Artificial intelligence (AI) has emerged as a potent ally in the fight against climate change and in tackling sustainability strategies. Researchers at Colorado State University are harnessing the power of AI to unravel the complexities of climate, weather, and land use.
This article delves into three key aspects of CSU's groundbreaking AI research, highlighting how it's paving the way for more accurate climate predictions and sustainable solutions.
Atmospheric Science Professor Elizabeth Barnes and her team employ machine learning to decipher the intricacies of climate science. The climate system, with its vast and intricate datasets, demands a tool capable of handling complexity, and machine learning fits the bill.
Machine learning allows researchers to delve deeper into climate relationships, providing a new level of understanding. Barnes emphasizes the importance of explainable AI, where understanding how AI reaches conclusions drives new insights into climate science.
Barnes and her team are committed to building interpretable AI models from scratch, ensuring transparency at every step. Such an approach, although challenging, promises a deeper trust in AI's predictions.
CSU's machine learning model, CSU-MLP (Colorado State University-Machine Learning Probabilities), has significantly bolstered forecasters' confidence in predicting severe weather. Developed over six years by a team led by Professor Russ Schumacher, this model can accurately predict excessive rainfall, hail, and tornadoes four to eight days in advance.
The CSU code now operates within the National Weather Service's operational computer systems, aiding forecasters in predicting hazardous weather events. Moreover, CSU researchers are working on making CSU-MLP forecasts more transparent to forecasters, further enhancing their utility.
Schumacher underscores that while AI holds immense promise for weather prediction, it should complement traditional methods, with each approach leveraging its strengths.
CSU, in collaboration with other universities, is part of the AI-CLIMATE Institute, aiming to develop climate-smart agriculture and forestry practices. With a $20 million grant from the National Science Foundation and the USDA National Institute of Food and Agriculture, the institute leverages AI techniques like deep learning and knowledge-guided machine learning.
CSU's expertise in measuring soil carbon stock changes and greenhouse gas emissions will be integrated with AI models. Led by University Distinguished Professor Keith Paustian, the AI-CLIMATE team seeks to optimize land use for carbon sequestration while considering biodiversity, water quality, and economic factors.
Paustian highlights AI's strength in tackling complex optimization problems, offering better decision-making tools. The resulting AI-guided model will assign likelihoods to various outcomes, empowering stakeholders to make informed choices.
AI is reshaping the landscape of climate science and sustainability work. CSU's research demonstrates the potential of AI to predict storms, unravel climate complexities, and optimize land use for climate mitigation. As AI continues to evolve, it promises to be an indispensable tool in addressing the challenges posed by climate change, which can help offer innovative ways of addressing sustainability initiatives as well.
