Air pollution is perhaps the greatest environmental risk to our health today. It is contributing to early deaths and is estimated to burden the global economy at $225 billion per year according to some recent assessments.
It is forecasted that 68% of all the global population will live in cities by 2050. As we increasingly move into cities, air quality becomes one of the major priorities to tackle.
|of the global population will live in cities by 2050|
Source: UN Department of Economic and Social Affairs
Where do we start?
The most important need is to be able to provide a positive dialogue with everyone living and working in a city when it comes to the options, decisions and actions taken to bring down air pollution. Air pollution is dangerously high in many of our cities today. The World Health Organization estimates 90% of people worldwide breathe polluted air.
Initiated by the Clean Air Act of 1956, London is now using a network of sensors to track air quality across the capital.
|of people worldwide breathe polluted air|
|Source: World Health Organization|
Achieving any dialogue needs some robust measuring and predictive tools. One has been developed by Siemens – called the City Air Management Tool (CyAM). What makes this valuable and perhaps different to other tools is something I want to touch upon here.
The whole focus of CyAM is based upon collected data for the (harmful) gases to show the degree of air pollution. It is also a forecasting analytical and predictive models that build understanding through its intelligent software to predict the levels of air pollution several days in advance. This can allow for some preventive measures to reduce “pollution peaks” before they happen.
To achieve a level of constant improvement, analytical and predictive technology uses a method called “recurrent neural networks”. It trains reinforced learning to pick up on the latent information (previously not as well understood) about the air pollution-causing factors of traffic, the hot spots, the industries and agricultural pollutants. The consistency of use, over time, allows the program to steadily reduce differences between its forecasts and the actual levels by ‘learning’. This triggers the program behind CyAM to change the weighting of the individual parameters and enable it to become this predictive and dynamic tool, which is needed to understand ‘cause and effect’ over time.
All air quality initiatives rely on air quality sensor data, both historical and real-time, to build an initial database. It needs levels of consistency of sensors at similar quality levels ─ stationary and clearly set up to monitor the same indicators. Having twelve months of historical data begins to build a picture to base predictions upon and then it can take three to four months to train the system (along with the parameters set) to give it reliable results and offer the potential for future predictions.
What does it set out to measure?
Mostly this is carbon-related, focusing on the key pollutant KPI’s of NOx, PM10 and PM2.5 Ozone measures can be included depending on the capturing parameters of the deployed sensors. As the CyAM solution is cloud-based it does have the ability to be modular and expandable to scale out. Its API and software are compatible with different types of sensors and as such, becomes technology-agnostic (plugs into everything). In other words, it can connect the city.
Bringing predictive options into play.
This can become a valuable planning tool used for modelling factors within a city, such as assessing buildings and energy systems. It can help look at their impacts and help in forecasting how changing technologies and infrastructures will alter air quality, by how much and when, based on sets of particular actions, or even by knowing what inactions will mean. It allows planners and city management to understand costs, benefits and impacts far more, which can better validate their assumptions. Having a tool like CyAM offers cities details that many today do not have, in collecting and visualizing the data to offer the level of detail to help map out the future, as well as restructure and question the present.
City Air Management’s design continues to advance. As nations and regions introduce stricter emission guidelines, focusing on particulate matter, photochemical oxidants and ground-level ozone, carbon monoxide, sulphur oxides, nitrogen oxides and lead, it becomes a significant analytical and predictive tool to help in this management of air quality. It is evolving as air quality increasingly becomes one major issue for all cities to manage.
Where’s the value and to whom?
We come back to all those who live and work in the city. The need is to be aware, to build trust and this positive dialogue on essential aspects of city life. Air quality is one of the most important to understand.
Becoming aware that different investment decisions are based on different technology and their sustainability impact on GHG and NOx emissions becomes increasingly relevant to all involved in city life. As cities make transportation decisions the need is to focus on reducing fossil fuels used by internal combustion engines, and the shift to electrification needs moving towards its generation in greener, more efficient ways.
As cities move towards a cleaner underlying energy mix coming from renewable sources, establishing more energy-efficient buildings and transformation and working towards a real modal shift in its transportation, all have direct impacts on air quality indicators. Each city needs to build an inventory of its emissions and to plan its future one. Having a tool like the Siemens CyAM helps support and inform decisions. It can also aid the dialogues needed within cities between planners, investors and people that live and work within the city.
The decision for sharing information is always with the city, it has that responsibility. The CyAM model provides information to the city and the city determines how to translate this and communicate it. The normal way for data to be provided is through dashboards to key users, not via text alerts. This dashboard can deliver a comprehensive, rich set of options within a visualization dashboard and can then allow “interrogation” or the basis for “what if’s”. The city takes the action not the software. The city creates a dialogue environment and leads on the changes needed based on the data they have available.
This dynamic tool can help inform city decision-makers on the impact of their decisions and aid them in knowing what, where and when has a changing impact on air quality.