Re-imagining the purpose of conferences in a time of isolation

It is now trite to say that the COVID-19 pandemic has impacted many aspects of routine life, from our personal to our professional realities. Every part of academic life has been touched by the pandemic, reducing all aspects of our research endeavours to virtual platforms, from coursework and student mentoring to faculty meetings and conferences.  Zooming in and out of meetings has become the norm for all of us. While there are obvious restrictions to life on an e-platform, I see an opportunity for us to use it to our advantage, to increase the impact and extent of how we communicate our science.  
 

The obligatory Zoom lab meeting screen capture.

I’ve been involved with a number of e-activities including giving departmental seminars, giving a conference talk and helping to organize a weekly on-line seminar series (Ecology Live). These experiences have led me to think quite a bit about new opportunities for giving talks and sharing ideas and findings.

One obvious casualty of COVID-19 restrictions has been conferences -large gatherings are simply untenable even if some regions are starting to reopen some activities. Some conferences were simply canceled outright early on while others have switched to online formats. These e-conferences seem like the best-case scenario allowing for scientists to share their findings while observing gathering and travel restrictions. I gave a talk in an organized oral session in an e-conference and have been contemplating signing up for another.  But I have mixed feelings. Let me be clear, the decision to move to e-format is the best decision for these conferences that have had to respond to these unprecedented changes, but moving forward are there other ways to facilitate interaction and communication? To me, the answer is yes.

The cons of the e-conference:

1. Spontaneous conversations

I don’t think fitting a traditional conference into an e-format works all that well. The point of a conference, to me, is more about the random meetings and discussions with friends and collaborators, rather than the extensive back-to-back talks, for which I have a rather low limit that I can actively listen to.  These sporadic encounters, which often amount to valuable research outputs and collaborations, are lost in the e-conference.

A mock debate at the last conference I attended before the pandemic

2. Child-care

Physical conferences have become better about providing childcare options for attendees. But, with e-conferences, the parents stuck at home with children might not have childcare options, making it difficult to attend whole sessions, and remain fully focused on the science. Added to this, is that the e-conference format with multiple concurrent sessions over the whole day is not that convenient for people at home, even without children.

3. Fees and funding

In my experience thus far, e-conferences appear to still be limiting attendance with still rather steep paywalls. The one I spoke in still had substantial fees to attend even though they were using what appeared to be university Zoom accounts. I totally understand that there are expenses, but the hefty fees limit an amazing opportunity to reach a broader audience.

Related to this, conferences traditionally are quite exclusive. Fees, travel, housing, visas and immigration all exclude people from different parts of the world, especially those who don’t have access to the same level of funding as researchers in North America and Europe. E-conferences can change this, but they have yet to. More than this, many of us are accustomed to being at institutions that bring in weekly seminar speakers, and again, people in other parts of the world have no opportunity to access these.

A route forward?

1. Seminars open for all

Working with the British Ecological Society to bring the weekly Ecology Live seminar series has been an incredible experience. Firstly, the BES has been amazingly supportive of this idea and helping to make it work. More than 3000 people have registered for this series, and from all over the world. The response from people has been phenomenal.

The lesson I take from this experience is that there is a demand for high-quality talks and that there are numerous colleagues from the global south who jump at the opportunity to hear from cutting-edge researchers. Many of these people are excluded from traditional, and likely online, conferences. If we are moving to an online format, accessibility and inclusion should be a motivating factor.

Obviously, there are expenses with delivering online content, but costs can be covered in other ways. Traditional conferences have sponsors and companies advertising their products in the main halls. These groups can still be engaged and in fact access to online audiences around the world and in permanent on-demand formats could be quite attractive to sponsors. We’ve now started including advertisements on the opening and closing slides of Ecology Live to keep the webinars free to watch.

2. Freed from time restrictions to conference length

Traditionally conferences are restricted to 3-5 days but switching to an online format means that societies are no longer subject to a conference structure. Without time limitations, e-conferences do not have to conform to sessions occurring simultaneously. By spreading talks over time, perhaps grouping by thematic topic areas, researchers would be able to attend far more talks than they would normally be able to in a traditional conference setting. I’d watch four 15-min talks on a specific area every couple of weeks.

3. A permanent record accessible by all, always

Many ecologists are quite overcome by a deluge of webinars, zoom meetings, etc. Taking the time to spend days in an e-conference is a daunting choice. Even if they are unable to watch talks live, conference organizers could make them permanent, searchable, and linkable. We post the Ecology Live talks to YouTube afterward and some of our earlier seminars have been viewed thousands of times. There is a general move towards open and transparent science, and free online talks that are permanently available is another step towards this. 

We live in a world where access to new ideas and hearing about cutting edge research is divided between the have and have-nots. Despite the limitations of COVID-19, given some planning, e-conferences can provide a powerful means to connecting the ecologists across the world, but there might be better ways forward to use these recent moves to on-line formats to better engage diverse audiences in a much more inclusive way.

 

---

Have you attended an e-conference recently, or plan on attending one soon? Let me know your thoughts and opinions down below.

Reclaiming contaminated land through manipulating biodiversity

Contents of this post originally appeared on the Applied Ecologist, but with expanded thoughts here.

Five years ago I spent my sabbatical in China and worked closely with a lab in Guangzhou. While there, I built meaningful collaborations and friendships that have continued to advance the science I'm involved with. While in China, I accompanied my friend, Jin-tian Li to a biodiversity field experiment on contaminated post-mining lands in Hunan province, and our discussions led to the just-published paper (please e-mail me if you want a copy) in the Journal of Applied Ecology, first-authored by a former PhD student in my lab, Pu Jia.

Why do we care about degraded lands?

According to the IPBES report on land degradation, the degradation of productive lands and intact habitats is a major threat to sustainability, biodiversity and ecosystem functioning globally, which reduces the resiliency of ecological and economic systems. In many emerging economy countries, environmentally harmful practices that result in contamination render lands and habitats seriously degraded. In many circumstances, the restoration of contaminated habitats to original conditions is not an option because the capacity for these habitats to harbor intact native ecosystems is greatly compromised. In these cases, we need management options that allow us to reclaim contaminated and degraded lands (Nathanail & Bardos 2005), and preferably ones that increase biodiversity and ecosystem function (Rohr et al. 2016).

The potential role of biodiversity in reclaiming contaminated lands

While the ecological literature on the linkages between biodiversity and ecosystem function is vast and rich (e.g., Tilman, Isbell & Cowles 2014), the application of this field of research to reclaiming contaminated lands has been strangely depauperate, and so there’s little guidance on whether we should be planting diverse plant assemblages on contaminated lands, or if we ought to simply plant the most productive species or those that provide efficient phyto-removal of contaminants. This question is of fundamental importance to places like China, where rapid development and industrialization through the 1970s-1990s resulted in severe contamination of lands near mining and mineral processing facilities (Li et al. 2019), and now with China’s commitment to improving it’s environmental health, biodiversity research has the ability to impact policy and management at a national scale.

Our paper

We evaluated whether more diverse plantings increased reclamation and ecosystem functioning of a mine wasteland in Hunan Province, China, which had been severely contaminated with cadmium and zinc over decades. We sowed plots with 1-16 species and these were selected from the herbaceous species that grew around contaminated sites in the region, and more diverse assemblages produced more biomass and were more stable over time. Further, there was less heavy metal contamination of leaf tissues in the more diverse plantings, which reduces the impact on herbivores.

Importantly though, plant diversity spurred plant-soil feedbacks (PSFs) that appeared to drive the increased ecosystem functioning. Higher plant diversity supported higher soil bacterial and fungal diversity. Importantly, higher plant diversity was accompanied with more soil cellulolytic bacteria that exude enzymes that degrade cellulose and so drive decomposition and nutrient cycling, which are essential components of a functioning ecosystem. 




Furthermore, the multi-species assemblages also performed better because these high diversity treatments harboured fewer soil fungal pathogens (and by extension more beneficial soil fungi). This appeared to be driven by the fact that high plant diversity supported a greater diversity of soil chitinolytic bacteria that produce anti-fungal enzymes that degrade the chitin in cell walls of soil-borne plant-pathogenic fungi.

In the search for efficient ways to reclaim contaminated lands, sowing high-diversity plant assemblages appear to be an effective tool. The key for reclamation is to ensure that soil processes like decomposition and nutrient cycling are able to support a self-sustaining ecosystem, and higher plant diversity can ensure this. The next steps will be to field test this in real reclamation projects and to see this research work its way into best practices.

Li, T., Liu, Y., Lin, S., Liu, Y. & Xie, Y. (2019) Soil pollution management in China: a brief introduction. Sustainability, 11, 556.

Nathanail, C.P. & Bardos, R.P. (2005) Reclamation of contaminated land. John Wiley & Sons.

Rohr, J.R., Farag, A.M., Cadotte, M.W., Clements, W.H., Smith, J.R., Ulrich, C.P. & Woods, R. (2016) Transforming ecosystems: when, where, and how to restore contaminated sites. Integrated Environmental Assessment and Management, 12, 273-283.

Tilman, D., Isbell, F. & Cowles, J.M. (2014) Biodiversity and ecosystem functioning. Annual Review of Ecology, Evolution and Systematics, 45, 471-493.

---

Publication Partners: a COVID-19 publication assistance program in conservation science

Researchers around the world are trying to keep up on work duties and responsibilities while being required to stay at home. For some people this means caring for young children or other family members, devising homeschooling, switching courses to online delivery, scheduling meetings with team members, receiving new duties from superiors, and perhaps worrying about job security. It is natural that these people may feel overwhelmed and that routine tasks, like checking references or proofreading manuscripts, might seem insurmountable.

However, for others, COVID-19 lockdowns have resulted in more time to push projects to completion and clear out backlogs. There is then inequality in the impact of COVID-19 restrictions on individuals.

These COVID-19 impacts on individuals not only have these unequal impacts on mental wellbeing and career trajectories but are on top of the desperate necessity of conservation science to continue. We win by having a greater diversity of experts communicating with one another.

Publication Partners is an attempt to address some of this COVID-19 impact inequality and to ensure that conservation science is still being published by assisting people with their manuscript preparation. This is a match-making service of the conservation community to bring researchers struggling with their current working conditions together with those that feel that have extra capacity and are willing to help others in this difficult time. The partner might be asked for publication advice, to assist with manuscript editing, help sorting and checking references, organizing tasks for revisions or preparing figures.

The idea is that the Publication Partners would normally be contributing less than would be expected for authorship and thus will be listed in the acknowledgments of the resulting paper. Publication Partners will match volunteers with those requesting support.

To volunteer or request a partner, please see this document with contact instrucitons.

As a journal editor, I see this a valuable and much needed assistance strategy. And I’m not alone. Many of the most important conservation journals have signaled their support and welcome submissions using this service. The journals support Publication Partners includes (please note that the list of journals is being updated and so will change over time):

Avocetta - Journal of Ornithology 

Biological Conservation

Conservation Biology

Conservation Evidence

Conservation Letters 

Conservation Physiology

Conservation Science and Practice

Ecological Solutions and Evidence

Endangered Species Research

Functional Ecology 

Global Ecology and Conservation

IBIS– International Journal of Avian Science

Journal of Animal Ecology

Journal of Applied Ecology

Journal of Ecology

Journal of Insect Conservation

Methods in Ecology and Evolution

Mires and Peat

Natural History Sciences 

Oryx

People and Nature 

Restoration Ecology

RIO - Rivista italiana di ornitologia - Research in Ornithology

Tropical Conservation Science

Wildfowl

 *Thanks to Bill Sutherland for sharing his thoughts on this post.

Can skipping the peer-review process be a legitimate way to communicate science?

Science is an approach to inquiry and knowledge production that provides an unsubstitutable approach to evaluating empirical claims. And it is a specific and particular thing. Beyond the experiments and data collection, science must be communicated in order to impact knowledge and inform humanity’s understanding of the world around us and potential solutions to problems of our own making. The gold standard for communicating scientific findings is through peer-review. Peer review is the process by which research articles are assessed by other experts and these reviewers are the gatekeepers that determine if papers should be published and how much revision they require. These reviewers look for flaws in logic, methodology and inference, and ensure that findings are set in their proper context.

While peer-review is a generally robust process, it can be corrupted. There are plenty of examples, including: fraudulent data appearing research articles, corrupted analyses to support personal agendas, journals realizing economic benefits by not adhering to peer-review best practice, or non-scientific agendas creeping into peer-reviewed journals.   

So, peer-review is not perfect, but it is necessary and can always be improved. However, there is another question: is it always needed? Are there legitimate reasons for a scientist to skip the peer-review process?

To me, there could be reasons to skip the peer-review process, but the goals should be clear and we need to acknowledge that conclusions and inferences will always be in doubt. Yet, impacting the scientific understanding of some phenomenon and communicating it to other experts might not be the goal. Like this blog post, for example. There can be other communication objectives that do not necessarily require peer-review.

Here are three non-peer reviewed communication pathways that I’ve personally pursued, and I’m not including blogs and other social media here, because I think they differ in their goals and objectives, but these are communication approaches you might want to consider:

1-    You might want to capture a broader public readership, to tell a story in a way that captures a non-specialist audience. For example, you might want to extend your science to a call for policy or societal change or to draw attention of the public and policymakers to a critical issue. I was recently a co-author on several papers that attempted to do this, for example, one on the need to protect the Tibetan Plateau, and another on the globally uneven distribution of the readership and submissions of applied ecology papers.

2-    You might want to target a specific audience that does not need to access peer-reviewed literature. Especially for agencies and NGOs that need specific guidance and summary of best practice. The grey literature is a rich and diverse set of communication pathways, which is not well captured in journals nor permanently available (something with the British Ecological Society that we’ve been trying to overcome!).

3-    You may desire to publish information or findings that are desperately needed and extremely time-sensitive. I recently decided to skip the typical peer review pipeline to get out analyses showing that governmental responses to COVID19 quickly resulted in significant drops in air pollution, across six different air pollutants for those cities impacted in February. I published the findings in this blog and posted the manuscript to EarthRQiv.

Why would I do this, especially when I am reporting the outcomes of hypothesis tests and data analysis? I did submit the manuscript to Science and it was quickly rejected, and I’m sure legitimate biogeochemists and atmospheric chemists are already submitting better analyses. However, I told myself before submission that if it was rejected, I would immediately go to plan B, which I did. I felt that the need to engage in this conversation and to shine the light on policy decisions that would lead to reduced pollution were too important for me to pursue the lengthy peer-review process, especially one that is not in my area of research. So, my plan B was to post to a preprint server and blog it. My hope is that it will spur more discussion and further analyses.

In some ways, these alternative vehicles for communicating science have been an experiment for me, but I have the luxury to do this given that I now have a mature research program and rather large group. Its is important to evaluate how we value non-peer reviewed material, or more importantly, how you use these to tell the story about your contributions to society and your impact. While we clearly need to distinguish peer-reviewed and non-reviewed material, and that there is no replacing the impact of peer-review, we should view non-peer-reviewed material more positively and as a way for knowledge mobilization and engage other communities in discussion. As scientists, we need to think carefully about when and how to communicate and the value of this communication to both society and to our careers. But certainly, these alternative forms of scientific communication can help make the broader impact statements on grant and tenure applications more compelling.

We are ultimately evaluated primarily on our peer-reviewed science, as it should be, we can better tell our story about our contribution with a complementing minority of other communication types. I would go so far as to say that a scientist who only publishes peer-reviewed articles might be missing important opportunities to share their knowledge and have an impact on societally important issues.

Excluding blog posts and tweets, about 30% of my contributions are not peer-reviewed. If I include blog posts, then I’d guess I’m at about a 1:1 ratio, peer-reviewed to not. But I am at the stage in my career where this is less risky to do. Pursuing alternative communication forms needs to be non-linear, you need more peer-reviewed articles upfront to establish your credibility which then frees you to pursue other intellectual endeavours and modes of communication. But perhaps more importantly, you’ve established that you are knowledgeable and a trusted authority, meaning that your non-peer-reviewed writings have greater impact.

Regardless, many of us got into this business to expand our collective understanding of the world around us or to make the world a better place. Neither of these goals is achievable if we are not communicating to non-scientists.

Early evidence that governmental responses to COVID-19 reduce urban air pollution

There is no doubt that the global spread of COVID-19 represents the defining crisis of the last decade. Governments around the world have scrambled to try to reduce person-to-person spread and deal with pressures on public health infrastructure. Regions with community spread have almost universally faced restrictions on travel, business and social activities. These restrictions are designed to reduce the exponential spread of COVID-19 (that is, to flatten the curve), these restrictions will also have a large number of other economic, social and environmental repercussions. Here, I ask a simple question: Has reductions in economic activity and movement caused by governmental responses to COVID-19 improved air quality in cities? I compare February 2019 and 2020 air quality measures and show that six cities that were impacted early by government restrictions in response to COVID-19 show consistent declines in five of six major air pollutants compared to cities that were impacted later (the text in this post has been modified from Cadotte 2020).

One of the most pernicious and inevitable consequences of urbanization and industrialization is the release of air pollutants. The WorldHealth Organization (WHO) estimates that about 90% of urban residents experience air pollution that exceeds WHO guidelines and that air pollution is responsible for more than four million premature deaths annually (World Health Organization 2018). Air quality is adversely affected by the aerosol release of a number of chemical compounds from agriculture, manufacturing, combustion engines and garbage incineration, and is usually assessed by measuring the atmospheric concentrations of six key pollutants: fine particulate matter (PM_2.5), course particulate matter (PM₁₀), ground-level ozone (O₃), nitrogen dioxide (NO₂), sulfur dioxide (SO₂), and carbon monoxide (CO). These pollutants have a number of serious human health impacts (Table 1). Reducing inputs of these pollutants into urban areas requires a combination of technological advancement and behaviour change that can be stimulated by governmental regulations and incentives.

Table 1: The six commonly measured air pollutants in cities and their human health impacts.

Alterations of human, transport and industrial activity are usually the result of long-term economic and behavioural change and difficult to legislate under normal situations. However, the recent emergence of the global COVID-19 pandemic has had clear epidemiological impacts with, as of March 25, 2020, almost half a million confirmed infections and close to 20,000 deaths (World Health Organization 2020). This pandemic has resulted in emergency measures attempting to reduce transmission rates that limit activity, movement and commerce in jurisdictions around the world. While these emergency measures are critically important to limit the spread and impact of the coronavirus, they also provide a glimpse into how governmental calls for behavioural change can alter air pollution levels in cities.

Early evidence reveals that pollution levels have dropped in places that have undergone COVID-19 shutdowns. As Marshall Burke showed in a blog post,  PM_2.5 and PM₁₀, levels are lower than expected in parts of China. Here I examine January and February 2020 AQI levels for the six pollutants in Wuhan to what would be expected under normal circumstances. I further compare the change in February air pollution levels over the past two years in six cities that instituted emergency measures by the end of February (early impacted cities) to 11 cities that did not declare states of emergency until March (later impacted cities) using freely available air monitoring data (World Air Quality Index Project 2020) -see Table 2 for a list of cities.

Table 2: The eleven cities used in this analysis, the month that emergency measures were enacted and two- to six-year AQI averages of the pollutants

City-data come from monitoring agencies listed at the end of this post

Wuhan, China was the epicenter for the December 2019 emergence and the first person-to-person spread of the novel coronavirus.  In response, authorities initiated a series drastic measures limiting human movement and activity in Wuhan and large parts of Hubei province by the end of January. Three air pollutants: PM_2.5, PM₁₀ and NO₂ all showed substantial January and February declines in Air Quality Index (AQI) (U.S.Environmental Protection Agency 2014) values over 2019 levels for those months and what would be expected from long-term trends (Fig. 1). These long-term declining air pollution trends do reveal that China’s recentpollution reduction and mitigation efforts are steadily paying off, but the government-enforced restrictions further reduced pollution levels. The expected air pollution values predicted by temporal trends (red dashed lines in Fig. 1) are all substantially higher than the observed levels, with observed values being between 13.85% lower than expected for January PM_2.5 and 33.93% lower for January NO₂. Further, the reductions in the pollutants shown in Fig. 1 increased the number of days where pollutant concentrations were categorized as ‘good’ (0 < AQI < 50) or ‘moderate’ (51 < AQI < 100) according to the AQI. The three other pollutants: SO₂, O₃ and CO, all showed idiosyncratic or non-significant changes, mostly because their levels have already reduced significantly over time or appear quite variable (Fig. 2). 

Fig. 1. Temporal patterns of Air Quality Index (AQI) PM_2.5, PM₁₀ and NO₂ values in Wuhan, China. Both January and February, 2020 values show significant declines compared to 2019 levels and to that predicted from long-term trends (red dashed line).

Fig. 2. Temporal patterns of Air Quality Index (AQI) SO₂, O₃ and CO values in Wuhan, China.

Once COVID-19 moved to other jurisdictions, and confirmations of community spread emerged in February 2020, emergency measures, like those in Hubei province, were instituted to limit human movement and interaction. The cities subjected to February restrictions include, in addition to Wuhan, Hong Kong, Kyoto, Milan, Seoul and Shanghai, and the AQI values from these cities were compared to other cities that did not see the impacts of the novel coronavirus or have emergency restrictions in place until well into March. Log-response ratios between the air concentrations of pollutants observed in February 2020 to those from February 2019 reveal that all air pollutants except O₃ show a decline in the 2020 values for the early impacted cities (Fig. 3). For later impacted cities, there is no overall trend in changes in the concentrations of pollutants between 2020 and 2019 and the individual cities in this group showed less consistency in the differences between years (Fig. 3). 

Fig. 3. Log response ratios for Air Quality Index (AQI) PM_2.5, PM₁₀, NO₂, O₃, SO₂ and CO values between February 2019 and February 2020 values. Negative values indicate a decline in 2020. The green symbols indicate values from an assortment of cities that did not have emergency measures in place until March, 2020 (later impacted cities) and orange symbols are for cities that were impacted by the end of February.

These results indicate consistent air pollution reduction in cities impacted early by the spread of the novel coronavirus. However, the analyses presented here require further investigation as governments increasingly restrict activity world-wide, and some are discussing the possibility of prematurely lifting restrictions in order to spur economic growth. Further, the data analyzed here present point estimates of air quality but air pollution impacts are not homogeneous through urban landscapes and is influenced by spatial variation in industrial activities and transportation (Adams & Kanaroglou 2016). Thus, as higher resolution spatial air pollution data become available, it would be valuable to see how reduced activity affects air quality in different parts of cities.

This analysis of early data indicates that governmental policies that directly reduce human activity, commercial demand and transportation can effectively and quickly reduce urban air pollution. While the COVID-19 pandemic represents a serious risk for health and wellbeing of populations globally, especially those living in high density urban areas, the impacts of air pollution are equally consequential. If governments are willing to expend trillions of dollars in direct funding and indirect economic costs to combat this disease, then why do these same governments permit or even subsidize activities that emit air pollution? Maybe the lessons learned with COVID-19 can serve as the impetus for further action. Perhaps mandating changes to economic or transportation activity or investing in clean technology would better protect human health from the effects of air pollution.

Cited sources

Adams, M.D. & Kanaroglou, P.S. (2016) Mapping real-time air pollution health risk for environmental management: Combining mobile and stationary air pollution monitoring with neural network models. Journal of environmental management, 168, 133-141.
Cadotte, M. W. (2020) Early evidence that COVID-19 government policies reduce urban air pollution. Retrieved from eartharxiv.org/nhgj3

Cesaroni, G., Forastiere, F., Stafoggia, M., Andersen, Z.J., Badaloni, C., Beelen, R., Caracciolo, B., de Faire, U., Erbel, R. & Eriksen, K.T. (2014) Long term exposure to ambient air pollution and incidence of acute coronary events: prospective cohort study and meta-analysis in 11 European cohorts from the ESCAPE Project. Bmj, 348, f7412.

Fann, N., Lamson, A.D., Anenberg, S.C., Wesson, K., Risley, D. &Hubbell, B.J. (2012) Estimating the National Public Health Burden Associated with Exposure to Ambient PM2.5 and Ozone. Risk Analysis, 32, 81-95.

Greenberg, N., Carel, R.S., Derazne, E., Bibi, H., Shpriz, M., Tzur, D. & Portnov, B.A. (2016) Different effects of long-term exposures to SO2 and NO2 air pollutants on asthma severity in young adults. Journal of Toxicology and Environmental Health, Part A, 79, 342-351.

Kampa, M., & E. Castanas. (2008) Human health effects of air pollution. Environmental Pollution, 151, 362-367.

Khaniabadi, Y.O., Goudarzi, G., Daryanoosh, S.M., Borgini, A., Tittarelli, A. & De Marco, A. (2017) Exposure to PM 10, NO 2, and O 3 and impacts on human health. Environmental science and pollution research, 24, 2781-2789.

Raaschou-Nielsen, O., Andersen, Z.J., Beelen, R., Samoli, E., Stafoggia, M., Weinmayr, G., Hoffmann, B., Fischer, P., Nieuwenhuijsen, M.J. & Brunekreef, B. (2013) Air pollution and lung cancer incidence in 17 European cohorts: prospective analyses from the European Study of Cohorts for Air Pollution Effects (ESCAPE). The lancet oncology, 14, 813-822.

U.S. Environmental Protection Agency (2014) AQI: Air Quality Index. Office of Air Quality Planning and Standards, Research Triangle Park, NC.

World Air Quality Index Project (2020) https://waqi.info/.

World Health Organization (2018) Ambient (outdoor) air pollution: https://www.who.int/news-room/fact-sheets/detail/ambient-(outdoor)-air-quality-and-health.

World Health Organization (2020) Coronavirus disease 2019 (COVID-19), Situation Report –65.

City air quality monitoring agencies:

¹ Division of Air Quality Data, Air Quality and Noise Management Bureau, Pollution Control Department, Thailand (http://aqmthai.com).

² Delhi Pollution Control Committee (http://www.dpccairdata.com).

³ Hong Kong Environmental Protection Department (http://www.epd.gov.hk).

⁴BMKG | Badan Meteorologi, Klimatologi dan Geofisika (http://www.bmkg.go.id).

⁵South African Air Quality Information System - SAAQIS (http://saaqis.environment.gov.za).

⁶ Japan Atmospheric Environmental Regional Observation System (http://soramame.taiki.go.jp/).

⁷ UK-AIR, air quality information resource - Defra, UK (http://uk-air.defra.gov.uk).

⁸ South Coast Air Quality Management District (AQMD) (http://www.aqmd.gov/).

⁹ INECC - Instituto Nacional de Ecología y Cambio Climático (http://sinaica.inecc.gob.mx).

¹⁰ Agenzia Regionale per la Protezione dell'Ambiente della Lombardia (http://ita.arpalombardia.it).

¹¹ CETESB - Companhia Ambiental do Estado de São Paulo (http://cetesb.sp.gov.br).

¹² Department of Public Health of the Sarajevo Canton (http://mpz.ks.gov.ba/).

¹³ Air Korea Environment Corporation (http://www.airkorea.or.kr).

¹⁴ Shanghai Environment Monitoring Center (http://sthj.sh.gov.cn).

¹⁵ Israel Ministry of Environmental Protection (http://www.svivaaqm.net).

¹⁶ Air Quality Ontario - the Ontario Ministry of the Environment and Climate Change (http://www.airqualityontario.com/).

¹⁷ Wuhan Environmental Protection Bureau (http://www.whepb.gov.cn/).