Last year while researching marine protected areas in Tonga we discovered an error in many of the online databases of marine reserves that had led to mistaken reporting on a large paper park in Tonga. The Ha’apai Conservation Area (HCA) was a planned MPA that was supposedly going to cover the entire central island group in Ha’apai, and area of 10,000km2. While it appears that initial plans were put in place to gazette this area, after doing some digging it was clear that it was never fully implemented. While it may exist on paper, the proof of a park is in whether there is any difference in management inside it boundaries, compared to outside. Having sailed through this area for the past three years, it is clear that this was not the case.
We have now published a short note in the journal Marine Policy that highlights this critical error and suggests ways to avoid situations like this in the future. The error was first highlighted in 2009 by Dr. Hugh Govan, and at that time this “paper park” accounted for 40% of the entire marine park coverage in the South Pacific!
Large-scale misrepresentations of conservation progress, in this case by two orders of magnitude, represent serious threats to protecting the ocean. While there are checks and balances in place to prevent these sorts of errors, clearly any system that the HCA was not subject to removal need refining. We recommend that checking should focus on larger parks first, or those with the greatest potential impact. Finally, this assessment also gave rise to the more general concern about measuring conservation progress by the area of parks. In theory, this should be one of the more easy ways to measure progress. However, given the complications revealed in monitoring even the geographical extent of an MPA, serious concerns may arise relating to whether it will be possible to accurately quantify their impact and equability.
Last week I had the opportunity of visiting Hong Kong for the National Geographic Sciencetelling boot camp and was able to share the work we’ve been doing in Tonga. We got to attend workshops on public speaking, photography and videography from experts in their fields, as well as learning more about National Geographic’s message. It was an amazing experience and I was able to meet a whole community of people doing amazing work from across the Asia Pacific.
Thank you so much for the amazing opportunity!
Overfishing in Vava’u
During our research last year into the efficacy of community-based marine management we were surprised to find additional evidence suggesting that the reef fish fishery in Vava’u may be depleted or collapsed. During the course of our research we built a model of relative fishing pressure for the island group of Vava’u from information on fishing in the 2016 national census. We used this fishing pressure metric as one of our variables to model the biomass of target reef fish species across the fishing grounds of Vava’u.
As fishing pressure increased (see below), fish biomass declined, which in of itself is expected. However once fishing pressure reached a certain level (units are the abundance of commercial and subsistence fishers targeting reef fish in each village, weighted by proportional differences in catch and overlaid across the fishing grounds of Tonga using decay kernels), it was no longer able to explain any difference in fish biomass. In other words, increasing fishing pressure lowers target reef fish biomass until it reaches a point at which it hits the ‘bottom of the barrel’. Spatially the values of fishing pressure above this point correspond to the inner islands of Vava’u. This provides the first clear evidence that the reef fish fishery in Vava’u may be over-exploited or collapsed.
Importantly both the fishing pressure model and the fish biomass model contain assumptions which are worth exploring. While this is part of our ongoing work, these findings are still important because they highlight the fragility of Vava’u’s reef ecosystem, which as we have seen below is already in poor health.
We recently published our first research article on the Special Management Area program in Tonga in Conservation Letters. Tonga’s SMA program is driven from the ground up by communities and we asked the question: when planning the configuration of a protected area, how well does the community do compared to the best possible configuration? Specifically we asked how much potential recovery is lost by letting communities decide how protected areas are configured. Typical approaches to this question focus on how well different configurations cover a proportional representation of different habitats. However this fails to address impact, or the actual difference made by the protected area. We found that the predicted impact, or recovery of target species biomass, of the community based configuration of protected areas in Vavau was 84% of the best case systematic configuration which aimed to maximize recovery. The similarity in predicted recovery was likely because communities tend to configure reserves close to the village, for ease of enforcement, which I historically where fishing pressure has been highest. These results provide strong support for community based marine management in regions where there is little scope for systematic reserve design.
During the 2017 survey trip through Vava’u we were excited to have met Dominique Serafini. Dominique was an artist for Jacques Cousteau and did many of his comic book series. Before he left he surprised us with a portrait of our work on Chaveta!
From August to November 2017 we completed our first series of surveys in the Vava’u island group. Vava’u is a stunningly beautiful series of islands, and a sailboat paradise full of sheltered anchorages and warm clear water. In total we collected data from 100 sites. Our work in Vava’u was supported by James Cook University, the Australian Research Council Centre for Excellence in Coral Reef Studies (ARC CoE), the Tongan Ministry of Fisheries and McIntyre Adventure. We also had enormous help from the Vava’u Environmental Protection Agency (VEPA).
Unfortunately, while above water Vava’u is very beautiful, the reefs across the island group appear to be in very poor condition. Average live coral cover was only 7%, extremely low, and this was inflated by a number of inshore, turbid, sites that had 20-30% cover of Porites rus, a single hardy coral species. If you factor these out live coral cover was ~4% across the rest of the area surveyed. In addition we also found evidence that that reef fish stocks around the inner islands are either depleted or collapsed. Both of these results are extremely concerning and we have met with the ministry of fisheries to discuss our findings.
While the depletion of the fish stocks is likely due to overfishing, it’s unclear why the reefs themselves are in such poor condition. We don’t know whether this is a recent event or if they have been like this for a long time. One hypothesis is that the Vava’u island group is very sheltered, with lots of bays and areas protected from the open ocean currents and swell, which could result in limited flushing from the ocean. Any pollutants and excess nutrients would therefore stick around. However more importantly there would also be very limited resilience to warm water bleaching events, as pockets of warm water likely stick around and very little cooler oceanic water may enter the system. This highlights the importance of ongoing monitoring.
There have been several other groups that have surveyed the coral reefs around Vava’u, although Ha’apai and Tongatapu remain largely unstudied. Most recently surveys were completed by VEPA, the WAITT Institute, the Ministry of Fisheries (funded through the Asian Development Bank), BioRAP and the Khaled Bin Sultan living ocean foundation. Combined there are approximately an additional 60-70 sites which can be combined with the 100 we surveyed to compile a single data set for the region.
We are excited to announce that we have been granted an OpenROV Trident by the SEE initiative! This will be a huge benefit to us in our future expeditions and we are looking forwards to putting it to good use. Thank you to everyone who has supported our project and we will be continuing to post regularly about our findings from this past trip.
For more information about S.E.E. visit: https://openexplorer.nationalgeographic.com/initiatives/see
And for more information about the Trident ROV visit: https://www.openrov.com/products/trident/?aff=openexplorer
Before we move onto the expedition itself, and the results of our project, I want to spend a few moments discussing how we are able to determine whether a marine reserve is doing its job.
Determining the impact of Tonga’s marine reserve program is a little trickier than just measuring reef health metrics inside the reserves, or even inside and outside (control-intervention), and even if you follow these changes over time (before-after-control-intervention). At the outset, we want to compare what we observe inside the reserve to what is called its counterfactual condition. The counterfactual condition is what the area which is now a reserve would have looked like if it had never been protected (i.e. fishing had continued).
Counterfactual thinking is a subtle, yet extremely important component of impact evaluation that is often missed and has been slow to be embraced by protected area research. Nearby sites outside of the reserve may have less fish than inside the reserve, but they don’t necessarily represent the counterfactual condition of the reserve (i.e. if the reserve was never implemented) unless we are able to control for all of the differences we may find between those sites. For example, if control sites have different wave energy, or habitat than the reserve, then measurements we make may be biased. While these are fairly obvious difference that are often accounted for haphazardly, there are also many more subtle differences that must be considered. Factors like distance from human populations, pollution sources, or fishing pressure must also be considered. Control sites in places with high fishing pressure only represent an accurate counterfactual if the marine reserve would have had equally high fishing pressure in the absence of protection.
Our research in Tonga is trying to use cutting edge impact evaluation techniques to address these caveats. One approach is to pull control sites from a pool of those available that are statistically matched in similarity to the reserve sites across a range of metrics. These include habitat, wave energy, fishing pressure, coral cover, distance to deep water etc. Another approach which we are aiming to develop is to use machine learning tools (boosted regression trees) to build models that accurately predict the counterfactual of a reserve sites using several hundred control sites spread across the country. This technique has not been used before and while at this stage we are unsure if it will work, it is an exciting area of research.
For more information on these techniques please refer to these two articles Ferraro (2009) and Ahmadia et al. (2015) :
How do we quantify reef health, or determine the impact of a marine reserve?
The methods we are using to address both these questions involves conducting underwater visual surveys (UVS). One diver swims along a 30 x 5 m belt transect and records the size and species of all large roving fish that are observed swimming in that area. The same diver then returns along the same transect and records the size and species of all smaller site attached fish across a two meter width. A second diver swims behind and captures 1 x 1 m photos of the reef. These are subsequently analyzed on a computer to determine the percent cover and species composition of the corals and invertebrates found on the reef. At each site this is repeated four times. We have now surveyed over 300 sites around the country, making this the most comprehensive picture of Tonga’s coral reefs that exists.
Many metrics can be used to examine how healthy the coral reefs are. One of the most common is coral cover, which is the proportion of live coral from the photo quadrats. However while a healthy reef has lots of live coral, its diversity is also important, which is why coral diversity is another metric that can be used. While coral cover is important, many communities that rely on fishing for their livelihoods are also interested in the status of the reef fish stocks. This is why we also examine the abundance and diversity of the reef fish, as well as the biomass of key target species. All of these metrics are quantified at each of the many sites we have across Tonga, and put together a picture will start to emerge showing large-scale patterns of reef health across the whole country.
The Special Management Area (SMA) program is Tonga’s version of a locally managed marine protected area network. In exchange for putting aside an area as a permanent no-take reserve, each SMA community is granted exclusive fishing access to the marine area around their village. The SMAs are therefore divided into two parts, the SMA itself, where only members of the community are allowed to fish, and the Fish Habitat Reserve (FHR), which is a portion of the SMA permanently closed to all fishing.
Each SMA is managed by a committee from the village that includes both men and women, as well as youth members. They are responsible for controlling fishing activities, monitoring and enforcement and raising community awareness about fisheries conservation and management. While this program is driven by the Tongan Ministry of Fisheries, it is the responsibility of each community to manage and look after their local reefs.
The SMA program started in 2006 with the village Ou’a in Ha’apai. Between 2006 and 2015 eight new SMAs were implemented. However, in the past three years the program has exploded, with 40 new SMAs being created all across the country. The goal is for every coastal community in the country to have an SMA by 2025.
Ultimately the success of the SMA program can only be measured by the difference it makes, or its impact. Despite its rapid expansion and popularity, it is unclear whether or not they are having a positive ecological or socioeconomic impact. It is expected that stocks of target fish should increase (particularly inside the no-take zone), which should translate into increased food security for the communities and healthier reefs. However thus far the main focus has been on implementation, and very little emphasis has been placed on evaluating their impact. It is unclear which SMAs are working and why, and understanding what makes some succeed can help ensure the program has a lasting effect.
This is why a large part of our research is also focusing on the impact of the SMA program. Mapping the health of Tonga’s coral reefs cannot be done effectively unless we also know whether our conservation actions are effective.
Many people in Tonga rely heavily on fishing for their livelihoods and food security. For remote communities the local reefs provide an excellent source of fish which, if it can be harvested sustainably, should provide food and income for years to come. However, sustainable fishing practices are only useful if everyone is on board and follows the rules.
Unlike many other Pacific Island nations, in Tonga there is no traditional marine tenure, which makes it very difficult for villages that want to sustainably manage their local fisheries. This open access system arose in the 1850’s, when King Taufa’ahau Topou I united all the Tongan islands into a kingdom, establishing himself as the sole owner of the land and sea. During this time, a proclamation by the king stated that i) all Tongans had equal fishing access to all Tongan waters, and ii) any traditional claims to the local management authority over fishing areas were abolished.
Fisheries management in Tonga is now at a crossroads. It is clear that while an open access fishery may have worked reasonably well in an era of subsistence fisheries, modern commercial realities make it unsustainable.
The Tongan Ministry of Fisheries is therefore in the process of redesigning fisheries management in Tonga, and reinstating a new form of community based tenure.
SV Chaveta is a 1974 fiberglass sloop that I first sailed to Tonga from Tahiti. Although old, she has been well maintained, and we have now spent three years sailing and exploring the islands around Tonga. While at 28 ft she is at the lower limit of an ocean going yacht, on board we can comfortably sleep three and still have room for all our gear.
Chaveta has also been kitted with everything needed for a fully functioning research boat. We have a dive compressor and four tanks, as well as all the scientific and safety equipment we need to conduct our research. Fully equipped, we are able to run research trips for two to three weeks before needing to resupply.
Thus far we have completed ecological surveys for almost three hundred sites and are now putting together a baseline data set for the whole country. Once collated, this will provide the first national baseline on the health of Tonga’s coral reefs and reef fish fishery.
The Kingdom of Tonga is a small oceanic country in the heart of Polynesia. Scattered across the sea are 170 islands, as well as thousands of coral reefs that are the backbone of this country. Despite their extensive area, there are increasing pressures being placed on this ecosystem from fishing, tourism, pollution and climate change. In addition, the coral reefs of Tonga remain largely unexplored and the health status of both the reefs and reef fish fishery are unclear.
Despite this significant challenge, the Tongan government has implemented a national program designed to both preserve its natural heritage and increase local food security. The Special Management Area (SMA) program is a Ministry of Fisheries led project that is implementing locally managed marine protected areas throughout the country. While the first SMA was implemented in 2006, the last three years have seen the program explode. Since 2016 over 40 new SMAs have been implemented, covering a large portion of Tonga’s reef environment and creating an extensive network of no-take zones. The goal is for every coastal community to be included by 2025.
We are working in collaboration with the Ministry of Fisheries and various other groups in Tonga to address two key research gaps. The first is simply to explore Tonga’s coral reefs, discovering what is there and mapping their overall health. The second is to determine the impact of their marine reserve network, and make recommendations on how it could be improved.
Our research is being conducted from S.V. Chaveta, a 1974, 28 ft sloop that has been overhauled into a fully functioning, yet tiny, research boat. Our goal is to demonstrate that new ground can still be covered by a small group on a budget.
We have recently returned from an incredibly successful six months in the field and will be posting regularly to share our stories, concerns, results and publications as they trickle in.
Contribute to this expedition
Thank You for Your Contribution!