Department of
Data Science

Amit Apte is an applied mathematician working on data assimilation and dynamical systems in earth sciences. Before joining IISER Pune, he was a faculty member at the International Centre for Theoretical Sciences (ICTS-TIFR) and the Centre For Applicable Mathematics (TIFR-CAM). He obtained his PhD at the University of Texas at Austin and was a postdoctoral fellow at The Statistical and Applied Mathematical Sciences Institute (SAMSI), UNC-Chapel Hill, and Mathematical Sciences Research Institute (MSRI).

Research

Dynamical systems and data assimilation

Three distinct themes of research in Prof. Amit Apte's group are (i) data assimilation problems, (ii) Hamiltonian dynamics, and most recently, (iii) dynamics of the Indian monsoon. In each of these themes, techniques from dynamical systems, statistics, and probability are used to elucidate physical phenomena, particularly in earth sciences.

Data assimilation (DA) refers to the powerful and versatile methodology for combining partial, noisy observational data of a nonlinear, chaotic, complex systems with its dynamical model, which is generally imperfect and incomplete, to generate estimates of the state of the system and also estimates of the associated uncertainty. Prof. Apte's work on DA is in the following themes: (i) to understand the mathematical and statistical foundations, and the consequent limitations and strengths, of various data assimilation methods; (ii) to search for new methods to overcome these limitations. The group has worked extensively with the problem of Lagrangian data assimilation (LaDA). The focus is on using tools from dynamical systems theory to gain insights that help address the core issues in DA.

Another emerging theme is the work on developing basic dynamical system models as well as data based stochastic models of the Indian summer monsoon, in order to provide mathematical and theoretical underpinning to the description of the monsoon dynamics.

Earlier research on Hamiltonian system focused on nontwist maps, which are area-preserving maps that violate the twist condition. The group studied the reconnection-bifurcation phenomena and also developed a renormalization group framework to describe the breakup of invariant circles.

Selected Publications

A. Mitra, A. Apte, R. Govindarajan, V. Vasan, S. Vadlamani. 2018. Spatio-temporal patterns of daily Indian summer monsoon rainfall. Dynamics and Statistics of the Climate System Vol.3: dzy010. doi:10.1093/climsys/dzy010

M. Bocquet, K.S. Gurumoorthy, A. Apte, A. Carrassi, C. Grudzien, C.K.R.T. Jones. 2017. Degenerate Kalman filter error covariances and their convergence onto the unstable subspace. SIAM/ASA Journal on Uncertainty Quantification, Vol.5: 304-333. doi:10.1137/16M1068712

L. Slivinski, E.T. Spiller, A. Apte, and B. Sandstede. 2015. A hybrid particle-ensemble Kalman filter for Lagrangian data assimilation. Monthly Weather Review Vol.143: 195–211. doi:10.1175/MWRD-14-00051.1

A. Apte, C.K.R.T. Jones. 2013. The impact of nonlinearity in Lagrangian data assimilation. Nonlinear Processes in Geophysics Vol.20: 329–341. doi: 10.5194/npg-20-329-2013

A. Apte, C.K.R.T. Jones, A. M. Stuart. 2008. A Bayesian approach to Lagrangian data assimilation. Tellus A: Dynamic Meteorology and Oceanography Vol.60: 336-347. doi:10.1111/j.1600-0870.2007.00295.x

A. Apte, M. Hairer, A. M. Stuart, J. Voss. 2007. Sampling the posterior an approach to non-Gaussian data assimilation. Physica D: Nonlinear Phenomena Vol.230: 50-64. doi:10.1016/j.physd.2006.06.009

A. Wurm, A. Apte, K. Fuchss, P.J. Morrison. 2005. Meanders and reconnection-collision sequences in the standard nontwist map. Chaos Vol.15: 023108. doi:10.1063/1.1915960