Western Sydney University and Eratos Reduce Urban Heat Island Effect by 10°C Using a Scalable, Sustainable AWS-Supported Solution

In cities across the globe, heat islands—urban areas that experience higher temperatures than outlying areas—are increasing energy costs, exacerbating air pollution, and triggering heat-related illness and death. The urban heat island (UHI) effect, caused by buildings, roads, and other infrastructure that absorb and re-emit the sun’s heat, is a major problem for many countries, including Australia where metropolitan regions like Western Sydney experience extreme heatwaves regularly. Dr. Sebastian Pfautsch, professor of urban management and planning at Australia’s Western Sydney University, explains, “More people die from heat in Australia than from all other natural disasters combined. In addition to driving up power usage and increasing the carbon footprint of cities, the UHI effect, together with climate change, is greatly impacting urban communities by costing human lives.”

Dr. Pfautsch and his fellow researchers have been seeking technological solutions to the UHI problem. “We’re trying to find ways to make our cities cooler, and we were interested in using green infrastructure as a natural cooling mechanism,” he says. To support this effort, the university’s researchers sought to integrate disparate data sets, workflows, and machine learning models into a single platform.

Western Sydney University reached out to Eratos, an AWS Partner that holds the AWS Smart City  Competency, which recognizes partners for providing best-in-class smart city solutions. Eratos offers a data and AI enabling platform based on Amazon Web Services (AWS) technologies. “We needed to find a partner that could help us with data management, data storage, and data visualization, and Eratos met these requirements,” recalls Pfautsch.

Collaborating with Western Sydney University, Sydney Olympic Park Authority, Sydney Water, and the New South Wales government, Eratos played a key role in creating Smart Irrigation Management for Parks and Cool Towns (SIMPaCT)—a cloud-based Smart Cities solution that uses embedded smart technologies, a digital twin, and machine learning to manage irrigation and maximize cooling. SIMPaCT uses the Eratos platform to host a digital twin of Bicentennial Park, a 42-hectare parkland within Sydney Olympic Park. The digital twin integrates near real-time observations, forecast data, biophysical modeling, and machine learning to issue smart commands to the park’s irrigation system.

The core digital ecosystem of SIMPaCT—which runs on Amazon Elastic Container Service (Amazon ECS), Amazon Elastic Kubernetes Service (Amazon EKS), and Amazon Elastic File System (Amazon EFS), plus Amazon Simple Storage Service (Amazon S3) for file and data storage—takes in data from over 250 sensors deployed across the parkland. These include soil probes, numerous air temperature sensors that use LoRa (a long-range, low power wireless communication technology), and weather stations. Machine learning algorithms, built and hosted on the Eratos platform, create commands that tell 193 irrigation zones throughout the park when to water. Over time, SIMPaCT learns how best to respond to current and forecasted environmental conditions to maintain optimal soil moisture levels and cooling, with minimal water usage.

In creating SIMPaCT, Eratos also benefited from participating in the AWS Think Big for Small Business (TBSB) program which offers small and/or minority-owned public-sector organizations unique opportunities to grow with AWS. TBSB program gives AWS Partners access to AWS Partner Programs, financial incentives, and additional visibility with customers and AWS teams. Odette Young, chief of staff at Eratos, says, “Through TBSB, we got the chance to improve our marketing collateral, attend three additional high-profile conferences, and arrange high-value meetings that resulted in new strategically important relationships with customers and partners.”