June 2024, Vol. 251, No. 6


Geospatial Analytics: Predictions, Trends, and A.I. Applications for Pipeline Management

By Sean Donegan, President and CEO, Satelytics 

Oil and gas companies have hundreds of thousands of miles of pipelines and infrastructure in remote areas that they must survey, to ensure their infrastructure and operations are secure.

A technology known as geospatial analytics has been instrumental in helping these companies locate events across these vast geographies in a timely manner, ranging from potential leaks to encroachment from buildings and vegetation to spotting chemicals on land and in water. 

Geospatial analytics is an efficient and cost-effective technology for oil and gas companies, which allows for multispectral and hyperspectral imagery, gathered from satellites, UAVs, planes and fixed cameras, to be analyzed in a process that ultimately can provide both alerts and qualitative results for oil and gas companies. 

Geospatial analytics allows customers to constantly monitor their infrastructure and then render images of results into interactive displays, alerts and visualizations that provide critical information quickly after data capture, facilitating customer action. 

Over the past few years, the use of artificial intelligence (A.I.) and specific algorithms have moved the technology of geospatial analytics to new heights — helping organizations keep their infrastructure, the surrounding communities and the environment safe. 

As technology advances, so do its use cases. 

Pressure to eliminate methane emissions will increase in 2024. Today’s oil and gas and energy companies need more sound data to locate small leaks that could escalate. 

Energy and oil and gas companies are facing increased pressure to limit their methane emissions, and, as such, these organizations are responding to demands from investors who prioritize environmental responsibility alongside profit. 

Additionally, insurance and investment organizations face increased pressure from shareholders to hold energy companies — such as those in oil and gas — accountable for the perceived lack of performance on environmental issues. 

While many energy companies are working to shore up their infrastructure, to prevent such events as methane leaks that can harm the environment and the communities where their customers reside, newer approaches are required to locate smaller and harder-to-detect issues. 

In 2024, organizations must ensure their infrastructure is sound, which means finding harder-to-detect issues. However, with infrastructure spread over thousands of miles — often in remote locations — the ability to pinpoint an issue or leak before it becomes a major event is extremely challenging. 

A.I.-powered geospatial analytics allows for the analysis of terabytes of data, producing actionable alerts that help a company pinpoint potential issues and allowing the company to limit negative consequences. A.I. offers the ability to analyze data quickly, learn from misidentified threats and provide accurate early warnings for energy companies. 

For example, methane measurement algorithms detect and measure plume concentrations and flow rates, in addition to indicating the leak’s source, all with accuracy. 

Early detection and alerts — with specifics about location and severity — minimize risk, ease environmental impacts, avoid escalating costs, and lessen the toll of public exposure. Today’s technology can direct a company to the source of the problem with specificity, saving time and money and directing valuable resources to where their expertise is critical to resolution. 

‘Forever Chemicals’ 

“Forever chemicals,” including perfluoroalkyl and polyfluoroalkyl substances (PFAS), and their effects on human and fauna health will be reported with increasing frequency as more is learned about how widespread they are in our environment and their specific negative effects on humans and other organisms. 

Consumers can find trace amounts of these chemicals in drinking water, beauty and home products and more, including the linings of fast-food boxes, non-stick cookware and fire-fighting foams. 

The persistence of forever chemicals in the environment and their prevalence countrywide makes them a unique water quality concern. High concentrations of PFAS may lead to adverse health risks in people, according to the U.S. Environmental Protection Agency (EPA). 

A recent U.S. Geological Survey found at least 45% of the nation’s tap water is estimated to have one or more types of PFAS. As a result, the EPA released its proposed rule seeking to set the first enforceable national drinking water standards for PFAS. 

In 2024, the onus falls on the industries and organizations that have released these chemicals to the environment to have better tools to detect the presence of forever chemicals. 

Organizations need to be able to survey waterways and land for the presence of forever chemicals. While an environmental scientist can go out and survey parts of a waterway for these forever chemicals with a handful of test tubes, science teams are not getting the whole picture, as these tests will not fully determine the concentration of PFAS, its extent or its fate and transport in the water or on soil. 

Only recently have A.I.-powered algorithms been developed to analyze satellite imagery and accurately pinpoint traces of PFAS in soil and water. When organizations have thousands of miles of waterways or land to test, a geospatial analytics approach with A.I. is necessary. 

Negative Attention 

Another major pipeline leak will again bring negative attention to an already-constrained pipeline industry, resulting in increased federal and state regulatory oversight. A.I.-powered geospatial analytics can stem the tide. 

Given the prevalence of oil and pipeline leaks in the past two years — wreaking havoc on the environment and communities and costing the companies operating these assets hundreds of millions of dollars — 2024 promises more pressure than ever from regulatory agencies on oil and gas and utility companies. 

The Keystone pipeline, which spans 2,600 miles, leaked an estimated 14,000 barrels of oil — more than half a million gallons — into a creek in Washington County, Kansas, on Dec. 7, 2022. Many pipeline leaks start small. The Keystone oil pipeline leak was found to be primarily due to a progressive fatigue crack that originated during the construction of the pipeline. 

One of the major issues is that organizations have pipelines spread across remote areas spanning hundreds and thousands of miles, making it difficult to regularly monitor the infrastructure with traditional leak detection technologies, such as SCADA systems, as these technologies are not optimized to identify smaller leaks. 

Unfortunately, pipeline leaks are sometimes detected first by people working on the land. This aspect adds to the negative press and public opinion toward these midstream companies, making it increasingly difficult for them to accomplish the job of energy transportation for our nation. 

Today’s satellite technology and geospatial analytics, powered by A.I., can be used for the early detection, location and quantification of operational issues within a company’s infrastructure. 

Therefore, it is time for pipeline operators to look to technology to help offset the considerable challenges of having vast infrastructure that cannot always be watched with the human eye. Oil and gas and utility companies need to utilize these technologies to understand the state of their pipelines better. 

The upshot is that finding any event, such as a leak, in its infancy and locating the source and extent of the risks is critical in minimizing costly consequences. With A.I., geospatial analytics offers the ability to analyze petabytes of data, comprising thousands of individual aerial or satellite images, to detect events such as potential leaks. 

Early detection and alerts, with specifics about location and severity, minimize risk, avoid escalating costs and impacts on the environment and lessen the toll of public exposure. 

Related Articles


{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}