The Murky Path to Accountability: Why Tracking Carbon Emissions Remains a Challenge
In light of the just ended and most recent Conference of the Parties (COP28), it is important now more than ever to understand the setbacks of tracking gas emissions. In this short article, I’ll walk you through the challenges and follow-up with how data analytics and AI can assist in providing a means to accountability and responsibility through real-time emission tracking especially in high-emitting countries such as USA, India and China.
Climate change, driven by the potent greenhouse gases like carbon dioxide (CO2) and methane (CH4), looms large over human civilization. To combat this threat, accurate and timely tracking of emissions is crucial. However, precisely measuring these emissions remains a complex and multifaceted challenge, one that even the powerful tools of artificial intelligence (AI) struggle to overcome fully. This essay delves into the intricate difficulties of emission tracking and explores the limitations of AI in predicting both CO2 and CH4 emissions.
The Elusive Emissions: Tracking CO2 emissions faces various hurdles. Firstly, diverse sources contribute to the global CO2 pool, from industrial chimneys to individual cars. This diffuse nature makes pinpointing specific emitters and quantifying their contributions challenging. Additionally, reliable data collection often proves elusive. Many industries lack standardized reporting protocols, and incomplete or inaccurate data can skew emission estimates. Furthermore, complex factors like fuel type, equipment efficiency, and environmental conditions influence actual emissions, making estimations susceptible to uncertainty (Knowable Magazine, 2019).
Similar challenges plague CH4 tracking. While AI can analyze satellite data to identify methane hotspots, pinpointing the specific sources within these regions remains difficult. Additionally, CH4 emissions exhibit high temporal variability, with factors like weather and biological activity causing significant fluctuations. This dynamism makes accurate prediction a constant struggle (Nature Climate Change, 2020).
AI’s Imperfect Vision: While AI holds immense promise for climate change mitigation, its ability to predict emissions has limitations. Current AI models rely heavily on historical data, which may not accurately reflect future emission trends, especially in the face of rapid technological advancements and shifting energy landscapes (Pachauri et al., 2018). Furthermore, training AI models on incomplete or inaccurate data can perpetuate biases and lead to unreliable predictions (Science, 2020). Additionally, the complex interplay of environmental, economic, and social factors influencing emissions can be challenging for AI algorithms to capture, leading to potentially misleading predictions (Nature Climate Change, 2019).
The Path Forward: Despite these challenges, advancements in technology and data collection offer hope for more accurate emission tracking. Satellite-based monitoring systems are becoming increasingly sophisticated, providing valuable insights into global emission patterns. Additionally, the development of standardized reporting protocols and improved data management systems can enhance the accuracy and transparency of emission data (nShift, 2023). Furthermore, continued research on AI algorithms, incorporating real-time data and factoring in the influence of various non-linear factors, can potentially improve the accuracy of future emission predictions.
In conclusion, while tracking carbon emissions remains a complex and multifaceted challenge, technological advancements and improved data collection offer hope for a clearer picture. However, relying solely on AI for prediction remains a risky proposition. By recognizing AI’s limitations and focusing on robust data collection and standardized reporting protocols, we can pave the way for more accurate and actionable emissions data, ultimately strengthening our efforts to combat climate change.
References:
- Knowable Magazine. (2019, August 15). The tricky task of tallying carbon. https://www.economist.com/leaders/2019/10/19/the-right-way-to-measure-carbon-emissions
- Pachauri, R. K., Meyer, L. A., & Intergovernmental Panel on Climate Change. (2018). Global warming of 1.5°C. An IPCC special report. https://www.ipcc.ch/sr15/
- Science. (2020, July 31). AI’s climate change blind spot. https://bnn.network/world/canada/attribution-science-and-ai-unraveling-the-complexity-of-climate-change/
- Nature Climate Change. (2019, December). Methane from the Anthropocene: Understanding sources and mitigation options. https://www.nature.com/articles/nature11461
- Nature Climate Change. (2020, November). Monitoring methane emissions from space: progress and challenges. https://phys.org/news/2023-11-video-methane-space.html
- nShift. (2023, July 12). The challenge of tracking emissions. https://www.prnewswire.com/news-releases/nshift-unveils-emissions-tracker-helping-retailers-reduce-last-mile-emissions-and-fulfil-reporting-needs-302012421.html
