A study to assess the applicability of using remote sensing to minimize service interruption of Canadian port physical infrastructure

Remote sensing can be an effective tool for providing early warning of structural deterioration that is difficult or impossible to detect with visual inspections, which can allow infrastructure owners and operators to prioritize key areas for maintenance and assist in more rapid recovery from incidents that cause service interruptions. Canadian Port Authorities (CPAs) do not currently use remote sensing techniques to assess the condition of physical infrastructure despite their application to similar operations such as bridges, pipelines, and dams. Interviews with key informants indicate that there are no technical or scientific barriers to using remote sensing to assess port physical infrastructure, and that while the presence of heterogeneous materials and surfaces, restricted air space, and a complex web of interconnected port partners and tenants complicates its use, these are surmountable. Instead, interviews suggest that the primary barrier stopping CPAs from implementing remote sensing is that there is no system-wide formal chain of responsibility and authority due to a lack of public sector agency leadership resulting from the commonly used “landlord model” for port operations, which generates fragmented stakeholder interests and inhibits collaboration. Given the emerging challenges of aging infrastructure, climate change-related impacts, and the demands of system-wide business continuity, a private and public sector collaborative pilot project is needed to test the feasibility of adopting remote sensing for port physical infrastructure assessment.


Introduction
Infrastructure of all types is affected by disruptions such as natural disasters, malevolent attacks, human-made accidents, and failures due to age, condition or interdependence.While risk planners were previously concerned with protection and prevention, their Extended author information available on the last page of the article 5 Page 2 of 18 interest is increasingly on the ability of infrastructure to withstand and recover from disruptions (Hosseini and Barker 2016).
Port infrastructure condition and resilience is of paramount importance for supply chain continuity with widespread potential consequences from disruption (Hsieh et al. 2014;Kochersberger et al. 2014;Thekdi and Santos 2016;Loh and Thai 2016;Duo et al. 2018).Port infrastructure is vulnerable to natural disasters due to coastal exposure and intentional attacks due to critical socioeconomic functions (Cao and Lam 2019).Mutombo and Ölçer (2017) point out that port infrastructure is broadly defined, but physical port infrastructure, the subject of this study, can be grouped into six categories: berthing structures, protection barriers, port structure, channels and harbours, road, and rail networks.The term port physical infrastructure in this paper refers to these six categories of assets.
Canadian Port Authorities (CPAs) are responsible for facilitating the movement of goods through their ports, and while they have a large network of partners for emergency management, CPAs are responsible for their own efforts related to quickly recovering and continuing service following a disruption.CPA physical infrastructure is distributed widely across Canada, and some are relatively remote.These port facilities may be difficult or dangerous to access even after an emergency has been declared over, so assessing the condition of physical infrastructure can be time consuming, thus extending the period that port business is interrupted.
CPAs are financially self-sufficient and operate at arm's length from the federal government, but with boards of directors partially controlled by the federal Governor in Council political appointment process.CPAs receive no operational federal funding, and they finance capital projects through a combination of their own revenues, partnerships with the private sector, debt financing, and certain federal grants related to infrastructure, the environment, or security.Individual CPAs are landlords and lease their port operations to private enterprise, a model that began with the 1995 National Marine Policy that led to the 1998 Canada Marine Act, which specified a goal of promoting a more competitive, effectively managed, and sustainable port system in Canada.The National Marine Policy's key principles are accountability to users and the public, business discipline, and self-sufficiency, shifting the burden of port operations from the Canadian taxpayer to port users.In a review of the impact of port reform in Canada, Ircha (2001) warned that landlord CPAs cannot take a passive role in merely providing facilities to private operators but must assist them in enhancing throughput.In Chen's (2009) review of the evolution of port authority administrative structure worldwide, the model most prominent prior to the 1980s whereby ports provided both infrastructure and services was found to "impede achievement of objectives" such as being commercially oriented entities.Since the 1980s the most common operating framework internationally has become the "landlord model", in which ports provide land and foundational infrastructure, but lease out operations to the private sector; this model was found to provide financial and operational benefits.Verhoeven (2010) notes the pressure on landlord ports worldwide whereby powerful private operators, such as carriers and terminal operators, gain control, which results in a "lack of coherence in the port community".The study also reports on a global phenomenon whereby landlord port authorities are the focus of criticism when things go wrong even for things outside of their formal responsibility (Verhoeven 2010).In Canada, Walker et al. (2015) observe that the Federal government, as the central authority, should have a more supporting function and perform specific harbour and port related tasks that cannot or are not performed effectively at the local level.
Traditional physical infrastructure condition assessment has focused on periodic engineering site visits with visual inspection of signs of scouring, corrosion, fatigue, and deterioration, sometimes using in situ instruments such as accelerometers (Pereira et al. 2016).These inspections combined with engineering judgment are necessary and invaluable, but are also time consuming, potentially infrequent, sometimes difficult, dangerous, or impractical to complete, and may not enable visual detection of deterioration in an early or timely manner (Ayele et al. 2020;Jofré-Briceño et al. 2021;Liu et al. 2020;Tang et al. 2020;Alshammari and Mohammed 2022).Remote sensing technology helps engineers to be better equipped to predict and prevent structural failure and enable better inspection, operation, and maintenance of physical infrastructure whereby sensors mounted on remote platforms provide objective and consistent measurement of deterioration and other forms of vulnerability (Jofré-Briceño et al. 2021;Liu et al. 2020;Tang et al. 2020;Alshammari and Mohammed 2022;Singhroy 2020).Remote sensing, whether by satellite, plane, or Uninhabited Aerial Vehicle (UAV), provides a synoptic view of a large geographic area and, depending on the type of platform and sensors used, provides early warning of the structural condition of steel and concrete elements, enabling quicker situation assessment and decision-making (Singhroy 2020).
Remote sensing technology could provide flexible, fast, and objective information about the condition of physical infrastructure at CPAs with high spatial resolution data and frequent revisit periods customized to the need and situation.It could provide guidance on where to focus more intense ground-based instrumentation and observation for continuous measurement and monitoring of key areas.It could be used when conditions are difficult, uncertain, or dangerous and could take on roles that emergency responders and human-operated vehicles cannot or should not.The question is, since other critical infrastructure owners use remote sensing operationally, why not CPAs?What barriers need to be removed to enable CPAs to incorporate remote sensing to take advantage of the technology to assess port physical infrastructure condition before and after a disruption to optimize condition assessment and minimize service interruption?
Remote sensing of infrastructure condition Ellingwood (2005) was an early proponent of bringing the science of remote sensing to the field of infrastructure management and specifically discussed the benefits for condition assessment of civil infrastructure.He emphasizes that decisions about maintenance, rehabilitation, and continued use of physical infrastructure need to be supported by quantitative evidence about factors such as structural strength or stiffness deterioration, and that remote sensing technology can provide information to supplement engineering judgment (Ellingwood 2005).Hausamann et al. (2005) also recognize the value of remote sensing for pipeline physical infrastructure, a sector that was an early adopter of the technology for operational condition and threat 5 Page 4 of 18 assessment.Fekete et al. (2015) discuss the value of remote sensing for critical infrastructure disaster risk governance in terms of how it helps with multiple stages of the disaster management cycle: mitigation, preparedness, response, and recovery.They emphasize the importance of having a time series of remotely sensed data such that pre-and post-disaster conditions can be compared (Fekete et al. 2015).In the context of natural disasters, Gomez and Purdie (2016) claim that society's expectations of immediately available critical infrastructure puts a lot of pressure on emergency and continuity managers, who would benefit from rapid collection of highresolution geospatial data provided by remote sensing.
While we cannot likely improve our ability to predict the timing and impact of incidents, Schweizer et al. (2018) suggest that we can improve our response and preparation, and that remote sensing is uniquely valuable since it enables safe access to critical physical infrastructure post-event and provides a spatially comprehensive, synoptic view.They argue that rapid deployment is required in order to maximize the benefits of the technology's unique abilities, which in turn needs personnel to have already established their competence and invested in the systems and processes; only then will people be able to rapidly process and disseminate appropriate remote sensing products for efficient and informed decision-making (Schweizer et al. 2018).Stow et al. (2018) identify the factors that hyper-critical infrastructure owners must understand to maximize the benefit of remote sensing: 1. the elements of the system that are indeed critical; 2. the probable failure modes and how hazards and threats will impact the critical infrastructure; and 3. the types of decisions that need to be made and how remote sensing results can inform those decisions.Singhroy (2020) provides an overview and introduction to current applications of remote sensing of physical infrastructure with a focus on monitoring transportation, energy, and mining physical infrastructure with radar, multispectral and hyperspectral optical, thermal-infrared, and LiDAR (Light Detection and Ranging).Singhroy (2020) covers the increased use of remote sensing to monitor physical infrastructure and provides examples of the technology being a powerful tool because of its increasingly high spatial and temporal resolution and ease of integration with other data to improve infrastructure monitoring.There is a recurring theme in Singhroy (2020) of the value of remote sensing in its ability to aggregate cascading hazards that could interact to magnify a disaster, something that individual discrete groundbased measurements and observations might not be able to do as well.

UAV-based remote sensing
One of the fastest growing UAV-based remote sensing applications is critical infrastructure monitoring for early warning of vulnerabilities, especially for roads, bridges, and tunnels (Kerle et al. 2019).Neocleous et al. (2016) refer to the challenges of corrosion of coastal urban structures and examined UAV-based remote sensing for its ability to provide early warning for deterioration of reinforced concrete of buildings.Case studies in Sharma (2019) demonstrate the use of UAV-based remote sensing for a range of applications, including physical critical infrastructure, and Shakhatreh et al. (2019) review the many uses of UAV-based remote sensing for civil applications, including real-time monitoring, search and rescue, delivery of goods, and precision agriculture.Politis et al. (2020) undertake a network-level pavement condition assessment using UAV-based remote sensing to build an index to score the overall condition of road network segments.An application of UAVbased remote sensing discussed in Singhroy (2020) was for detecting rail network changes over time through the use of repetitive, flexible, and inexpensive data acquisition to monitor the speed and acceleration of ground movement, thereby enabling the early identification of displacement-related risk.Pulella and Sica (2021) conduct a study of airport situational awareness with UAV-based remote sensing noting the value of spatial-temporal data to provide insights regarding airport physical infrastructure conditions.Besada et al. (2018) discuss the continuous market growth of UAV usage due to cost reductions and familiarity with the value of remote sensing, but highlight how UAV-based physical infrastructure inspection, maintenance, and security applications are complex and challenging.Regardless, because of the increased popularity and demand, more UAV-based remote sensing tools and systems are available to help define mission scope, select appropriate sensors, and support the integration of remotely sensed data with other relevant data in a geographic information system (Besada et al. 2018).Notwithstanding the complex mission planning and execution, UAV-based remote sensing provides high resolution data and flexibility for damage mapping and has evolved from simple visual descriptive overviews to more sophisticated sensors and data analysis with multi-temporal and multi-perspective methods (Kerle et al. 2019).Notably, a practical challenge highlighted in the literature is related to training or hiring staff with appropriate skills and experience to implement UAV-based remote sensing for infrastructure monitoring and assessment (Sharma 2019).Colomina and Molina (2014) recognize this challenge early and discusses the importance of devoting time and expertise to UAV mission planning with respect to sensor choice, flight path and trajectory, and wind conditions.The operational importance of regular practice and innovation with UAV mission planning is stressed by Nagasawa et al. (2021), as it is crucial to avoid delays when critical infrastructure is disrupted following an incident.

Remote sensing of physical port infrastructure
There are very few references in the literature to the use of remote sensing of physical port infrastructure.Recently, Tang et al. (2020) evaluate UAV-based remote sensing for cranes used in a port environment for their flexible maneuverability with respect to assessing the vertical structures for cracks and rust.Both Tang et al. (2020) and Liu et al. (2020) report that cranes are good candidates for UAV-based remote condition assessment because they are vulnerable due to the huge loads, heavy use, and aggressive coastal conditions related to wind, salt, and seawater dynamics; both papers report that remote sensing is attractive since traditional assessment is dangerous, time consuming, and labour-intensive.
5 Page 6 of 18 Jofré-Briceño et al. (2021) note that facility management and maintenance in port environments can be difficult due to the harsh coastal dynamics and progressive and variable deterioration of physical infrastructure, so they propose a method for creating a tool that is able to incorporate digital surveys collected via UAV for a baseline elevation model.Alshammari and Mohammed (2022) use UAV-based remote sensing to monitor deformation and stability of a breakwater protecting harbour entry and exit channels and determined that the technology provides crucial information for early warning of vulnerabilities.They also note that the technology is helpful for gaining a better understanding of failure mechanisms, especially in the context of changing ocean conditions with climate change (Alshammari and Mohammed 2022).
After surveying port sector executives regarding their intention to use UAV technology for maritime transportation applications, Yang (2019) conducts a factor analysis to identify the top expected applications as environmental issue detection and delivery of goods to offshore ships.Poling (2021) provides a comprehensive overview of remote sensing for marine shipping activity surveillance and activity detection.

Remote sensing of bridge infrastructure
The paucity of work published related to remote sensing of physical port infrastructure indicates that it is an underdeveloped area of application.Bridges, however, have been a prominent focus for development of methodologies, selection of appropriate sensors, and experimentation with data integration and analysis.Endsley et al. (2012) discuss the value of UAV-based remote sensing for asset management and structural health monitoring of bridges to evaluate indicators for early warning and diagnosis.Harris et al. (2016) suggest that bridge physical conditions could be assessed more frequently, comprehensively, and with less service interruption by fusing remotely sensed measurements from multiple sources with traditional surveys.According to Harris et al. (2016) some sensors, such as thermal-infrared, radar and acoustic, can assess the internal condition of bridge components, including the location of rebar deterioration and corrosion, deformation, and delamination, that are not visible in traditional surveys.Omar and Nehdi (2017) and Mac et al. (2019) further examine thermal-infrared technology for bridge condition assessment via UAV and note that delamination due to the corrosion effect of steel rebar in reinforced concrete is a dangerous form of deterioration because it is not usually visible, so this type of remote sensing gives early warning and enables proactive maintenance to address infrastructure vulnerability.Mac et al. (2019) claim that cement delamination up to 4 cm below the surface was detectable using UAVbased remote sensing.
In their paper published on crack inspection using UAV-based remote sensing, Ayele et al. (2020) conclude that the technology reduces time and cost of bridge infrastructure inspection.After reviewing 65 journal and conference papers on remote sensing of bridge physical infrastructure, Feroz and Dabous (2021) summarize specific methodologies and sensors used for bridge condition assessment.Aliyari et al. (2022) focuses on the use of UAV remote sensing for hazard identification and risk assessment for bridge inspections, while Cusson et al. (2021) explore satellite-based remote sensing of physical infrastructure using Interferometric Synthetic Aperture Radar (InSAR) and examine integration of the data for 3D visualization and early warning of unexpected bridge displacements.The authors also study the concept of thermal sensitivity and its use to monitor bridges for early warning of abnormal displacements to help avoid failures (Cusson et al. 2021).
In their investigations of the value of radar remote sensing for monitoring bridge infrastructure, Cusson et al. (2021) suggest that bridges have the benefit of having well defined structures, materials, and designs.There are elements of port infrastructure that may be more heterogeneous than bridges, which could complicate remote sensing for condition assessment, especially as related to port foundation, moisture content, closeness to a water body, and type of substrate.Structurally, however, ports are not dissimilar to bridges in that there is extensive and complex use of steel and concrete; it has been demonstrated with bridges that not only can remote sensing technology pinpoint defects like cracks in steel and fractures in concrete, but it can do so earlier than possible with visual inspections.

Methodology
The aim of this research is to understand the barriers that explain why CPAs have not embraced remote sensing technology for physical infrastructure condition assessment.Remote sensing technology is a well-established operational tool for application to physical infrastructure as confirmed by the literature review, so the focus of this research is on understanding attitudes about policy challenges to implementing the technology.

Research design
The following hypotheses informed the decisions about how data should be collected and analyzed: 1.There are no scientific or technical barriers at ports that cannot be easily overcome for operational use of remote sensing for physical condition assessment.2. There are some practical and regulatory barriers such as airspace restrictions and personnel expertise that can be overcome.3.There are legal landlord-tenant and landlord-partner agreements that present nontrivial policy barriers and constrain collaboration.4.There are no fundamental or philosophical objections to evaluating the feasibility of remote sensing technology for condition assessment of CPA physical infrastructure.
5 Page 8 of 18 The above hypotheses guided the qualitative research methods employed in this study, which explored the opinions and attitudes of key stakeholders regarding the feasibility of the following hypothetical proposal: 1.A pilot project can be facilitated by some or all the following: Transport Canada, National Aeronautical Surveillance Program, National Research Council, the Canada Centre for Remote Sensing, a private sector UAV-based remote sensing company, Defence Research and Development Canada, and a volunteer CPA. 2. A pilot project can demonstrate the logistical requirements, including hardware, software, and internal versus external human resources and essential knowledge can be transferred to CPA operational staff.3.With feasibility and value demonstrated through the pilot project, relevant parties will be motivated to explore practical modifications to their legal agreements to enable a collaborative system-wide approach to physical infrastructure condition assessment on an operational basis such that it can be immediately and seamlessly utilized in the case of an incident to minimize service interruption.

Sampling strategy
The data collection took place over the course of six months and can therefore be considered cross-sectional as opposed to longitudinal.The justification for using this time horizon for data collection is that it is current attitudes that are of concern, not change over time.Since the aim was to gather the opinions and attitudes about the use of remote sensing technology at CPAs, personal interviews with individuals, one-at-a-time was the selected approach rather than an impersonal survey.The justification for conducting personal interviews rather than surveys or group interviews was to maximize the likelihood of candid, uncensored, and free flowing conversations and minimize any element of uncertainty if unfamiliar with the technology used by critical infrastructure peers.The population from which data was sampled is loosely defined as those who are in key positions to make, influence, or control decisions related to the use of remote sensing technology for physical infrastructure condition assessment at CPAs.The population was broken down into the following key stakeholder groups of relevant parties: • The regulator The population was sampled using practical methods of personal connections, networking, and online open-source research.To maximize candour during interviews, participants were promised anonymity.The individuals interviewed are highly educated, seasoned professionals with deep expertise and broad experience and insights into one or more element of this research.

Data collection
With the premise that some interviewees would appreciate an introduction to remote sensing technology and all interviewees would appreciate an explanation of the context, purpose, and scope of the research, a 1-page memo was prepared for interviewees to read prior to the interview.With the premise that it is not the case that CPAs purposefully and willfully avoid the use of remote sensing technology for physical condition assessment, but rather it is the case that CPAs have not considered the technology, open-ended, conversational, probing-type interview questions were prepared.To secure and conduct interviews, the following methodology was followed: 1.A memo was emailed to potential interview participants with a request for an interview to be conducted remotely either online or by phone.

Interviews were conducted during which open-ended conversational questions
were used to invite discussion of attitudes about the content of the memo and a potential pilot project.3. Responses were critically analyzed, and solutions were considered regarding next steps to overcome the objections and barriers mentioned in the interviews.

Methodological limitations
A significant limitation of the methods chosen is that the results may not be replicable with a different sample of interviewees or with a different sampling strategy.This limitation could be tested with a second set of interviews with people at the same institutions with the null hypothesis that there is no difference in the attitudes expressed.This limitation could also be tested with a second research phase during which a larger number of relevant organizations are sampled, and interviews conducted with the same null hypothesis.Another limitation of the methods is the small number of interviews conducted (eight) with 13 interviewees in total; their opinions and attitudes may not be representative of their firms and agencies.This limitation was mitigated in several instances by the interviewees themselves: unprompted, perhaps because the topic was unfamiliar to them, several people prepared for the interviews by speaking with colleagues beforehand, thus potentially increasing the likelihood that their comments are more representative of their organization and not exclusively their own opinions.Regardless, this limitation could be overcome in future research by expanding the number of interviews with the null hypothesis that there is no 5 Page 10 of 18 difference between the findings from the current sample and the future, larger, sample.Another way to address this limitation with future research would be to distribute a written survey to a large number of people at the same organizations.

Results
This section presents and describes the key qualitative findings from the interviews.Results are categorized according to stakeholder group and subcategorized by theme, maintaining the confidentiality of the interviewees.

Stakeholder group #1: regulator perspective
A. No System-wide Authority willing to Take Charge: Transport Canada is the regulator of CPAs, so the interviewee considered it a conflict of interest to be involved in an effort aimed at assessing the very physical infrastructure that it regulates.The conflict of interest could be removed, the interviewee said, for a pilot project that focuses on the processes and requirements to implement a remote sensing component to infrastructure assessment.The larger issue, according to the interviewee, is that a system-wide assessment of the condition of port physical infrastructure is outside of Transport Canada's mandate.B. No Technical Barrier Expected: The interviewee saw no technical barriers to using remote sensing as a tool to assist with CPA condition assessment and mentioned that CPAs can apply for a waiver of UAV air space restriction via Navigation Canada, which has taken over the role of approving waiver requests from Transport Canada.C. Cooperative Collaboration Desirable: The interviewee suggested that the National Aeronautical Surveillance Program might be able to be involved in a pilot project to help CPAs establish their remote sensing and mission planning competence.

Stakeholder group #2: government agency perspective
A. No System-wide Authority willing to Take Charge: The interviewees expressed surprise that Transport Canada claims that a conflict of interest as the regulator precludes their involvement in a research pilot project.The interviewees expressed concern about system-wide approvals required with respect to unauthorized condition assessment of physical infrastructure.

Stakeholder group #3: private sector user and service provider perspective
A. No System-wide Authority willing to Take Charge: An interviewee from a private sector port terminal tenant believes that Transport Canada should lead the effort because they have the authority and responsibility to assess the entire transportation infrastructure network.An interviewee from a private sector port terminal operator suggested that CPAs have a vested interest in continuing to download more of their physical infrastructure monitoring and maintenance responsibilities and liabilities to their tenants.B. No Technical Barrier Expected: Interviewees from private sector remote sensing service providers expressed keen interest in applying UAV-based remote sensing for condition assessment at CPAs and said that there are no foreseeable technical barriers.Interviewees commented that they have mission planning and execution expertise and can provide human, hardware, and software resources for the effort.
Interviewees observed the importance of including connected rail infrastructure in the CPA network.C. Cooperative Collaboration Desirable: An interviewee from a private sector marine transportation firm said that the correct 'product -market' fit for UAV assisted monitoring would be the entire network of physical infrastructure, disruption of which would significantly impact the supply chain.The interviewee gave an example of a rail bridge that controls marine traffic to oil terminals, which are the source of most of the petroleum products for the region; the impact of vessel wakes and unusually large tidal movements related to climate change would all have an impact on the structural condition of the rail bridge, which could further be affected by a barge or vessel making contact with the bridge structure when passing through.The interviewee also discussed the value of including vessel mooring within the scope of the port network of physical infrastructure because 5 Page 12 of 18 when large vessels are moored to the terminals, wire ropes from the vessel are hooked on to bollards and bitts on the dock structure; bollards and bitts are rated for a certain 'Safe Working Load', but if the dock structure itself is in poor condition, it can cause the vessel to break mooring and potentially cause a pollution incident and human injury.The interviewee said that the impact of increasing size of vessels, powerful engines, wind, wave, and sea conditions would all have an impact on the port physical infrastructure and mentioned impacted stakeholders of terminal owners, cargo owners, local municipalities and first nation communities, all of whom he suggested would provide support to a solution to network-wide condition assessment.

Stakeholder group #4: port authority perspective
A. No System-wide Authority willing to Take Charge: One interviewee said there is 'fragmentation' in jurisdictions around the world where ports are combination of public and private ownership and suggested that because of the intense competition amongst elements of port physical infrastructure, collaboration is missing.This interviewee shared a comment from a British Columbia Provincial Cabinet Minister that 'we made a mistake when we downloaded flood management to BC municipalities because there is no systemwide approach'; the interviewee said that this comment resonated with him, and he thinks it applies to ports in that the downloading of responsibilities to CPAs and other related partners is problematic since there is no system-wide, collaborative approach to investment in and application of innovation.The interviewee said that 'everyone wants to see digitization happen, but no one feels like they have the authority'.B. No Technical Barrier Expected: One interviewee believes strongly that remote sensing technology has great promise for ports and that the reason for slow/lack of use is only because of other priorities and a busy workforce.The interviewee expected that it is not a budget issue because even through the pandemic, revenue and profits were high and growing, but did recognize that this may not be the case for all ports in Canada.The interviewee thinks that if momentum and awareness can be built, ports will embrace and integrate remote sensing technology to the betterment of their physical condition assessment and asset management.C. Cooperative Collaboration Desirable: One interviewee expressed a 'fear that we are talking ourselves into a false sense of security that there is a collaborative and coordinated approach to port emergency management' when that is not entirely true.The interviewee thinks that the Provincial and Federal governments need to incentivize innovation through monetary investment but demand greater collaboration in exchange.

Summary of findings
Three themes were consistently expressed by interviewees from each of the stakeholder categories when asked for their opinions on the barriers that might explain why Canadian ports have not adopted remote sensing for their physical infrastructure condition assessment.The first theme expressed was that there is no system-wide authority willing to take charge of condition assessment of the interconnected web of critical physical infrastructure of CPAs.The second theme expressed was that there are no expected technical barriers to using the technology to assess the condition of port infrastructure.The third theme expressed was a desire for cooperative collaboration between the multiple parties across jurisdictions.

Discussion
Remote sensing technology using a variety of sensors and platforms, everything from thermal sensors mounted on UAVs to radar sensors mounted on satellite platforms, is used operationally for all forms of critical physical infrastructure monitoring in Canada and globally -except ports.Why not ports?There appears to be no technical or scientific limitations to the use of remote sensing technology for monitoring port infrastructure.What are the policy and practical barriers that might explain why Canadian ports have not adopted remote sensing for their physical infrastructure condition assessment?Are these barriers surmountable, and if so, what is the next step to overcome these barriers and adopt remote sensing technology operationally to reduce service interruption in the case of an incident?
Interpretation of theme A: no system-wide authority willing to take charge The interview results suggest that there is no organization that has the authority to conduct a system-wide assessment of the interconnected and interdependent network of port physical infrastructure.The interview results suggest that the regulator of marine transportation, Transport Canada, does not have the mandate to coordinate a system-wide approach to assessing physical infrastructure condition on an ongoing basis even though interviewees acknowledged that holistic situational awareness would increase resilience and efficiency of recovery after an incident, thus reducing service interruption.
The interview results confirm that, beyond managing shipping channels, CPAs do not have the mandate to coordinate system-wide port physical infrastructure condition assessment and that they are not motivated or incentivized to act outside of their mandate on this topic or for this purpose.An insightful comment by one interviewee that 'everyone wants to see digitization happen, but no one feels like they have the authority' suggests that without system-wide leadership and coordination, advancement in using digital tools like remote sensing that other types of critical infrastructure benefit from is not likely to happen.
The analysis of interview results identified "fragmentation" amongst infrastructure owners in jurisdictions around the world where ports are landlords as the explanation for the lack of collaboration between port elements even though they know they are dependent on one another for their success.The results suggest that the landlord port model encourages intense competition amongst elements of port physical infrastructure and explains why collaboration is missing.
5 Page 14 of 18 The interview results indicate a conflict in the understanding of the role of Transport Canada as the agency with authority to lead the effort because they have the responsibility to assess the entire transportation infrastructure network.It is not likely that the Marine Act will be revised to expand Transport Canada's mandate beyond the transportation waterways to include the complete network of port physical infrastructure, and nor does it seem likely based on the interview responses that Transport Canada will take action to lead this effort outside of their mandate.

Interpretation of theme B: no technical barrier expected
While interview results indicate a range of experience and awareness of the operational use of remote sensing technology by operators of critical infrastructure other than ports, all indicated that the technology has great promise for ports and that the reason for slow or lack of use is due to practical and policy barriers.Furthermore, the interview results suggest that there are no significant budget constraints that would explain the lack of operational use.Interview results indicate excitement about the prospect and that 'if momentum and awareness can be built, ports will embrace and integrate the technology to the betterment of their condition assessment and asset management'.

Interpretation of theme C: cooperative collaboration desirable
The interview results suggest that all parties have positive ideas about the opportunity for specific collaborations and cooperation with one expressing 'fear that we are talking ourselves into a false sense of security that there is a collaborative and coordinated approach to port emergency management'.Interview results identified several potential groups for collaboration including the National Aeronautical Surveillance Program of Transport Canada, several diverse research groups at the National Research Council, Defence Research and Development Canada, and supply chain participants inside and outside the immediate port system such as oil terminals, rail bridges, vessel mooring structures, cargo owners, local municipalities and first nation communities.

Recommendations
Policy barriers stop Canadian port networks from operationally benefiting from remote sensing technology, so system-wide condition assessment of physical port infrastructure does not exist.Without situational awareness of the baseline condition and vulnerabilities of the broad and diverse network of physical infrastructure, service interruption is likely to be longer after a natural disaster, accident, or security incident.If there was system-wide leadership to coordinate a collaborative approach using remote sensing for physical infrastructure condition assessment, then business continuity efforts following a disruption could seamlessly build on the existing baseline situational awareness.In this case, valuable time would be saved because the system-wide effort could immediately prioritize and address physical infrastructure deficiencies and minimize critical service interruption.
A pilot project to build awareness and understanding of the available tools and processes would enable leaders within the broad port physical infrastructure network to determine the logistical details and human and technical investments required.A suggested pilot project includes the following steps: 1.One CPA agrees to conduct a pilot project guided by the mission planning and instrumentation expertise of collaborators.2. National Research Council (NRC), Construction Research Centre and other government research scientists with expertise on remote sensing for physical infrastructure condition assessment agree to collaborate with the CPA on the pilot project.3. The CPA prioritizes specific elements of the physical infrastructure within their port network that is within their authority and responsibility to monitor and maintain.4. Collaborators match sensor and mission specification to the expected failure mechanisms of the prioritized physical infrastructure elements.5. Collaborators determine data capture requirements, data processing requirements (hardware, software, people, data storage), and data analysis requirements.6. Collaborators determine the specific requirements for 'product' creation that will be useful for port system-wide decision makers so they can integrate results into their physical condition assessment, business continuity, and emergency planning workflows and processes.7. Collaborators document pilot project lessons learned, limitations of the technology, benefits of the technology, and recommendations for next steps.

Conclusions
In the 1990s, Canada joined the global trend to shift port administration to a landlord model to encourage the competitive facilitation of trade by increasing private sector involvement.The financial performance reported in CPA annual reports suggest that this change has been a success when measured against the intention to facilitate trade.Analysis of the interview results of this study, however, suggests that this success has come at the expense of collaboration and system-wide approaches to addressing problems that cross jurisdictional boundaries such as natural disasters, aging interdependent physical infrastructure, and intentional threats.
The results of this study indicate that too much responsibility, but no system-wide authority, has been downloaded to a fragmented group of port physical infrastructure owners and operators and society will pay the price with unnecessarily long service interruptions in the case of an incident.The recommended next step is a pilot project for system-wide physical infrastructure condition assessment via remote sensing to supplement and fill the gaps inherent in traditional visual inspection and engineering judgment.The pilot project will identify the human and technical resources required for operational use of remote sensing for ongoing condition assessment across jurisdictions throughout the port network, which is essential for seamless application in the case of an incident.Leadership is required and the results of this study indicate that it will need to emerge from outside of formal existing mandates based on a motivation and vested interest in improving safety and security of the complex and interconnected system-wide port physical infrastructure network.

Page 16 of 18
Authors' contributions Heather Holden prepared the research proposal and ethics application, undertook the interviews, and wrote the paper draft.Maha Abdallah and Dane Rowlands reviewed and advised on the proposal, ethics application and supervised the research process.Dane Rowlands edited the paper and prepared it for submission.
Funding There was no funding to this research.Heather Holden undertook the research as a graduate student at Carleton University, and received financial assistance as a student, but not for the project.Maha Abdallah and Dane Rowlands received no funding specific to this research project.
B. Cooperative Collaboration Desirable: Interviewees confirmed that remote sensing of port physical infrastructure falls within the mandated scope of the National Research Council (NRC) and pilot projects involving industry are encouraged.Interviewees suggested that expertise from other NRC groups could provide interdisciplinary collaboration -referencing an NRC researcher with UAV mission planning and data processing expertise.C. Cooperative Collaboration Desirable: An interviewee from Defence Research and Development Canada confirmed that CPAs and NRC both qualify for pilot project funding if the CPA proposes to address public safety challenges with innovative application of science and technology.