Timezone: »

 
Controllable Generation for Climate Modeling
Moulik Choraria · Daniela Szwarcman · Bianca Zadrozny · Campbell Watson · Lav Varshney

@
in Tackling Climate Change with Machine Learning »
Event URL: https://www.climatechange.ai/papers/neurips2022/61 »

Recent years have seen increased interest in modeling future climate trends, especially from the point of view of accurately predicting, understanding and mitigating downstream impacts. For instance, current state-of-the-art process-based agriculture models rely on high-resolution climate data during the growing season for accurate estimation of crop yields. However, high-resolution climate data for future climates is unavailable and needs to be simulated, and that too for multiple possible climate scenarios, which becomes prohibitively expensive via traditional methods. Meanwhile, deep generative models leveraging the expressivity of neural networks have shown immense promise in modeling distributions in high dimensions. Here, we cast the problem of simulation of climate scenarios in a generative modeling framework. Specifically, we leverage the GAN (Generative Adversarial Network) framework for simulating synthetic climate scenarios. We condition the model by quantifying the degree of ``extremeness" of the observed sample, which allows us to sample from different parts of the distribution. We demonstrate the efficacy of the proposed method on the CHIRPS precipitation dataset.

[ Topia] 

Author Information

Moulik Choraria (University of Illinois at Urbana-Champaign)
Daniela Szwarcman (IBM-Research Brazil)
Bianca Zadrozny (IBM Research)
Campbell Watson (IBM Research)

I'm an atmospheric scientist at IBM Research where my research spans climate, weather and water. I was a postdoc at Yale University with Prof. Ron Smith, and completed a PhD at the University of Melbourne with Prof. Todd Lane. Currently leading AI for Climate initiatives with the Future of Climate at IBM Research.

Lav Varshney (Salesforce Research)

More from the Same Authors