Why Did We Spend Our Past Year Preparing for This Trip?

Today the researchers are converging on Palmyra Atoll by land and by sea, the culmination of a year’s planning and work. Excitement runs high, fueled by a touch of anxiety as minds race to remember what has been forgotten, misplaced, left behind. Time to pause and think about why all this effort.


by Stuart Sandin, head researcher on the Fish Team

About two years ago a bug landed in my ear. Not a literal bug, but an academic bug. Our research in the Line Islands and elsewhere had told us quite a bit about how coral reefs used to look before humans changed the underwater landscape in most locations on the planet. We had determined conclusively that human activities systematically reduce the number of large fish, favor fleshy seaweeds (algae) on the substrate over long-lived corals, and create an environment that supports more microbes. But we still did not know what these consistent findings meant for the future of coral reefs.

The services that coral reefs offer to humanity are many, probably the two most important being the harvests from their productive fisheries and their role in shoreline protection. The greatest value of the fisheries lies in the fish that can be caught and eaten. While it is interesting to know how many fish currently live on a reef, the fishing community cares more about how many fish will be there tomorrow and the day after. In other words, the asset of interest is not the current density of fish, but rather the productivity of the fishery. Similarly, from the standpoint of shoreline protection, what is important is not only how much coral exists today, but the rate of reef construction by the coral and other reef builders in the benthos. Our previous research measured what was living on the reef today (and how the organisms present correlated with local human activities), but we had no direct measure of the productivity of these organisms. Were the fish and the corals growing rapidly, or were they just lazing around, passing the time, maintaining, but not growing a bit?

Given this question—my academic bug—we started thinking about a shift in our research focus. Perhaps we should be asking: Is a coral reef that is ‘pristine’ also very productive? Or might a slightly altered coral reef actually be producing much more fish, and perhaps more coral, thereby offering a functional and more valuable ecosystem for human use? To answer these questions, we turned again to the natural experiment provided by the Northern Line Islands.

The Northern Line Islands comprise an archipelago of five islands (six if you include Jarvis that lies less than one degree south of the equator) that span a broad range of human population, and thus of human influence on their surrounding coral reefs. Kingman, Palmyra, and Jarvis are uninhabited, and thus their reefs are experiencing no local human influences. These islands serve as our baseline. In contrast, Washington, Fanning, and Christmas are all inhabited islands within the Republic of Kiribati. Their coral reefs serve as the main source of protein for the local populations (approximately 1000, 3000, and 8000 people, respectively). Thus, within this one archipelago in the central Pacific we have a ready-made experiment comparing three inhabited islands with three that are not.

Well, not exactly perfect. There is a reason that we said a natural experiment. Experiments conducted in the laboratory, in contrast, are set up so as to study the effect of only one factor at a time on the system of interest. In these lab experiments we do not allow other factors to influence the results, i.e., in the terminology of science, we use control experiments to eliminate the influences of additional factors from our results. In our natural experiment measuring the impact of human disturbances on coral reefs, we would want to control for the effects of other factors that might be different at each reef, factors such as their oceanography, geology, and climatology. In other words, we would want all of these non-human influences to be identical across all of the studied reefs. Unfortunately (or fortunately!), the Earth is much more interesting than this and the islands are not perfect oceanographic, geological, and climatological replicates of one another. Instead, as one travels from north to south in the archipelago, the water becomes slightly cooler, nutrient levels in the water increase, and the effects of El Niño grow more profound.

In our earlier 2005 expedition to the Line Islands, we focused our study on only four of the islands: Kingman, Palmyra, Fanning, and Christmas. These four provided a strong north-to-south gradient of human population (increasing from Kingman to Christmas), but they also exhibited  a gradient of water temperature (decreasing) and of El Niño activity (increasing). We could not rigorously distinguish the effects of human activity from those due to oceanography or other factors as required for proper science. Our study design was confounded. Our data demonstrated the reduced number of sharks and big fish around the southern islands, but we could not conclusively say how much of this was the result of people versus how much was due to changing physical conditions.

We have learned from this shortcoming. The past 5 years have been spent gathering more data from more regions to robustly separate the effects of these various factors on reef ecosystem structure—their effects on the fish, the benthos, and the microbes. During this cruise, in order to avoid those confounding influences, we have added two more islands to our study: Washington, an inhabited island in a mild oceanographic region, and Jarvis, and uninhabited island in a much more challenging oceanographic region. With these additions, we will be better able to separate the effects of people from those due to oceanographic conditions in our natural experiment.

And to satisfy that original bug, we will use a suite of both established and innovative field techniques to measure the productivity of the reef community. For the fish, our measurements will show the rate of production of new fish biomass. For the benthic organisms, we will estimate how quickly they are growing today and how much their patterns of growth have changed in the past. Lastly, we will monitor the microbes to see how rapidly they are respiring and growing. Perhaps most importantly, we will be doing all three at the same time in the same locations. This approach will give us a measure of reef productivity from a broad ecosystem perspective and also enable us to test whether increases in productivity in one group are coupled with decreases in another. We have hypotheses regarding what we may find, but we will save these for future posts.