Home

Welcome to the Papers wiki!

This is a place to record papers, notes, documentation, etc. that aren't transient in issues.

General DA papers

• A recent review of all DA schemes and advances, Nice figures to get a sense of many techniques and issues!
• A more advanced paper from Alex that is cited above, as an approach to deal with high dimensions.
• Gordon, 1993, Novel approach to nonlinear/non-Gaussian Bayesian state estimation
• Local Ensemble Transform Kalman Filter: An Efficient Scheme for Assimilating Atmospheric Data
• Fearnhead and Künsch, 2018. Particle Filters and Data Assimilation. Annual Review of Statistics and Its Application
• Malleson et al, 2019. Simulating crowds in real time with agent-based modelling and a particle filter
• Verge et al, 2015, On parallel implementation of sequential Monte Carlo methods: the island particle model
• Van Leeuwen, 2009, Particle Filtering in Geophysical Systems
• Van Leeuwen, 2019, Particle filters for high-dimensional geoscience applications: A review
• Jain et al, 2018, Dynamically adaptive data-driven simulation of extreme hydrological flows
  • Paper on tsunami simulation with PDAF
• Yang et all, 2019, Tsunami Wavefield Reconstruction and ForecastingUsing the Ensemble Kalman Filter

Volna-OP

• Reguly et al, 2018, The VOLNA-OP2 tsunami code (version 1.5)
• Dutykh et al, 2011, The VOLNA code for the numerical modelling of tsunami waves

Maeda Code (TDAC)

• Maeda et al, 2015, Successive estimation of a tsunami wavefield without earthquake source data: A data assimilation approach toward real‐time tsunami forecasting
• Gusman et al, 2016, Tsunami data assimilation of Cascadia seafloor pressure gauge records from the 2012 Haida Gwaii earthquake

GPU Implementations of Particle Filter

• Hendeby et al, 2010, GPU implementation of a particle filter (with OpenGL)
  • Main conclusion: Main bottlenecks are random number generation and cumulative sum of the CDF. These can be done by libraries modern GPUs.
  • Possibly an allgather of particles is also required?
  • Time attributed to generation of random numbers is in fact time spent doing data copy because random numbers are generated on CPU.
• Gelencsér-Horváth et al, 2013, Fast, parallel implementation of particle filtering on the GPU architecture (with CUDA)
• Cumulative sums in Cuda, Nvidia GPU Gems, Chapter 39