Zeguo Zhang

Deep learning methods on the application of climate field reconstruction

My research mainly focus on the climate field reconstruction/predictions by using deep learning methods. Global surface temperature reconstructions of past millenniums temperature variability offer insights into climate sensitivity and feedback mechanisms, and also provide a useful source of information about the variability and sensitivity of the global climate system. Systematic instrumental temperature records only extend back to the nineteenth century, our knowledge of temperatures in earlier times therefore mainly relies on indirect recorders of temperature information: temperature‐sensitive proxy records.

We will use the state-of-the-art deep learning methods to make the climatic variables reconstruction by using sparse observations in order to improve our knowledge of temperature variations over the past millenniums and better understanding of global climate change both in the past and future.