Data (alone) won't save our forests. Learnings from local restoration heroes
For the past few months, our team at Thinking Machines has been exploring how we can use geospatial data science to support, scale, and accelerate the restoration and conservation of tropical forests. While data science can do some amazing things—like mapping deforestation drivers, or prioritizing areas for mangrove restoration—we know these approaches will only have an impact if they are useful to "frontliners" on the ground doing the hard and (literally) dirty work of guarding and replanting the forests.
So, to better understand these challenges, our team recently went on field visits to two ecosystem restoration projects—a mangrove restoration site in La Union managed by Oceanus Conservation, and an upland reforestation site in Alaminos, Laguna owned by Ayala Land, Inc. and managed by its partners PUNLA and Center for Conservation Innovations. By understanding the work that these organizations do, we hope to eventually develop data tools to help scale and accelerate their impact.
Understanding mangrove restoration with Oceanus Conservation
We recently visited Oceanus’ mangrove restoration site in Aringay, La Union to observe how Oceanus and local communities practice mangrove restoration. Oceanus Conservation is an environmental non-profit organization focusing on the Blue Carbon Initiative, a global program to mitigate climate change by restoring coastal and marine ecosystems. Oceanus’ current blue carbon projects revolve around mangrove restoration through grassroots community engagement in La Union, Misamis Oriental, and Surigao del Sur. Their project in Aringay is an experiment that follows science-based methods to restore a 0.5 hectare abandoned fishpond back to healthy mangrove forests!
We learned from Camille Rivera, Co-Founder of Oceanus Conservation, that mangrove restoration projects can fail if the wrong species of mangrove are planted in the wrong place.
There are 50-60 species of mangroves worldwide, of which over 35 can be found in the Philippines! Each species of mangrove is unique. Some help prevent erosion, while others increase the supply of crabs and fish. At the site in Aringay, we encountered three to four different mangrove species and learned the unique benefits they provide for nearby communities' protection and livelihoods. In 2021, Oceanus received support from Microsoft and Accenture to develop a tool to support mangrove restoration site selection. Now, Oceanus and Thinking Machines are partnering to use data and machine learning to accelerate the process of selecting appropriate sites and recommend correct mangrove species in different areas of the country. “Right now, it requires specialized scientific expertise to identify mangrove species based on the leaves and flowers. Technology could help make this information more readily available to anybody working on the ground,” says Pia Faustino, Thinking Machines’ Director for Social Impact and Sustainability.
Oceanus recognizes the importance of community engagement, which is why they partner with locals to plant mangroves, harvest and nurture seedlings, and collect data on the mangroves’ survival and growth rates. Lailah Lacadue, one of the local community members, runs a mangrove nursery and even does the planting herself. For Ate Lailah, this provides an additional source of income aside from raising bangus (milkfish) and shrimp in the fishponds surrounding the restoration site. Over the past few decades, aquaculture has been one of the key drivers of mangrove loss and degradation not only in the Philippines but around the world. Conservationists like Oceanus need to cooperate with local communities to ensure that conservation and restoration projects benefit not only the mangroves but the livelihood and well-being of fishing communities.
Monitoring the restoration areas is also crucial and painstaking. In the early stages of a mangrove restoration project, seedling survival needs to be monitored monthly. Currently, Oceanus and its community partners regularly visit the site to take photos and measurements of every seedling using a mobile application. Of course, this takes a lot of time and work! During our visit, we took drone shots to see if these could be used to remotely monitor seedling growth. We’ve learned that drone imagery can be useful in a ton of ways in conservation and restoration because of its super high resolution and absence of cloud cover, which is often a challenge when working with satellite imagery. We look forward to further exploring how to process drone imagery to support organizations like Oceanus in monitoring various mangrove features over time, such as tree survival, growth rates, canopy cover, and carbon sequestration.
Terrestrial forest restoration in Alaminos, Laguna
Our team also visited Ayala Land’s 132-hectare carbon forest in Alaminos, Laguna. We were able to witness the rigorous labor that goes into forest restoration—from rescuing fallen seeds from trees and potting new seedlings in the nursery, to removing overgrown weeds and thick vines from young trees.
Another vital part of the restoration process is the meticulous monitoring of the trees. The ALI community uses a mobile application to record the locations and planting dates of the trees. This helps them keep track of their progress and address challenges.
The carbon forest initiative is one of Ayala Land’s programs toward achieving carbon neutrality by 2022. ALI has dedicated 586 hectares of its land bank as forests located in five provinces in the country. Their community partner PUNLA, a people’s organization based in Alaminos, hosts volunteer programs for the public, where groups can explore the forest and help their partners' local rehabilitation efforts on the ground.
The program is open to groups of 10-20 from Mondays to Saturdays at Php1,000 per head. It includes tour guides, snacks, tools for the reforestation activities, and a plant that participants can take home. To join, sign up through this link or contact Crissa Caponpon of PUNLA at (+63) 923-021-5721.
Enhance site selection and monitoring with spatial analysis
Organizations interested in ecological restoration must evaluate a wide range of factors when selecting specific sites. This includes geospatial features like historical forest cover, land tenure, hydrological factors, carbon stocks, exposure to climate risks, and the presence of naturally occurring species. Regular data collection on seedling survival and tree growth rates is also critical for organizations looking to monitor existing sites. With our partners, we’re looking for ways to bring together multiple streams of information, from open data to high-resolution satellite and drone imagery, to aid in planning and monitoring forest carbon sites. And when open data is not enough, we can use computer vision techniques to generate datasets to fill in gaps.
If your organization is working on forest conservation in Southeast Asia, let’s connect! We’d love to learn more and add value to your restoration and conservation work. Send us a message at [email protected]
Stay tuned for more updates as we explore innovative remote sensing solutions for forests!