Elías D. Niño-Ruiz, Ph.D.

Short bio

Forecasts are wrong. Observations are noisy. The interesting part is what you do next.

I hold a Ph.D. in Computer Science and Applications from Virginia Tech (2015), where I worked in the Computational Science Laboratory under Prof. Adrian Sandu. Before that I completed a B.Sc. in Systems Engineering, an M.Sc. in Systems Engineering and an M.Sc. in Industrial Engineering at Universidad del Norte, in Barranquilla.

Along the way I spent research stays at Argonne National Laboratory (2013) and Lawrence Livermore National Laboratory (2014), and taught Numerical Methods at Virginia Tech in 2015.

Since 2016 I have been at Universidad del Norte's Department of Computer Science, now as Associate Professor and Chair, and as director of the Applied Math and Computer Science Lab (AML-CS). My group works on ensemble-based data assimilation, covariance matrix estimation, and the numerical optimization that holds both together.

Since 2021 I have served on the Early Career Advisory Board of Control Engineering Practice (Elsevier), and I review regularly for journals across atmospheric science, applied mathematics and computing.

Research lines

Four problems I keep coming back to

Data assimilation

Ensemble Kalman filters, 4D-Var without an adjoint, and non-Gaussian formulations for highly non-linear models.

Covariance estimation

Modified Cholesky decomposition and shrinkage estimators that recover structure from far fewer ensemble members than variables.

Numerical optimization

Line-search and trust-region methods in reduced spaces, derivative-free frameworks, and surrogate models.

Combinatorial optimization

Metaheuristics based on deterministic finite automata, tabu search and simulated annealing, where my research started.

Selected work

Recent and most-cited papers

A dozen out of 70+ items. The full record lives on ORCID and Google Scholar.

2026 PyTEDA-web: A FastAPI platform for interactive data assimilation benchmarking with real-time streaming and persistent experiment trackingSoftwareX, Elsevier.
2025 TEDA: A lightweight Python framework for educational data assimilationSoftwareX 31, 102297.
2025 Improved Rosenbrock method with error estimator and Jacobian approximation using complex stepPérez Rivera, Turizo, Niño-Ruiz & Montoya. Results in Applied Mathematics 27, 100629.
2025 Statistical package for computing precision covariance matrices via modified Cholesky decompositionSoftwareX 30, 102125.
2024 A 4D-EnKF method via a modified Cholesky decomposition and line search optimization for non-linear data assimilationNiño-Ruiz & Díaz-Rodríguez. Atmosphere 15(12), 1412.
2023 A stochastic covariance shrinkage approach in ensemble transform Kalman filteringPopov, Sandu, Niño-Ruiz & Evensen. Tellus A 75(1), 159–171.
2023 Ensemble based methods for leapfrog integration in the SPEEDY primitive-equation dynamics modelNiño-Ruiz, Consuegra Ortega & Lucini. Quarterly Journal of the Royal Meteorological Society 149(751), 573–587.
2021 An efficient ensemble Kalman filter implementation via shrinkage covariance matrix estimation: exploiting prior knowledgeLópez-Restrepo, Niño-Ruiz, Guzmán-Reyes, Yarce, Pinel & Heemink. Computational Geosciences 25(3), 985–1003.
2020 An adjoint-free four-dimensional variational data assimilation method via a modified Cholesky decomposition and an iterative Woodbury matrix formulaNiño-Ruiz, Guzmán-Reyes & Beltrán-Arrieta. Nonlinear Dynamics 99(3), 2441–2457.
2018 An ensemble Kalman filter implementation based on modified Cholesky decomposition for inverse covariance matrix estimationNiño-Ruiz, Sandu & Deng. SIAM Journal on Scientific Computing 40(2), A867–A886.
2015 An efficient implementation of the ensemble Kalman filter based on an iterative Sherman–Morrison formulaNiño-Ruiz, Sandu & Anderson. Statistics and Computing 25(3), 561–577.
2015 Ensemble Kalman filter implementations based on shrinkage covariance matrix estimationNiño-Ruiz & Sandu. Ocean Dynamics 65(11), 1423–1439.

All journal papers · Conference papers · Peer reviews

Works: 72
Citations: 806
h-index: 15
i10-index: 24
Reviews: 90

Source: ORCID and Google Scholar. Update these by hand when they move.

Software

Code you can actually run

Everything below is open source and documented in a SoftwareX paper.

Teaching toolbox · Python

TEDA

An object-oriented toolbox for teaching ensemble-based data assimilation: stochastic EnKF, dual formulations, Cholesky and modified-Cholesky variants, B-localization, with Duffing, Lorenz-63 and Lorenz-96 as test models and built-in error diagnostics.

About GitHub Paper

Benchmarking platform · FastAPI

PyTEDA-web

A web platform for interactive data assimilation benchmarking, with real-time streaming of experiment output and persistent experiment tracking so runs can be compared later.

Paper

Research package · Python

AMLCS-DA

Data assimilation for atmospheric general circulation models. This is the package the lab uses for SPEEDY experiments at operational-ish resolutions.

Paper AML-CS

Statistical package

Modified Cholesky

Precision (inverse covariance) matrix estimation via modified Cholesky decomposition, packaged for reuse outside our own experiments.

Paper

Teaching

A free graduate course on data assimilation

When I moved from combinatorial optimization into data assimilation, the hardest part was that the field sits between computer science, meteorology and applied mathematics, and no single source explained it that way. This course is the one I wish I had: full lecture notes, Python notebooks for every method, and toy models chaotic enough to be interesting.

It is free, it needs no registration, and it is used in the Ph.D. program in Systems Engineering at Universidad del Norte.

Open the course

Course listings: graduate · undergraduate · talks and keynotes

News

What's happening

June 2026
PyTEDA-web is out in SoftwareX. Data assimilation benchmarking in the browser, with live streaming of results and experiments you can come back to. Read the paper.
April 2026
New collaboration outside my usual field: Language development and inequality in early childhood: a study in Caribbean Colombian contexts, in Vulnerable Children and Youth Studies. Read it.
September 2025
TEDA now has its own SoftwareX paper as a lightweight Python framework for educational data assimilation. Paper · Code.
May 2025
The modified Cholesky library I kept promising is finally released: a statistical package for computing precision covariance matrices. Paper.
November 2024
A 4D-EnKF method combining a modified Cholesky decomposition with line-search optimization for non-linear data assimilation, in Atmosphere. Paper.
2022
Joined the Early Career Advisory Board of Control Engineering Practice, Elsevier.
June 2017
Best Workshop Paper Award at ICCS 2017 in Zurich, for A surrogate model based on mixtures of Taylor expansions for trust region based methods.

Field notes

Some good memories

Sep 2019 · First International Workshop on Data Assimilation for Decision Making, Barranquilla

ISDA 2019 at the RIKEN Center, Japan — Jan 2019 · 7th International Symposium on Data Assimilation, RIKEN Center, Japan

IMA program on machine learning and computational methods for geoscience — Oct 2018 · IMA program on Machine Learning and Computational Methods for Geoscience, USA

Mar 2022 · Supporting the national government in Misión de Sabios

Elsewhere

Profiles and contact

enino@uninorte.edu.co ORCID Google Scholar Scopus GitHub AML-CS Lab CV (PDF)

But they who wait for the LORD shall renew their strength; they shall mount up with wings like eagles; they shall run and not be weary; they shall walk and not faint. Isaiah 40:31

Elías D.Niño-Ruiz, Ph.D.