4

Arecibo Telescope’s Management and Oversight

Hurricane Maria struck the Arecibo Telescope during the period in which the National Science Foundation (NSF) would have been considering proposals in response to its solicitation for the Management and Operations of the Arecibo Telescope, due on May 4, 2017 (NSF-538), and the award to the University of Central Florida (UCF). Following the award of the Arecibo Telescope operating contract to UCF, the full management contract started on April 1, 2018.¹ The transition period of Arecibo Observatory (AO) to UCF lasted from February 2018 through June 2018, supported by an award for $913,935.² SRI International was the managing organization until the end of March 2018, and the organization received an extension to support close out until May 31, 2018, more than 6 months after Hurricane Maria. There were several meetings among NSF, SRI, and UCF during the facility’s turnover. The AO site director, as well as the majority of the AO staff, remained throughout this transition. The $20 million grant to fund UCF’s operations and maintenance (O&M) award was not received until April 2018,³ 7 months after Maria had struck the Arecibo Telescope. The UCF proposal separated the Arecibo Telescope’s science and maintenance functions, with the responsibility “for facilities infrastructure, engineering, operations, maintenance, information technology and support services, as well as logistics and security services” to be assumed by Yang Enterprises, Inc.⁴ While the announcement stated that “Yang Enterprises, Inc. will introduce new technologies and cutting-edge tools to support Arecibo’s requirements and modernize operations,”⁵ the committee did not find evidence of significant change in the Arecibo Telescope’s inspection methods and maintenance practices after UCF assumed control of operations. During the transition period, UCF management planned to implement a computer maintenance management system. Most maintenance records, however, were kept onsite as hard copies.

UCF assumed the Arecibo Telescope’s operation in the face of a budget that had been declining over the previous 5 years, and NSF announced a decision in 2017 (a few weeks after Maria) that “operations at the observatory

The Design Engineer or Engineer of Record (EOR) for the original design or significant modifications/upgrades is most knowledgeable concerning the risks, technologies, assumptions, redundancies, weaknesses considered in the design, and the potential threats to overall structural integrity. The EOR generally has knowledge of proof testing, field validation, and quality control in design and construction, what critical structural components need to be monitored, locations to inspect, and the limits of adequate performance of the critical components within the context of their design scope. Unfortunately, the original EOR is often not involved in significant changes or upgrades, particularly when there is a significant time lag involved. In these cases, the EOR for the changes must assume this role for all parts of the structure that are impacted by modifications, even if they were not modified.

The most recent EOR should provide a written narrative for the facility records describing the final conceptual design and its expected performance, as well as identify the critical components affecting the structure’s performance and how these should be monitored. This step is doubly important for a facility whose planned operation is by competitively bidding contractors who will have had no role in the facility design. Furthermore, the EOR should be retained to develop an inspection and monitoring manual for the ultimate facility operator, clearly outlining the procedures to monitor the performance of the most critical components affecting the facility’s structural integrity. To the extent possible, the prescribed monitoring should not require engineering knowledge or judgment from the inspectors and instead prescribe physical monitoring of non-judgmental parameters, such as counted wire breaks or measured cable pullouts. The inspection manual should also clearly identify the acceptable limits for the facility’s operation and inspection actions to be taken when the operation, or degradation such as a cable pullout, exceeds the acceptable limits, including safety measures to ensure the safety of the personnel, employees, scientists using the facility, and the visiting public. Online structural monitoring and health management, condition monitoring, condition-based maintenance, and reliability-centered maintenance are typical of today’s monitoring techniques.

The inspection, monitoring, maintenance, and repair program of the structural portion of the facility developed by the EOR should include clear guidelines for reporting and evaluating the information in the reports and records from both the Level One and the Level Two programs. Of utmost importance is the fact that the inspection, monitoring, maintenance, and repair program should include guidelines for reporting and assessing the criticality of any structural findings, especially those considered unusual or unexpected. The Arecibo Telescope lacked the proper inspection, monitoring, maintenance, and repair manuals addressing critical structural considerations that only the EORs in the original design and the upgrade knew. This deficiency is most significant when considering that the expected service life for the facility was 10 years.¹⁰ The Arecibo Telescope underwent several changes, which significantly increased the applied loads. At the time of the collapse, it had been in service for 57 years. NASA concluded, “The design did not explicitly consider the time-dependent effects of creep and cyclic loading on design capability, nor did it set service life inspection intervals with pass/fail inspection criteria. It also did not specify an end-of-life capability requirement associated with service life degradation.”¹¹ All of these deficiencies resulted in a facility that had exhibited unheeded signs of distress for several years.

The collapse on December 1, 2020, resulted from the lack of recognition of the significance of the cable pullout by the experts contracted to assess the Arecibo Telescope and to develop a response plan after Hurricane Maria. There was inadequate funding for the repairs or replacements needed because of the cable pullout. The operating entity and its managers failed to consider all the findings from the facility inspections to ensure the safety of its employees, scientists, staff, and the public visiting the facility.

The following findings and recommendations are relevant to any large structural facility owned by NSF with contract operators.

Finding: A Level Two inspection, maintenance, monitoring, and repair manual was not developed by the EOR or NSF, nor was such a policy implemented during the management of the Arecibo Telescope by its last two operators.

Conclusion: A good practice would be to have separate accounting and funding, independent of facility operations, for inspection, maintenance, monitoring, and repair of facilities. Otherwise, inspection, maintenance, monitoring, and repair may suffer in financial competition with the science operations that can gain the facility further funding.

Recommendation: Facility owners should enforce compliance with contract requirements through independent auditing of inspection, monitoring, maintenance, and repair records.

The Arecibo Telescope’s condition monitoring, while spotty throughout its history, demonstrably degraded after the 1997 upgrade and further degraded after 2011. The first comprehensive Ammann & Whitney inspection of the telescope after the completion of the Gregorian upgrade in 1997 was in 2003. The following comprehensive inspection was not until late 2010. While there were continuous inspections, apart from broken wire counts, these were generally only visual inspections, and photographic documentation is irregular. In the obsolete system, there was little codification or organization of the data with which to observe any trends. There appear to have been no specified procedures as to who should be contacted if there were any serious or discrepant findings. After supervisors were informed, there was no process to involve the chief engineer for additional analysis. There was also no central repository of the maintenance and inspection data at the facility, as shown when UCF began management of the facility, and UCF contemplated visiting Cornell University to review the archives.¹² Even though the Arecibo Telescope was aging, inspection methodologies did not change and seem to have become less frequent, as indicated by the reduced available inspection data discussed previously. By the time extensometers were considered, it was too late. Cable and socket maintenance consisted of applying sacrificial zinc paint; epoxy paint was used everywhere else.