“IT WAS LIKE a constant adrenaline rush—we knew we could die any second—it was the climax of our struggle, and all our secret activities were suddenly out in the open.” February 22, 1986: The People Power Revolution was under way in the Philippines. Early that morning, Deng called Daniel. “I won’t be coming into the office today. Don’t ask me where I’m going,” she said. She joined the human chain that was blocking the entrance to a military camp. “Part of Marcos’s army had defected, so we went to protect the soldiers with our bodies, because they’d had the courage to say no. We lasted four days, and in the end, the whole army abandoned Marcos, and he fled.”
The first peaceful revolution in the Philippines was a success: Corazon Aquino took power, the first woman president in all of Asia. “I get teary-eyed just thinking about it,” Jay Maclean tells me. “My wife, Margie, was up on the front lines, helping out with the election and then going up to stand in front of the tanks. It was very emotional, scary, the prospect of bombs and tanks firing on you.” This popular uprising meant the end of a long struggle, one that had been exacerbated by the assassination of Ninoy Aquino on August 21, 1983. “The Philippines has a martyr,” one French newspaper announced. A longtime opposition leader who was imprisoned, then later exiled to the United States, Ninoy braved great danger by returning to his home country on a commercial China Airlines flight, traveling with journalists and television cameras. When they arrived in Manila, Marcos’s thugs handcuffed him, hauled him off the plane, slammed the door shut on the other passengers and the media, made Ninoy get down on his knees, and put a bullet in his head before throwing his body onto the tarmac from the jet bridge. His wife, Corazon, left her life as a proud housewife behind to take up her husband’s torch and challenge Marcos’s reelection in early 1986, causing the dictator’s downfall and making her the shining light of a popular uprising in which thousands of Filipinos took to the streets.
“It was only then that Daniel, like a lot of people I knew, understood how involved I was,” remarks Deng, who had worked with him for four years at that point. “When I came back, he opened up to me, talking about his life and what a dilemma it was for him to be a leftist working in an international research center with an expatriate’s salary. But we agreed that it was through our work that we were going to change things,” she concludes.
And Daniel did work, even under fire. His friends and family recall that, from his rooftop terrace, he would watch the dance of military helicopters and fighter jets over the city, evaluating the risk of taking his family through Manila’s different neighborhoods. His system wasn’t infallible, however. “During one of the military coups in the late eighties, the ICLARM board was meeting in the Mandarin Hotel,” recalls Jay. “We went out to eat, and I remember saying, ‘I don’t think we should be here’—there were people shooting up and down the streets in Makati. But Daniel insisted we get back to the office and continue the meeting through the chaos. He was the driving force.”
Manila was a political and climatic hotbed, but the family escaped for a few weeks a year to vacation in California with Sandra’s family. Their flight usually stopped in Hawaii, which they also visited, and Daniel, of course, went to meet the local researchers. Among them was Jeffrey Polovina—a tall, slim math whiz from the Honolulu Laboratory of the National Marine Fisheries Service. Jeff had joined this unit of the National Oceanic and Atmospheric Administration in 1978. In the early 1980s, his supervisor, Richard Shomura, asked him to create a mathematical model of an entire coral reef, the French Frigate Shoals, called Kānemiloha‘i in Hawaiian. The English name for the atoll, located in the middle of the North Pacific 560 miles northwest of Honolulu, honors the French explorer Jean-François de La Pérouse, who, while navigating the shoals in 1786, nearly lost two ships, the Boussole and the Astrolabe. La Pérouse had come from California and was en route for Macao, Korea, Sakhalin, and Kamchatka, before going to his eventual doom in the Solomon Islands of the South Pacific.
The French Frigate Shoals is now part of the largest marine protected area in the world, Papahānaumokuākea. The atoll measures eight miles in diameter and has a landing strip at its northern end that dates from the Second World War, as well as a marine ecology research station, both of which are situated only a few yards above the waves. These sites have to be evacuated regularly during storms and tsunamis. Over the decades, scientists have studied every aspect of the local ecosystem: the coral, the reef’s tiny inhabitants, the fish in the lagoon, the ocean sharks, the multitude of marine birds who nest and feed in the area, a sizeable population of sea turtles, and the world’s largest group of Hawaiian monk seals. In this faraway island paradise, each researcher tended to obsess over a single species or group of species, each so unique and extraordinary that no one had bothered to work on the big picture yet. The scientists knew that the coral ecosystems are one of the most incredibly productive in the world, that they are an oasis of life in the midst of the mostly clear and nutrient-poor tropical oceans, but they also knew that their coastal fisheries were largely in decline.
Jeff Polovina wanted to understand the big picture of how coral reef ecosystems function in order to help with their conservation. He was familiar with models developed by colleagues in Denmark (for the North Sea)1 and Seattle (for the Bering Sea),2 and went to see them. He returned deflated, certain that their highly complex simulations, which require a huge amount of very precise information from various sources, could never work for the coral reef ecosystem of the French Frigate Shoals, nor for other marine systems for which much less information is available.
But Jeff was pragmatic and creative. He simplified his colleagues’ models and ended up with the following structure: rather than considering species individually, he grouped them together according to their role in the ecosystem, like different professional categories in human society. All the phytoplankton made up one functional group, the zooplankton another, and so on for the reef fish, the crabs and crayfish, the carnivorous fish, the birds, and the marine mammals. Polovina drew a diagram that looked like a subway map: each station, marked by a box, represents one functional group, and the lines between the stations indicate who eats whom. From there, you need to know, for each box over a one-year period, the total mass of all the organisms in that functional group, the total mass of food ingested (which has been pulled from one or more of the other boxes), total weight gain for each member of the group, and the mass caught by humans (if the group is exploited by fisheries). At the NOAA laboratory in Honolulu, everyone lent a hand to provide the data needed to complete Polovina’s giant puzzle. Once that was done, multiple equations connected the boxes to one another, calculating the total mass and the total production of the ecosystem in terms of biomass for the year in question.
In an ecosystem-wide snapshot like this one, balance is the main concern: there has to be enough plankton to feed the higher levels, from the reef grazers to large predators like sharks. At the base of the whole thing is energy from the sun, which feeds microscopic algae through photosynthesis. In order to test his model, Jeff Polovina worked with Marlin Atkinson and Richard Grigg, who measured coral respiration. Though it seems incredible, the two researchers managed to enclose whole heads of coral and measure the concentration of oxygen in the water around them over time. The more intense photosynthesis is, the higher the rate of oxygen production, and the more the coral grows and produces biomass that will feed the other inhabitants of the reef. Thanks to their elegant coral trapping operation, Atkinson and Grigg were able to estimate the biomass produced by photosynthesis. Their results were similar to those predicted by Polovina, validating his model. “No one was more surprised than I,” he wrote ten years later.3 Jeff named his computational routine Ecopath. Once it was used to model an ecosystem, it could also predict the characteristics of any missing boxes. Ecopath thereby made it possible to estimate the biomass of fish “produced” by an ecosystem, and potentially made available to fisheries, each year.
Always modest, Jeff Polovina published his results without much fanfare: “I was busy elsewhere and I had doubts about Ecopath’s acceptance, given some criticism that the model was overly simplistic,” he says. Daniel certainly didn’t think it was overly simplistic, though, and explained to Polovina that, on the contrary, a simplified system was exactly what everyone was looking for. “He told me that if I made Ecopath user-friendly and wrote a user’s manual, he would see that it was applied around the world,” Polovina recalls. “I did my part, and he certainly did his. For several years, the requests I received for Ecopath material often arrived in batches by country; thus, I could track the locations of Daniel’s seminars on Ecopath.”4
“Thanks to Ecopath, Daniel reached another level,” comments Jay Maclean. Daniel had been dreaming about an ecosystem-wide approach since Sakumo, and since San Miguel Bay, he had been working to make it real by studying the writings of Danish scientists Erik Ursin and Per Sparre, who developed multispecies models for fish populations. This was also a logical follow-up to his previous work: the various versions of ELEFAN made it possible to quickly estimate the biomass of each group of species that made up Ecopath, the two routines fitting together like Russian dolls. Daniel mentioned Ecopath enthusiastically in the book he coauthored with Alan Longhurst on the ecology of tropical oceans and in the document he produced for his habilitation to direct research.
In fact, while Daniel was working on his habilitation in Kiel in 1984, Silvia Opitz came knocking on his office door. She had met him at the oceanographic institute in the 1970s. Since then, Silvia had spent several years in Brazil before returning to Germany via the Caribbean, where she had become fascinated with coral reefs. And now she was looking for a thesis topic. “I have something for you,” Daniel said right away, pulling out Polovina’s freshly written article. And so Silvia became the test pilot for Pauly’s brand of Ecopath. The duo set their sights on the coral reef ecosystem of the British Virgin Islands because, as Daniel notes, “we knew John Randall had collected a huge pile of data on the fish there in the 1960s.” Daniel sketched a network of boxes connected by lines, an Ecopath model suited to the coral reefs of the Caribbean. To feed this needy brainchild, which would later become the most detailed model ever constructed of a coral reef ecosystem, Silvia drew on Randall’s publications, but also on the FAO’s archives in Rome. “Cornelia Nauen worked there at the time, and she was a big help,” Silvia remembers. “My husband (who had come to assist me) and I stayed with her.” After her Roman holiday, Silvia, who was soon to be a mother, returned to Kiel. “Other than the fish, it was tough to find all the necessary information for the other organisms in the coral reef ecosystem,” Silvia tells me. “But Daniel was persistent and infinitely creative, and I became an expert at extracting data from the scientific literature myself.”
Silvia started a family while she was working on her thesis. She tells me about those golden years over lunch on a snowy day in Kiel in early 2016. Silvia has a charming Berliner accent, fast-flowing and low-pitched. “With Daniel, we worked like equals—there wasn’t a big age difference, and he’s always gotten along well with his female colleagues. He’s nothing like the stereotype of the dried-up scientist. He’s kind of out there, but besides being a great researcher, he’s also a great human being, very warm and funny.”
Once he’d finished his habilitation, Daniel made only short visits to Kiel. In a way, this worked out well for Silvia because their scientific conversations were so intense that she needed time to process everything. During her thesis, tons of documents and manuscripts shot back and forth between Kiel, Manila, and Daniel’s many other destinations via fax or post. “The few times Daniel was actually in Kiel, he edited my thesis chapters at impossible speeds, advising me on the scientific content but also on the writing. His worries about research funding, his colleagues’ jobs, etc., none of that entered the conversation—he stayed focused on the science.” Silvia would have liked to focus on science, too, but she also had to make a living. “It’s hard for women, even in Germany,” she admits. After a rich but somewhat tortuous career, she is still an untenured associate researcher at Kiel’s oceanographic institute.
During the second half of the 1980s, Daniel wasn’t into playing games of power and scientific strategy like his mentor Gotthilf Hempel, who recruited armies of young colleagues by cultivating his political connections. Influenced by the feminist movement of the 1960s, Daniel was, however, perfectly aware of the lack of support for women in research. He published an essay on the subject, which began with some insightful remarks on the influence of researchers’ sex on their interpretations of ecological phenomena.5 In particular, Daniel relates that all the male scientists consulted on the subject believe that male sea lions actively form their harems, whereas the only woman* to have studied the behavior of these amiable creatures concluded that it was the females who chose to share a limited number of males because all they do is lie around consuming fish without contributing to the care of the young. As Daniel concludes, “The problems which man(?)kind faces are simply too big to lose one half of the world’s potential scientists just because some people confuse sex (a biological fact) with gender.” He cites a large body of literature, highlighting the scientific basis of his claims, but also talks about his daughter, Angela, whom he has always told, “Yes, women can and should become professional divers, scientists and airplane pilots.” Daniel mentions that “Special encouragement is thus needed, such as the fellowship created exclusively for female scientists by the Canadian government.” This short text, not even four pages long, would resonate strongly with many of Daniel’s female colleagues all over the world. “I thought Daniel was kind of macho,” comments Coleen Moloney, professor of oceanography at the University of Cape Town in South Africa, “so I was very surprised.”
Daniel continued tinkering with the first version of Ecopath through the late 1980s. But it was only after Villy Christensen’s arrival in Manila in early 1990 that all the ingredients came together. Villy was born in Skagen, in Denmark’s far north, where the waters of the North Sea and the Baltic mix together in spectacular fashion. Both sides of his family had been fishermen for generations. Villy grew up in the port of Hirtshals, thirty miles southwest of Skagen. “It looked like this,” he says, showing me an aerial photograph: the port full of hundreds of boats, including Villy’s father’s, which bears the registration number S211. The young Christensen didn’t become a fisherman, though, and he was the first member of his family to go to college, studying biology and mathematics at Aarhus University, then marine biology in Copenhagen. Hired by the Danish Institute for Fisheries Research in 1980, he worked there for ten years, mainly on herring larval growth. In late 1989, the development arm of the Danish Ministry of Foreign Affairs offered to assign him to ICLARM. Villy was aware of the political situation in the Philippines, which was undergoing a series of attempted coups d’état—and the risks to his family—but he decided to accept: “I had met Daniel in 1989 in Kiel, then again in the Netherlands, and we had very interesting discussions—that was my main reason for going.”
As soon as he arrived, Villy took up the frenetic rhythm of work at ICLARM. He began by reprogramming Ecopath. “Polovina’s version was a minefield: if you tried to do anything unexpected, it all fell apart. That whole generation of models was unstable, and it wasn’t unusual for them to just not work at all.” Villy’s priority was to make the new version of Ecopath more flexible, “so it could adapt to any ecosystem.” Additionally, the computer language had to be easy to use by a large number of people. Daniel and Villy also wanted to use Ecopath to assess the condition of the ecosystems they studied, notably the trophic level of the fisheries therein. Trophic level refers to the position of a group of species on the food chain or pyramid. The trophic level of plants is set at one, two for herbivores, three for the herbivores’ predators, et cetera. Classifying food chains by trophic level was suggested in 1942 by Raymond Lindeman, but the idea was later rejected because species, or groups of species, often feed at multiple trophic levels. This is the case with omnivores, including us, Homo sapiens: we would have a trophic level of two if we were all vegetarian and be classified as level six superpredators if we consumed only marine mammals. Three decades later, William Odum and Eric Heald suggested recalculating trophic levels not as whole numbers, but as decimals.6 Humanity as a whole therefore has a trophic level of about 2.21,7 similar to the anchoveta, but with strong disparities between countries: in Burundi, where the diet is 96.7 percent plant-based, the mean trophic level is around 2.04, whereas Icelanders, whose diet is 50 percent meat and fish, have a mean trophic level of 2.57.
Daniel and Villy added a model to Ecopath that made it possible to calculate the mean trophic level of each species and the trophic level of the whole ecosystem. The duo also drew heavily on the work of Robert Ulanowicz to come up with an indicator for ecosystem health. “Daniel was fascinated by Ulanowicz’s work,” Villy recalls, “and especially by his 1986 book on growth and development,8 despite how tough it was to get through.” At the time, Ulanowicz was a professor and ecologist at the University of Maryland, where he studied the organization of flows of matter and energy within ecosystems. He developed a measure of ecosystem performance called “ascendency,” which combines the diversity of groups of species present in an ecosystem, the degree of specialization of those groups, the intensity of the links between each group, and the speed with which energy and matter flow through the ecosystem. An intact natural system will have higher ascendency than an ecologically degraded one. Much impressed by this theoretical leap forward, which was also being used in cognitive psychology and economics, Villy and Daniel included a measurement of ascendency in the new version of Ecopath. “A very productive combination,” Villy comments soberly.
With these additions, Ecopath II was born. The open-access program was made available to the scientific community in July of 1990 and would also be the subject of an article coauthored by Christensen and Pauly,9 the first in a long series and a great classic in the domain of ecosystem modeling. In the wake of their publication, the ICLARM researchers organized a conference with the International Council for the Exploration of the Sea (ICES )on Ecopath II in Copenhagen in October of 1990. “There were twenty or thirty posters describing different ecosystems—in just a few months, the number of models available had doubled worldwide,” recalls Daniel.
A photo shows Villy and Daniel—more than a little proud—standing on either side of one of those posters. The author is one A. D. Pongase of ICLARM. They often joked about him back in Manila. In fact, Pongase—roughly “put yourself there!” in Spanish—was a fictional colleague that Jay Maclean had invented while he was working in Australia. Jay had brought him along when he came to work at ICLARM, where Pongase signed reports and evaluations of scientific articles by colleagues abroad, especially when the work in question was lacking and the criticism particularly acerbic. The Copenhagen conference was the high point of the enigmatic Pongase’s scientific career: he received a letter from the director of ICES congratulating him on his work. On a more serious note, the great Ulanowicz himself soon showered them with praise: “One day a parcel appeared in my mailbox containing some 50 or more quantified food webs, replete with accompanying ascendencies. It was perhaps the most startling and gratifying moment of my professional career. . . It may not be much of an exaggeration to say that the realm of ecosystems is being opened to us by Polovina, Pauly and Christensen through their ‘ecoscope.’ For that is what ECOPATH II and its associated analyses represent—a macroscope through which to view the structure and functioning of entire ecosystems.”10
Villy and Daniel spent the next two years editing the scientific articles that resulted from the conference. Contributions came from every continent, covering lakes, rivers, aquaculture systems, coastal zones, coral reefs, and offshore areas. The final product, a four-hundred-page volume published in 1993, quickly made its mark on the whole research community in marine ecology.11 When I complimented Villy on the work’s scope, calling it the “Copenhagen conference proceedings,” he stopped me right away: “No, they aren’t ‘proceedings’—it’s a book, that’s very important.” Somewhat incongruously, the “black book,” as Villy and Daniel call it (a reference to its dark cover), contains several models of aquatic ecosystems in France: Aydat Lake, the Garonne River, and the Étang de Thau. These three chapters are written by Deng Palomares, who is also responsible for modeling Lake Victoria, Lake Tanganyika, and Lake Chad. But what was Deng doing in France?
“Daniel was working with Jacques Moreau of the National School of Agronomy* in Toulouse,” Deng tells me. “I wanted to do a thesis, and in 1988, I had been awarded a scholarship to study in Kiel—I had even started learning German at the Goethe-Institut in Manila. But just before I was supposed to leave, in spring of 1989, Daniel told me that Moreau was looking for doctoral students.” Deng didn’t have anything against the switch, and Daniel explained that Jacques Moreau knew and applied all his methods, so she wouldn’t feel lost. Never short of arguments, he added that French was spoken much more widely than German, especially in Africa. “Four weeks later, I landed in France,” Deng tells me, laughing. “I learned French in three months in Vichy. I liked the class—we also learned about French culture and cuisine.”
In September, Jacques Moreau—a specialist in freshwater African fish and a somewhat whimsical personage, the very picture of a French colonialist—welcomed her to Toulouse. “Despite all that, Daniel was his god,” Deng tells me. Indeed, Moreau later translated into French and edited a set of Daniel’s method papers, covering everything from his thesis to Ecopath II.12 Moreau was kind to Deng, if a bit odd. Her first week there, she was in student housing. Moreau came to see her, sat down on her bed, and asked for a cup of coffee. Deng answered, “That is my bed, that is a chair. My bed is for me, the chair is for visitors. If you want coffee, we can go to a café. I don’t make coffee for anyone, not even Daniel.” Laughing, she admits that, even in Manila, everyone made coffee for Daniel without his needing to ask, simply because he didn’t have those kinds of sexist attitudes. Moreau’s main job was to protect Deng from racist colleagues, particularly one Monsieur Gilles, a perpetual doctoral student who whiled away his time at the Toulouse laboratory mumbling about “these foreigners who take the French government’s money.” When he became violent, Moreau sheltered Deng in his office, but Gilles managed to sneak in and destroy the floppy disks on which Deng had saved her thesis manuscript. “Luckily, I had hidden copies everywhere! Moreau was in a fit, and Daniel called the student to tell him that he was going to break his face.”
Deng’s memories of her time in Toulouse aren’t all sunshine and rainbows, but when it came to her research, everything went smoothly. “One of the goals of my thesis was to show that our methods of analysis for fish growth were just as valid for freshwater fish. It turned out that salinity doesn’t make a difference, which confirmed the importance of temperature.” As always, Deng was one step ahead. “When I arrived in Toulouse, my thesis was already well on its way. I had a hypothesis, the bibliography, the introduction, methodology, some of the derivations. All that was left were the experiments and the modeling. I spent a year measuring fish growth, and in eighteen months, it was done.”13 Deng used Ecopath, with which she had become familiar in Manila. “At ICLARM, I worked with Astrid Jarre, who was good with numbers. She put together a series of models of the Peruvian upwelling ecosystem.” Deng, for her part, put all of Moreau’s beloved African lakes “in a box” before doing the same with the Étang de Thau and other French aquatic systems. She especially liked Auvergne and its Aydat Lake, as well as Lake Pavin, which was modeled by her colleagues in Clermont-Ferrand. Impressed by her work and her mastery of French, the Clermontois team offered her a teaching position at Blaise Pascal University—it was a tempting offer, but she still felt that going home was the right thing to do. She politely declined.
ECOPATH II MADE a splash in the 1990s, but it was still a static model, a snapshot of an ecosystem’s condition, whereas ecological processes are by nature constantly in motion. In order to create something more dynamic, Astrid Jarre, another of Daniel’s doctoral students, generated separate Ecopath simulations of the Peruvian upwelling system for each of the twelve months of the year—but it was hardly a perfect fix.
The next qualitative advance would be initiated by Carl Walters. I meet Carl for lunch in Vancouver in 2015. He seems happy with his status as emeritus professor at the University of British Columbia, which allows him to take his afternoons off to do yoga and visit with his granddaughter, whose photo he shows us proudly while we eat our sandwiches. A half day is more than enough for him to get everything done—especially since he wakes up at five in the morning. We have an interesting chat about animal physiology and then about the world economy. I like Carl a lot and I know he’s a force to be reckoned with in mathematics, the laureate of half a dozen big international prizes for his work. A native of Albuquerque, New Mexico, he studied at Colorado State University and defended his thesis there in 1969. At just twenty-five years old, he immediately found a job at UBC, where he would work his whole career while also consulting for the Canadian Department of Fisheries and Oceans.
When Daniel met him in the 1990s, Carl had recently published, among a slew of other scientific articles, a huge tome on the adaptive management of renewable resources,14 and an equally hefty volume on fishery management, coauthored with Ray Hilborn.15 In this latest book, he had forcefully criticized ecosystem models as too unreliable to be of any use. And yet, when Polovina, Pauly, and Christensen launched their first Ecopath training session in Vancouver in 1995, Carl, always curious, joined in quietly alongside the students. As the days passed, his visits became more frequent until, one morning, he arrived with a spreadsheet he’d cobbled together overnight. Carl had transformed Polovina’s linear equations into differential equations—Ecopath had acquired a temporal dimension, making it possible to simulate the dynamic evolution of ecosystems over time. Ecosim was born.16 This computer program, improved over the years, would also make it possible to test the impact of all kinds of fisheries management schemes: “What will happen in my study area if fishing is reduced year after year, or if, on the other hand, it doubles? How many fish will be available if global warming reduces the growth of phytoplankton by half in a ten-year period?”
Once Ecosim went online in the form of an open-access program, Villy and his colleagues began spreading the good news, hosting more than fifty training sessions in twenty-two countries. Ecopath with Ecosim (EwE) would be used by six thousand researchers, inspiring a whole community to generate not one but a whole family of models, not unlike the different climate models used by scientists at the IPCC.* In 2017 alone, EwE was the basis for sixty scientific publications, the latest of which is currently spread across my desk: this article tells the story of how Catalan scientist Marta Coll’s team collaborated with Villy Christensen to make a simulation of ecosystem dynamics in the Mediterranean Sea for the period between 1950 and 2011. Their work shows that there has been a strong downward trend in both the populations of fish and their nonhuman predators, as well as the cause of this decline: a combination of climate change and overfishing.17
* Daniel doesn’t mention her by name, but that woman is Patricia Majluf.
* The National School of Agronomy, or École nationale supérieure agronomique (ENSA )in French, is a highly selective and prestigious teaching and research institution. (TN)
* The United Nations’ Intergovernmental Panel on Climate Change.