and Emekwuru, N3

1UKCEH, 2STFC-UKRI, 3Coventry University

Each month we highlight one of our Scoping Studies, each funded through our Collaboration Building Workshops. This month we welcome a blog post from Dr Yuk Sim Tang and the team developing a next generation DELTA-Mi sensor that can be deployed in a wider range of settings, particularly in low-resource regions.

Nitrogen pollution, in particular increasing global emissions of ammonia (NH3), is a major driver of biodiversity loss in the 21st century. The reaction of NH3 with atmospheric acid gases also forms secondary inorganic ammonium (NH4+) aerosols (Figure 1) that contribute to fine-mode particulate matter (PM2.5) implicated in adverse health impacts and increased mortality.

Figure 1: Reaction scheme showing the emissions, atmospheric chemistry and fate of NH3, acid gases and secondary inorganic NH4+ aerosols [1].

Understanding chemical species composition is necessary to assess impacts on ecosystems and human health, and to direct and target options to mitigate poor air quality. Worldwide, observation data on the reactive gas and aerosol phase components are sparse. In this context, there is an ongoing need for cost-effective, easy-to-operate, time-integrated atmospheric measurement at sufficient spatial and temporal scales.

The UKCEH DELTA® system [2] is a leading diffusion denuder-filter pack method for speciated measurement of reactive gases and aerosols, implemented in high density air quality networks (e.g. UK [3] and China [4]) and recommended in EMEP Level I monitoring [5]. A study from a European DELTA® network (Figure 2) for example provided evidence for the dominance of NH3 gas and NH4NO3 aerosol in the inorganic atmospheric pollution load.

The DELTA-Mi project aims to upgrade the UKCEH DELTA® system with flow sensing capabilities and telemetry, to achieve miniaturisation and permit remote monitoring of air flow. We will modify and test components from a UAV-ready sensor” developed by STFC RAL Space UAV facility in a previous SAQN scoping study [6]. In parallel, we will also field test compact gas and aerosol sample trains, based on UKCEH Mini-ANnular DEnuders (MANDE), for further miniaturisation.

Figure 2: Atmospheric chemical species composition measured in a European DELTA® network [2].

Other key challenges are in optimising power requirements and to produce a DELTA-Mi with low unit cost, to maximise flexibility in deployment options and affordability, particularly in low-resource countries. Collaboration with Coventry University and their work in Sub-Sahara regions will identify pathways to build infrastructure in air quality monitoring to increase data availability for understanding air quality drivers and impacts in low-resource regions.

Find out more about all our Scoping Studies on our dedicated web pages.


[1] Tang, Y. S., Flechard, C. R., Dämmgen, U., Vidic, S., Djuricic, V., Mitosinkova, M., Uggerud, H. T., Sanz, M. J., Simmons, I., Dragosits, U., Nemitz, E., Twigg, M., van Dijk, N., Fauvel, Y., Sanz, F., Ferm, M., Perrino, C., Catrambone, M., Leaver, D., Braban, C. F., Cape, J. N., Heal, M. R., and Sutton, M. A.: Pan-European rural monitoring network shows dominance of NH3 gas and NH4NO3 aerosol in inorganic atmospheric pollution load, Atmos. Chem. Phys., 21, 875–914, https://doi.org/10.5194/acp-21-875-2021, 2021.

[2] https://www.ceh.ac.uk/services/delta-active-sampler-system

[3] Tang, Y. S., Braban, C. F., Dragosits, U., Simmons, I., Leaver, D., van Dijk, N., Poskitt, J., Thacker, S., Patel, M., Carter, H., Pereira, M. G., Keenan, P. O., Lawlor, A., Conolly, C., Vincent, K., Heal, M. R., and Sutton, M. A.: Acid gases and aerosol measurements in the UK (1999–2015): regional distributions and trends, Atmos. Chem. Phys., 18, 16293–16324, https://doi.org/10.5194/acp-18-16293-2018, 2018.

[4] Xu, W., Wu, Q., Liu, X. et al. Characteristics of ammonia, acid gases, and PM2.5 for three typical land-use types in the North China Plain. Environ Sci Pollut Res 23, 1158–1172 (2016). https://doi.org/10.1007/s11356-015-5648-3

[5] https://unece.org/fileadmin/DAM/env/documents/2009/EB/ge1/ece.eb.air.ge.1.2009.15.e.pdf

[6] Feature Scoping Study: A UAV-ready sensor package for rapid deployment during volcanic crisis (saqn.org)

Feature Scoping Study: A UAV-ready sensor package for rapid deployment during volcanic crisis

Each month we highlight one of our Scoping Studies, each funded through our Collaboration Building Workshops. This month we welcome a blog post from Dr Dan Peters and the team using RAL Space’s expertise in sensor development to expand our knowledge of air quality in volcanic plumes.

Volcanic hazards are varied in nature, and among these, issues relating to volcanic gas and particulates can be investigated from the perspective of air quality research.

Volcanic ash poses a threat on a range of scales, ranging from ashfall in local communities to regional risks to aviation. Volcanic gases pose a further threat to the surrounding area, stemming largely from emissions of sulphur dioxide. For example, “vog” – volcanic smog – is a recurring issue in Hawaii, formed as volcanic SO2 interacts with the surrounding atmosphere to produce sulphuric acid aerosols.

Small eruptions of the kind common at volcanoes such as Stromboli (Italy) and Kilauea (Hawaii) produce plumes reaching up to hundreds of metres in the atmosphere, which drift in the direction of the prevailing wind (Figure 1). Whilst plume dispersion models can predict the direction of travel of plumes, local measurements are needed to constrain the volume of pollutants present and verify the predictions of models.

Figure 1. Plumes of gas and ash during an eruption of Stromboli in September 2019. Image: Jean-François Smekens

Although many active volcanoes have established monitoring networks, a network of ground-based sensors is unlikely to be sufficiently dense to fully monitor the development of a moving plume; and monitoring networks may not be present at all around newly active volcanoes. Additionally, it is desirable to be able to measure pollutants at altitude within the plume and along its dispersal axis, as these may later affect communities further downwind.

To address these challenges Jean-François Smekens [1] joined the SAQN Collaboration Building Workshop, where the consortium expertise in volcanology, UAVs, aerosol and gas sampling were brought together. The consortium are aiming to develop a new instrument for monitoring particulates and sulphur dioxide in volcanic plumes, as well as carbon dioxide, another important measurement in volcanological studies. Our aim is to build an instrument utilising commercial off the shelf components suitable for mounting on a small Unmanned Aerial Vehicle (UAV). We will utilise the STFC RAL Space UAV facility [2], expertise in payload design and operation [3], and STFC’s expertise in aerosol and gas handling [4]. The instrumented UAV will be available to other researchers via STFC for air quality studies.

Whilst a number of published studies have been conducted using UAVs to monitor volcanic emissions, a remaining challenge is to combine unbiased sampling of particulates with simultaneous measurements of gas concentrations. Such co-located measurements are necessary to understand the interactions between gas and particles during transport, and to more accurately model the dispersal of both. To support this aim, our instrument is designed to be mounted on a fixed-wing UAV, which not only enables longer range than multirotor UAVs but also permits a relatively stable airflow across the airfoil. Using modelling and simulations capabilities from the Computational Engineering Group in the Scientific Computing Department [5] at STFC, we aim to configure an instrument for isokinetic sampling to enable us to simultaneously carry out gas measurements and unbiased sampling of particulates.

[1] https://www.earth.ox.ac.uk/people/jean-francois-smekens, http://www.jfsmekens.com/about/

[2] https://www.ralspace.stfc.ac.uk/Pages/Autonomous-Systems-Facilities.aspx

[3] https://www.lboro.ac.uk/departments/aae/staff/cunjia-liu/

[4] https://www.ukspacefacilities.stfc.ac.uk/Pages/RAL-Space-Molecular-Spectroscopy-Facility.aspx

[5] https://www.scd.stfc.ac.uk/Pages/Computational-Engineering.aspx

Feature Scoping Study: On Track of NH3

Cross-validation of satellite and ground-level measurements

Each month we highlight one of our Scoping Studies, each funded through our Collaboration Building Workshops. This month we welcome a blog post from Dr Anna Font (Imperial College London), who is working with STFC colleagues in RAL Space and the Hartree Centre to improve our understanding of ammonia concentrations through satellite and ground-level measurements.

Ammonia (NH3) is mostly emitted to the atmosphere from agricultural and farming activities including the use of fertilizers, manure, cattle and dairy farming, among others. NH3 is an important precursor of fine particles due to its reaction with available acids (i.e. nitric and sulfuric acid) to form ammonium nitrate and ammonium sulphate. Airborne fine particles are a significant human health thread and are associated with cardiovascular and respiratory diseases. Further, gaseous ammonia and ammonium compounds are deposited into the ecosystems damaging sensitive habitats. In the UK, agricultural activities represent more than 80% of the atmospheric ammonium emissions and there are no regulations in place to limit these emissions. The Clean Air Strategy published in 2018 aims to reduce NH3 emissions in the UK making available a code of good agricultural practice (COGAP).

Hotspots of ammonia have been identified by means of satellite measurements from both CrIS and the IASI instruments onboard the Suomi NPP and MetOp satellites, respectively. In Europe these include areas in the UK and neighbouring countries such as France, Belgium and The Netherlands. NH3 emissions from agricultural fields in north-west Europe have been associated with particle episodes which accumulate on a regional scale especially during springtime in south-east England. In the UK, hotspots of NH3 are observed in the intense agricultural regions in southern England. 

In the UK, a network of 85 sites distributed in the country have been measuring ammonia at the surface level since mid-1990s. Despite the network provides a good coverage of the UK land, there is a large heterogeneity of concentrations related to the large variation in emission sources at ground level as it can be seen in Figure 1. Concentrations are reported as monthly means. Recently, on-line concentrations of NH3 are available at two rural locations in the country comprising one agricultural site in Scotland and one in south-east England; and also in urban areas comprising London, Manchester and Birmingham. On-line techniques report high time resolve data of NH3 concentrations, usually at hourly basis.

Figure 1. Monthly concentrations of NH3 gas as measured by the Defra UKEAP National Ammonia Monitoring Network in 2019. Only active samplers (delta) are shown in this graph. Data from UK-air website.

Satellite observations are column-integrated and have larger footprints compared to the ground-level measurements. Little is known if satellite retrievals of NH3 concentrations are representative of ground-level measurements in the UK and how they are related. Despite the mistime and misdistance errors which are introduced by comparing measurements of a very reactive species such as NH3 that are not perfectly collocated in time and space, such comparison of column-integrated and ground-level measurements is still important to validate satellite products.

The SAQN workshop on last autumn 2020 was the perfect platform to establish collaboration links between scientists working with satellite data, in-situ observations and big data specialists. The project is establishing links between STFC RAL space, STFC Hartree Centre and Imperial College London. With this project we are aiming to evaluate the ability of satellite observations of NH3 to reproduce temporal variability of surface air concentrations across the UK to before estimate long-term changes in air pollution to assess the efficacy of air quality policies in recent years. The integration of satellite data and ground-level measurements will help us identify hot-spots of NH3 in the UK and monitor possible changes over time.

Featured Scoping Study: AIREFUNITS

Our guest blog this month details the AIREFUNITS project, a collaboration between Coventry University, University of Cambridge and STFC’s RAL Space.

Attention is increasingly being paid globally to the air quality state in regions where continuous air quality monitoring devices hardly exist. This lack of infrastructure is mainly because the resources to acquire such monitoring equipment are lacking, especially in regions such as sub-Saharan Africa, where over 1.1 billion people are affected.

The data acquired so far from these regions indicate the presence of poor air quality levels and the need to devise intervention measures to improve the air quality and reduce the resultant health and socio-economic consequences. However, due to their paucity of resources, these regions largely lack robust reference air quality monitoring units that are used to underpin the data collected from diverse low- and medium-cost monitoring units that many use in these areas. These reference units are essential as we need confidence in the air quality data collected from these monitoring units to formulate the appropriate intervention policies.

This issue was addressed at the recent SAQN collaboration building workshop, and researchers from Coventry University, the University of Cambridge and STFC RAL Space were funded to investigate the development of a portable reference/calibrating unit for air quality sensor networks.

The AIREFUNITS project will utilise the STFC RAL Space Spectroscopy Group’s capability for developing cost-effective, highly sensitive, reliable, portable laser-based gas sensors.

This study will advance the development of a robust portable reference air quality unit that can be used to calibrate the various monitoring units in these low-resource regions, thus enabling the construction of effective intervention measures to mitigate the impacts of poor air quality.


Dr Nwabueze Emekwuru, Coventry University. Dr Lekan Popoola, University of Cambridge. Dr Thomas Wall, STFC RAL Space.

Featured Scoping Study: Emulating the Chemical mechanism through Machine Learning to speed up the real time Air Quality Prediction (ECheMLAQ)

This month we welcome a guest blog from a team using STFC expertise in Machine Learning to improve Air Quality Forecasting Systems. The scoping study is led by Dr Pushp Raj Tiwari.

Air Quality Forecasting Systems (AQFS) have an integrated chemistry module and simulations through them are computationally very expensive. Here we demonstrate the concept of replacing one module from CTMs with a novel Machine Learning (ML) framework to achieve orders of magnitude speed up in chemistry simulations, which mostly through traditional methods are slow and tend to suffer from numerical instability.

The recent SAQN workshop provided a unique opportunity to collaborate and address this problem. The three experts* from different organisations will be developing and investigating the potential for Machine Learning to reproduce the behaviour of a chemical mechanism, yet with reduced computational expense by working closely with each other.

In the period of January to June 2021, this pilot study aims to:

  • Develop the ML algorithm and test it against the model generated result
  • Replace the chemical mechanism with ML and perform simulations

The ML technique developed and results from this initial study will be used as proof of concept to allow this group to continue developing the next generation Air Quality Forecasting Systems (AQFS). Once the ML based system is sufficiently developed, it will enable scientists and prediction centres to implement it in their AQFS and achieve orders of magnitude speed up in prediction with reduced computational time and cost.

* Dr. Pushp Raj Tiwari, Centre for Atmospheric and Climate Physics Research, University of Hertfordshire; Dr. Vera He, UK Centre for Ecology and Hydrology, Dr. Barry Latter, STFC.

Guest blog: Modular Relaxed Eddy Covariance sensor for Air Quality – MOREC-AQ

Our featured Scoping Study this month is from the team led by Dr Lekan Popoola, examining the relationship between ammonia (NH3) and particulate matter (PM).

The Clean Air Strategy (Defra, 2019) sets out an ambitious, stringent target to cut emission of major air pollutants by 2020 and 2030. A significant air pollution challenge is the shift in the relative importance from a relatively small number of major emission sources to many minor sources (such as intensive agriculture, wood burning from homes and smaller industrial sites). The impact of COVID-19 restrictions and various lockdowns have created changes in mobility behaviour, with increasing importance of residential emissions as many of us work from home. Evaluation of the impact of these emission sources requires evidence-based scientific methods and data.

During the recent SAQN collaboration building workshop, research scientists from STFC-RAL Space (Thomas Wall), Cranfield University (Zaheer Nasar), and the University of Cambridge (Lekan Popoola) were successful in getting funding from SAQN to develop a proof-of-concept for a cost-effective Modular Relaxed Eddy Covariance (MOREC-AQ) measurement approach to fluxes/source characterisation and a miniaturised cost-effective NH3 instrument to incorporate into MOREC-AQ (see schematics below).

The specific objectives include: (1) feasibility studies for a portable high-resolution NH3 sensor; (2) design and characterisation of a prototype MOREC-AQ unit; (3) explore additional funding opportunities to further develop and optimise the prototype MOREC-AQ unit.

Monitoring and quantifying atmospheric emissions and their drivers is important to investigate the interplay between gaseous pollutants and PM, informing and evaluating the impacts of air quality interventions. This proof of concept study will allow scoping out the development of cost-effective, reliable emission monitoring solutions for air quality management, particularly in the context of the NH3/PM relationship.

System schematics for the MOREC-AQ

If you have comments or questions about this project, you can share them on our discussion board.

Guest Blog – Sources, behaviour and mitigation strategies influencing indoor air quality: A pilot study

Each month, we welcome a guest blog post from one of our current Scoping Studies, funded through our recent Collaboration Building Workshop. This month we’re pleased to hear from a team examining different influences on indoor air quality.

The average person in the UK spends more than 90% of their time indoors, and indoor air quality (IAQ) related emissions can contribute significantly to total air pollution exposure. Despite this, relatively few studies focus on IAQ compared to outdoor air quality. 

A recent SAQN workshop brought together experts on building design and ventilation, indoor air quality measurement, occupant behaviour, and computational fluid dynamics (CFD). The 9 experts* from academia and industry will be utilising the domestic energy systems and technologies incubator based at The University of Chester (designed to represent a kitchen space), to characterise IAQ related events using equipment donated by the participating organisations.

In the period of January to June 2021, this pilot study aims to:

  • Define occupant behaviours (e.g., cooking, cleaning, etc) based on the UK Time Use Survey.
  • Carry out physical behaviours in different ventilation conditions, e.g. cooking & cleaning.
  • Measure air quality factors (including particulates and microbes) and relevant environmental factors (e.g., temperature and air movement) before and after the activities occur.
  • Develop a CFD simulation and utilise STFC high performance computing facilities to improve its relationship to actual data. 

The data from this initial study will be used as proof of concept to allow this group to continue developing and validating the CFD simulation against physical results, as well as to facilitate future deployments into real-world indoor environments. Once the CFD simulation is sufficiently developed, it will enable faster exploration of many behavioural activities and environmental settings.

If you have comments or questions about this project, you can share them on our discussion board.

* Dr Vicki Stevenson, Cardiff University; Dr Archit Mehra, University of Chester; Dr Zaheer Nasar, Cranfield University; Dr Stefano Rolfo, STFC; Dr Stephanie Gauthier, University of Southampton; Dr Alejandro Moreno Rangel, Lancaster University; Dr Jo Zhong, Nottingham Trent University; Dr Rob Ferguson, University of Essex; Dr Douglas Booker, NAQTS