Scientific Programme

Day 1: Monday 22 July 2024

Computational Physics Tools, Data Analysis and Machine Learning I

08h30 - 09h00: Registration and get together

09h00 - 09h40: Welcome to EPIC4 and a short introduction to the Physics Without Frontiers Programme.

09h40 - 10h20: Plenary Talk: Decoding Zebrafish development with AI and Soft Matter Physics (Andrés Morales-Naverrete, Konstanz)

10h20 - 11h00: Coffee Break

11h00 - 13h00: Tutorial 1: Modelling Fluids with Python (Wladimir Banda-Barragán, Yachay Tech)

In this tutorial we will look into modelling fluids (gases) by using numerical techniques in python. Specifically, we will build a python routine to study the evolution of subsonic waves and shocks in 2D meshgrids.

13h00 - 15h00: Lunch Break

15h00 - 17h00: Tutorial 2: Star formation in galaxies using SDSS data (Helga Dénes, Yachay Tech)

Day 2: Tuesday 23 July 2024

Computational Physics Tools, Data Analysis and Machine Learning II

08h30 - 09h10: Outreach talk: La dilatación del tiempo en la relatividad especial de Albert Einstein: conferencia teórico experimental (Nelson Herrera, UIDE)

09h10 - 10h40: UIDE Campus & Laboratory Tour

•⁠ ⁠Tour around the Technical Sciences facilities.

•⁠ ⁠Tour around the Health Sciences Campus

•⁠ ⁠Tour around the Elon Musk Building

•⁠ ⁠Sports Tour

10h40 - 11h00: Coffee Break

11h00 - 13h00: Tutorial 3: Analysis and visualisation of adaptive mesh refinement simulations (Wladimir Banda-Barragán, Yachay Tech)

In this tutorial, which is a follow-up of tutorial 1, we will discuss structured and unstructured meshgrids used in computational fluid dynamics. We will look at techniques and methods used to analyse and visualise several simulation data files, written in both grid types.

13h00 - 15h00: Lunch Break

15h00 - 17h00: Tutorial 4: Introduction to deep learning for bioimage analysis (Andrés Morales-Naverrete, Konstanz)

This hands-on tutorial provides an introductory overview of deep learning techniques tailored for bioimage analysis. Participants will learn fundamental concepts of deep learning, including neural networks, convolutional neural networks (CNNs), and their applications in the analysis of biological images. The tutorial emphasizes practical implementation, guiding attendees through the process of preparing image datasets, training deep learning models, and evaluating their performance. By the end of the session, participants will have gained the skills necessary to apply deep learning methods to their bioimaging projects

Day 3: Wednesday 24 July 2024

Advanced topics on Computational Physics Tools and Machine Learning

08h30 - 10h00: Student contributed talks (7 min + 3 min each)

•⁠ 08h30 - 08h40: Atomistic study of the effect of composition on the structural and thermodynamic properties of binary Al-Pd nanoalloys (Dario Gonzalo Escriba, UNMSM)

Intermetallic nanoparticles (NPs) are of great interest due to the unique properties brought by their composition and structure. This is reflected, for example, in their higher thermal stability and the potential to enhance their catalytic activity by varying the stoichiometry and size of the NPs. In particular, Al-Pd intermetallic NPs have diverse applications, from biomedical applications due to their ultra-low magnetic susceptibility, to applications in renewable energies due to their larger surface area, providing excellent selectivity in catalysis processes. However, this system has been quite controversial due to the complexity of its phase diagram and its limits. For these reasons, the objective of this work is to calculate the thermal and structural properties of Al-Pd NPs for various sizes and concentrations ranging from 10% to 90% Al with a 10% step, using atomistic simulation with the molecular dynamics method, starting from a random distribution of Al and Pd atoms in an NP with an initial FCC structure. To achieve this, the heating and cooling stages of the system were simulated using the LAMMPS package with the interatomic potential of the embedded atom method (EAM). Additionally, for visualization and determination of certain properties, various tools of the OVITO software were used, such as the common neighbor analysis (CNA) technique and the radial distribution function (RDF). The preliminary results obtained show, from the calculation of heat capacity, for each composition, the directly proportional relationship of the melting temperature with the nanoparticle size, which tends toward the macroscopic value. Furthermore, for the AlPd nanoalloy, an ordered intermetallic compound has formed, where each Al atom is surrounded by Pd atoms and vice versa, whose lattice parameter and theoretically calculated periodic structure are in agreement with experimental results. On the other hand, the thermal hysteresis curve for each composition and size is well defined only after reheating, indicating that the optimized structure after cooling is suitable and characteristic of the system. Finally, part of the system characterization includes the analysis and graphs obtained from the Lindemann index data, Mean Square Displacement (MSD), and radius of gyration provided by LAMMPS. As well as the calculation of phonon dispersion, which helped determine the dynamic stability of the phases to be simulated.

•⁠ 08h40 - 08h50: Analysis of the Folding process of a small polypeptide using molecular dynamic simulations (Tatiana Cabrera, EPN)

Biomolecules naturally fold into precise three-dimensional structures under physiological conditions to perform their biological functions. The process of folding is an interdisciplinary problem of great interest because the mechanisms behind it remain unclear. In this study, we investigated this process using molecular dynamics (MD) simulations of the deca-alanine system in water. The simulations were carried out using the software NAMD, followed by visualisation and data analysis with the software VMD and the TCL language. A preliminary simulation was conducted to sample the configuration space between the folded and unfolded states. This data was then analysed using cluster analysis in R, which identified eleven groups that characterise the regions of configurational space visited by deca-alanine during folding. This information was subsequently used to perform umbrella sampling of the folding process, with the representative configuration of each group used to realise MD simulations of sampling. This resulted in a probability distribution of the end-to-end distance accessed by the polypeptide. The distribution was then employed to calculate the potential of mean force (PMF) utilising the weighted histogram analysis method implemented in a C++ program. The calculated PMF is consistent with the spontaneous nature of the folding process and describes a trajectory in configuration space that differs from those previously reported.

•⁠ 08h50 - 09h00: Role of Charging in Organometallic Overlayer On-Surface Self-Assembly (Yordan Solórzano, YT)

Recent developments in microscopic methods such as Scanning Tunneling Microscopy (STM) and CO-functionalized non-contact Atomic Force Microscopy (nc-AFM) provide a direct insight into the chemical composition of technologically relevant self-assembled overlayers formed on the surface. An in-depth comprehension of the intermolecular associations occurring on a surface holds the potential to anticipate the arrangement, alignment, and electrical conductivity characteristics of films. Moreover, it facilitates the customization of novel molecular compositions that are specifically engineered for a desired purpose. In this work, we will combine state-of-the-art STM and nc-AFM measurements with theoretical models based on Density Functional Theory (DFT) and STM/nc-AFM simulations to provide insight into the driving mechanisms behind the formation of different organometallic overlayers.

•⁠ 09h00 - 09h10: Brief analysis of rainfall series measured in Quito (Luis Gualco Centeno, EPN)

•⁠ 09h10 - 09h20: Efectos sísmico-electromagnéticos en la detección de rayos cósmicos (James Peñafiel, USFQ)

El siguiente proyecto se enfoca en evaluar la viabilidad de detectar, a nivel de superficie terrestre, las variaciones en el flujo de partículas secundarias de rayos cósmicos atribuibles a modificaciones del campo geo-electromagnético provocadas por la actividad sísmica. Para ello, se ejecutarán diversos ambientes de simulaciones computacionales y se analizarán los datos obtenidos enfocándose en el estudio de incertidumbres sistemáticas y estadísticas, para determinar la viabilidad de replicar estas mediciones en entornos reales.

•⁠ 09h20 - 09h30: Scalar and Tensor Power Spectrum of the Higgs inflationary model using Numerical Solution (Mariannly Marquez, YT)

In this research project, we aim to analyze the predictions of the Higgs inflationary model by comparing them with observational data from the Planck 2018 satellite. We intend to solve the scalar and tensor perturbations equations using numerical integration approaches. The objective is to evaluate the effectiveness of these numerical integration methods in reproducing the scalar power spectra, the scalar spectral index, and the tensor-to-scalar ratio. We expect to assess the consistency of the Higgs inflationary model with the latest cosmological observations. This study aims to provide a deeper understanding of the model’s viability and contribute to the ongoing research in cosmological inflation theories.

•⁠ 09h30 - 09h40: Some inflationary models under the light of Planck 2018 results (Jordan Zambrano, YT)

In this work we study four well–known inflationary scenarios that are reported by the most recent Planck observations: Natural inflation, Hilltop quartic inflation, Starobinsky inflationary model, and Large field power–law potentials V (ϕ) ∼ ϕ^{p}, considering p = 2/3, 4/3. The analysis is done using both the slow–roll approximation and the numerical solution to the background and perturbation equations. We show that the numerical solution improved the precision of these models with respect to the contour plot r vs. nS, having a lower r in each model compared to the value calculated from the slow–roll approximation.

•⁠ 09h40 - 09h50: Study of acausally displaced vertices from Lee-Wick particle decays with LHC open data (Jonathan Joel Sánchez Jácome, YT)

We investigated the viability of reconstructing and identifying acausally displaced vertices predicted by the Lee-Wick (LW) standard model (LWSM). This theory correct divergence problems in the standard model with the addition of new partner fields with negative metric. The model predicts a phenomenon termed as wrong displaced vertices (WDV) because it seems to defy micro causality. In this work, we studied the decay of Lee-Wick partners of the electron (LW-electrons) emitted from a Z boson decay, presumably the most probable channel. We generated the model with mass values of the LW-electron 200-500 GeV using Feynrules and Madgraph according to LHC Run-1 conditions. The highest calculated cross-section (5.97 fb) corresponds to 200 GeV, with an average flight distance of 2.7e-2 mm—greater than the CMS (Compact Muon Solenoid) tracker system resolution. We found that the signals for mass values of 300, 400, and 500 GeV are less likely to be observed. The immediate decay consists of a pair of highly energetic electrons and a pair of Z-bosons emerging from a pair of wrongly displaced vertices. The detection and reconstruction were simulated in the CMS-software-5.3.32. We characterized the final topology, identifying two energetic electrons and partially identifying four jets emerging from Z boson decay. Other final topologies are being studied. For the first time, we defined the quantity parallelity as the dot product between total momentum of the decaying products and propagation vector of the mother particle. Asymmetries in the parallelity distribution suggest the feasibility of measuring wrong displaced vertices if they are present in experimental data.

•⁠ 09h50 - 10h00: Estudio de la disipación de la energúa durante las medidas en los procesadores cuánticos debido al principio de Landauer (Jorge Cedeño, ESPOCH)

Las aplicaciones de puertas lógicas cuánticas no ideales provocan disipaciones de energía que están acotadas desde abajo por la variación en la entropía de von Neumann debido al principio de Landauer cuántico, así, con el fin de comprender como evoluciona la información cuántica al ser eliminada, el objetivo de esta investigación fue estudiar la disipación de energía al ejecutar medidas en los procesadores cuánticos. La metodología implementada tuvo un enfoque cualitativo, llevando como principal factor de aplicación la mitigación de errores cuánticos, generando dos grupos de circuitos cuánticos al mitigar y no mitigar los resultados. Además, la información contenida en los qubits, denominados sistema y reservorio, fue reconstruida mediante la tomografía de estados cuánticos antes y después de aplicar el protocolo de eliminación. Todos estos circuitos cuánticos fueron ejecutados a través de una simulación computacional utilizando la librería Qiskit de IBM. Mediante esta metodología, el incremento del parámetro de entrelazamiento entre los qubits provocó un aumento en la entropía y una disminución de la información inicial contenida en el sistema, a la vez que posicionó la cantidad de energía disipada al reservorio a niveles inferiores, iguales y superiores al límite de disipación clásica de Landauer. Adicionalmente, al implementar la mitigación se redujo las entropías iniciales, aumentando la variación de entropía, sin embargo, la información del sistema fue trasladada a niveles más bajos de fidelidades, provocando un proceso irreversible de pérdida de información. En este contexto se concluye que el límite clásico de Landauer, en el mundo cuántico, funge como una aparente frontera entre los procesamientos de información reversibles e irreversibles. Por otro lado, aunque la mitigación provocó un acercamiento a la irreversibilidad, su aplicación nos llevó a la definición del defecto de medida, que proporciona una herramienta matemática que caracteriza la calidad de las mediciones en determinados procesadores cuánticos. Link en el Dspace: http://dspace.espoch.edu.ec/handle/123456789/21732

10h00 - 10h40: Plenary Talk: Aplicaciones del aprendizaje automático: algoritmos de clasificación (Paulina Vizcaíno, UIDE)

El aprendizaje automático ha revolucionado numerosos campos al permitir el desarrollo de sistemas que aprenden de los datos y realizan predicciones. Entre las diversas tareas de aprendizaje automático, la clasificación es una de las fundamentales, ya que el objetivo es clasificar los datos en clases predefinidas. Este charla proporciona una visión general de los algoritmos de clasificación más populares, incluidos los árboles de decisión, la regresión logística, los vecinos más cercanos y las redes neuronales.

10h40 - 11h00: Coffee Break

11h00 - 13h00: Tutorial 5: Building Generative AI Apps (Manuel Eugenio Morocho Cayamcela, Yachay Tech)

In this tutorial, we’ll explore some tools and frameworks to help you build your first generative AI application.

13h00 - 15h00: Lunch Break

15h00 - 17h00: Tutorial 6: Introduction to Physics-informed Neural Networks (Andrés Morales-Naverrete, U. Konstanz)

This hands-on tutorial offers an introduction to Physics-Informed Neural Networks (PINNs), a cutting-edge approach that integrates physical laws with deep learning techniques. Participants will explore the fundamental principles of PINNs, including their architecture, training process, and applications in solving partial differential equations (PDEs). The tutorial will cover practical aspects such as formulating physical constraints, implementing PINNs using popular deep learning frameworks, and evaluating their accuracy and efficiency.

18h00 - 19h00: Special Dinner

Day 4: Thursday 25 July 2024

Model based design and Data Analysis

08h30 - 10h30: Tutorial 7: Model based design with Matlab – Simulink (Vladimir Bonilla, UIDE)

Model-based design with MATLAB and Simulink is a powerful approach for developing complex control systems and embedded software. This tutorial provides a guide on how to create, simulate models, enabling rapid prototyping and efficient system verification. Discover the integration of MIL, SIL and PIL testing and code generation to streamline your development process.

10h30 - 11h00: Coffee Break

11h00 - 13h00: Tutorial 8: Programming a neural network from scratch, without using libraries (Benjamín de Zayas, Yachay Tech)

In this session, we will delve into the intricate world of programming a neural network from scratch, bypassing the convenience of pre-built libraries. Our journey will emphasize the mathematical foundations underpinning neural networks, providing a thorough understanding of how these powerful models function at a granular level. We will explore key concepts such as linear transformations, activation functions, forward and backward propagation, and gradient descent. By the end of the class, attendees will have a concrete understanding of the inner workings of neural networks and will be equipped with the skills to implement them using only basic programming constructs. This hands-on approach will demystify the algorithms and techniques that drive modern artificial intelligence, fostering a deeper appreciation for the complexity and elegance of neural network design.

13h00 - 14h30: Lunch Break

14h30 - 15h10: Plenary Talk (Best TICEC 2023 Paper Prize): Explorando el desempeño del aprendizaje profundo en física de altas energías (Daniela Merizalde, USFQ)

Los datos en física de altas energías tienen un volumen inmenso y son complejos. Los algoritmos de aprendizaje automático pueden ofrecer una solución prometedora para el análisis de estos datos. Por tanto, esta investigación evaluó el rendimiento de las redes neuronales convolucionales para clasificar el resultado de las colisiones de partículas de de altas energías utilizando los datos del Portal de Datos Abiertos del CERN.

15h10 - 15h50: Plenary Talk: La ruta de la Ciberseguridad ¿Qué camino puedo seguir? Daniel Tenorio (BSmart, https://www.bsmart.ec)

15h50 - 17h00: Dorothy Coding Challenge

Day 5: Friday 26 July 2024

EPIC 4 Student Presentations

08h30 - 10h00: Student group presentations (7min + 3min per group)

•⁠ 08h30 - 08h40: Group 1

•⁠ 08h40 - 08h50: Group 2

•⁠ 08h50 - 09h00: Group 3

•⁠ 09h00 - 09h10: Group 4

•⁠ 09h10 - 09h20: Group 5

•⁠ 09h20 - 09h30: Group 6

•⁠ 09h30 - 09h40: Group 7

•⁠ 09h40 - 09h50: Group 8

•⁠ 09h50 - 10h00: Discussion

10h00 - 10h40: Plenary Talk: 2D Nanoelectrods for single module junctions (Henrry Osorio, Escuela Politécnica Nacional)

Molecular electronics, which applies individual molecules located between two or three fixed electrodes to construct an electronic device, has the potential to offer novel technological opportunities. Noble metal electrodes, especially gold, have been extensively used to comprehend the charge transport at metal-molecule-metal junctions. However, there are potential drawbacks of using gold electrodes such as non-compatibility with complementary metal-oxide-semiconductor (CMOS) technologies, surface mobility and high price. In this context, the use of non-metallic electrodes has attracted growing attention as a result of the emergence of new properties and to avoid the use of rare, expensive and potentially toxic materials. Nevertheless, recently used non-metallic novel electrodes in this field, graphene or GaAs, need expensive methods for deposition or to provide optimal electrical contact. In this talk, I will describe our efforts to form and characterize non-symmetric molecular junctions in a simple way using scanning probe microscopy-based methods. Electrical properties of graphene-molecule-gold and MoS2-molecule-gold junctions will be discussed. In both cases, simple exfoliation techniques were used to obtain the non-metallic electrode. In summary, this study paves the way for the use of nanoelectrodes based on 2D-materials to construct single molecule junctions and evaluate charge transport at the molecular level.

10h40 - 11h00: Coffee Break

11h00 - 11h40: Plenary Talk: Scattering solutions of the DKP equation with the cusp potential barrier (Clara Rojas, Yachay Tech)

In this work, we solve the Duffin–Kemmer–Petiau (DKP) equation in the presence of the cusp potential for spin–one particles. We derived the scattering solutions in terms of the Whittaker functions. We show that transmission resonances are present.

11h40 - 12h20: Women in STEM Forum

12h20 - 13h00: Dorothy Coding Challenge Prizes, Concluding Remarks & Wrap-up

Good-Bye Lunch (13h00 - 15h00)

About our speakers:

Henrry Osorio, Escuela Politécnica Nacional

BSc in Physics from the Escuela Politécnica Nacional (EPN), Ecuador, in 2010. Master’s Degree in Nanostructured Materials for Nanotechnological Applications from the University of Zaragoza, Spain, in 2012. Ph.D, Cum Laude, in Physical Chemistry from the University of Zaragoza, Spain, in 2015. Since 2016, Assistant professor in the Department of Physics at Escuela Politécnica Nacional and member of the electronic and magnetic materials group (GI-MEM), Department of Physics at Escuela Politécnica Nacional. Current research interest is mainly focused on the study of electrical properties of two-dimensional materials, individual molecules and molecular assemblies using tunnel effect microscopy and atomic force microscopy. Research in collaboration with International Laboratories such as the Institute of Nanoscience and Materials of Aragón (INMA), Spain, and the Department of Chemistry of the University of Liverpool, United Kingdom.

Daniela Merizalde, Universidad San Francisco de Quito

Me gradué de física en la USFQ. Actualmente estoy estudiando el másterado en física de altas energía dentro de la USQ. Me interesa el campo de relatividad general y averiguar cómo los modelos de machine learning se pueden aplicar a la física.

Benjamín de Zayas, Universidad Yachay Tech

Benjamin de Zayas is a nuclear physicist with experience in medical physics. He holds a master’s degree in mathematics, where he researched Clifford algebras and quantum field theory. Currently, he is pursuing a master’s in artificial intelligence and a PhD in applied mathematics and artificial intelligence, focusing on the inverse problem of imaging. His research interests include geometric algebras, quantum field theory, and machine learning methods applied to inverse imaging problems. Benjamin’s work integrates mathematics, physics, and AI to address complex scientific challenges

Fernanda Paulina Vizcaíno Imacaña, UIDE

Líder de la Escuela de Ciencias de la Computación de la UIDE con un profundo interés en la intersección entre la tecnología y el aprendizaje. Con un doctorado en Educación y una formación inicial como ingeniera informática, tiene interés en los campos cómo la programación, la ciencia de datos y la inteligencia artificial aplicada.

Andrés Morales-Navarrete, Universität Konstansz

Andres is scientist in computational biology and biophysics, specialized in deep learning, tissue modelling and active matter physics. He holds a PhD from the Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany. He also has an MSc and BSc in Physics from Escuela Politécnica Nacional, Quito, Ecuador. Andre’s professional experience includes postdoctoral research at the University of Konstanz and the Max Planck Institute. His contributions to computational embryology were recognized as part of the Method of the Year 2023 by Nature Methods. His current research interests include bioimage analysis and deep learning applications in developmental biology.

Vladimir Bonilla, UIDE

I am Vladimir Bonilla, a professor at the School of Mechatronics at UIDE. I specialize in the design of medical equipment and have a passion for control systems of these machines. I focus on the design of embedded systems, particularly for controlling prostheses or exoskeletons.

Eugenio Morocho, Universidad Yachay Tech

Full-time professor at Yachay Tech University for the courses Computer Vision, Artificial Intelligence, and Networks. His currest research interests are using AI-based tools to improve engineering processes.

Clara Rojas, Universidad Yachay Tech

Short biography: Ph. D: Instituto Venezolano de Investigaciones Cientificas IVIC (Venezuela), Postdoctorado e Investigadora: Instituto Venezolano de Investigaciones Cientificas IVIC hasta 2017, Docente e Investigadora: Universidad Yachay TECH desde 2017 hasta la actualidad. Research interest: Cosmologia Inflacionaria y Mecanica Cuantica Relativista.

Diego Morales, Universidad Yachay Tech

I am a mathematical engineer working on theoretical approaches for modelling spatio-temporal non-gaussian data. I am particularly interested in working on the theoretical and methodological development of space-time stochastic models for non-Gaussian data. I completed an undergraduate program in mathematical engineering at Escuela Politécnica Nacional in Quito-Ecuador. Then, I worked as a researcher and statistician in public and private companies in Ecuador for a few years. In the beginning of 2022, I obtained a Ph.D. degree in Statistics at the Department of Statistics of Pontificia Universidad Católica de Chile. As a doctoral researcher under the supervision of Prof. Mauricio Castro and Prof. Moreno Bevilacqua, I was working on developing novel approaches for modelling spatial and spatio-temporal count data. Additionally, I was part of the Center for the Discovery of Structures in Complex Data MiDas as a PhD student and guest researcher at the Department of Biology at the University of Konstanz. Currently, I am a research professor at Yachay Tech University.

Helga Dénes, Universidad Yachay Tech

Helga Dénes has a Masters degree in Astronomy from the Eötwös Lórand University in Budapest, Hungary. She obtain her PhD in Astrophysics in 2015 from Swinburne University of Technology in Melbourne, Australia. She has been a postdoctoral researcher at CSIRO Astronomy and Space Science in Sydney, and at ASTRON, The Netherlands Institute for Radio Astronomy. At the moment she is a Professor at Yachay Tech University in the School of Physical Sciences and Nanotechnology.

Helga’s main scientific interest is how the gas content of galaxies influences their evolution. She works on both extragalactic and Galactic astrophysics topics and is a member of several large radio astronomy survey teams, such as the Apertif Imaging Surveys, the WALLABY Survey and the GASKAP Survey.

Nelson Herrera, UIDE

Nelson Herrera Aráuz nació en Quito octubre de 1950. Estudios secundarios Colegio Central Técnico estudios superiores en Ingeniería Electrónica en la Escuela Politécnica Nacional; Estudios de Mathematica Pura, en la Escuela de Ciencias, de la Universidad Central del Ecuador, Electromecánica; Instituto José Antonio Echeverria La Habana Cuba, Física Universidad de Montpellier Francia. Investigaciones y experiencias científicas en Praga y Varsovia. Estudio teórico - experimentales documentado sobre fenómenos y perturbaciones magnéticas causadas por el eclipse solar del 11 de julio de 1991. Profesor Emérito en Cátedra sustentada por 25 años a la fecha de Matemáticas, Física y Circuitos Digitales en la Escuela de Ingeniería mecánica automotriz, de la Universidad International del Ecuador.

Wladimir Banda-Barragán, Universidad Yachay Tech

Wladimir is a computational astrophysicist working at the boundary between astrophysical gas dynamics and magnetohydrodynamics. Wladimir did his undergraduate studies at the Physics Department of Escuela Politécnica Nacional in Quito, and completed his PhD at the Research School of Astronomy and Astrophysics of the Australian National University in Canberra. Wladimir has worked as a physics lecturer and astrophysics researcher at several institutes, including Ecuador’s Meteorology Office (INAMHI), Observatorio Astronómico de Quito, Universidad Técnica de Ambato, the Australian National University in Canberra, Hamburg Observatory of Hamburg University, and more recently he joined Universidad Yachay Tech in Urcuquí.