Re-electrification in Post Disaster Relief: Lessons Learned from PV System Installation in Puerto Rico after Hurricane Maria
Written by Florence Dou and Elin Nybonn, and edited by Trent Dillon
Solar energy can be a valuable tool to support areas struck by natural disasters, but, like most community-based projects, the keys to success depend on trust, flexibility and communication.
As climate change continues and extreme weather events intensify, the importance of disaster relief grows. A vital element of this relief is electricity access. When natural disasters such as hurricanes, earthquakes or tsunamis hit, fallen trees, high winds, or heavy rain can decimate electrical infrastructure, and it can take months before power can be restored.
Meanwhile, a variety of relief strategies, such as telephone communication, lighting and visibility, and, perhaps most importantly, healthcare, depend on electricity. Critical buildings like hospitals can use generators, but many households that need to power medical equipment or refrigerate medicine can end up without any access to electricity, which can have deadly consequences.
When Hurricane Maria hit Puerto Rico in 2017, Dr. Lilo Pozzo — a professor at the University of Washington with personal connections to Puerto Rico — knew that residential re-electrification in rural areas could be life-saving. She assembled our organization, GRID, to install 150 W solar energy systems in Jayuya, Puerto Rico, targeting households that might be reliant on medical devices or medicine and without electricity for a prolonged period.
Five years following the onset of our installation efforts, we’re reflecting on and sharing the lessons we learned from our installation efforts, hoping to highlight some of the most important considerations for similar disaster-relief efforts. Through interviews with our 2017 installation team we’ve identified five key stages of project execution: assessment, logistical planning, design and installation, data collection, and sustainability.
Assessment
Our first step when considering the task of re-electrification in Jayuya was to assess community needs: How many people are dependent on electricity? Who is most vulnerable? What devices do people need to survive? Can they be powered from a solar panel? The more information that can be acquired up front, the more hurdles can be avoided later on.
For this project, our team did this assessment during an initial trip to Puerto Rico, about 2 months after the hurricane. The team went door-to-door to speak with community members, leaders and health professionals such as Jayuya’s mayor, director of the water authority, clinicians at the central clinic, and directors of assisted and non-assisted living health facilities. From this, they learned that many members of the community had medical conditions that were treated with electrically-powered home devices. It was clear that this was a need that was critical to community health and recovery, and also a manageable load for 150 W solar panels.
Due to the urgent nature of this project, the team primarily performed their assessment via word of mouth. When this is the case, care must be taken to avoid biases or accidentally exclude certain populations. Nevertheless, verbal assessment is extremely powerful, as community members themselves will always know their own needs most intricately, and verbal communication is one of the most accessible forms of information-gathering.
This highlights how necessary it is for those providing relief to have a strong relationship with the communities they intend to support. Working with a local partner or someone with established ties to the community may be one of the best ways to build trust, demonstrate that your team has genuine intentions, and also avoid a “white savior complex” that often accompanies foreign aid. In our case, Prof. Pozzo had strong personal ties to Jayuya.
Logistical Planning
The project team estimated that the budget for the installations would be about $31,000 USD. The team applied for funding internally from UW’s Clean Energy Institute (CEI) and Office of Global Affairs’ Global Innovation Fund, and externally from the Resilient Power project of the Clean Energy Group. They also received funds from individual donors.
Typically, administrative processes are slow and bureaucratic, which is something to always keep in mind when applying for funds or planning fundraisers. This being said, the urgency of re-electrification propelled the project team through administrative clearance, enabling them to travel to Puerto Rico within just two months of the hurricane.
The bulk of the funding was spent on flights, hotel lodging, and solar panel supplies. The team avoided burdening local residents by not relying on them for accommodation. However, they did have a resident help with driving. The area was very difficult to navigate due to unsafe road conditions from the hurricane, which included fallen debris and landslides.
Design and Installation of PV Systems
We based PV system design on the household electricity needs that we identified during our initial assessment. However, there were a few other considerations that motivated our design choices. For example, an important thing to keep in mind when estimating power usage is that households might use power in unplanned ways, such as watching TV or powering devices brought to the household by a neighbor or family member. Additionally, during our assessment interviews, we identified interest in solar energy systems that can eventually be grid-tied, to allow households to continue to use the system for a long time.
Due to shipping restrictions and supply chain delays, any imports had to first go through the US mainland before being shipped to Puerto Rico, which delayed arrivals and made them hard to predict. Therefore, the team attempted to limit design components to supplies available on the island. Batteries in particular were too expensive and time-consuming to ship due to their bulkiness and handling conditions, so the project team bought batteries locally, partnering with a local solar company.
The team did the bulk of their installations during a second trip to Jayuya, about 6 months after the hurricane. When doing the installations, the team ran into a number of technical issues such as loose wiring and roofs that were somewhat unsafe to access. Looking back, had the project team examined the roof type, house height, and tree coverage at each house during the assessment stage, they could have minimized unexpected hurdles. This may have also allowed the team to plan alternate setups for roofs that were especially difficult to install on.
In our interviews with the project team, we identified that powering a community clinic, as opposed to individual households, may have been a more equitable way to ensure that everyone who needs power can access it. However, this may have presented different and unanticipated hurdles, such as transporting larger components.
Data Collection
The project team made use of both data acquisition systems and Institutional Review Board (IRB) approved interviews to learn more about the efficacy of our installations. This information was acquired during a third and final trip to Puerto Rico, about 4 months after the installation.
Along with solar panels, the project team installed data loggers that track energy consumption, load profiles, and battery degradation over time. During the third installation trip, GRID collected data from the loggers and conducted interviews with each household. Through data analysis, the project team was able to deduce which devices were being used at specific times and how much power was needed.
Data collection was helpful for reviewing how well each system performed under the energy needs of each household over time. However, it would have been better if the loggers were set up to include live, remote tracking of load profiles and tracking when specific devices were being used. This way, problems with the panels and batteries could be addressed as they arose, rather than simply recording the information.
The project team’s interviews relied heavily on households’ memory for data collection, which could be a source of error, as a few months passed between trips. It is also a large burden to ask families to record their daily activities. Having remote data collection could have allowed more space in interviews to ask qualitative and holistic questions about the needs, concerns, and hopes of the community, and how they could be further supported. This being said, the third trip to Puerto Rico was absolutely critical for maintaining a relationship with the community and checking in with their well-being.
Maintenance, Sustainability, and Longevity of Systems
During disaster recovery it can take anywhere from a couple weeks to years for power to return, so it is important that any re-electrification system has longevity.
One way to improve longevity is to select an appropriate battery, preferably with minimum replacement or maintenance needs. For this reason it is helpful to use batteries that do not need to be refilled with water, and to encourage users to stay above the safety limits for discharge to maximize battery life.
During installations in Puerto Rico, each household was given a manual explaining how to use the systems, as well as a verbal walkthrough. However, the team had limited time per installation, so it was difficult to ensure the systems were fully functional before moving on to the next. Along with the manuals, the team provided contact information to help with troubleshooting. Team members occasionally received phone calls, but it was difficult to help over the phone. This was another equity concern, as many households do not have the resources to make international phone calls.
Another longevity problem that arose was student turnover. The contact person for troubleshooting eventually left UW and was unable to continue helping. This emphasizes the need for creating a sustainable system for households to be able to continue independently maintaining and using their panel for many, many years. Another option would be to establish a local contact for help with maintenance, especially in cases of older people or patients who were not capable of troubleshooting by themselves.
Conclusion
In all stages of installations in Puerto Rico, our work depended on having strong and healthy connections with the communities in need of relief. Building relationships with community members helps establish trust and shows good intentions, and it also opens opportunities for collaboration. Having locals who can give information on community needs or help with long-term maintenance is of tremendous benefit. Because of the fast pace of disaster relief work it is hard to plan a project perfectly, but simple things like spending time and attention on trust and interpersonal communication can help make sure the project serves the community as best as possible.
This lesson, and others we have described in this article, are good to consider when planning disaster-relief projects, but, there will likely be many other unforeseen obstacles that arise along the way. So, it is important to remember that the success of the project depends not just on avoiding potential problems, but also on the ability of your team to work together with each other and with community members to solve them.
The team published two papers on the results of this project, if you’d like to learn more, check out the links below:
