Hi, My name is Pankaj. I am a post-doctoral researcher in Aerosol and Reactive Tracer Modelling at Karlsruhe Institute of Technology (KIT) Germany. I am currently working on the development of ICON-SmART (a unified modeling system for both numerical weather forecasts and climate predictions in seamless global to regional model configurations) alongside the acceleration of its various components using Machine Learning. My research interests include Numerical Modeling and Causal Machine Learning (broadly Explainable AI). I develop open source tools for modeling and analysis in python.
During my PhD, I studied the interactions between chemistry and weather/climate using Chemistry Transport Models (CTMs) and Machine Learning as a member of ATMOS lab at IIT Kharagpur.
Open Source Projects
MieAI
A neural network based emulation of Mie scattering theory for calculating optical properties of internally mixed aerosols in atmospheric models.
Go to Project MieAI-Fortran Nature arXivbias-correction
A python library to perform bias correction of datasets in order to remove cross-dataset biases. It supports various popular parametric and non-parametric approaches.
Go to ProjectHyTraj
A python package for trajectory modeling and analysis using HySPLIT. It supports receptor modeling for pollution sources analysis alongwith trajectory mining using various machine learning techniques.
Go to Project
pyvortex
A python package for polar vortex analysis. It includes functions to calculate equivalent latitude and edge of polar vortex using Nash criteria (important for polar ozone hole studies).
Go to Project
News Highlights
_logo.png){kind=logo}
Paper Highlights
-
Pankaj Kumar, Heike Vogel, Julia Bruckert, Lisa Janina Muth, and Gholam Ali Hoshyaripour: MieAI: A neural network for calculating optical properties of internally mixed aerosol in atmospheric models, npj Climate and Atmospheric Science, 2024 [link].
-
Pankaj Kumar, Jayanarayanan Kuttippurath and Adway Mitra: Causal discovery of drivers of surface ozone variability in Antarctica using a deep learning algorithm, RSC Environmental Science: Processes & Impacts, 2022 [paper, summary].
-
Pankaj Kumar, Jayanarayanan Kuttippurath, Peter von der Gathen, Irina Petropavlovskikh, Bryan Johnson, Audra McClure-Begley, Paolo Cristofanelli, Paolo Bonasoni, Maria Elena Barlasina, and Ricardo Sánchez: The increasing surface and tropospheric ozone in Antarctica and their possible drivers, Environmental Science & Technology, 2021 [paper, summary]
-
J. Kuttippurath, P. Kumar, P. J. Nair, and P. C. Pandey: Emergence of ozone recovery evidenced by reduction in the occurrence of Antarctic ozone loss saturation, npj Climate and Atmospheric Science, 2018 [paper, summary].
