RIDNAIC_Vol. 19-20

REVISTA INTERNACIONAL DESASTRES NATURALES, ACCIDENTES E INFRAESTRUCTURAL CIVIL ISSN 1535-0088 Volumen 19-20 ● Número 1 ● Diciembre 2020 ©Observatorio de Arecibo/UCF/AFP Terremotos y réplicas 2020 UNIVERSIDAD DE PUERTO RICO, RECINTO UNIVERSITARIO DE MAYAGÜEZ

REVISTA INTERNACIONAL DE DESASTRES NATURALES, ACCIDENTES E INFRAESTRUCTURA CIVIL PRESIDENTE COMISIÓN EDITORIAL BENJAMÍN COLUCCI-RÍOS Catedrático y Director, Cátedra Abertis-Puerto Rico Director, Centro de Transferencia de Tecnología en Transportación Universidad de Puerto Rico, Mayagüez, Puerto Rico MIEMBROS FUNDADORES Y EDITORES EMÉRITOS LUIS E. SUÁREZ Catedrático, Universidad de Puerto Rico, Mayagüez, Puerto Rico LUIS A. GODOY Profesor Titular Plenario, Universidad Nacional de Córdoba, Argentina COMISIÓN EDITORIAL ALESANDRA MORALES Universidad de Puerto Rico, Mayagüez, Puerto Rico SERGIO M. ALCOCER EUGENIO OÑATE Instituto de Ingeniería, Universidad Nacional Universidad Politécnica de Cataluña, Barcelona, Autónoma de México, CDMX España MIGUEL CANALS SILANDER GUSTAVO PACHECO-CROSETTI Universidad de Puerto Rico, Mayagüez, Universidad Politécnica de Puerto Rico Puerto Rico (Representando Argentina) CARLOS M. CHANG ISMAEL PAGÁN-TRINIDAD Florida International University, USA Director, Departamento de Ing. Civil y (Representando Perú) Agrimensura, Universidad de Puerto Rico, ALBERTO M. FIGUEROA MEDINA Mayagüez, Puerto Rico Universidad de Puerto Rico, MIGUEL A. PANDO Mayagüez, Puerto Rico Drexel University, Philadelphia, Pennsylvania, USA CARLOS HUERTA LÓPEZ FRANCESC ROBUSTÉ Universidad de Puerto Rico, Mayagüez, Universidad Politécnica de Cataluña, Puerto Rico (Representando México) Barcelona, España SANGCHUL HWANG ANDRÉS RODRÍGUEZ Texas State University, USA Universidad Nacional de Córdoba, Argentina (Representando Corea del Sur) ARTURO TENA COLUNGA Universidad Autónoma Metropolitana, XIAOPENG LI México, CDMX DIDIER M. VALDÉS DÍAZ University of South Florida, USA Universidad de Puerto Rico, (Representando China) Mayagüez, Puerto Rico (Representando Colombia) RICARDO R. LÓPEZ JORGE A. VANEGAS Universidad de Puerto Rico, Mayagüez, Presidente, Academia Panamericana de Ingeniería Puerto Rico (API), Texas A&M University, USA CARLA LÓPEZ DEL PUERTO (Representando Colombia) Universidad de Puerto Rico, Mayagüez, Puerto Rico

REVISTA INTERNACIONAL DE DESASTRES NATURALES, ACCIDENTES E INFRAESTRUCTURA CIVIL COLABORADORES ACADÉMICOS SECTOR PRIVADO Y FEDERAL HERNÁN O. FERNÁNDEZ ORDÓÑEZ Profesor Emérito, Universidad del Cauca, Colombia JOSÉ DOMINGO PÉREZ Presidente Electo 2022-2024, Academia Panamericana de Ingeniería (API) JUAN CARLOS RIVERA Administración Federal de Carreteras, (FHWA-USDOT) CARLOS E. RUIZ U.S. Army Engineer Research and Development Center, Vicksburg, Mississippi, USA DENNIS TRUAX Presidente Electo 2022, Asociación Americana de Ingenieros Civiles (ASCE) La Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil está indexada en la base de datos LATINDEX, ULRICH, DIALNET, EBSCO y DOAJ (Directory of Open Access Journals). ISSN 1535-0088

DEDICATORIA Dr. Luis E. Suárez-Colche Dr. Luis A. Godoy Fundadores y Editores Eméritos “Visionarios divulgando investigación de excelencia en Iberoamérica contribuyendo a reducir la vulnerabilidad de la infraestructura ante los fenómenos naturales extremos para el bienestar de la humanidad y el planeta”.

Revista Internacional de Desastres Naturales, Accidentes e i-ii Infraestructura Civil 1-6 7-10 Volumen 19-20, Número 1, Diciembre 2020 11-12 13-20 ISSN 1535-0088 21-42 43-51 Contenido: 52-64 Mensaje del Director del Departamento de Ingeniería Civil y Agrimensura UPRM 65-83 Ismael Pagán-Trinidad 84-104 Mensaje del Presidente Comisión Editorial Internacional RIDNAIC 105-115 Benjamín Colucci Ríos La Primera Etapa de RIDNAIC Luis Suárez Colche Infraestructura 2030 Juan F. Alicea Flores The Need for Resilient Infrastructure Dennis D. Truax Collapse of the Arecibo Observatory in Puerto Rico: Reflections from a Structural Engineering Perspective Juan C. Morales, Luis Suárez Colche Fleet and Fuel Strategies for Transportation Resilience Alexander Kolpakov, Austin Marie Sipiora, Xiaopeng Li, Caley Johnson y Erin Nobler Evaluating Risk from a Holistic Perspective to Improve Resilience: A Subnational Level Evaluation in Colombia Paula Marulanda Fraume, Omar Darío Cardona, Mabel Cristina Marulanda y Martha Liliana Carreño Integración entre Científicos, Ingenieros y las Comunidades afectadas sobre Inquietudes del Impacto de Terremotos y Tsunamis en Puerto Rico Ricardo R. López Rodríguez, Wilson R. Ramírez Martínez, Victor A. Huérfano, Christa Von Hillebrandt-Andrade y Ernesto F. Weil Machado Earthquake Experience in Puerto Rico and the Caribbean: Lessons and What We Have Learn from Them in the Last Two Decades Carlos I. Huerta López, José A. Martínez Cruzado y Luis Súarez Colche Caracterización de Sismos de Campo Cercano Jorge A. Rodríguez Ordoñez

Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil Volumen 19-20, Número 1, Diciembre 2020 ISSN 1535-0088 Contenido: Comportamiento de Edificios Industriales Típicos que Utilizan Techos Prefabricados 116-122 Tipo Doble T Durante los Terremotos Ocurridos en Enero y Mayo 2020 en Puerto Rico José O. Guevara Mejoras a la Red Sísmica de Movimiento Fuerte a Raíz de los Huracanes 123-128 Irma y María y de la Secuencia Sísmica José A. Martínez Cruzado Análisis de Fallas Estructurales en Infraestructura de Transportación 129-145 para Estimar las Velocidades de Viento Durante el Huracán María en Puerto Rico Gustavo E. Pacheco Crosetti y Héctor J. Cruzado Evaluación Forense y Lecciones Aprendidas de Fallas de Poste de Hormigón 146-163 Pretensado en Puerto Rico Durante el Huracán María Felipe J. Acosta Costa, Omar Esquilin Mangual, Didier Valdés, C. Kennan Crane, Stephanie G. Wood y Wendy R. Long Nuevos Códigos de Construcción 2018 y Educación para Mejorar 164-174 la Resiliencias de la Infraestructura Costera en Puerto Rico Carla López del Puerto, Ismael Pagán Trinidad, Luis D. Aponte Bermúdez y Stuart Adams Landslide Science in Puerto Rico: Past, Present and Future 175-187 K. Stephen Hughes y Alesandra C. Morales Vélez Classical Stress Analysis of a Fuse Plate that Failed during 188-201 Hurricane Maria in Puerto Rico Juan C. Morales, Jey N. Sánchez, Seira N. De Jesús y Jorge A. Caraballo RISE-UP: Una Herramienta Educativa Interdisciplinaria para la Generación 202-210 de Infraestructura Sostenible y Resiliente Carla López del Puerto, Humberto Cavallín, José Perdomo, Jonathan Muñoz, Marcelo Suárez y Drianfel Vázquez Investigaciones Innovadoras que Contribuyen a la Seguridad, Sostenibilidad 211-230 y Resiliencia en los Sistemas de Transporte Benjamín Colucci-Ríos, Alberto M. Figuera Medina, Didier M. Valdés Díaz

Revista Internacional de Desastres Naturales, Accidentes 231-239 e Infraestructura Civil 240-256 257-274 Volumen 19-20, Número 1, Diciembre 2020 275-286 ISSN 1535-0088 287 Contenido: Sistemas de Gestión de Pavimentos: Pasado, Presente y Futuro Carlos M. Chang Uso de Inteligencia Artificial para el Análisis del Comportamiento de Conductores y la Generación de Accidentes de Tránsito Wilson Arias Rojas, Jorge Eliecer Córdoba Maquilón y Germán Jairo Hernández Pérez Un Concepto No Convencional Basado en “Flutter” para Cosechar Energía Martín E. Pérez Segura, Emmanuel Beltramo, Bruno A. Roccia, Marcelo F. Valdez, Marcos L. Verstraete, Luis R. Ceballos y Sergio Preidikman Evaluación de Precipitaciones Mensuales Estimadas con TRMM para su Uso en Estudios de Sequías Meteorológicas Leticia del Valle Vicario, Carlos Marcelo García y Francina Domínguez Congresos y Seminarios

RIDNAIC: REVISTA CON VISIÓN DE INVESTIGACIÓN Y EDUCACIÓN EN CIENCIAS E INGENIERÍA COMO LEGADO DE LATINOAMÉRICA PARA EL MUNDO Ismael Pagán Trinidad, Director Departamento de Ingeniería Civil y Agrimensura, UPRM La Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil (RIDNAIC) se estableció en mayo de 2001 en el Departamento de Ingeniería Civil y Agrimensura del Recinto Universitario de Mayagüez de la Universidad de Puerto Rico. Su visión fue crear y desarrollar un foro científico y técnico abierto a profesionales e investigadores para divulgar en los idiomas español, portugués e inglés los avances de sus conocimientos a la comunidad internacional en “sistemas de ingeniería que dan apoyo y sirven para el diseño de la infraestructura civil, y los desastres naturales y accidentes de origen humano que puedan afectar esta infraestructura”. Los doctores Luis A. Godoy y Luis E. Suárez Colche, editores I. Pagán- fundadores de RIDNAIC y distinguidos catedráticos en Ingeniería Estructural Trinidad reconocidos en la comunidad científica internacional, forjaron la travesía de nuestro departamento en la red científica mundial por medio de la divulgación del nuevo conocimiento. Dos décadas en esta travesía han plantado en su camino 30 publicaciones con la participación de más de 500 autores procedentes de 20 o más países de América y Europa. Estamos seguros que RIDNAIC ha alcanzado miles de lectores y que muy bien le ha servido en sus investigaciones, su enseñanza o en la práctica profesional. La ingeniería y la ciencia enfrentan los retos de los riesgos naturales y tecnológicos alrededor del mundo. Los efectos de los fenómenos naturales extremos como calentamiento global, cambio climático, modificación del ambiente, aumento del nivel del mar, envejecimiento de la infraestructura, inundaciones urbanas y costeras, vientos extremos, deslizamientos de terrenos, terremotos y muchos otros eventos continúan siendo de enigmas sin resolver y de primera prioridad para ser mitigados disminuyendo la exposición y la vulnerabilidad de la comunidad universal. La infraestructura construida de edificaciones, transportación, agua, comunicaciones, electricidad, puertos, y otros elementos civiles continúan expuestos y vulnerables a la naturaleza. Estamos comprometidos con la investigación, el desarrollo y la educación formal y profesional que produzca los avances necesarios para establecer el nivel de resiliencia que permita proteger nuestra gente de los desastres que vivimos recurrentemente. La divulgación del conocimiento de los avances científicos internacionales a través del vehículo de RIDNAIC contribuyen a proveer la comunidad con la infraestructura sostenible y resiliente necesaria contra eventos extremos que causan los desastres ha sido una de las grandes contribuciones educativas que nuestra institución ha tenido el privilegio de auspiciar por las últimas dos décadas. A través de RIDNAIC, la academia amplía su impacto más allá de sus aulas y provee a nuestra comunidad la sublime bondad de educar nuestra audiencia internacional en ciencias e ingeniería. Es nuestra expectativa que estas contribuciones impacten directamente el desarrollo económico, la calidad ambiental y ecológica, el bienestar público y el estado sostenible y resiliente de la infraestructura civil y la comunidad afectada. Nuestra institución y Puerto Rico quedamos en extraordinaria deuda de gratitud con nuestros editores fundadores, los doctores Godoy y Suárez, por su visión, el trabajo y las contribuciones que por las pasadas dos décadas nos han dejado como legado para el futuro. Por ellos les reconocemos y le admiramos. Igualmente reconocemos y agradecemos a los autores, auspiciadores, lectores y todo el personal que de una u otra forma apoyó a RIDNAIC durante sus primeras dos décadas, en especial mención al Recinto Universitario de Mayagüez y Scipedia. En la celebración de dos décadas de éxito con la revista RIDNAIC, renovamos nuestro compromiso institucional a continuar expandiendo nuestra audiencia, las contribuciones y el alcance de nuestra revista. Agradecemos y felicitamos al nuevo Presidente de la Comisión Editorial Internacional, el Dr. Benjamín Colucci-Ríos, Catedrático en el campo de la Ingeniería en Transportación en este Recinto, quien aceptó el reto de desarrollar la nueva generación de RIDNAIC, ampliando el alcance al idioma inglés, con el fin de alcanzar una audiencia más amplia y diversa en el futuro. Bienvenidos a una nueva década de conocimiento y educación a través de RIDNAIC para un mundo más seguro y resiliente. ¡Enhorabuena!

MENSAJE DEL PRESIDENTE COMISIÓN EDITORIAL RIDNAIC EDICIÓN ESPECIAL CONMEMORATIVA DEL VIGÉSIMO ANIVERSARIO DE RIDNAIC Benjamín Colucci Ríos1 La Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil (RIDNAIC) es una publicación científica que se fundó en mayo de 2001 por iniciativa de los doctores Luis A. Godoy y Luis E. Suárez Colche del Departamento de Ingeniería Civil y Agrimensura del Recinto Universitario de Mayagüez de la Universidad de Puerto Rico. Es un inmenso honor y privilegio dedicar la Edición Especial Conmemorativa del Vigésimo Aniversario de RIDNAIC a estos dos extraordinarios colegas, catedráticos e investigadores de excelencia y fundadores de esta prestigiosa Revista a nivel Internacional. Esta Edición Especial Conmemorativa de RIDNAIC contiene 21 artículos técnicos de 57 autores que presentan trabajos de investigación relacionados a desastres naturales e infraestructura civil en Puerto Rico, a nivel internacional, y las lecciones aprendidas de estos eventos. Comenzamos esta edición con la historia y reflexión de los primeros 20 años de RIDNAIC. El Dr. Suárez describió de manera magistral los comienzos de esta iniciativa en conjunto con el Dr. Godoy en su propósito fundamental de documentar los hallazgos científicos relevantes y de interés a la comunidad ingenieril de Iberoamérica para reducir la vulnerabilidad de la infraestructura a los fenómenos naturales extremos y accidentes. El segundo artículo, Infraestructura 2030 del Ing. Juan F. Alicea Flores, Presidente del Colegio de Ingenieros y Agrimensores de Puerto Rico (CIAPR), y el tercero, La Necesidad de una Infraestructura Resiliente del Dr. Dennis Truax, Presidente entrante de la Asociación Americana de Ingenieros Civiles (ASCE) en el 2022, enmarcan la pertinencia de una infraestructura resiliente en esta década, y los cambios de paradigma y acciones afirmativas para alcanzar estas metas. Los restantes 18 artículos nos presentan una radiografía de los diferentes daños y vulnerabilidad de la infraestructura civil, recomendaciones para diseños de obras más resilientes, lecciones aprendidas e iniciativas comunitarias que contribuyen a orientar y preparar a nuestra ciudadanía en el proceso de adaptación en la eventualidad de futuros desastres naturales. El colapso del observatorio de Arecibo en diciembre del 2020, lecciones aprendidas de los terremotos y réplicas durante el año 2020, los daños a la infraestructura civil relacionados a los huracanes Irma y María en el 2017 debido a los vientos, inundaciones y deslizamientos son descritos en varios artículos en esta Edición Especial Conmemorativa del Vigésimo Aniversario de RIDNAIC. Además, se presentan artículos relacionados a tecnologías emergentes e innovadoras de simulación de vehículos de conducción y realidad virtual para atender la seguridad y aspectos de congestión y equidad en el área de infraestructura de transporte. En esta Edición Especial de RIDNAIC se presentan además, artículos técnicos relacionados a la inteligencia artificial para analizar el comportamiento de conductores, cosecha de energía, generación de infraestructura 1 Catedrático, Director del Centro de Transferencia de Tecnología en Transportación y Cátedra Abertis, Portavoz de la Década de Acción para la Seguridad Vial, Departamento de Ingeniería Civil y Agrimensura y pasado Decano Interino de la Facultad de Ingeniería, Universidad de Puerto Rico, Mayagüez, Puerto Rico 00681-9000. Email:[email protected] Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 1

sostenible y resiliente, evaluación holística del riesgo, fallas de poste de hormigón pretensado, comportamiento de edificios industriales que usan vigas de hormigón pretensado tipo doble-T, nuevos códigos de construcción 2018, placas fusibles, sismos de campo, fallas estructurales en infraestructura de transportación, mejoras a la Red Sísmica de Movimiento Fuerte, estrategias de flota y combustible para una transportación resiliente, sistemas de gestión de pavimentos, e inquietudes del impacto de terremotos y tsunamis, entre otros de interés a la comunidad científica Iberoamericana. En el contexto de esta Edición Especial Conmemorativa del Vigésimo Aniversario de RIDNAIC, es menester compartir a continuación los testimonios de pasados estudiantes graduados y colegas de la comunidad científica sobre la calidad, profesionalismo e impacto que han tenido estos dos extraordinarios visionarios en las diferentes disciplinas en la nueva generación de investigadores en la infraestructura civil Iberoamericana. El Dr. Gustavo Pacheco-Crosseti, catedrático e investigador de la Universidad Politécnica de Puerto Rico (UPPR), egresado del programa doctoral del RUM y exalumno resaltó “…la cultura general, su conocimiento, su tremenda capacidad docente, y su rigurosidad científica”. Señaló además, que “…fue un placer ser su discípulo en la Universidad Nacional de Córdoba, Argentina, y en cursos graduado en Estabilidad Estructural y Placas y Cáscaras en Mayagüez. Sin duda ha sido una gran influencia en mi vida académica, y un modelo a seguir. Gracias Dr. Godoy por su entrega, y el legado de RIDNAIC”. En el caso del Dr. Luis A. Suárez-Colche, nos comentó que “…el Dr. Luis Suárez fue quien lo introdujo al análisis dinámico de estructuras,…, fue mi profesor de Dinámica Estructural, Dinámica de Suelos, y Consejero de Tesis”. Afirmó que “…todo lo que sé de dinámica estructural y de suelos se lo debo a él, y que considero sus libros y apuntes son los mejores en el área”; y concluye que “…siempre ha sido un modelo por su enorme conocimiento, su capacidad de transmitirlo, su dedicación y meticulosidad, su rigurosidad, su humildad,…, su sensibilidad y empatía, y su disponibilidad a tender una mano en momentos difíciles”. El Dr. Carlos I. Huerta, considera al Dr. Suárez un “…excelente académico y educador por vocación”. El Dr. Ricardo López, pasado Director del Centro de Infraestructura Civil y de Estudios Graduados del Departamento resalta “los atributos y contribuciones de los Luises”, a los cuales conoce por espacio de prácticamente tres décadas. En cuanto al Dr. Luis Godoy “… es un profesional brillante que ha aportado significativamente con sus conocimientos e ideas a una gran cantidad de profesionales a nivel mundial en la mecánica estructural y siempre se ha distinguido por su eficiencia y objetivos claros en su desempeño. Su contribución en a través del libro sobre la historia de la mecánica estructural es una publicación invaluable a nivel internacional, además de los cientos de artículos técnicos sobre estabilidad de la mecánica estructural y como editor de un sinnúmero de revistas arbitradas.” En cuanto al Dr. Luis Eduardo Suárez Colche, el Dr. López resalta la capacidad intelectual y logros, “… es un profesional brillante que se ha destacado grandemente por hacer todo su esfuerzo académico e investigación por ayudar a sus estudiantes a nivel sub-graduado y graduado a entender la dinámica estructural, y a tener una experiencia grata como estudiantes al cursar sus estudios en el Recinto Universitario de Mayagüez de la Universidad de Puerto Rico, y prepararlo para los retos profesionales en esta disciplina”. Dr. López culmina su testimonio indicando que “…ambos compañeros han contribuido al estado del arte con sus publicaciones en la dinámica estructural, Dr. Suárez y a la mecánica estructural, Dr. Godoy”. Por último, la Dra. Alesandra Morales, colega del Dr. Luis Suárez reconoce que a pesar de que nunca tuvo la oportunidad de tomar clases con el compañero, sus estudiantes le han comentado como “…deja una huella e inspira a cada estudiante que toca…, y tiene el don de saber convertir la ciencia de las Estructuras en el Arte de las Estructuras”. En su opinión, resalta que “…esa pasión y ese romanticismo ante la Ingeniería Estructural es la que lo ha llevado a ganarse el respeto y admiración de cientos de estudiantes y colegas no solo en Puerto Rico, sino alrededor del mundo”. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 2

Muchos colegas compartieron similares testimonios de los Luises. Por último, comparto un testimonio en particular del Dr. Suárez que es digno de emular en esta coyuntura histórica “…su tremendo humor, y su manera peculiar de interpretar el mundo”. Gracias, muchas gracias “a los queridos y muy respetados Luises” a nombre de los demás colegas de la comunidad científica y de este servidor por su entrega y el legado a las primeras dos décadas de la Revista RIDNAIC. Al momento de redactar este mensaje el 24 de diciembre de 2020, en víspera de Navidad, Puerto Rico nuevamente fue estremecido por dos sismos de magnitud 4.8 y 4.7 en la escala Richter en apenas 47 minutos de diferencia. Estos movimientos telúricos nos recordaron nuevamente los eventos traumáticos que pasó la ciudadanía el pasado 7 de enero de 2020 con un sismo de magnitud 6.4. Estos fenómenos naturales extremos, el efecto adverso del cambio climático y la nueva norma de COVID-19 a nivel mundial, emplaza a los ingenieros y a la comunidad científica a reflexionar y ser visionarios sobre los futuros diseños y rehabilitación de la infraestructura civil, de tal manera que sea más resiliente y sostenible para mejorar la calidad de vida y los procesos de adaptación de la presente y futura sociedad civil. Agradezco a los pasados Miembros del Comité Editorial RIDNAIC, cuyos conocimientos y sabiduría, seleccionando a los mejores trabajos de investigación en Iberoamérica durante las primeras dos décadas, contribuyeron en la publicación de una Revista Internacional de excelencia, aportando significativamente a reducir la vulnerabilidad de la infraestructura ante los fenómenos naturales extremos para el bienestar de la humanidad y el planeta. Ciertamente agradecemos la confianza brindada por el Prof. Ismael Pagán-Trinidad, Director del Departamento de Ingeniería Civil y Agrimensura del RUM, en presidir la Comisión Editorial Internacional de esta Edición Especial Conmemorativa al Vigésimo Aniversario de RIDNAIC. Esta Edición Especial Conmemorativa al Vigésimo Aniversario de RIDNAIC ciertamente es una realidad gracias al titánico esfuerzo y dedicación de la Srta. Ciara Toro-Rosario, Asistente Administrativa del Centro de Transferencia de Tecnología en Transportación. Desde un principio, la Srta. Toro dijo presente y estuvo mano a mano con el Editor y equipo de apoyo en todo el proceso de transición, delineando el Plan de Trabajo para los próximos dos años de RIDNAIC, establecer contactos con la plataforma Scipedia, en el proceso de entrevistas de los nuevos miembros de la Comisión Editorial, coordinación de traducciones y todo lo que conlleva una Revista Internacional del calibre de RIDNAIC. Eternamente agradecido. La Srta. Toro Rosario y este servidor estamos en deuda con la Sra. Irmalí Franco-Ramírez, Oficial Administrativo por el apoyo y sabiduría en la logística del desarrollo del “website” de RIDNAIC, el arte del logo y portada y el apoyo técnico durante la transición al “website” Scipedia para la publicación electrónica. Agradecemos además, a la Srta. Natalia Alzate Pérez y a la Sra. Grisel Villarrubia-Echevarría por el apoyo incondicional para completar con éxito esta Edición Especial de RIDNAIC. Esperamos que esta Edición Conmemorativa del Vigésimo Aniversario de RIDNAIC cumpla con las expectativas de la comunidad científica Iberoamericana en contribuir a reducir la vulnerabilidad de la infraestructura ante los fenómenos naturales extremos para el bienestar de la humanidad. Con mi mayor aprecio y admiración, Benjamín Colucci Ríos, PhD, PE, PTOE, F.ASCE, F.ITE, JD, PAE Presidente Comisión Editorial RIDNAIC Edición Especial Conmemorativa - 20vo Aniversario Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 3

MESSAGE FROM THE PRESIDENT OF THE RIDNAIC EDITORIAL COMMITTEE SPECIAL EDITION COMMEMORATIVE OF THE TWENTIETH ANNIVERSARY OF RIDNAIC Benjamín Colucci-Ríos1 The International Journal of Natural Disasters, Accidents and Civil Infrastructure (RIDNAIC) is a scientific publication that was founded in May 2001 by initiative of Dr. Luis A. Godoy and Dr. Luis E. Suárez-Colche of the Department of Civil Engineering and Surveying of the University of Puerto Rico at Mayagüez. It is an immense honor and privilege to dedicate the RIDNAIC’s Twentieth Anniversary Special Commemorative Edition to these two extraordinary colleagues, professors and researchers of excellence and founders of this prestigious International Journal. This Special Commemorative Edition of RIDNAIC contains 21 technical articles by 57 authors that present international research related to natural disasters and civil infrastructure in Puerto Rico, and the lessons learned from these events. We begin this edition with the history and reflection of the first 20 years of RIDNAIC. Dr. Suárez masterfully described the beginnings of this initiative in conjunction with Dr. Godoy in his fundamental purpose of documenting relevant scientific findings and of interest to the Latin American engineering community to reduce the vulnerability of infrastructure to extremes natural phenomena and accidents. The second article, Infrastructure 2030 by Ing. Juan F. Alicea-Flores, President of the College of Engineers and Surveyors of Puerto Rico (CIAPR), and the third article, The Need for a Resilient Infrastructure by Dr. Dennis Truax, incoming President of the American Society of Civil Engineers (ASCE) in 2022, frame the relevance of a resilient infrastructure in this decade, and the paradigm shifts and affirmative actions to achieve these goals. The remaining 18 articles present a radiography of the different damages and vulnerabilities of civil infrastructure, recommendations for more resilient work designs, lessons learned and community initiatives that help guide and prepare our citizens in the adaptation process in the event of future natural disasters. The collapse of the Arecibo observatory in December 2020, lessons learned from the earthquakes and aftershocks during 2020, the damage to civil infrastructure related to hurricanes Irma and María in 2017 due to winds, floods and landslides are described in several articles in this Twentieth Anniversary Special Commemorative Edition. In addition, articles related to emerging and innovative technologies for driving vehicle simulation and virtual reality are presented to address safety, congestion, and equity issues in transportation infrastructure. This Special Edition of RIDNAIC also presents technical articles related to artificial intelligence to analyze the behavior of drivers, energy harvesting, generation of sustainable and resilient infrastructure, holistic risk assessment, prestressed concrete pole failures, industrial building behavior utilizing double-T type prestressed concrete beams, new 2018 building codes, fuse plates, field earthquakes, structural failures in transportation infrastructure, improvements to the Strong Motion Seismic Network, fuel and fleet strategies for resilient 1 Professor; Transportation Technology Transfer Center and Abertis Chair Director, Spokesperson of the Decade of Action for Road Safety, Civil Engineering and Surveying Department and past Interim Dean of the Engineering Faculty, University of Puerto Rico, Mayagüez, Puerto Rico 00681-9000. Email: [email protected] Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 4

transportation, pavement management systems, and concerns about the impact of earthquakes and tsunamis, among others of interest to the Ibero-American scientific community. In the context of this RIDNAIC’s Twentieth Anniversary Special Commemorative Edition, it is necessary to share the testimonies of past graduate students and colleagues from the scientific community about the quality, professionalism, and impact that these two extraordinary visionaries have had in the different disciplines in the new generation of researchers in the Ibero-American civil infrastructure. Dr. Gustavo Pacheco-Crosseti, professor and researcher at the Polytechnic University of Puerto Rico (PUPR), graduated from the RUM doctoral program and former student highlighted \"... the general culture, his knowledge, his tremendous teaching capacity, and his scientific rigor\". He further noted that “… it was a pleasure to be his disciple at the National University of Córdoba, Argentina, and in graduate courses in Structural Stability, and Plates and Shells in Mayagüez. He has certainly been a great influence on my academic life, and a role model. Thank you, Dr. Godoy, for your dedication, and the legacy of RIDNAIC”. In the case of Dr. Luis A. Suárez-Colche, he told us that “… Dr. Luis Suárez introduced him to the dynamic analysis of structures, …, he was my professor of Structural Dynamics, Soil Dynamics, and Thesis Advisor”. He stated that “… everything I know about structural dynamics and soils I owe to him, and I consider that his books and notes are the bests in the area”; and concludes that \"... he has always been a model for his enormous knowledge, his ability to transmit it, his dedication and meticulousness, his rigor, his humility, ..., his sensitivity and empathy, and his willingness to lend a hand in difficult times.\" Dr. Carlos I. Huerta considers Dr. Suárez an \"... excellent academic and educator by vocation.\" Dr. Ricardo López, past Director of the Center for Civil Infrastructure and of the Graduate Studies in the Department, emphasized on “the attributes and contributions of the Luises”, whom he has known for almost three decades. As for Dr. Luis Godoy “… he is a brilliant professional who has contributed significantly with his knowledge and ideas to many professionals worldwide in structural mechanics and has always distinguished himself for his efficiency and clear objectives in his performance. His contribution through the book on the history of structural mechanics is an invaluable international publication, in addition to the hundreds of technical articles on stability of structural mechanics and as editor of countless peer-reviewed journals”. As for Dr. Luis Eduardo Suárez-Colche, Dr. López also appointed his intellectual capacity and achievements, “… he is a brilliant professional who has greatly stood out for making all his academic and research efforts to help his students at the undergraduate and graduate level to understand structural dynamics, and to have a pleasant experience as students when studying at the Mayagüez Campus of the University of Puerto Rico, and prepare them for the professional challenges in this discipline”. Dr. López concludes his testimony by stating that “… both colleagues have contributed to the state of the art with their publications on structural dynamics, Dr. Suárez, and on structural mechanics, Dr. Godoy”. Finally, Dr. Alesandra Morales, Dr. Luis Suárez's colleague, recognizes that although she never had the opportunity to take classes with her colleague, her students have commented how “… he leaves a mark and inspires each student he meets... and has the gift of knowing how to transform the science of Structures into the Art of Structures”. In her opinion, she brought to attention that \"... that passion and romanticism towards Structural Engineering leads him to earn the respect and admiration of hundreds of students and colleagues not only in Puerto Rico, but around the world”. Many colleagues shared similar testimonies from the Luises. Finally, I share a particular testimony of Dr. Suárez that is worth emulating at this historical juncture \"... his tremendous humor, and his peculiar way of interpreting the world.\" Thank you, thank you very much \"to the dear and highly respected Luises\" on behalf of the other colleagues of the scientific community and of this server for their dedication and the legacy to the first two decades of RIDNAIC Journal. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 5

At the time of writing this message on December 24, 2020, on Christmas Eve, Puerto Rico was once again shaken by two earthquakes of magnitude 4.8 and 4.7 on the Richter scale, barely 47 minutes apart. These telluric movements reminded us again of the traumatic events that citizens went through on January 7, 2020 with an earthquake of magnitude 6.4. These extreme natural phenomena, the adverse effect of climate change and the new norm of COVID-19 worldwide, calls on the engineers and the scientific community to reflect and be visionaries about the future designs and rehabilitation of civil infrastructure, in such a way that is more resilient and sustainable to improve the quality of life and the adaptation processes of the present and future civil society. I am grateful to the past Members of the RIDNAIC Editorial Committee, whose knowledge and wisdom, selecting the best research works in Ibero-America during the first two decades, contributed to the publication of an International Journal of excellence, contributing significantly to reducing the vulnerability of infrastructure in the face of extreme natural phenomena for the well-being of humanity and the planet. We certainly appreciate the trust given by Prof. Ismael Pagán-Trinidad, Director of the Department of Civil Engineering and Surveying of the UPRM, in presiding the International Editorial Commission of this Special Edition Commemorating the Twentieth Anniversary of RIDNAIC. This Special Commemorative Edition to the Twentieth Anniversary of RIDNAIC is certainly a reality thanks to the titanic effort and dedication of Ms. Ciara Toro-Rosario, Administrative Assistant of the Transportation Technology Transfer Center. From the beginning, Ms. Toro-Rosario was present and worked hand in hand with the Editor and support team throughout the transition process, outlining the Work Plan for the next two years of RIDNAIC, establishing contacts with the Scipedia platform, in the process of interviews of the new members of the Editorial Commission, coordination of translations and everything that an International Journal of the caliber of RIDNAIC entails. Eternally grateful. Ms. Toro-Rosario and this server are indebted to Mrs. Irmalí Franco-Ramírez, Administrative Officer, for the support and wisdom in the logistics of the development of the RIDNAIC website, the art of the logo and cover, and the technical support during the transition to the Scipedia website for electronic publishing. We also thank Ms. Natalia Alzate Pérez and Ms. Grisel Villarrubia-Echevarría for their unconditional support for a successful completion of this RIDNAIC Special Edition. We hope that this Commemorative Edition of the Twentieth Anniversary of RIDNAIC meets the expectations of the Ibero-American scientific community in helping to reduce the infrastructure vulnerability by encountering extreme natural events for the well-being of humanity. With my greatest appreciation and admiration, Benjamín Colucci-Ríos, PhD, PE, PTOE, F.ASCE, F.ITE, JD, PAE President, RIDNAIC Editorial Committee Special Commemorative Edition - 20th Anniversary Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 6

LA PRIMERA ETAPA DE RIDNAIC Dr. Luis E. Suárez-Colche, Catedrático UPRM y Co-fundador de RIDNAIC L. Suárez- La Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Colche Civil se comenzó a publicar en mayo del año 2001 por iniciativa del Dr. Luis A Godoy, en ese entonces profesor del área de Ingeniería Estructural en el Departamento de Ingeniería Civil y Agrimensura de la Universidad de Puerto Rico en Mayagüez. El profesor Godoy, investigador, autor y docente mundialmente reconocido, me invitó a participar como uno de los dos editores, lo que acepté con mucho entusiasmo. Debido a los nombres de los dos editores, algunos colegas comenzaron a llamar afectuosamente a la publicación “la revista de los Luises”: nosotros la llamábamos RIDNAIC. Debemos reconocer que desde el comienzo siempre tuvimos el apoyo incondicional del director de nuestro departamento, el Profesor Ismael Pagán Trinidad, y en mayor o menor grado, de los Rectores de la institución. En ese entonces, pensamos que la publicación llenaría un vacío en la literatura en español sobre la temática de la revista contribuyendo a la divulgación de información técnica que podría ser muy útil para la comunidad ingenieril de Iberoamérica para reducir la vulnerabilidad de la infraestructura a los fenómenos naturales extremos y accidentes. El alcance de la nueva publicación comprendería a los sistemas de ingeniería que dan apoyo y sirven para el análisis y diseño de la infraestructura civil, y a los desastres naturales y accidentes de origen humano que pueden afectar esa infraestructura. El término infraestructura civil se usaría para designar al conjunto de instalaciones físicas que permiten movilizar a personas, bienes, materias primas, agua, residuos, energía e información. En general, se incluirían aquí carreteras, puentes, puertos, canales, aeropuertos, ferrocarriles, sistemas de tránsito urbano, líneas de comunicación y energía, tuberías, represas y plantas de tratamiento de aguas. La frase desastre natural hacía referencia a las acciones que se traducen en pérdidas de vidas humanas y materiales ocasionadas por fenómenos naturales, como terremotos, huracanes, tornados, inundaciones, tsunamis, deslizamientos de tierra y licuación de suelos, fuegos y sequías. Asimismo, RIDNAIC se enfocaría en temas relacionados con accidentes y eventos producidos por causas humanas, incluyendo fallas por diseño o construcción, colisiones, explosiones y otras. Hay puntos de vista en la comunidad científica que postulan de que no existe tal cosa como un desastre “natural”, que si un fenómeno de la naturaleza causaba estragos en la población e infraestructura no deberíamos acusarla de causar el “desastre”. El argumento se basa en que un evento natural extremo, como un sismo fuerte o un huracán, si bien es el desencadenante del desastre, no su causa de fondo. A modo de ejemplo, si ocurre un terremoto intenso en un desierto o un huracán en una isla deshabitada, estos no van a causar pérdida de vidas humanas ni de propiedad, y por lo tanto, no son un desastre. El desastre ocurre cuando el terremoto ocurre en, o cuando el huracán pasa por, una zona en donde habita una población que no está preparada, establecida en zonas inseguras por falta de planificación adecuada, que vive en estructuras precarias que no cumplen con las normas y especificaciones vigentes en su jurisdicción, con infraestructura precaria, y por lo tanto, vulnerables a estos fenómenos. De todas maneras, debido al uso extendido de la frase “desastres naturales” y a que formaba parte del nombre reconocido de la revista, decidimos continuar con el mismo. Antes de comenzar a publicar la revista, nos dedicamos a reunir un grupo de distinguidos académicos y profesionales para que formaran parte de la Junta Editorial. Mis viajes a presentar trabajos en diversos países de nuestra Latinoamérica ayudaron a establecer nuevos contactos, dar a conocer la publicación, invitar a potenciales autores a publicar en la misma y renovar la Junta Editorial. Tuvimos así la fortuna de reunir a un elenco de profesionales y científicos de primer orden a nivel mundial que nos serviría de consejeros y le daría credibilidad y prestigio a la revista. A todos ellos vaya nuestro más sincero agradecimiento y respeto. Desde sus comienzos nos fijamos la meta que los artículos fueran revisados en forma anónima por dos expertos en el tema, siguiendo la tradición de las mejores revistas científicas (“peer-reviewed journals”). Conseguir los revisores de la talla y nivel al que aspirábamos no fue una tarea fácil, dado que muchos de los potenciales árbitros académicos eran personas muy ocupadas y competíamos por su tiempo con prestigiosas publicaciones en inglés de larga tradición y fama. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 7

El objetivo en la primera etapa de RIDNAIC no era competir con las numerosas revistas técnicas existentes en esos momentos, en idioma inglés y de gran prestigio y raigambre. Por esta razón, y por una decisión editorial que se discutió mucho, decidimos que los únicos idiomas oficiales de la revista iban a ser el español (o castellano) y el portugués. Además, pretendíamos que los artículos que se publicaran no solo tuvieran un nivel apropiado, sino que sirvieran de utilidad para profesionales de la ingeniería. No queríamos, usando la frase de nuestro distinguido Editor para Sudamérica, el Dr. Carlos Prato, contribuir al “paper pollution”. Todo número de RIDNAIC empezaba con un editorial sobre un tema relevante para los lectores. Estos editoriales eran por invitación solamente, y me place afirmar que muchos de ellos eran muy interesantes y profundos, a pesar de la limitación en el número de páginas. Además, siempre tratamos de que hubiera un balance entre las áreas de los trabajos que se publicaban en RIDNAIC (Geotecnia, Hidráulica, Hidrología, Ingeniería Ambiental, Gerencia de Construcción, Transportación y Estructuras, entre otras). Reconozco que esta tarea no siempre era sencilla, dado que la experiencia académica y profesional de los dos editores era en Ingeniería Estructural, pero creemos que pudimos alcanzar esta equidad. Una manera de lograr el balance consistió en la publicación de ediciones especiales dedicadas a un área específica, como Geotecnia y Transportación, con distinguidos editores invitados para esas ocasiones. Desde que comenzamos la publicación de RIDNAIC hace 20 años, el mundo de las revistas técnicas ha sufrido grandes cambios, y en mi opinión, muchos de ellos negativos. Los que estamos dedicados a la academia conocemos de numerosos “journals” que han proliferado en los últimos años gracias a la ubicuidad de la internet. Estoy seguro de que a muchos lectores les llegan, a través de diversas plataformas, invitaciones a leer artículos que ni siquiera tienen el nivel de un proyecto de una clase graduada. Para colmo de males, muchos de ellos tampoco parecen preocuparse porque el material que publican sea original. RIDNAIC comenzó a crecer y tuvo vasta aceptación entre la comunidad de ingeniería civil y afines en Latinoamérica y la comunidad hispana en Estados Unidos. Durante la primera etapa, recibimos trabajos de Argentina, Chile, Colombia, Cuba, Brasil, Ecuador, España, Estados Unidos, México, Perú, República Dominicana, Venezuela y Puerto Rico. Durante varios años, RIDNAIC se publicaba en papel y en forma electrónica. Con cada número se distribuían alrededor de 1,500 ejemplares libre de costo a suscriptores de todo el continente y España. Con el paso de los años y la popularización de la internet, la revista comenzó a enfocarse cada vez más en su versión en línea. Los crecientes costos de impresión, y sobre todo los gastos de envío, junto con una situación económica complicada en la Universidad de Puerto Rico, obligaron primero a suspender los envíos al exterior, y luego dentro de Puerto Rico. La versión en papel continuó, pero solo se repartía a bibliotecas, agencias del gobierno local relacionadas a la infraestructura civil, y a universidades. Posteriormente se canceló totalmente la versión impresa. Esta tendencia a abandonar la tradicional versión en papel es hoy en día compartida por la gran mayoría de las revistas técnicas. Por iniciativa del Dr. Luis Godoy, la versión en línea de RIDNAIC se transfirió a la plataforma electrónica de publicaciones académicas Scipedia, físicamente localizada en Barcelona, España. Scipedia es una iniciativa promovida por el Centro Internacional de Métodos Numéricos en la Ingeniería (CIMNE) de la Universitat Politècnica de Catalunya para impulsar la publicación de artículos técnicos en el formato “Open Access”. Esta decisión le garantizaba mayor difusión a RIDNAIC a través de la internet. Me place notificar que en la nueva etapa RIDNAIC continuará asociada a Scipedia, que ha adquirido reconocimiento a nivel mundial por la calidad y profundidad de sus publicaciones. Otro acontecimiento importante en la historia de la publicación fue el retiro del Dr. Luis Godoy de la Universidad de Puerto Rico en Mayagüez y el regreso a su país natal, Argentina. Aunque el Dr. Godoy continuó trabajando desde su Córdoba natal, su ausencia física se hizo notar. Debido a mis otras responsabilidades en la Universidad que ameritaban prioridad y con la conformidad del Dr. Luis Godoy, tomamos la decisión de poner a RIDNAIC en una pausa temporal. En los momentos en que estoy escribiendo esta reflexión, el mundo entero ya lleva muchos meses pasando por un periodo de extrema dificultad por la pandemia del COVID-19, aunque aparecen esperanzas en el horizonte. Sin embargo, esta tribulación mundial nos ha hecho despertar a una nueva realidad y tomar más conciencia de la fragilidad de nuestro planeta. El cambio climático global, que algunos “líderes” irresponsables insisten en negar, finalmente nos arropa. Fenómenos como el aumento en el nivel del mar, el incremento en la severidad y frecuencia de los eventos atmosféricos y otras consecuencias nefastas requieren de nuestra atención inmediata. Los avances tecnológicos en las ciencias e ingeniería y la unidad de propósito de nuestra comunidad ingenieril pueden contribuir a resolver estos problemas urgentes de pertinencia global: debe ser nuestro compromiso. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 8

THE FIRST STAGE OF RIDNAIC Dr. Luis E. Suárez-Colche, UPRM Professor and Co-founder of RIDNAIC L. Suárez- The International Journal of Natural Disasters, Accidents and Civil Infrastructure Colche began to be published in May 2001 at the initiative of Dr. Luis A Godoy, at that time a professor in the area of Structural Engineering in the Department of Civil Engineering and Surveying of the University of Puerto Rico in Mayagüez. Professor Godoy, a world renowned researcher, author and professor, invited me to participate as one of the two editors, which I accepted with great enthusiasm. Due to the first names of the two editors, some colleagues began to affectionately call the publication \"The Luises Journal\": we called it RIDNAIC. We must recognize the unconditional support that from the beginning we always had of the director of our department, Professor Ismael Pagán Trinidad, and to a greater or lesser degree, the Chancellors of the institution. At that time, we thought that the publication would fill a gap in the Spanish literature on the subject of the journal, contributing to the dissemination of technical information that could be very useful for the Ibero-American engineering community to reduce the vulnerability of the infrastructure to the extreme natural and accidents phenomena. The scope of the new publication would include the engineering systems that support and serve the analysis and design of civil infrastructure, and the natural disasters and man-made accidents that can affect that infrastructure. The term civil infrastructure would be used to designate the set of physical facilities that allow the movement of people, goods, raw materials, water, waste, energy and information. In general, roads, bridges, ports, canals, airports, railways, urban transit systems, communication and power lines, pipelines, dams, and water treatment plants would be included here. The phrase natural disaster referred to actions that result in loss of human and material life caused by natural phenomena, such as earthquakes, hurricanes, tornadoes, floods, tsunamis, landslides and soil liquefaction, fires and droughts. Likewise, RIDNAIC would focus on issues related to accidents and events produced by human causes, including design or construction failures, collisions, explosions and others. There are points of view in the scientific community that postulate that there is no such thing as a \"natural\" disaster, that if a phenomenon of nature wreaked havoc on the population and infrastructure we should not accuse it of causing the \"disaster\". The argument is based on the fact that an extreme natural event, such as a strong earthquake or a hurricane, although it is the trigger for the disaster, not its root cause. As an example, if an intense earthquake occurs in a desert or a hurricane on an uninhabited island, they will not cause loss of human life or property, and therefore, they are not a disaster. Disaster occurs when the earthquake occurs in, or when the hurricane passes through, an area inhabited by a population that is not prepared, established in unsafe areas due to lack of adequate planning, living in precarious structures that do not comply with the regulations and specifications in force in your jurisdiction, with precarious infrastructure, and therefore, vulnerable to these phenomena. However, due to the widespread use of the phrase \"natural disasters\" and because it was part of the recognized name of the journal, we decided to continue with it. Before publishing the journal, we dedicated ourselves to bringing together a group of distinguished academics and professionals to serve on the Editorial Board. My trips to present works in various countries of our Latin America helped to establish new contacts, promote the publication, invite potential authors to publish in it and renew the Editorial Board. Thus, we were fortunate to bring together a cast of world-class professionals and scientists who would serve as advisors and give the journal credibility and prestige. Our most sincere thanks and respect go to all of them. From the beginning, we set the goal for the articles to be reviewed anonymously by two experts on the subject, following the tradition of the best scientific journals (“peer-reviewed journals”). Getting reviewers of the stature and level to which we aspired was not an easy task, since many of the potential academic referees were very busy people and we competed for their time with prestigious publications in English of long tradition and fame. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 9

RIDNAIC first stage objective was not to compete with the numerous journals that existed at that time, in English and of great prestige and roots. For this reason, and due to an editorial decision that was discussed a lot, we decided that the only official languages of the journal would be Spanish (or Castilian) and Portuguese. In addition, we wanted the articles to be published not only at an appropriate level, but also useful for engineering professionals. We did not want, using the phrase of our distinguished Editor for South America, Dr. Carlos Prato, to contribute to “paper pollution”. Every issue of RIDNAIC began with an editorial on a topic relevant to readers. These editorials were by invitation only, and I am pleased to say that many of them were very interesting and insightful, despite the limited number of pages. Furthermore, we always try to achieve a balance between the areas of the works that were published in RIDNAIC (Geotechnical, Hydraulics, Hydrology, Environmental Engineering, Construction Management, Transportation and Structures, among others). I recognize that this task was not always easy, given that the academic and professional background of the two editors was in Structural Engineering, but we believe that we were able to achieve this equity. One way to achieve balance consisted in the publication of special editions dedicated to a specific area, such as Transportation and Geotechnical, with distinguished guest editors for those occasions. Since we began publishing RIDNAIC 20 years ago, the world of technical journals has undergone major changes, and in my opinion, many of them negative. Those of us who are dedicated to academia know of numerous “journals” that have proliferated in recent years thanks to the ubiquity of the internet. I am sure that many readers get, through various platforms, invitations to read articles that do not even have the level of a project of a graduating class. To add insult to injury, many of them don't seem to care that the material they post is original either. RIDNAIC began to grow and was widely accepted by the civil engineering and related community in Latin America and the Hispanic community in the United States. During the first stage, we received jobs from Argentina, Chile, Colombia, Cuba, Brazil, Ecuador, Spain, the United States, Mexico, Peru, the Dominican Republic, Venezuela, and Puerto Rico. For several years, RIDNAIC was published on paper and electronically. With each issue, around 1,500 copies were distributed free of charge to subscribers throughout the continent and Spain. Over the years and the popularization of the internet, the journal began to focus more and more on its online version. The rising costs of printing, and especially shipping costs, together with a difficult economic situation at the University of Puerto Rico, forced first to suspend shipments abroad, and then within Puerto Rico. The paper version continued, but was only distributed to libraries, local government agencies related to civil infrastructure, and universities. The printed version was later completely canceled. This tendency to abandon the traditional paper version is today shared by the vast majority of technical journals. At the initiative of Dr. Luis Godoy, the online version of RIDNAIC was transferred to the electronic platform for academic publications Scipedia, physically located in Barcelona, Spain. Scipedia is an initiative promoted by the International Center for Numerical Methods in Engineering (CIMNE) of the Polytechnic University of Catalunya to promote the publication of technical articles in the “Open Access” format. This decision guaranteed greater dissemination to RIDNAIC through the internet. I am pleased to announce that RIDNAIC will continue to be associated with Scipedia, which has gained worldwide recognition for the quality and depth of its publications. Another important event in the history of the publication was the retirement of Dr. Luis Godoy from the University of Puerto Rico in Mayagüez and the return to his native country, Argentina. Although Dr. Godoy continued working from his native Córdoba, his physical absence was noted. Due to my other responsibilities at the University that merited priority and with the agreement of Dr. Luis Godoy, we made the decision to put RIDNAIC on temporary pause. At the time I am writing this reflection, the whole world has already been going through a period of extreme difficulty due to the COVID-19 pandemic for many months, although hopes are on the horizon. However, this global tribulation has made us wake up to a new reality and become more aware of the fragility of our planet. Global climate change, which some irresponsible \"leaders\" insist on denying, finally clings to us. Phenomena such as the rise in sea level, the increase in the severity and frequency of atmospheric events, and other dire consequences require our immediate attention. Technological advances in science and engineering and the unity of purpose of our engineering community can contribute to solving these pressing problems of global relevance: it must be our commitment. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 10

INFRAESTRUCTURA 2030 Juan F. Alicea Flores, PE Presidente, Colegio de Ingenieros y Agrimensores de Puerto Rico Puerto Rico lleva más de una década de inmerso en una depresión económica. En los últimos siete años el tema de mayor discusión es sobre la deuda pública y del ajuste que hay que hacer en los gastos para poder pagar la deuda misma. Sin embargo, muy poco se ha discutido sobre acciones afirmativas a considerarse que permitan reactivar la actividad económica de nuestro Puerto Rico. Si no se logra reactivar la actividad económica, cualquier deuda será imposible de pagar no importa lo reducida que sea. J. Alicea- Por otro lado, tenemos que reconocer que la infraestructura de Puerto Rico en Flores los años sesenta facilitó un desarrollo económico sostenible que generó las riquezas y la calidad de vida de la cual todavía disfrutamos hoy. Sin embargo, no hay que ser un experto para saber que la infraestructura de los servicios esenciales no está en su mejor momento. Un informe publicado por la Sociedad Americana de Ingenieros Civiles (ASCE), otorgó una clasificación general de (D-) a ocho categorías de infraestructura en Puerto Rico para el año 2019 (ASCE, 2019). La agrupación de dicha infraestructura en D+, D, D- y F es la siguiente: condición de puentes, represas y manejo de aguas negras (D+); condición infraestructura de agua potable e infraestructura de puertos (D); Condición de las carreteras y manejo desperdicios sólidos (D-); y la condición infraestructura de energía (F). Los efectos del terremoto del 7 de enero del 2020 confirmaron que un número significativo de nuestras edificaciones utilizadas para ofrecer servicios esenciales tales como: escuelas, hospitales, aeropuertos, parques de bombas, cuarteles de la policía, represas de aguas, centrales generatrices de energía eléctrica, y una gran cantidad de viviendas privadas son altamente vulnerables a la actividad sísmica. Cabe resaltar que una publicación del corriente año 2020 sobre la inversión de 168 países alrededor del mundo en su infraestructura, reflejó que Puerto Rico, contrario a los años sesenta, donde Puerto Rico ocupaba una posición competitiva es su inversión capital destinado a la infraestructura, en el 2007 ocupó la posición 157 de 168 de inversión realizaba en su infraestructura esencial con solamente 13% de su PIB y la posición 168 de 168 (última posición) en el 2017, con solamente un 8% de su PIB (“The Global Economy”, 2020). Estas alarmantes estadísticas confirman que el deterioro significativo en la condición de la infraestructura en Puerto Rico no es casualidad, sino el efecto del resultado de políticas públicas y procesos de planificación deficientes. La interrupción del servicio de energía eléctrica a casi la mitad de los clientes de la Autoridad de Energía Eléctrica (AEE), tras el paso de una tormenta de baja intensidad y una cantidad de lluvia moderada, es indicativo que la infraestructura de energía eléctrica en Puerto Rico está en su peor condición en las últimas cinco décadas. La condición de la infraestructura de agua potable no está muy lejos de la de energía eléctrica debido a que, aun cuando los meses de enero y febrero de 2020 son los de mayor actividad de precipitación en la isla, al no llover suficiente durante los meses de marzo, abril y mayo requirió implantar planes de racionar el agua potable debido a la falta de agua almacenada. Este comportamiento es uno similar a países de bajo desarrollo económico. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 11

La condición evidente de las vías públicas principales, junto al enmarcado y la rotulación, demuestra que el sistema de infraestructura vial está en su peor estado en las últimas tres décadas. Por otro lado, los controles administrativos para reducir las posibilidades de contagios, adoptados por el gobierno de Puerto Rico en las operaciones esenciales requirieron mayor utilización de la tecnología y controles remotos. Esto reflejó que la infraestructura de comunicaciones no está a la altura que creíamos estaba. Las deficiencias en la informática son similares en el servicio público como sector privado. Sin embargo, ante esta triste realidad, tenemos una oportunidad única de reconstruir toda nuestra infraestructura utilizando los objetivos de desarrollo sostenible recomendados por las Naciones Unidas. En nuestro caso, tenemos una responsabilidad directa bajo los objetivos #3 Salud y bienestar, #7 Energía asequible y no contaminante, #9 Industria, innovación e infraestructura, #11 Ciudades y comunidades sostenibles y la #13 Acción por el clima. (Naciones Unidas, 2020) Por otro lado, el Gobierno Federal se ha comprometido con el Gobierno de Puerto Rico realizando asignaciones para reconstruir esta infraestructura con una cantidad que supera los $70,000 millones de dólares. Es pertinente que el Gobierno de Puerto Rico elabore un plan ordenado y estructurado, considerando la economía global, la última versión de códigos de edificación y las mejores prácticas de la industria. Puerto Rico cuenta con el peritaje técnico para dirigir y desarrollar este proceso de reconstrucción de la infraestructura del país de una manera sostenible, segura y resiliente durante esta década. El Colegio de Ingenieros y Agrimensores de Puerto Rico (CIAPR), como parte de la iniciativa Infraestructura 2030, en conjunto con alianzas estratégicas con centros de investigación y desarrollo, y otras iniciativas de las universidades públicas y privadas del país, continuará aportando a este desarrollo estratégico, y seguirá siendo facilitador del Gobierno de Puerto Rico ante estos retos y desafíos con el firme propósito de proveer mejor calidad de vida y bienestar a presentes y futuras generaciones. REFERENCIAS ASCE (2019). 2019 Report Card for Puerto Rico’s Infrastructure: American Society of Civil Engineers Puerto Rico Section. https://www.infrastructurereportcard.org/wp-content/uploads/2019/11/2019-Puerto-Rico- Report-Card- Final.pdf. The Global Economy (2017). Por el piso la inversión en infraestructura en Puerto Rico. https://public.flourish.studio/visualisation/1610221/?utm_source=showcase&utm_campaign=visualis ation/1610221 Naciones Unidas (2020). Objetivos de desarrollo sostenible. https://www.un.org/sustainabledevelopment/es/objetivos-de-desarrollo-sostenible/ Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 12

THE NEED FOR RESILIENT INFRASTRUCTURE Dennis D. Truax, Ph.D., P.E., DEE, D.WRE, F.ASCE, F.NSPE Director of the Richard A. Rula School of Civil and Environmental Engineering, James T. White Chair, Department Head, and Professor Mississippi State University D. Truax There is little dispute that we live in a time of change, a world of change. The increase in frequency and severity of natural events and human aggression have repeatedly resulted in natural and man-made disasters on our society. (Olsen, 2015) These events have also repeatedly demonstrated the vulnerability of infrastructure designed when such events were not considered significant enough to include in the design of these systems. As a result, those who plan, design, construct, maintain and operate the infrastructure, which is so critical to the quality of life and socioeconomic condition in our communities, must address infrastructure vulnerability in order to make it safer, secure, sustainable, and more resilient. The systems upon which we rely to provide safe transportation, clean water, secure structures, and reliable energy and telecommunications are increasingly frail as they reach the end of their design life. However, this need comes at a time when natural and monetary resources have become increasingly limited. Furthermore, many of the methods and materials traditionally used have an adverse environmental impact which only serve to exacerbate issues of resource scarcity and a changing climate. It is not enough to simply replace the old built infrastructure with similar appurtenances. Limitations on natural and fiscal resources, a changing climate, and expectations of modern infrastructure to have a low life-cycle cost and increased longevity are making the ways of the past obsolete. Today’s engineers need to consider innovative design and construction approaches incorporating new materials and emerging technologies. We must ensure the next generation of infrastructure systems, not only meets nominal expectations, but has the resilience to perform during, and recover rapidly after exposure to disastrous conditions. Recognizing this reality, the American Society of Civil Engineers (ASCE) has established the Grand Challenge as a way of emphasizing the need for resilient infrastructure upon which our society depends. In support of this initiative, ASCE has established roughly two dozen Policy Statements (PS) related to infrastructure development and disaster mitigation. PS 500 - Resilient Infrastructure Initiatives (ASCE, 2020a) was adopted to “…support initiatives that increase resilience of infrastructure against man-made and natural hazards through education, research, planning, design, construction, operation and maintenance. Development of performance criteria and uniform national standards that address interdependencies and establish minimum performance goals for infrastructure is imperative.” This policy goes on to note that “…an all-hazard, comprehensive risk assessment that considers event likelihood and consequence, encourages mitigation strategies, monitors outcomes, and addresses recovery and return to service should be routinely included in the planning and design processes for infrastructure at all government levels.” Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 13

Similarly, PS 518 – Unified Definitions for Critical Infrastructure Resilience (ASCE, 2020b) establishes consistent terminology to allow a clear characterization of critical infrastructure, hazards, and resilience. In this vocabulary, resilience refers to “…the capability to mitigate against significant all-hazards risks and incidents and to expeditiously recover and reconstitute critical services with minimum damage to public safety and health, the economy, and national security.” In continued response to the need to promote innovative programs leading to planning, design, construction, operation, and maintenance of robust infrastructure systems at the national and international arena, ASCE formed research consortia under the Committee for Technical Advancement (CTA). These collaborative groups include the Infrastructure Resilience Division (IRD), the Cold Regions Engineering Division (CRED), and the Committee on Adaptation to a Changing Climate (CACC). The charge for these groups is to develop innovative products and services to advance practices related to resilient civil infrastructure and lifeline systems. Others are beginning to recognize the need for change in the way infrastructure is conceived. In 2016, California’s governor appointed and established the Climate-Safe Infrastructure Working Group (CSIWG) to bring together the state’s scientific experts, engineers, and architects from multiple scientific and infrastructure disciplines to examine how climate change impacts can be included in infrastructure planning, design, and implementation processes. The working group published a report entitled “Paying It Forward: The Path Toward Climate-Safe Infrastructure in California” (California Natural Resources Agency, 2018). CSIWG’s deliberations resulted in a broad-reaching set of recommendations to chart a path toward investment in climate-safe infrastructure. The report addresses the infrastructure that was built decades, even a century, ago – from historical bridges to major dams, highways, and buildings – and the infrastructure that will be built in the coming years and is meant to last for many decades to come. In starting this undertaking, it was thought they might solve the challenges of incorporating forward-looking climate information into infrastructure design. However, they found that this goal has been something engineers and architects have struggled with for decades. As a result, CSIWG discovered that the science challenge in moving toward climate-safe infrastructure may be significant, but it is not obdurate and substantive progress has already been made. However, they also recognize that a more problematic challenge is a required change in the society’s paradigm regarding corporate values, community thinking, priority setting, and policy commitments (California Natural Resources Agency, 2018). Those professionals actively working on delineating appropriate design and operation constraints are attempting to provide guidance for, and contributes to, the developing or enhancing of methods for infrastructure analysis and design for a world in which risk profiles are changing. These changes can be projected with varying degrees of uncertainty requiring an innovative design philosophy to meet the challenge. The underlying approaches are based on probabilistic methods for quantitative risk assessment and a focus on identifying and analyzing adaptive strategies to make a project more resilient while minimizing the collateral impacts of construction and operation (Ayyub, 2018). This process requires an overview of the forces and hazards associated with changing infrastructure design requirements. One must become aware of the potential for extreme events, natural and man-made. Finally, to develop truly resilient infrastructure systems, one must employ adaptive design and risk management in the context of life- cycle cost analysis (LCCA) based on the triple bottom line (the true cost in terms of economics, society, and environment). The result of these efforts will emphasize the requirement to use new sustainable materials and design approaches. As one example, the application of “big data” to the planning, design, operation, and maintenance of infrastructure will help create the “smart city” of the future aligned with the United Nation’s Sustainable Development Goals (SDGs) (UN,2020). Already underway, researchers are exploring ways of exploiting the power of big data toward enhancing the community resilience against extreme events. One lesson learned has been the need for agencies to interact and freely exchange data to assist decisionmakers and stakeholders. This is integral to the data revolution. Otherwise, data is siloed, archived, and useless to those who must make change happen in developing the next generation of resilient infrastructure systems (NSPARC, 2019). Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 14

This is equally important and pertinent in managing and protecting people and infrastructure in extreme events. For a truly resilient infrastructure design to be possible, there are four critical questions to be answered: (1) how to anticipate the event, (2) how to prepare for the event, (3) how to respond to the event, and (4) how to recover from the event. The importance of using big data to assist in answering these questions was emphasized by the National Science Foundation (NSF) as part of “Harnessing the Data Revolution,” one of 10 Big Ideas for future investments. (NSPARC, 2019) To achieve this will require an integrated framework to identify, harness, synthesize, and distribute pertinent data so it can assist decision-making in multiple sectors before, during, and after an extreme event. From healthcare to transportation, data communicated in a timely and consistent manner can be utilized to mitigate the impacts of an impending disaster. As an example, wild fires threaten the lives and property of individual, but also challenge municipal infrastructure systems. The heat from these fires has the potential to adversely impact PVC water lines resulting in a weakened water delivery system needed to attach the fire causing the problem. This has been obvious, but what is equally noteworthy is that, after the fire is extinguished, the potable water it will deliver is likely to contain a multitude of organic chemicals which pose short-term and long-term health risks. Another example would be the destruction of cell phone towers during intense fires. In the short-term, this will adversely impact local communications, community warning systems, and the ability of those attaching the fire to exchange information. Long-term, it will adversely impact communications access in remote areas, limiting services to those dependent on this modern telecommunications system for safety and commerce. Furthermore, after the fire with grass and trees destroyed, there is an increase probability of landslides causing the destruction of transportation infrastructure and isolating communities in need or resulting in inordinate increases in travel times for extensive periods. In creating resilient infrastructure systems, the decision-makers must recognize that each event cascades in a downstream effect. Hence, it must be a design paradigm to reduce the chain of consequences precipitated by an event though design. One can no longer be satisfied in solving the immediate problem without considering subsequent impacts. For example, the creation of the “wrong side of the tracks” was a process of sound engineering logic focused on one issue and a few collateral considerations. It only seems logical that placing transportation corridors through the center of a community minimizes capital cost of the systems when compared to the alternative of going around the population. A side benefit is that access to the system is “balanced” for those throughout the area. What is forgotten in this design approach is that the system represents a physical barrier which divides communities. The city of Boston recognized this and spent billions of dollars to replace the other “Big Green Monster” of bridges on interstate system I-93 with a central artery tunnel. (Mass.gov, 2020) However, one must also recognize that those on the leeward side of this infrastructure system will be subjected to more pollution. The resulting particulate matter, air pollution, and noise adversely impacts the quality of life. Those having the socioeconomic standing to move from the area do so, and those who do not are forced to stay. Subsequently, an infrastructure system that should have been of great benefit to all creates a class-based society. So, in summary, Darwin’s “Origin of Species” asserts that it is not the strongest of the species, nor the most intelligent, that survives. Rather, the species that survives is the one that is most adaptable to its changing environment. This have never been truer than today. The infrastructure systems upon which we rely to keep us safe, support our economy, and establish the quality of life desired, must be resilient in the face of a changing planet. Robust systems that survive the extreme events imposed on them by nature or man must be a goal. These systems must be rapidly repaired and returned to service without causing a cascade of subsequent problems for the community or the other infrastructure systems upon which they rely. They must have longer design lives, lower life-cycle costs, as defined by the triple bottom line. These infrastructure systems need to conserve resources and be flexible enough to meet the demands placed on them by escalating populations and increasingly intense disaster events. Our 21st century infrastructure must be resilient, or it will fall well short of our needs and expectation. REFERENCES Ayyub, B. M. (2018). Climate-Resilient Infrastructure: Adaptive Design and Risk Management, Manual of Practice 140, ASCE, Reston, VA, ISBN: 9780784415191, doi.org/10.1061/9780784415191 (2018) ASCE (2020a). “Policy Statement 500 - Resilient Infrastructure Initiatives”. https://www.asce.org/issues-and- advocacy/public-policy/policy-statement-500---resilient-infrastructure-initiatives/. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 15

ASCE (2020b). “Policy Statement 518 - Unified Definitions For Critical Infrastructure Resilience”. https://www.asce.org/issues-and-advocacy/public-policy/policy-statement-518---unified-definitions-for- critical-infrastructure-resilience/. California Natural Resources Agency (2018). Paying It Forward: The Path Toward Climate-Safe Infrastructure in California, https://resources.ca.gov/Initiatives/Building-Climate-Resilience. Mass.gov (2020). “The Big Dig: Background Information, Logistics, and Statistics for the Central Artery Project,” https://www.mass.gov/the-big-dig. NSPARC (2019). “The Need for Integrated Data in Disasters”, NEXUS, Vol. 2, No. 1, pp. 16-20. https://www.nsparc.msstate.edu/wp- content/uploads/2019/04/NEXUS-SPRING- SCREEN- SPREAD.pdf Olsen, J. R. (2015). Adapting Infrastructure and Civil Engineering Practice to a Changing Climate, ASCE, Reston, VA, ISBN: 9780784479193, 2015. UN (2020). “17 Goals to Transform Our World,” https://www.un.org/sustainabledevelopment. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 16

LA NECESIDAD DE UNA INFRAESTRUCTURA RESILIENTE Dennis D. Truax, Ph.D., P.E., DEE, D.WRE, F.ASCE, F.NSPE Director, Colegio de Ingeniería Civil y Ambiental Richard A. Rula Director Cátedra James T. White, Catedrático y Director del Departamento Universidad Estatal de Mississippi Hay poca disputa en que vivimos en un tiempo de cambio, un mundo de cambio. El aumento en la frecuencia y severidad de los eventos naturales y la agresión humana han resultado repetidamente en desastres naturales y causados por el hombre en nuestra sociedad. Estos eventos también han demostrado repetidamente la vulnerabilidad de la infraestructura diseñada cuando dichos eventos no se consideraban lo suficientemente significativos como para incluirlos en el diseño de estos sistemas. Como resultado, quienes planifican, diseñan, construyen, mantienen y operan la infraestructura, que es tan crítica para la calidad de vida y la condición socioeconómica de nuestras comunidades, deben abordar la vulnerabilidad de la infraestructura a fin de hacerla más segura, protegida, sostenible y más resiliente. Los sistemas de los que dependemos para proporcionar transporte seguro, agua limpia, estructuras seguras y energía y telecomunicaciones fiables son cada vez más frágiles a medida que llegan al final de su vida de diseño. Sin embargo, esta necesidad llega en un momento en que los recursos naturales y monetarios se han vuelto cada vez más limitados. Además, muchos de los métodos y materiales tradicionalmente utilizados tienen un impacto ambiental adverso que sólo sirve para exacerbar los problemas de escasez de recursos y el cambio climático. No basta con reemplazar simplemente la infraestructura construida obsoleta con accesorios similares. Las limitaciones de los recursos naturales y fiscales, un clima cambiante y las expectativas de que la infraestructura moderna tenga un bajo costo de ciclo de vida y una mayor longevidad están haciendo obsoletos los caminos del pasado. Los ingenieros de hoy en día necesitan considerar enfoques innovadores de diseño y construcción que incorporen nuevos materiales y tecnologías emergentes. Debemos asegurarnos de que la próxima generación de sistemas de infraestructura, no sólo cumpla con las expectativas nominales, sino que tenga la resiliencia necesaria para funcionar durante, y recuperarse rápidamente después de la exposición a condiciones desastrosas. Reconociendo esta realidad, la Sociedad Americana de Ingenieros Civiles (ASCE) ha establecido el Gran Reto como una forma de enfatizar la necesidad de una infraestructura resiliente de la que nuestra sociedad depende. En apoyo de esta iniciativa, la ASCE ha establecido aproximadamente dos docenas de declaraciones de política (PS) relacionadas con el desarrollo de la infraestructura y la mitigación de desastres. La PS 500 - Iniciativas de Infraestructura Resiliente (ASCE, 2020a) fue adoptada para \"...apoyar las iniciativas que aumenten la resiliencia de la infraestructura contra los peligros naturales y causados por el hombre mediante la educación, la investigación, la planificación, el diseño, la construcción, el funcionamiento y el mantenimiento. Es imperativo desarrollar criterios de rendimiento y normas nacionales uniformes que aborden las interdependencias y establezcan objetivos mínimos de rendimiento para la infraestructura\". Esta política continúa señalando que \"...una evaluación de riesgos comprensiva y para todos los peligros, que considere la probabilidad y las consecuencias del evento, fomente las estrategias de mitigación, supervise los resultados y aborde la recuperación y el retorno al servicio, debería incluirse de forma rutinaria en los procesos de planificación y diseño de la infraestructura en todos los niveles de gobierno\". Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 17

De forma similar, la PS 518 - Definiciones Unificadas para la Resiliencia de las Infraestructuras Críticas (ASCE, 2020b) establece una terminología consistente para permitir una clara caracterización de las infraestructuras críticas, los peligros y la resiliencia. En este vocabulario, la resiliencia se refiere a \"...la capacidad de mitigar contra los riesgos e incidentes significativos de todo tipo y, de recuperar y reconstituir rápidamente los servicios críticos con un mínimo de daños a la seguridad y la salud pública, la economía y la seguridad nacional\". En respuesta continua a la necesidad de promover programas innovadores que condujeran a la planificación, diseño, construcción, funcionamiento y el mantenimiento de sistemas de infraestructura robustos en el ámbito nacional e internacional, la ASCE formó consorcios de investigación bajo el Comité para el Avance Técnico (CTA). Estos grupos de colaboración incluyen la División de Resiliencia de la Infraestructura (IRD), la División de Ingeniería de Regiones Frías (CRED), y el Comité de Adaptación a un Clima Cambiante (CACC). La encomienda de estos grupos es desarrollar productos y servicios innovadores para fomentar las prácticas relacionadas con la infraestructura civil resiliente y los sistemas vitales. Otros están comenzando a reconocer la necesidad de un cambio en la forma de concebir la infraestructura. En el 2016, el gobernador de California nombró y estableció el Grupo de Trabajo de Infraestructura Segura para el Clima (CSIWG) para reunir a los expertos científicos, ingenieros y arquitectos del estado de múltiples disciplinas científicas y de infraestructura para examinar cómo los impactos del cambio climático pueden ser incluidos en los procesos de planificación, diseño e implementación de la infraestructura. El grupo de trabajo publicó un informe titulado \"Paying It Forward: The Path Toward Climate-Safe Infrastructure in California\" (California Natural Resources Agency, 2018). Las deliberaciones del CSIWG dieron lugar a un conjunto de recomendaciones de amplio alcance para trazar el camino hacia la inversión en infraestructura segura para el clima. El informe aborda la infraestructura que se construyó hace décadas, incluso un siglo - desde puentes históricos, hasta grandes represas, carreteras y edificios - y la infraestructura que se construirá en los próximos años y que se pretende que dure muchas décadas más. Al iniciar este emprendimiento, se pensó que podrían resolver los desafíos de incorporar la información climática de futuro en el diseño de la infraestructura. Sin embargo, encontraron que este objetivo ha sido algo con lo que ingenieros y arquitectos han luchado durante décadas. Como resultado, el CSIWG descubrió que el desafío científico de avanzar hacia una infraestructura segura para el clima puede ser significativo, pero no es obstinado y ya se han hecho progresos sustanciales. Sin embargo, también reconocen que un desafío más problemático es un cambio necesario en el paradigma de la sociedad en lo que respecta a los valores corporativos, el pensamiento de la comunidad, el establecimiento de prioridades y los compromisos políticos (California Natural Resources Agency, 2018). Los profesionales que trabajan activamente para delinear restricciones apropiadas del diseño y funcionamiento intentan proporcionar guías y que contribuyan al desarrollo o la mejora de los métodos de análisis y diseño de infraestructuras en un mundo en el que los perfiles de riesgo están cambiando. Estos cambios pueden proyectarse con diversos grados de incertidumbre y requieren una filosofía de diseño innovadora para hacer frente al desafío. Los enfoques subyacentes se basan en métodos probabilísticos para la evaluación cuantitativa de los riesgos y se centran en la identificación y el análisis de estrategias de adaptación para que un proyecto sea más resiliente y al mismo tiempo se minimicen los impactos colaterales de la construcción y el funcionamiento (Ayyub, 2018). Este proceso requiere una visión general de las fuerzas y peligros asociados a los cambios en los requisitos de diseño de la infraestructura. Hay que tomar conciencia del potencial de los fenómenos extremos, tanto naturales como provocados por el hombre. Por último, para desarrollar sistemas de infraestructura verdaderamente resilientes, se debe emplear el diseño adaptativo y la gestión de riesgos en el contexto del análisis del costo del ciclo de vida (LCCA) basado en el triple resultado final (el costo real en términos económicos, a la sociedad y al ambiente). El resultado de esos esfuerzos pondrá de relieve la necesidad de utilizar nuevos materiales y enfoques de diseño sostenibles. Por ejemplo, la aplicación de \"big data\" a la planificación, el diseño, el funcionamiento y mantenimiento de la infraestructura ayudará a crear la \"ciudad inteligente\" del futuro alineados con los Objetivos de Desarrollo Sostenible de las Naciones Unidas (SDGs) (UN, 2020). Ya en marcha, los investigadores están explorando formas de Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 18

explotar el poder de los grandes datos para mejorar la resiliencia de la comunidad contra los eventos extremos. Una de las lecciones aprendidas ha sido la necesidad de que los organismos interactúen e intercambien libremente datos para ayudar a los encargados de la adopción de decisiones y a las partes interesadas. Esto es parte integral de la revolución de los datos. De lo contrario, los datos se almacenan en silos, se archivan y son inútiles para quienes deben hacer que el cambio se produzca en el desarrollo de la próxima generación de sistemas de infraestructuras resilientes (NSPARC, 2019). Esto es igualmente importante y pertinente en la gestión y protección de las personas y la infraestructura en los eventos extremos. Para que sea posible un diseño de infraestructuras verdaderamente resilientes, hay cuatro preguntas críticas que deben ser respondidas: 1) cómo anticipar el evento, 2) cómo prepararse para el evento, 3) cómo responder al evento y 4) cómo recuperarse del mismo. La importancia de utilizar grandes datos para ayudar a responder a estas preguntas fue enfatizada por la Fundación Nacional de Ciencias (NSF) como parte del \"Aprovechamiento de la revolución de los datos\", una de las 10 grandes ideas para futuras inversiones. (NSPARC, 2019) Para lograrlo se necesitará un marco integrado para identificar, aprovechar, sintetizar y distribuir los datos pertinentes de manera que puedan ayudar a la toma de decisiones en múltiples sectores antes, durante y después de un evento extremo. Desde el punto de vista del cuidado de la salud hasta el transporte, los datos comunicados de manera oportuna y consistente pueden utilizarse para mitigar los efectos de un desastre inminente. Por ejemplo, los incendios forestales amenazan la vida y la propiedad de las personas, pero también desafían los sistemas de infraestructura municipal. El calor de estos incendios tiene el potencial de impactar negativamente las líneas de agua de PVC, lo que resulta en un sistema de suministro de agua debilitado, necesario para sujetar el fuego que causa el problema. Esto ha sido obvio, pero lo que es igualmente digno de mención es que, una vez extinguido el fuego, es probable que el agua potable que entregará contenga una multitud de productos químicos orgánicos que plantean riesgos para la salud a corto y largo plazo. Otro ejemplo sería la destrucción de las torres de telefonía móvil durante los incendios intensos. A corto plazo, esto afectará adversamente a las comunicaciones locales, a los sistemas de alerta de las comunidades y a la habilidad de los que se encargan del fuego para intercambiar información. A largo plazo, afectará adversamente al acceso de las comunicaciones en zonas remotas, limitando los servicios a quienes dependen de este moderno sistema de telecomunicaciones para su seguridad y comercio. Además, después del incendio con la vegetación y árboles destruidos, aumenta la probabilidad de que los deslizamientos de tierra causen la destrucción de la infraestructura de transporte y aíslen a las comunidades necesitadas o den lugar a aumentos desmesurados de los tiempos de viaje durante periodos extensos. Al crear sistemas de infraestructura resiliente, los encargados de la adopción de decisiones deben reconocer que cada acontecimiento tiene un efecto en cascada. Por lo tanto, debe ser un paradigma de diseño para reducir la cadena de consecuencias precipitadas por un evento a través del diseño. Ya no se puede estar satisfecho con la solución del problema inmediato sin considerar los impactos posteriores. Por ejemplo, la creación del \"lado equivocado de los rieles\" fue un proceso de sólida lógica de ingeniería centrado en una cuestión y unas pocas consideraciones colaterales. Sólo parece lógico que la colocación de corredores de transporte a través del centro de una comunidad minimiza el costo de capital de los sistemas cuando se compara con la alternativa de ir alrededor de la población. Un beneficio colateral es que el acceso al sistema está \"balanceado\" para los que están en toda la zona. Lo que se olvida en este enfoque de diseño es que el sistema representa una barrera física que divide a las comunidades. La ciudad de Boston reconoció esto y gastó miles de millones de dólares para reemplazar el otro \"Gran Monstruo Verde\" de puentes en el sistema interestatal I-93 con un túnel que forma parte de la arteria central. (Mass.gov, 2020) Sin embargo, también hay que reconocer que los que están en el lado de sotavento de este sistema de infraestructura estarán sujetos a más contaminación. Las partículas resultantes, la contaminación del aire y el ruido afectan adversamente a la calidad de vida. Los que tienen la capacidad socioeconómica para desplazarse de la zona lo hacen, y los que no lo hacen se ven obligados a quedarse. Posteriormente, un sistema de infraestructura que debería haber sido de gran beneficio para todos crea una sociedad basada en clases. Así que, en resumen, el \"Origen de las especies\" de Darwin afirma que no es la más fuerte de las especies, ni la más inteligente, la que sobrevive. Más bien, las especies que sobreviven son las que se adaptan mejor a su entorno cambiante. Esto nunca ha sido más cierto que hoy en día. Los sistemas de infraestructura de los que dependemos para mantenernos seguros, apoyar nuestra economía y establecer la calidad de vida deseada, deben ser resilientes ante un planeta cambiante. Sistemas robustos que sobrevivan a los eventos extremos impuestos por la naturaleza o el hombre deben ser una meta. Estos sistemas deben ser rápidamente reparados y puestos de nuevo en servicio sin causar Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 19

una cascada de problemas posteriores para la comunidad o los otros sistemas de infraestructura de los que dependen. Deben tener vidas de diseño más largas, menores costos de ciclo de vida, como se define en el triple resultado final. Estos sistemas de infraestructura deben conservar los recursos y ser lo suficientemente flexibles como para satisfacer las demandas que les imponen el aumento de la población y los desastres cada vez más intensos. Nuestra infraestructura del siglo XXI debe ser resiliente, o se quedará muy corta en relación con nuestras necesidades y expectativas. REFERENCIAS Ayyub, B. M. (2018). Climate-Resilient Infrastructure: Adaptive Design and Risk Management, Manual of Practice 140, ASCE, Reston, VA, ISBN: 9780784415191, doi.org/10.1061/9780784415191 (2018) ASCE (2020a). “Policy Statement 500 - Resilient Infrastructure Initiatives”. https://www.asce.org/issues-and- advocacy/public-policy/policy-statement-500---resilient-infrastructure-initiatives/. ASCE (2020b). “Policy Statement 518 - Unified Definitions For Critical Infrastructure Resilience”. https://www.asce.org/issues-and-advocacy/public-policy/policy-statement-518---unified-definitions-for- critical-infrastructure-resilience/. California Natural Resources Agency (2018). Paying It Forward: The Path Toward Climate-Safe Infrastructure in California, https://resources.ca.gov/Initiatives/Building-Climate-Resilience. Mass.gov (2020). “The Big Dig: Background Information, Logistics, and Statistics for the Central Artery Project,” https://www.mass.gov/the-big-dig. NSPARC (2019). “The Need for Integrated Data in Disasters”, NEXUS, Vol. 2, No. 1, pp. 16-20. https://www.nsparc.msstate.edu/wp- content/uploads/2019/04/NEXUS-SPRING- SCREEN- SPREAD.pdf Olsen, J. R. (2015). Adapting Infrastructure and Civil Engineering Practice to a Changing Climate, ASCE, Reston, VA, ISBN: 9780784479193, 2015. UN (2020). “17 Goals to Transform Our World,” https://www.un.org/sustainabledevelopment. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 20

COLLAPSE OF THE ARECIBO OBSERVATORY IN PUERTO RICO: REFLECTIONS FROM A STRUCTURAL ENGINEERING PERSPECTIVE1 Juan C. Morales2 and Luis E. Suárez-Colche3 ABSTRACT: The suspended platform of the Arecibo Observatory collapsed on December 1, 2020, after 57 years of illustrious service to the scientific community. This article reflects on the collapse from a structural engineering perspective. It explores issues that most likely contributed, including the cable terminations and their performance under loading, corrosion, fatigue, safety factors, the dynamic demands on the remaining cables when one of them suddenly snaps, and the dynamics of the suspended platform during its pendular swing. The article is based on publicly available reports from the news media, photographs, and published video recordings of the collapse sequence. These were expanded with relevant studies collected from the literature and with publicly available technical reports from structural engineering consultants that evaluated the structure during its final months. It is organized as a sequence of events, starting with the first cable failure on August 10, 2020, the second failure on November 6, 2020, and the final collapse. Despite the limitations, it is concluded that corrosion does not seem to have played a significant role, and that the failure of the first cable was most likely due to flaws during the fabrication of the cable/socket connection. These will be reviewed and updated once the forensic analyses of the failed elements are concluded and made public. Keywords: Arecibo Observatory, collapse, fatigue, corrosion, socket COLAPSO DEL OBSERVATORIO DE ARECIBO EN PUERTO RICO: REFLEXIONES DESDE UNA PERSPECTIVA DE INGENIERÍA ESTRUCTURAL RESUMEN: La plataforma suspendida del Observatorio de Arecibo colapsó el 1 de diciembre de 2020, luego de 57 años de ilustre servicio a la comunidad científica. Este artículo reflexiona sobre el colapso desde una perspectiva de ingeniería estructural. Explora los problemas que probablemente contribuyeron, incluyendo las terminaciones de los cables y su rendimiento bajo carga, la corrosión, la fatiga, los factores de seguridad, las demandas dinámicas de los cables restantes cuando uno de ellos se rompe repentinamente, y la dinámica pendular de la plataforma suspendida durante el colapso final. El artículo se basa en informes disponibles públicamente de los medios de comunicación, fotografías, y grabaciones públicas de video de la secuencia del colapso. Estos se ampliaron con estudios relevantes recopilados de la literatura y con informes técnicos disponibles públicamente de consultores de ingeniería estructural que evaluaron la estructura durante sus últimos meses. Está organizado como una secuencia de eventos, comenzando con la primera falla del cable el 10 de agosto de 2020, la segunda falla el 6 de noviembre de 2020, y el colapso final. A pesar de las limitaciones, se concluye que la corrosión no parece haber jugado un papel significativo, y que la falla del primer cable probablemente se debió a fallas durante la fabricación de la conexión del cable con su terminación. Éstos serán revisados y actualizados cuando se concluyan y se hagan públicos los análisis forenses de los elementos fallidos. Palabras clave: Observatorio de Arecibo, colapso, fatiga, corrosión, terminación INTRODUCTION The suspended platform of the iconic Arecibo Observatory collapsed catastrophically on December 1, 2020, shortly before 8:00 am (Drake, 2020). The observatory, also known as the National Astronomy and Ionosphere Center (NAIC), generated a long list of illustrious scientific accomplishments since it was inaugurated on November 1, 1963, most notably the 1993 Nobel Prize in Physics that was awarded to Russell Hulse and Joseph Taylor. Their radio astronomy work in Arecibo with a binary pulsar provided a strict test of Einstein’s Theory of General Relativity and revealed the first evidence for the existence of gravitational waves (NAIC, 2020a). 1Article received on December 22, 2020 and accepted for publication on December 27, 2020. 2Professor and Head of Mechanical Engineering, Ana G. Méndez University, Gurabo Campus, PO Box 3030, Gurabo, Puerto Rico 00778. Email: [email protected] 3Professor, Civil Engineering and Surveying Department, University of Puerto Rico-Mayaguez (UPRM), Puerto Rico 00681-9041. Email: [email protected] Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 21

The objective of this article is to reflect on the collapse from a structural engineering perspective. It explores issues that most likely contributed, including the cable terminations, corrosion, fatigue, safety factors, the dynamic response of the cables when one of them suddenly snaps, and the dynamics of the suspended platform during its pendular swing. It is based on limited information, mainly photographs, videos, and reports from the news media. These were expanded with relevant information collected from the literature and publicly available reports from structural engineering consultants of the Arecibo Observatory. The authors are familiar with the structure. The second author served as the advisor of the first author on his PhD dissertation that investigated the dynamic properties and seismic response of the Arecibo Observatory using modern computational techniques (Morales, 2006). The article begins with a brief background that identifies the main components of the structure. Then it sequentially addresses the failure of the first cable on August 10, 2020, the failure of the second cable on November 6, 2020, and the final collapse on December 10, 2020. Relevant information is introduced within each of these three events. The article ends with a conclusions section. The conclusions will be reviewed and updated in the near future, once the forensic analyses are completed and made public. BACKGROUND OF THE STRUCTURE The selected site for the observatory was a sinkhole in the mountains of Arecibo, Puerto Rico, where the limestone had collapsed almost perfectly to accommodate the 1000 ft-diameter primary reflector. Three towers were erected around the perimeter of the sinkhole and were numbered based on their position around an imaginary clock – 4, 8 and 12 – with tower 12 on the North (Figure 1(a)). The triangular platform was hoisted approximately 500 ft and suspended with cables extending from the towers. Finally, the spherical primary reflector was assembled in the sinkhole. The platform housed the electronic instrumentation that emitted and received radio waves that bounced off the 1000 ft-diameter dish. A suspended catwalk bridge and waveguide (Figure 1(b)), located near tower 12 and the control room, provided access to the platform for personnel and a path for power and instrumentation cables (NAIC, 2020b; NAIC, 2020c). The three towers reached the same height; however, tower 8 was 120 feet taller because it was founded at the foot of a hill (Figure 1(c)). The towers had a cruciform cross section which gave them two equally strong axes, one of them aligned with the cables. The cross-sectional dimensions changed approximately every 60 feet in height as the towers were stepped in by approximately 3 feet on all sides while the thickness remained constant at 6 feet. The 900-ton platform was suspended by 18 cables (6 per tower) whose tension was counteracted by 21 backstay cables (7 per tower). Four main cables (3.0-inch diameter) extended from a saddle on each tower to a corner of the platform. The main cables and the saddle dated from 1963. In addition, two auxiliary cables (3.25-inch diameter) extended from a collar on each tower to the 2/3 points on the sides of the platform. The auxiliary cables and collar, which dated from 1996, assisted in supporting the weight added during the Gregorian Dome upgrade and provided lateral support. Figures 1(b) and 1(d) show the arrangement of the main and auxiliary cables. Also, there was one service cable extending from each tower that connected near (not attached to) the platform and provided a support for the upper end of the catwalk bridge. The service cable is identified as “SC” in Figure 1(b). Figure 1(b) also shows the tuned mass dampers on the cables used to reduce vibrations that cause fatigue loading. Figure 3(a) shows the saddle and collar that anchor the cables on tower 4. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 22

Figure 1: Several views of the structure that identify its main components. Taken from Morales (2006) except (b) which was taken from https://ecommons.cornell.edu/handle/1813/33229. FAILURE OF THE FIRST CABLE ON AUGUST 10, 2020 The first press release from the Arecibo Observatory (NAIC, 2020d) reported that one of the auxiliary cables of Tower 4 broke on August 10 at 2:45 am. A later press release (NAIC, 2020e) reported that the break was caused by the auxiliary cable slipping out from its socket (anchorage) at tower 4. The structural engineering of record confirmed it: “…the cable pulled from its socket and fell…” (Thornton Tomasetti, 2020). The cable is identified as “Aux 1” in Figure 1(b). This section explores the corrosion state of the failed cable, the typical process to fabricate a cable/socket connection, the performance and capacity of sockets under loading, potential fatigue zones to be examined, and the dynamic consequences of the sudden failure on the remaining cables. Corrosion Figure 2(a) shows a photograph of the 3.25-inch diameter cable as it lay on the dish after the failure. It shows that the cable was fabricated by wrapping a large number of small-diameter galvanized wires in a single strand to the Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 23

desired cable diameter. The wires appear in very good condition in terms of corrosion. There are some regions of whitish corrosion products of the protective zinc coat that seem to classify between Stages 1 and 2 of the Hopwood and Havens classification system for corrosion of galvanized wire, shown in Figure 2(b). There are no brownish discoloration zones that would indicate corrosion of the steel inside the zinc coat. These would have classified as Stage 3 or 4 of corrosion of the wires, but none are detected from the photograph. Figure 2(a) also shows the cylindrical-shaped end of the cable near the top of the photograph. The length of the cylindrical end is approximately three times the cable diameter. It seems mostly intact and free of corrosion. The wires on the end appear to have been unwound and separated during the brooming process (addressed in the next section). Based on its cylindrical shape, the cable clearly appears to have failed by slipping out of its socket anchorage, as reported. However, there are also several unattached wires surrounding the broomed end which will be addressed later. Figure 2: (a) Photograph of the auxiliary cable after failure. Taken from Wall (2020). (b) Photographs of the four stages of galvanized wire corrosion based on the Hopwood and Havens classification system. Taken from Chavel and Leshko (2012). Identification of socket Figure 3(a) shows a photograph of the top of tower 4 at the instant that the remaining three main cables failed. It shows the socket of the remaining auxiliary cable below the word “Drone” and includes approximate dimensions that were scaled off using the 3.25-inch diameter cable as the basis. It was presumably identical to the socket involved in the failure; whose anchor hole is seen on the left of the photo. The failed socket had been removed and lowered for forensic analysis. The auxiliary cable entered the socket at the lower-left end, the load was transferred to the socket Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 24

in the ~12 in conical section, and the clevis joint transferred the load to the collar. Slippage occurred from the ~12- inch conical section of the socket. Figure 3(b) shows that typical sockets are also conical on the inside and this internal conical volume is called the “basket” of the socket. The federal specification RR-S-550 Rev. D (1980) specifies a basket whose length is 3.7 times the cable diameter, for the case of the largest diameter cable that they specify (2.625 inch). The updated specification (RR-S-550 Rev. F, 2018) still specifies the same ratio of basket length to cable diameter. If the 3.7 ratio was maintained for the 3.25-inch diameter auxiliary cable, then the basket length would be calculated at 12 inches which agrees with the scaled-off dimension shown in Figure 3(a). Judging from Figure 2(a), the end of the cable that went into the socket appears to be only 3.0 times the cable diameter, i.e., it appears to be shorter than the basket size of the socket. Figure 3: (a) Photograph of the top of tower 4 during final collapse. Taken from the NSF video (2020a). It shows the location of the sockets and their approximate dimensions (scaled off). (b) Typical geometry of a socket that shows its internal conical shaped basket. Taken from Tokyo (2020). Process to fabricate the cable/socket bond Manufacturing flaws in the process of bonding the cable to the socket may have played a part in the pull-out failure. The details of the Arecibo sockets, and the manufacturing process used to bond them to the cables, are unknown to the authors. To the best of the authors knowledge, the spelter (molten metal) used to bond the Arecibo sockets was molten zinc. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 25

Certex (2020, p. 4-11) provides a series of steps for zinc spelter sockets that are summarized below. The steps should be similar to the process used for the Arecibo sockets, and they are included to provide a basis to reflect on potential issues. 1. Measure the length of cable that will be inserted in the socket and seize it with wire at that location (to avoid unwinding the wires below the socket), in preparation for brooming. 2. Prepare the broom at the end of the cable, i.e., unwind and separate the wires all the way down to the seizing. (Figure 4(a)). The brooming procedure increases the surface area available for bonding because each wire participates in transferring load to the socket. The greater the surface area for bonding, the greater the resistance. For example, for a hypothetical 3.25-inch diameter cable fabricated from 76 individual wires with an average diameter of 0.274-inch, the bonding surface area increases by a factor of 6.4 when each individual wire is bonded versus simply bonding the outside diameter of the 3.25-inch cable. 3. Thoroughly degrease, clean, and dry the individual wire strands in the broom. 4. Orient the cable vertically, place it in a vise, and close the broom with seizing cable. Once opened, the broom must be closed temporarily so that it can fit through the hole in the socket. Ensure a “minimum vertical length of rope extending from the socket equal to about 30 cable diameters. This vertical length is necessary for cable balance. Premature wire breaks at the socket can occur if the cable is not balanced at pouring”. 5. Heat the socket to the appropriate temperature and insert the cable in the socket basket (internal cone). The socket is heated to prevent the molten zinc from cooling and solidifying too quickly when the zinc is eventually poured into the basket. The molten zinc must reach the bottom of the basket. “A word of caution: Never heat a socket after it is placed on a cable. To do so may cause damage to the cable”. 6. Untie the broom and separate the wires into the cone shape inside the basket. The wires must extend to the top of the basket (Figure 4(b)). “Use extreme care in aligning the socket with the cable’s centerline”. 7. Pour the molten zinc (spelter) into the socket basket (Figure 4(c)) at a temperature of 950-1000 degrees F. “A word of caution: Overheating of the zinc may affect its bonding properties”. “Pour the zinc in one continuous stream until it reaches the top of the basket and all wire ends are covered”. Allow the zinc to cool. Figure 4: (a) Broom at the end of the cable and the “seizing” (taken from Wirelock, 2017). (b) Socket after inserting the cable and re-separating the broom wires (taken from Wirelock, 2017). (c) Pouring the molten zinc into the socket (taken from Certex, 2020). Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 26

Discussion of manufacturing challenges during the bonding of the cable to the socket The manufacturing process summarized above points at several potential challenges. One of them is size. The Arecibo cable weighed approximately 30 lb/ft and the socket weighed another 200-300 lb. which would require mechanical equipment to handle it, even for the cleaning operations (step 3) which require good workmanship. Size also comes into play while setting up for the zinc pour. A vertical length of 30 cable diameters below the socket is recommended (step 4). For a 3.25-inch diameter cable, a minimum of 8 feet would be required. If four additional feet are added to provide a radius for the cable as it nears the floor, plus 3 feet for the overall length of the socket, the process had to be conducted at approximately 15 feet above the floor of the shop. The socket had to be heated before sliding it into the cable. Additional heat input is not allowed after the cable is inserted because it could damage the cable (step 6). Size comes into play once again. The socket had to be handled relatively quickly during the cable insertion to avoid unfavorable cooling. Then, the temporary seizing on the cable had to be removed, and the wires spread out once again into a cone shaped broom, while working within the hot socket. There seems to be good potential at this point for poor workmanship. If the re-formation of the broom ended up more cylindrical than conical in shape, slippage would be induced. The zinc pour also presents a challenge. If the zinc were overheated, it could have affected its bonding properties (step 7). Size also comes into play because the Arecibo socket had a ~12-inch-long basket that required a substantial volume of molten zinc. Step 7 requires a continuous pour from start to finish. This was not a trivial process. It would have required good planning, a good setup to work at a height of approximately 15 feet above the floor shop, and trials to ensure that the entire process could be conducted appropriately within a specified amount of time to avoid undue cooling of the socket and the molten zinc prior to the pour. Literature review of socket performance under loading Bradon and Ridge (2003) conducted single-wire pull-out laboratory tests for a molten lead-alloy spelter (Arecibo’s spelter was presumably molten zinc and would exhibit similar performance). The objective was to determine if the bonding between the wires and the spelter could be sufficient to hold the wire. They considered various lengths of embedment on a cylindrical (not conical) test piece. The results showed a high degree of scatter; however, in the case of an embedded length of 5.4 times the diameter of the cable, the wire broke before slipping in three out of the four tests that they conducted. They attributed the scatter to three possibilities: incomplete wetting of the entire wire; breakdown of the bond where the wire entered the spelter due to shrinking of the spelter upon cooling; and/or traces of contaminants that prevented full bonding from taking place. Due to the high scatter in the data, they concluded that bonding on its own, although significant, is not sufficiently reliable. In addition, they reported that the load is also transferred by the frictional force developed between the wire and the spelter due to the conical spelter’s wedging action against the confinement provided by the conical socket. They instrumented the outside of the sockets with strain gauges and determined that the load was transferred throughout the entire length of the socket for metal poured sockets. In one of the samples – sample 3 – there was incomplete wetting (bonding) at the entrance to the socket. It resulted in low strain gauge readings at the entry point but higher load transfer at the opposite end, i.e., it compensated. They also found that upon removal of the load, the hoop (circumferential) strains in the socket dropped only slightly, suggesting that “an irreversible ‘wedging’ action was taking place, with the spelter cone moving relative to the socket under load and remaining wedged in this position even when the load was removed”. They attempted to measure the movement (draw) of the spelter cone when loaded; however, their method proved unsuccessful due to “the very small draw values observed in the tests”. As a final note, they only increased the load to approximately 70% of the cable breaking strength since the objective was to investigate the mechanism of load transfer. On the other hand, Metcalf and Matanzo (1980) conducted tests up to failure. They wanted to determine the performance of nine of the best-known cable terminations under static and fatigue loading, including a zinc poured socket similar to the one at Arecibo. They fabricated several specimens of each of the nine types under controlled Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 27

shop conditions and included flawed specimens to evaluate the effect of poor workmanship on the strength of the cable termination. In the case of the zinc poured socket assembly, they investigated poor workmanship due to the elimination of the acid wash of the broomed end. On the pull test (static load) of a 2-inch diameter cable, the zinc poured socket achieved the highest efficiency among the nine terminations. Efficiency was defined as the specimen breaking load divided by the maximum breaking load. A 98% efficiency was achieved with the zinc poured socket (from Table 1 in the article). The failure modes observed for the static test were given in Table 7 of the article and are included here: ● 55% - Multiple strand breaks inside or at base of socket ● 40% - Multiple strand breaks in gage area (away from the socket) ● 5% - Rope pulled out of socket (This was the second-lowest percentage of pull-outs of all nine terminations, with the lowest at 4% and the highest at 9%) Service life tests (fatigue loading) were conducted on a vertical fatigue testing machine which, for the 2-inch diameter cable, applied an oscillating load at a frequency of 12 Hz that ranged over 25% of the Catalog Breaking Strength (CBS) of the cable. The cyclic load was superimposed over a mean stress of 17.5% of the CBS. “For the fatigue tests, failure was defined as twenty broken wires at the base of the termination fitting, or one million cycles, whichever came first”. The zinc poured socket’s efficiency was 99%, including the flawed specimens (from Table 2 in the article). The failure modes observed for the fatigue tests were given in Table 8 of the article and are included here: ● 100% - Multiple strand breaks inside or at base of socket In Table 6 they recommended frequent inspections to search for broken wires at the base of the zinc poured socket. They add: “It is imperative that an inspector have a clear-cut set of instructions to guide him in determining exactly when a particular wire rope termination must be taken out of service. The replacement criterion proposed is that replacement be required if the inspector detects cracks in the pressed sleeve or socket body, or ten (10) broken wires in the adjoining wire rope.” Detached wires around end of failed cable: a fatigue issue? Figure 5(a) zooms in on the broomed end shown in Figure 2(a) and shows approximately 20 detached wires surrounding the broomed end (circled in yellow). Metcalf and Matanzo (1980) observed that 100% of the wire fatigue failures were located either inside or at the base of the socket. These detached wires would therefore be at the predicted location for fatigue failures. If fatigue were the cause, the wire breaks would have taken place before the failure on August 10, 2020 and would have been identifiable. It is unknown to the authors if wire breaks were present before failure. Thornton Tomasetti (2020) mentions “a few documented wire breaks on the original cables over the years”. Another possible mechanism for these detached wires is that they became unbonded or that they broke off due to the impact of the fall from the tower (~500 ft). A forensic inspection of the fracture zones of the cables (the tips) with a Scanning Electron Microscope (SEM) would most probably be able to detect fatigue signatures such as crack initiation sites and “beach marks”. Another interesting area to conduct a forensic inspection with a SEM is the circled zone in Figure 5(b) which zooms in at the bottom of the broomed end. Wire breaks appear to have occurred in this zone, either by fatigue, or by the impact of the fall. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 28

Nevertheless, if fatigue had been the primary failure mechanism, then the cable should have failed, not by slipping from the socket, but by rupturing near the base of the socket. The fact that the broomed end seems almost intact (Figure 5(a)), points at slippage from the socket as the cause of failure, even if all of the 20 wires circled in yellow had failed by fatigue. Also, Metcalf and Matanzo (1980) observed that pull-out failures occurred only in cases of static loading. Therefore, it appears that the slip mechanism is mostly associated to the capacity of the socket assembly for static load resistance rather than a fatigue-cycling condition. Figure 5. (a) Detail of the broomed end showing detached tables (yellow circles) that could have failed either by fatigue or by detachment/breakage due to the impact of the ~500 ft fall. (b) Detail of the bottom of the broomed end showing a potential zone where cables broke off (red circle). Taken from Wall (2020). Is further research required on the length of the spelter cone? The Bradon and Ridge (2003) single-wire pull-out results showed that when the embedment length was 5.4 times the cable diameter, in 3 out of the 4 tests the wires broke instead of pulling out. As mentioned earlier, the basket length in the Arecibo socket was approximately 3.7 times the cable diameter. This satisfied the federal specification RR-S-550 Rev. D (1980) and its more recent Revision F of 2018 which also showed a ratio of 3.7 for the largest sized cable that they address (2.625-inch diameter). However, the German standard DIN 3092-1 (1985) specifies a minimum spelter cone length 5.0 times the cable diameter for wire ropes (“wire rope” is a type of steel cable fabricated similarly but not equally to the Arecibo cables: wires are first wound into intermediate-sized strands which are then wound to the final diameter of the wire rope. The Arecibo cables were wound into a single strand and is known as “strand” cable instead of “wire rope”.). Citing from Verreet (1997): “In accordance with DIN 3092, the opening angle of a wire rope socket should be between 5˚ and 18˚, and the length of the spelter cone should be at least five times the nominal rope diameter. Break tests have proved that, even with shorter spelter cone lengths, the full breaking strength of the rope can be transferred”. They note that shorter lengths have worked but the standard still specifies the minimum of 5.0 times the diameter of the cable. A 5.0 minimum ratio between the cone length and the cable diameter would add reliability to the socket. It would be very comforting to assume that a cable failure should take place in the cable itself by rupturing rather than by Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 29

slippage from the termination. It may be asked, is more research required on these types of terminations to increase the level of reliability? The research would need to consider the manufacturing implications because it would increase the required level of workmanship to properly manufacture it. As noted previously, it is already challenging to manufacture it at a 3.7 ratio. The cable that failed was apparently not the only one with slippage problems. Citing from Thornton Tomasetti (2020), “Furthermore, TT recommended the replacement of all of the auxiliary cables, since the one 3¼-inch auxiliary cable completely pulled from its socket and numerous other auxiliary cables exhibited unusual slip at their sockets”. In the case of the socket shown in Figure 3(a), the photograph shows that the remaining auxiliary cable did not appear to be one of the cables with slippage problems. Dynamic consequences of a cable failure on the remaining cables This section investigates the dynamic effects on the remaining cables of a multi-cable structure caused by the sudden loss of a single cable. It derives a dynamic amplification coefficient α assuming a single degree of freedom system. In the case of the Arecibo Observatory, assuming that the damping ratio of the cables was ξ = 0.05, the dynamic amplification coefficient is calculated at α = 1.85. This means that the load added by the sudden failure of the cable was momentarily increased by 85% over its static value. In addition, a dynamic redistribution coefficient β is also derived to distribute the dynamically amplified load into all the remaining cables. The factor β is a function of the number of cables and the previous coefficient α. In the case of Arecibo, two cases were considered to bound the results. The first case (lower bound) assumes that all the cables participate, and so N = 18 cables. In that case, β = 1.05. The second case assumes that only the cables in one tower participate and therefore N = 6. This second case provides an upper bound and seems appropriate given the uncertainties with the amount of damping offered by the platform and its 3D rotational degrees of freedom. For N = 6, the calculations result in β = 1.14. This means that the instantaneous dynamic force in the remaining cables of Tower 4 would have momentarily increased by between 5% - 14% over its static value. The analytical model was verified against a 2D, multi-degree of freedom model in the program SAP2000 and the error was less than 3%, which validates the analytical model. The derivation of the two coefficients α and β are provided next. Consider a simple model of a planar structure consisting of a weight W supported by N cables or springs. All the cables are assumed to have the same stiffness coefficient. Each cable supports a static force with amplitude Fcable = W/N. If one of the cables (springs) is suddenly removed, the structure will vibrate. This phenomenon may be confused with an initial displacement type of problem, but it is actually a forced vibration condition. Suppose that one cable is slowly removed. Its force will be distributed among the remaining N-1 cables. Evidently, the total new force acting on each of the N-1 cables is Fnew = W/(N-1). It is convenient to obtain an expression for the additional force Fadd caused by the loss of one support, as follows: Fnew W  Fadd  W  Fadd  N W N N 1  N 1 (1) However, if one of the cables is suddenly removed, the force Fadd is not slowly added to the original load W/N, but rather its time variation has the shape of a unit step function U(t) (sometimes referred to as the Heaviside function): F (t)  Fadd U (t) (2) Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 30

This force is shown in Figure 6 where a value of 10 kip was arbitrarily selected for Fadd. Figure 6: Step function load applied to the remaining cables. It can be shown that when a damped single degree of freedom oscillator is subjected to a dynamic load with a step-function variation and an amplitude Fadd, the internal force in its spring is given by:     1  e n t  cos d t  1 2 F (t ) Fadd  sin d t  (3) where ωd is the damped natural frequency, which can be expressed in terms of the damped natural period Td as: d  2 Td (4) The damped natural frequency is in turn defined in terms of the undamped natural frequency ωn and damping ratio ξ as: Td  2 n 1 2 (5) Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 31

The force F (t) attains its maximum value at a time t = Td/2. Substituting this time t, as well as ωd and Td in equation (3) one obtains     1 e 1 2 Fmax  Fadd   (6) Introducing a dynamic amplification coefficient α defined as:   (7)   1 e 1 2 The maximum force can be written as: Fmax   Fadd (8) For an undamped system (with ξ = 0), the maximum force Fmax in the spring is 2xFadd, i.e., twice the value as if the force F (t) were applied very slowly. For other values of the damping ratio, the ratio α between the maximum dynamic and static force varies as shown in Figure 7. Figure 7: Variation of the ratio α between the maximum dynamic and static force as a function of the damping ratio. Note that a damping ratio equal to 0.1 or higher is only achievable by adding external devices, such as viscous fluid dampers, treatment with damping layers, etc. For a naturally occurring energy dissipation and for a metal structure with the material below yielding, the values of ξ range from 0.005 to 0.02. For these values, the ratio α = Fmax/Fadd varies from 1.98 to 1.94. The value ξ = 0.05 is usually adopted when the material is at or above the yielding level. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 32

Based on equation (1), the new force acting on the remaining cables, which now we will refer to as Fstatic, is: Fstatic  W W  Fadd where: Fadd  W N 1 N N  N 1 (9) However, based on the previous demonstration, if one wants to obtain the maximum value of the force on each of the remaining cables that accounts for the dynamic effect, this new force is: Fdynamic W  Fmax where: Fmax  N W N  N 1 (10) which can be more conveniently expressed in terms of a dynamic redistribution coefficient as: F  Fdynamic static (11) where:  N  1 ; Fstatic  W N N 1 (12) We will apply now the formulas selecting first a 5% damping ratio. The dynamic amplification coefficient α is:   0.05 (13)   1 e 10.052  1.85 The dynamic redistribution coefficient is now only a function of the number of cables N:   N  0.8545 (14) N Figure 8 displays the variation of the dynamic distribution coefficient β that permits to calculate the maximum dynamic force by amplifying the static force as Fdynamic = β Fstatic: Figure 8: Variation of the dynamic redistribution coefficient β as a function of the number of cables for a 5% damping ratio. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 33

The objective of this section was to show with simple examples that the breakage of one of the cables in a cable- supported structure introduces dynamic effects in the response of the remaining cables, which translate into a maximum instantaneous axial force larger than the one predicted by a static analysis. This situation undoubtedly occurred in the cables of the Arecibo Observatory. The dynamic effects probably increased the size of any pre-existing fatigue cracks in the remaining cables. This would have weakened the cables even further and decreased their factors of safety below the values calculated by Thornton Tomasetti (2020). These are discussed next. State of the structure after the failure of the first cable Thornton Tomasetti (2020) indicated that the safety factor was a suitable 1.67 on the second cable that would eventually end up failing. It was reduced from 2.1, but a factor of 1.67 still provided a suitable safety margin. The calculation was based on the reported 1,044 kip breaking strength of the cable. Citing from Thornton Tomasetti (2020), “After the failure (of the first cable), observatory staff, TT, WJE and WSP inspected/reviewed the remaining structure for signs of distress and deterioration. Given the generally good appearance of the remaining elements; suitable factor-of-safety remaining in the platform elements, as shown through analysis; and adequate redundancy of the cable system, we believed the platform to be stable then and for some time forward. Our analysis had shown that the loss of another cable would not cause catastrophic collapse of the platform. Therefore, we believed work to stabilize the structure could begin, with continuous monitoring and safe operational procedures. The observatory procured materials and supplies and planned for installation.” FAILURE OF SECOND CABLE ON NOVEMBER 6, 2020 The second cable failed 88 days after the failure of the first cable. Following this second failure, the press release of the Arecibo Observatory (NAIC, 2020e) reported the following: “A main cable that supports the Arecibo Observatory broke Friday (November 6) at 7:39 p.m. Puerto Rico time. Unlike the auxiliary cable that failed at the same facility on August 10, this main cable did not slip out of its socket. It broke and fell onto the reflector dish below, causing additional damage to the dish and other nearby cables. Both cables were connected to the same support tower (tower 4)”. The cable is identified as M4 in Figure 1(b) and Figure 9. Figure 9 shows a photograph of the original saddle of tower 4 just prior to the final collapse. The wires within a red oval are the remnants of the second cable that failed (identified as M4 in Figure 9). This was confirmed by NSF (2020b). The remnants of the cable establish that the cable ruptured, instead of slipping out from its termination. The fact that the rupture occurred near its anchorage point is consistent with the findings of Metcalf and Matanzo (1980) that 100% of the fatigue failures occur near or inside the termination. The cable ends are the most vulnerable because they are the weakened the most during fatigue loading caused by the oscillations (vibration) of the cable. Figure 1(b) shows the tuned mass dampers that were used in the main cables to mitigate the vibration effects. Figure 9: Photo of main saddle on Tower 4 just prior to the final collapse. The cable identifiers are consistent with Figure 1(b). Taken from NSF (2020a). Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 34

The second cable failed at a load significantly lower than its breaking strength. The calculated safety factor was 1.67 as reported previously. According to Thornton Tomasetti (2020), the “weather at the time of failure was calm, with no unusual winds or ambient temperatures and no ground shaking. Failure was unexpected”. Accumulated fatigue damage throughout the 57 years of operation most likely played a role in the unexpected failure. Also, the dynamic demand of the sudden failure of the first cable would have shortened the remaining life of the cable even further, as discussed previously. Different cable terminations: “Blocks” instead of “Sockets” Figure 9 also shows the five cylindrical terminations of each of the five backstays. These are called “blocks” instead of “sockets”. The main cables M1-M4 used similar blocks that were located on the rear of the saddle and are hidden from view in Figure 9. These blocks beared against the saddle to resist the cable forces. The details of the blocks are unknown to the authors; however, similar blocks are used in some cable stayed bridges. In these, the outside is cylindrical, but they are conical on the inside; the wires are splayed at the entrance and terminate at a locking plate; and the internal conical section is filled with a bonding agent (National Academies, 2005, p. 11). There is some similarity with the sockets of the auxiliary cables except that, instead of a clevis joint, the load is transferred by bearing against a saddle. Another significant difference is that the ends of each wire are locked in a plate. The locking plate provides an additional load transfer mechanism against slippage. It is interesting that the blocks of the main cables dated from 1963 while the sockets of the auxiliary cables dated from 1996; yet the older termination properly anchored the main cable M4 in place. In fact, all 27 block terminations (9 per tower) held in place during the 57-year life of the structure. The authors reflect that this type of termination had the required level of reliability that would be warranted for the observatory’s design, i.e., that, in case of failure, the failure should take place in the cable itself by rupturing, not by slippage from the termination. State of the structure after the second cable failure Citing from Thornton Tomasetti (2020), “With the loss of two cables, there are now three original cables (of four) and one auxiliary cable (of two) connecting the platform to Tower 4. Should another of these three original cables fail, the two remaining original cables will undergo static force demands at or above the specified minimum breaking strength. A catastrophic failure would be very likely. These cables are not capable of handling the required dynamic demands of a sudden failure of an adjacent cable. The structural redundancy is no longer available and cannot be factored into determining safety”. Thornton Tomasetti (2020) constructed a finite element model of the complete structure and considered several scenarios on how to reduce the load on the remaining cables (all are detailed in the reference). These included relaxation of the backstays, installation of a temporary cable, removal of counterweights from the Azimuth arm, and cutting the M4 cable that was hanging from the platform, among others. However, the report states “We have noted wire breaks on the three remaining 3-inch-diameter original cables from Tower 4, which occurred during the November event. We continue to monitor the structure and continue to note wire breaks since the failure last week. Furthermore, there is no evidence that the existing original cables can achieve the specified minimum breaking strength and certainly evidence to the contrary, since one failed at 62% of this strength” (Thornton Tomasetti, 2020). At this point, Thornton Tomasetti recommended decommissioning and a controlled demolition. WSP (2020) concurred with this recommendation and mentioned that “Since we are observing additional wire breaks, this leads us to believe that there is additional degradation of the cables and therefore less capacity than expected”. On the other hand, WJE (2020) proposed stabilizing the structure. As a first step, they considered proof loading the structure, i.e., adding some load to prove that it could resist it. Citing from WJE (2020), “The key element in pursuing this path is reducing structural uncertainty to acceptable levels by demonstrating that key elements have the capacities needed to support the work that must be done.” Also, “Since the alternative to repair is demolition of the Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 35

facility, the risk of possibly collapsing the unoccupied structure during an attempt to save it may be acceptable.” They proposed to proof load the structure by tensioning the vertical tiedown cables on the corners of the platform. On November 19, 2020, a press release from the Arecibo Observatory announced the decision of NSF to decommission and demolish the structure (NAIC, 2020f). FINAL COLLAPSE ON DECEMBER 1, 2020 The structure survived for 25 days after the failure of the second cable. The wire breaks mentioned by Thornton Tomasetti (2020) were also reported in the news media; for example, nine days into this period, Coto (2020) reported that “university officials say crews have already noticed wire breaks on two of the remaining main cables”. Figure 9 confirms the breaks in cables M1 and M2. The breaks are easily noticed because the paint flakes off as a result of the sudden snap (NSF, 2020b). The wire breaks occurred near the termination, as predicted by Metcalf and Matanzo (1980) for fatigue loading. The paint on cables M3 and the service cable SC is still intact. In the case of the service cable, there was no overload since it only supported the catwalk bridge. Corrosion The loss of paint on cables M1 and M2 offers a glimpse of the cables to evaluate their corrosion state. Figure 9 shows that the cables had a grayish color which is consistent with no corrosion issues, perhaps in the Stage 1 of corrosion based on Figure 2(b). In addition, the saddle and the blocks that terminate the backstays also appear free of corrosion. Sequence of failures during collapse The drone video (NSF, 2020a) shows that the sequence of cable failures took place in approximately 1 second. Cable M2 was the first to break, followed by M1, and finally M3. The order of failure is consistent with observations of Figure 9, based on the amount of paint removed from the cable by wire breakage prior to the collapse, i.e., M2 had the most paint removed (the most wire breaks) while M3 had no paint removed (no wire breaks). Once M2 failed, the level of overload, plus the dynamic effects of the sudden failure of M2, were simply too great to be supported by the remaining cables. The collapse had been predicted by Thornton Tomasetti (2020), as cited previously. The service cable SC broke approximately 6 seconds after M3 as a result of the cables of Tower 12 impacting the catwalk bridge during the pendulum swing of the platform, thus overloading the service cables that supported it. Estimates of the platform dynamics during the pendular swing The system was modeled analytically as a pendulum (one degree of freedom) with the end of the pendulum string pinned to the middle point of an imaginary line between the tops of towers 8 and 12. The length of the pendulum string was calculated as L = 357.5 ft, using known dimensions from the observatory that were used by Morales (2006). It is also known that the lower corner of the platform (attachment points of main cables) was located 130.4 ft below the top of the towers, which sets the initial point for the pendulum swing. The point of interest in the calculations is at the bottom of the pendulum swing, i.e., the point at which the string becomes vertical. At this point, the speed and centripetal (normal) acceleration reach their maximum values. The non-linear differential equation of motion of the pendulum can be easily derived in polar coordinates and is included as equation (15). It assumes that there is no air resistance, no friction at the pinned end, and all the mass is lumped at the free end. The Mathematica software was used to solve the equation with numerical integration. The elapsed time to the bottom of the swing was calculated by Mathematica as t = 5.75 seconds. The control video (NSF, 2020a) shows an elapsed time of approximately 6 seconds from the start of the pendular swing to the bottom of the swing (instant where the catwalk bridge is impacted and breaks). The comparable results (~ 5% error) show that the assumptions are reasonable to conduct additional estimates. ������2������ ������ (15) ������������2 + ������ ������������������ ������������������ (������) = 0 Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 36

Assuming that the potential energy due to the height of the platform was fully transformed to kinetic energy at the bottom of the swing (principle of conservation of energy), and using the above assumptions, the speed of the platform at the bottom of the swing was calculated at v = 120.9 ft/s = 82.4 mi/hr (equation (16)). Exactly the same results were obtained from the solution of the non-linear differential equation with Mathematica. ������ = √2������ℎ = √2(32.2 ������������ − 130.4 ������������) = 120.9 ������������ = 82.4 ������������ (16) ������2)(357.5 ������������ ������ ℎ������ Also, the speed formula based on conservation of energy may be combined with the definition of normal (centripetal) acceleration while assuming a radius of curvature “������” equal to the pendulum string length. The resulting normal acceleration at the bottom of the swing would have been 1.27 g’s, where 1 g = 32.2 ft/s2. (equation (17)). ������2 (√2������ℎ)2 2������(357.5 ������������ − 130.4 ������������) = 1.27������ ������������ = ������ = ������ = (17) 357.5 ������������ Based on a free body diagram of the pendulum at the bottom of the swing, and ensuring dynamic equilibrium in the vertical direction, the vertical component of the total tension in the cables at the bottom of the swing would have been 2.27 times the platform weight (equation (18)). ������ = ������ + ������������������ = ������������ + ������(1.27������) = 2.27������������ = 2.27������ (18) The combined factors of the increased tension in the cables due to the dynamic effect of the pendular motion, the more inclined angle of the cables at the bottom of the swing, the fact that the cables were now re-oriented between the strong axes of the towers of cruciform cross-sectional shape, the inefficiency of the backstays in this new orientation of the cables, and the rupture of the service cables (the only remaining elements connecting the three towers) ended up collapsing the top portions of towers 8 and 12 toward each other. In the case of tower 4, the top portion collapsed backward because the backstays pulled it back once the service cable snapped. Also, the sudden release of stored elastic energy in the service cable contributed to the backward collapse of tower 4. It is interesting that in the case of the shorter towers 4 and 12, only the top portion collapsed while in the case of the taller tower 8, the upper two portions collapsed. Figure 10 shows the variation with time of the angle θ(t) obtained by solving the nonlinear equation of motion (15). The initial position is θ(0) = 68.6o. The green dot indicates the instant at which the pendulum passes through the vertical position. Figure 10: Time variation of the angle θ(t) of the pendulum. 37 Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1)

Figure 11 displays the tangential velocity of the bob of the pendulum as a function of time. The green dot indicates the instant at which the pendulum passes through the vertical position and the velocity attains a maximum. Figure 11: Time variation of the tangential velocity V(t) of the pendulum. Accelerogram recorded during the event Figure 12 shows an accelerogram (z-axis, vertical) during the final collapse. The accelerometer is located at the site of the observatory near tower 12. It was installed and operated by the Puerto Rico Seismic Network (PRSN) of the University of Puerto Rico at Mayagüez (UPRM) and had been operational for approximately 10 years. The signal was retrieved and processed by Dr. Elizabeth Vanacore, seismologist of the PRSN and associate professor at the Department of Geology of UPRM. A time scale was included as an inset to compare times versus the video footage (NSF, 2020a). The table that is presented below the accelerogram synchronizes times with the control room footage, the drone footage, and the time scale in the accelerogram. Both footages are available in NSF (2020a). The footage times are precise to the nearest second and there are some apparent discrepancies with the accelerogram time scale (1 second discrepancies) which are highlighted in red in the last column. The origin of the time scale (t = 0 s) was placed at the highest acceleration peak which, after synchronizing the events, most likely represents the instant at which the platform impacted the hill. This seems reasonable because the platform was massive (about 700 tons after shedding the dome and the Azimuth arm) and moving at a speed close to 80 mi/hr, as calculated in the previous section. Notes were placed within the accelerogram to indicate the most likely events, based on the table. The impact of the Azimuth arm and the dome are not noted because this instant was not captured by the cameras. However, it is estimated that they impacted the dish at just about the same time that the towers failed (t = -2 sec). Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 38

Figure 12: Accelerogram obtained during the collapse event. Provided by Elizabeth Vanacore of PRSN. The Control Room and Drone footage times were taken from the NSF video (NSF, 2020a). CONCLUSIONS Despite the limitations of time and available information, the following is concluded: 1. Corrosion does not seem to have played a significant role in any of the three events discussed in this article. 2. The slippage of the auxiliary cable from its socket termination (first cable failure) was most probably caused by manufacturing flaws during the fabrication of the cable/socket connection. These will be reviewed and updated once the forensic analyses are concluded and made public. ACKNOWLEDGMENTS The authors gratefully acknowledge the staff of the journal for their assistance in the preparation of this manuscript. We would also like to thank Dr. Elizabeth A. Vanacore, of the PRSN and Department of Geology of UPRM, and Dr. José A. Martínez Cruzado, of the Puerto Rico Strong Motion Program and Department of Civil Engineering of UPRM, for providing the accelerogram recorded during the collapse and for providing initial clues on its interpretation. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 39

REFERENCES Bradon, J. E., and Ridge, I. L. (2003). “Comparison of white metal and resin socket terminations for wire ropes”, The Journal of Strain Analysis for Engineering Design, Vol. 38, No. 2, pp. 149-160. Certex (2020). “Wire rope terminations”, Certex USA, Charlotte, North Carolina, https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=&cad=rja&uact=8&ved=2ahUKE wiI-MrmjO7tAhVOrFkKHTjCAsQQFjAAegQIARAC&url=https%3A%2F%2Fcertex.com%2Fwp- content%2Fuploads%2Fwire-rope-terminations.pdf&usg=AOvVaw0fUYtnvOReQB5PWWateMLr Chavel, B.W. and Leshko, B.J. (2012). “Primer for the Inspection and Strength Evaluation of Suspension Bridge Cables”, Report FHWA-IF-11-045, HDR Engineering Inc., Pittsburgh, Pennsylvania. Coto, D. (2020). “Cable failures endanger renowned Puerto Rico radio telescope”, Associated Press, November 15. https://apnews.com/article/technology-arecibo-observatory-science--053fbf8834e50d5e3ad88443a3fde550 DIN 3092-1 (1985). “Socketing of Wire Ropes - Casting in Metal - Safety Requirements and Testing”, Deutsches Institut fur Normung E.V. (DIN), Germany. Drake, N. (2020). “Iconic radio telescope suffers catastrophic collapse”, National Geographic, December 1. https://www.nationalgeographic.com/science/2020/12/arecibo-radio-telescope-in-puerto-rico-collapses/ Metcalf Jr, J. T., and Matanzo Jr, F. T. (1980). “Wire rope terminations, section, and replacement criteria”, Offshore Technology Conference, Houston, Texas. May 5-8. Morales, J. C. (2006). “Dynamic properties and seismic response of the cable structures and towers of the Arecibo Observatory”, PhD dissertation, Department of Civil Engineering and Surveying, University of Puerto Rico at Mayagüez, Puerto Rico. NAIC (2020a). “Astronomy”, Arecibo Observatory. https://www.naic.edu/ao/astronomy NAIC (2020b). “Telescope Description”, Arecibo Observatory. https://www.naic.edu/ao/telescope-description NAIC (2020c). “Arecibo Observatory Construction”, Arecibo Observatory. http://www.naic.edu/history_gal/historicgal.html NAIC (2020d). “Broken cable damages Arecibo Observatory”, Arecibo Observatory, August 11. https://www.naic.edu/ao/node/1169 NAIC (2020e). “A second cable fails at NSF’s Arecibo Observatory in Puerto Rico” Arecibo Observatory, November 8. https://www.ucf.edu/news/a-second-cable-fails-at-nsfs-arecibo-observatory-in-puerto-rico/ NAIC (2020f). “Arecibo Observatory Telescope to be Decommissioned After Second Cable Break” Arecibo Observatory, November 19. https://www.ucf.edu/news/arecibo-observatory-telescope-to-be- decommissioned-after-second-cable-break/ National Academies (2005). “Inspection and Maintenance of Bridge Stay Cable Systems”, The National Academies Press, Washington, DC: https://doi.org/10.17226/13689 NSF (2020a). “Footage of Arecibo Observatory telescope collapse”, National Science Foundation, December 3. https://www.youtube.com/watch?v=ssHkMWcGat4 NSF (2020b). “Media briefing: Arecibo Observatory 305-meter telescope update”, National Science Foundation, December 3. https://www.nsf.gov/news/special_reports/arecibo/2020_12_03_AreciboBriefing_Transcript_FINAL.pdf RR-S-550 Rev. D (1980). “Federal Specification Sockets, Wire Rope”, U.S. General Services Administration – Federal Acquisition Service, Washington, D.C. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 40

RR-S-550 Rev. F (2018). “Federal Specification Sockets, Wire Rope”, U.S. General Services Administration – Federal Acquisition Service, Washington, D.C. Thornton Tomasetti (2020). “Recommendation for Course of Action at Arecibo Observatory”, Memorandum to Ramón Lugo dated November 12, Thornton Tomasetti, Inc., New York, New York Tokyo (2020). “Wire Rope Catalog”, Volume 5. Tokyo Rope Manufacturing Co., LTD, Tokyo, Japan. Verreet, R. (1997). “Wire rope end connections”, Casar Wire Ropes, Casar Drahtseilwerk SAAR GMBH, Kirkel, Germany. Wall, M. (2020). “Arecibo Observatory in Puerto Rico suffers serious damage after cable breaks”, Space.com, August 13. https://www.space.com/arecibo-observatory-damaged-shut-down.html Wirelock (2017). “Technical Data Manual”, version 2-11/17, Millfield Enterprises (Manufacturing) Limited, Newcastle upon Tyne, United Kingdom. WJE (2020). “Arecibo Observatory Stabilization Efforts”, Memorandum WJE No. 2020.5191 to Ramón Lugo, dated November 12, Wiss, Janney, Elstner Associates, Inc., Indianapolis, Indiana. WSP (2020). “Recommendations for future efforts at Arecibo Observatory”, Memorandum to Ramón Lugo, dated November 11, WSP USA Solutions, Inc., New York, New York. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 41

AUTHORS BIOSKETCH Professor and Director of Mechanical Engineering (since 2003) in the Mechanical Engineering program at University of Turabo (now Ana G. Méndez University). Also served as the ABET Coordinator of the School of Engineering between 2003-2016, a period that covered four accreditation visits and resulted in the initial ABET-EAC accreditation of all five engineering programs. He holds BS and MS degrees in mechanical engineering and a PhD in civil (structural) engineering where he investigated the dynamic characteristics and seismic response of the Arecibo Observatory using modern engineering tools (PhD thesis). His research interests reside in machine design, structural engineering, and engineering education. J. Morales L. Suárez- Professor in the Civil Engineering & Surveying Department of the University Colche of Puerto Rico at Mayaguez (UPR-M) and the Coordinator of the Structural Engineering Area. His areas of research include structural dynamics, earthquake engineering and soil dynamics. He has taught graduate courses in Puerto Rico, USA, Colombia, Dominican Republic, and Argentina. Dr. Suárez graduated at the top of his class in the National University of Cordoba, Argentina, in 1981 with an engineering diploma in Mechanical and Electrical Engineering. In 1984, he received his M.Sc. degree and in 1986 his Ph.D. degree both in Engineering Mechanics from the Engineering Science and Mechanics Dept. at Virginia Tech. He published five books about Structural Dynamics, Matlab, Mathematica and SAP2000. Revista Internacional de Desastres Naturales, Accidentes e Infraestructura Civil. Vol. 19-20 (1) 42