7 minute read

Climate change is here. How can AI help us adapt?

Crippling heat waves, devastating floods and uncontrollable wildfires. The repercussions of climate change have arrived.

Climate change is here. How can AI help us adapt?

Crippling heat waves, devastating floods and uncontrollable wildfires. The repercussions of climate change have arrived.

As the United Nation’s Intergovernmental Panel on Climate Change report blatantly states, this is just the beginning. Temperatures will likely continue to rise around 1.5 degrees Celsius over the next 20 years, triggering more frequent sweltering heat waves, stronger storms, higher floods, severe droughts and drastic ecosystem shifts.

We have to continue our efforts to slow down climate change. But we also have to adapt to our rapidly changing environment.

This is why IBM has launched a new Environmental Intelligence Suite (EIS)—to help companies predict, prepare for and adapt to the increasingly severe risks of our warming climate. The technologies now available within the suite also help companies to identify and better understand how their own work impacts nature, as well as steps they can take to start mitigating their own carbon footprint.

This unique integration of AI with environmental, climate, and weather data is aimed at adapting to and mitigating climate change with accurate analysis of climate risk at scale and precise accounting of carbon emissions.

Climate risk and impact analytics

First, there is a new environmental and climate impact modeling framework, CIMF.

It relies on the leading and unique capabilities of IBM PAIRS++, a geospatial data and analytics platform that aggregates hundreds of layers of curated geospatial-temporal data in space and time. PAIRS++ collects, analyzes, and maintains massive amounts of heterogeneous and unstructured data from aerial imagery, maps, IoT infrastructure, drones, LiDAR, and satellites. And it continuously ingests data from major content providers such as The Weather Company, ECMWF, and USGS.

PAIRS++ enables CIMF to predict the risk and potential impact of upcoming climate and weather hazards in a much more efficient, standardized and integrated way than current methods. It addresses the most challenging barriers organizations face today when attempting to model climate risks, including:

  • Opening up access to large-scale computing power through the cloud. Climate risk models require large amounts of compute power to draw in and analyze tremendous volumes of large-scale data. Through the cloud, CIMF opens up and accelerates access to these computing resources that many currently do not have.
  • Streamlining data inefficiencies that are inevitable when collecting all of the unstructured, heterogeneous and disparate data that weather forecasting requires. Using capabilities built within PAIRS++, CIMF easily converts enormous volumes of images, IoT data, LiDAR and other sources into clean and usable information.
  • Standardizing weather forecasting models that are notoriously diverse. CIMF pulls these models into an accessible and easily interpretable framework to show where risks and impact may occur.

Organizations often lack access to specialized AI tools geared towards expediting and improving climate risk and impact analytics, such as weather generators, uncertainty propagation, and model calibrations. By combining the analytical and aggregation strength of PAIRS++ with hyper-local weather data, CIMF forms the critical engine of the Climate Risk Layer within EIS.

CIMF uses AI-driven hazard models to predict evolving climate and weather risks in specific regions, such as flooding, wildfire, heat waves, and more severe storms.

For example, a manufacturing company could use the climate risk analytics EIS to assess where high floods may be a risk in the future, and then decide where to maintain, build and move warehouses. The company could also overlay these future flood forecasting maps on top of current operational data, and determine where they should invest in additional weather-hardening to protect existing investments.

In other areas, CIMF within EIS could help utility companies to prepare for oncoming severe weather events and limit potential power outages and disruptions to service. And insurance companies could use this information to warn policyholders and response teams when severe weather is on the way, and then recommend steps consumers can take to prepare.

With CIMF now available through the Environmental Intelligence Suite, IBM scientists are opening up access to vital technologies that companies can use in the face of more severe climate and weather risks.

Pluvial flood climatological risk: model inputs and outputs.Pluvial flood climatological risk: model inputs and outputs.

Carbon performance

Then there is the harmful carbon dioxide and other potent greenhouse gases like methane and nitrous oxide that are being emitted into our atmosphere in ever-growing amounts causing global warming and its adverse effects.

Following the GHG (Green House Gas) protocols, IBM scientists have designed new AI-enhanced, general-purpose carbon footprint reporting, tracking, and optimization capabilities that help clients account, reduce, and optimize emissions from their business processes and supply chains. For example, IBM’s new carbon footprint APIs use AI and natural language processing algorithms to move carbon accounting and optimization from manual aggregation and measurement processes to an automated method. This helps solve for data quality issues, analyze hotspots to identify “super” emitters, and achieves multi-objective optimization that balances monetary and emission metrics, side by side.

The new carbon accounting APIs help automate data collection for the three different types of emissions defined by the GHG Protocol, namely Scope 1 and 2, as well as expanding capabilities in Scope 3. This information can be used to return emission details and calculations within minutes instead of months. These APIs were designed to:

  • Automatically update evolving reporting standards;
  • Use natural language processing techniques to capture nuances needed to accurately report on emissions, despite the data's country of origin;
  • Easily fold into existing enterprise resource planning (ERP) systems to process updates frequently and easily.

To better understand how such APIs could be used, consider the example of an organization with a large vehicle fleet, such as a shipping and logistics company. With the carbon accounting APIs within the Environmental Intelligence Suite, data could be collected at a granular level to track how much fuel each truck consumes.

Then, carbon accounting APIs could be applied to calculate mobile emissions into carbon equivalence. This individual fleet data would then be aggregated to generate an operational view to capture the fuel consumption of a full trucking fleet.

After this process, the operational fleet data would then be analyzed with other dimensions of carbon emissions accounting to produce an enterprise-wide view. These visualizations can help teams to better interpret source data, how these emissions were calculated and tracked, and where they can act to reduce emissions moving forward.

Beyond the example of emissions from a fleet of vehicles, these carbon performance APIs can also be applied to chemical usage, fuel consumption in industrial plants, energy usage for heating and cooling, process operations, and all types of transportation expanding to more categories.

IBM Environmental Intelligence Suite helps businesses address sustainability and climate risk

We can no longer reverse all of the inevitable effects of climate change before our planet begins to heal. But we can prepare and adapt to minimize infrastructure damage and risk. And we can take crucial steps forward to reduce our own impact as we work towards a more sustainable future, such as accurately measuring the carbon footprints of organizations and identifying ways to reduce them and optimise them across business processes ranging from asset management to infrastructure to supply chains across industries.