Addressing Climate Change with the Cloud
February 9, 2016 § Leave a comment
Across the world the effects of pollution and climate change are impacting the way people live. From increased traffic to large-scale industrial manufacturing, the tools that have enabled economic growth now create health hazards for the citizens they benefit. Traditionally, combating these harmful elements has required reactive efforts to decrease further pollution by either limiting or quantifying the impact of human actions on air quality. While this has helped the world understand the scope of pollution and its impact on climate change, proactive action has been limited by siloed data compiled based on limited variables and historical patterns. This does not include insights from unstructured data, which makes up 80% of the world’s data. Instead, what if real time analysis of sensors, traffic cameras, weather imaging, and even satellite photography was used to cognitively understand pollution and help users make better decisions?
This is something that is possible with cloud-based technologies with inputs from sensors and nodes that are a part of the internet of things (IoT). By leveraging data from multiple sources on a centralized cloud platform, advanced analysis and modeling can be used to develop patterns that track the impact of pollution in real time. As an example, a smart city may utilize ground sensors, weather forecasts, and satellite imaging to understand the movement of smog across cities. This data could be analyzed by a central platform that is shared across all entities within local governments to enable collaborative decision-making in advance. In this specific scenario, the model may predict that certain parts of the city will experience dangerous levels of air pollutants due to incoming smog and existing normal traffic patterns. The system may then suggest increasing toll rates on the most impacted routes and deploying additional capacity on existing mass transit systems to counterbalance the higher levels of smog in the air.
The key advantages provided by cloud computing models are economies of scale, robust analytic models based on consistent data, and the foundational elements of machine learning and cognitive capabilities. A centralized cloud platform enables organizations to share and model data based on new funding and revenue models optimizing investments. As an example, a pollution forecasting system could be installed by a national government and costs are recouped based on charging back local entities based on usage. The national government could also introduce APIs for their pollution patterns and allow third party organizations to use them for a fee; this could be mapping applications leveraging these patterns to suggest optimal “clean” paths to their users.
Additionally, information is no longer siloed behind different organizational boundaries as it shared on a single common platform. Data quality is also enhanced by non-traditional, unstructured data such as ground sensors and images captured from cameras that are connected to the central platform via the Internet of Things. This transparency ultimately enables more accurate analytic patterns and models and creates cognitive capabilities that understand, reason, and learn from these data sources to suggest insights. A real example of this is IBM’s Green Horizons Initiative (read more here) currently working in cities like Beijing, New Delhi, and Johannesburg. In Beijing, this was accomplished through an “advanced air quality forecasting and decision support system […] able to generate high-resolution 1km-by-1km pollution forecasts 72 hours in advance and pollution trend predictions up to 10 days into the future.”
In closing, while cloud-based systems will not solve challenges related to pollution and client change, they do provide users with the next generation of tools to help find an answer. These systems create a centralized platform for data collection and analysis while allowing advanced capabilities like cognitive analytics to utilize unstructured data. When used together real time insights are generated and shared easily allowing users to make smarter decisions, faster.