Will AI make carbon emissions tradable?

Trading carbon emissions is the El Dorado of climate-change management, a new market that will allow corporations to offset their presumed contribution to global warming and give governments a mechanism to balance economic and environmental interests. But carbon emissions are notoriously hard to measure, and the existing scorekeepers for emissions have run into withering criticism by scientific referees. 

Carbon-emitting businesses are supposed to purchase credits from entities that reduce carbon emissions, including reforestation, reducing emissions from landfills, and man-made removal of carbon dioxide from the atmosphere. Measurement is the main problem. 

The carbon market has a long way to go to catch up with the hype. The global consulting firm McKinsey claims that “demand for carbon credits could increase by a factor of 15 or more by 2030 and by a factor of up to 100 by 2050. Overall, the market for carbon credits could be worth upward of $50 billion in 2030,” and presumably US$350 billion by 2050.

Other estimates project a $22 trillion market by 2050 as artificial intelligence produces more reliable data.

An improvement in measurement techniques could make the difference between a stillborn project and one of the world’s largest markets. One startup claims to have made a radical improvement in measurement accuracy by applying AI to a suite of existing models of carbon emissions. If AI succeeds in creating a globally accepted measurement standard, the carbon trading market could become one of the world’s largest. 

Jizhaojia GCN says it has developed an algorithm that incorporates several of the most widely accepted models of carbon emissions, drawing data from a wide range of industrial enterprises and joint ventures. The firm was founded by the Chinese-American entrepreneur Bruno Wu, a shareholder in Asia Times’ holding company.

Stillborn attempts to trade carbon

The Chicago Mercantile Exchange began trading carbon offset futures two years ago, but investor interest has been minimal. An exchange-traded fund (ETF) that tracks carbon offset futures contracts, KraneShares Global Carbon Offset Strategy, has attracted just $1.2 million in investments – a microscopic amount by ETF standards – and trades just 7,000 shares a day on average. The carbon ETF topped Yahoo Finance’s list of 2023 ETF flops.

The CBL GEO (Global Emissions Offset) futures trade carbon emissions for aviation under the Carbon Offsetting and Reduction Scheme for International Aviation (CORSIA), a small market in which emissions are easy to measure. But only 202 contracts were traded on July 21, out of 16.3 million contracts that changed hands on the Chicago Mercantile Exchange.

Carbon emissions from passenger cars, let alone factories and farms, entail measurement problems far greater than aviation, which involves a limited number of vehicles whose fuel use is easy to calculate. 

Credits for reducing carbon emissions are generated by activities like rainforest reforestation. Verra, a non-profit organization that sets global standards for measuring emissions, has been under fire from investigative journalists and scientific critics.

joint investigation by the UK newspaper The Guardian, the German weekly Die Zeit, and SourceMaterial “has found that, based on analysis of a significant percentage of the projects, more than 90% of their rainforest offset credits – among the most commonly used by companies – are likely to be ‘phantom credits’ and do not represent genuine carbon reductions,” The Guardian reported on January 18. Verra has certified nearly $3 billion of carbon credits.

SourceMaterial wrote that the investigation “raises questions for the organizations that many of the world’s biggest companies, and the consumers who buy their products, rely on to set the standard for effective carbon offsetting – in particular the biggest of them, Verra.”

Verra is one of four carbon offset registries that measure the carbon reduction due to the reforestation of rainforests and allow carbon-emitting businesses to purchase credits from rainforest nations as an offset. 

The trading volume and price of carbon offset credits have both fallen this year in response to challenges to the integrity of the system.

A recent World Bank report notes, “The voluntary carbon market is difficult to measure. The cost of carbon credits varies, particularly for carbon offsets, since the value is linked closely to the perceived quality of the issuing company.

“Third-party validators add a level of control to the process, guaranteeing that each carbon offset actually results from real-world emissions reductions, but even so, there’s often disparities between different types of carbon offsets.”

World Bank added, “A measure of skepticism attends the use of credits in decarbonization. Some observers question whether companies will extensively reduce their emissions if they have the option to offset emissions instead. Companies would benefit from clear guidance on what would constitute an environmentally sound offsetting program as part of an overall push toward net-zero emissions.

“Principles for the use of carbon credits would help ensure that carbon offsetting does not preclude other efforts to mitigate emissions and does result in more carbon reductions than would take place otherwise.”

A World Bank official familiar with the institution’s activity in carbon trading says the Bank offered its own system of carbon emissions measurement as a global standard but failed to persuade constituent governments to adopt it.

Despite the glowing forecasts, carbon trading markets are stagnating or shrinking due to the poor quality of data.

Jizhaojia GCN says it can generate higher-quality carbon emissions data using a proprietary algorithm that links three technologies.

The first is Enabling Satellite-based Crop Analytics At Scale, or ECAAS. This analyzes satellite imagery and remote sensing to create large datasets for agriculture and forestry. ECAAS promises to dramatically reduce the cost of collecting on-the-ground data and generate artificial-intelligence training sets for machine learning.

According to a Jizhaojia GCN release, “The ECAAS platform is compatible with various devices in the fields of energy, transportation, industry and agriculture. Through the application of technology, the speed of market development and the efficiency of data capture and utilization are greatly improved. 

“ECAAS can create a network effect similar to Microsoft or the SWIFT funds transfer system to integrate digital technology into carbon emissions management.”

Second, Jizhaojia GCN applies the EPC (Engineering, Procurement and Construction) system for energy management, deriving its own data from more than 210 gigawatts of clean energy projects now under contract.

Third, Jizhaojia employs cloud-based energy management systems using Virtual Power Plant technology, which aggregates the output of different energy sources, trading electricity in order to maximize efficiency.

Jizhoajia’s AI algorithm integrates data learning sets from all three technologies to assign a digital identity to carbon products, making it possible to settle trades across the full range of emissions products, according to a company release.