Alfred C. Redfield's Remarkable Ratios


Ecologists have individualistic perspectives on how nature works because our personal interests in nature (plants, birds, soils) subsequently become modified by our encounters with teachers, historical events, individuals, and ideas. I suspect that, upon reflection, most of us could name a dozen or so persons, courses, papers or books that have had a special influence on our career trajectories. The ESA Bulletin's editor, through this Paper Trail series, has given us the opportunity to reflect upon particularly influential papers as a way for us to understand the intellectual fabric of our discipline. This is especially germane on the eve of ESA's Centennial. I describe here a seminal paper that shaped my thinking which has led, in turn, to other developments in ecology.

A little autobiography sets the scene. I received a classical plant ecology Ph.D. degree under Murray F. Buell at Rutgers University in 1964. But this background became overlain by an interest in ecosystem science, stimulated by my dissertation at Brookhaven National Laboratory with George M. Woodwell. This interest in ecosystems became further focused toward biogeochemistry by interactions with Eville Gorham at my first academic position, to be concentrated still more at my second position through interactions with Gene E. Likens at Dartmouth College.

Around 1968, Gene and I undertook the organization of a graduate course in biogeochemistry. There were no models, much less textbooks, for such a course, so Gene and I cobbled together materials from our own backgrounds. Gene came to the task with a very strong limnological and aqueous chemistry background, together with fresh insights from his leadership role in the Hubbard Brook project. I contributed from my terrestrial background in plants, soils, and geology. (As an historical aside, there happened to be a bright, Dartmouth freshman, William H. Schlesinger, working in my lab at that time who eventually wrote the world's most comprehensive biogeochemistry book.) It was in our integrating, course-building process that Gene introduced me to the paper that was to become so important to the rest of my career: Alfred C. Redfield's “Biological control of chemical factors in the environment” (Redfield 1958).

This paper was a “bell-ringer” for me. I had already been involved in “mineral cycling,” but as practiced in terrestrial ecology at the time, mineral cycling was a rather dreary business of identifying and quantifying pools and pathways. It was narrow in scope and lacked understanding of feedback controls. In fact, mineral cycling is a limited view of the broader, multi-disciplinary field of biogeochemistry, in which the powers of chemical, earth, atmospheric and biological sciences are brought to bear on understanding the distribution and transport of elements and other chemicals at all scales throughout the planet. Redfield's paper brought all of this to life for me with its wholly different view of process, environment, and scale.

Revelle (1995) provides a splendid biography of Redfield (1890–1983). Redfield was a Harvard-educated, Harvard professor who largely worked at the Woods Hole Oceanographic Institution. Initially, Redfield addressed physiological questions, but he moved readily into many different roles, ranging from oceanic biogeochemistry to development of anti-fouling paints, to paleoecology, and finally to the study of tides. This patrician scholar was president of ESA in 1946, president of the American Society of Limnology and Oceanography in 1956, and became a member of the National Academy of Sciences in 1958.

It probably was the combination of Redfield's physiological perspective with access to a growing body of chemical information about marine biota and marine waters around the globe that led to his primary discovery (initial publication in Redfield 1934). He observed that the molar ratio of P, N, and organic-C in protoplasm of plankton (1:15:105) is, on average, the same across all of the oceans, and that the P:N ratio is generally the same as that of seawater itself. Carbon concentration is stoichiometrically higher in seawater because of its abundant inorganic forms. He also observed that the O2 content of epilimnetic seawater is approximately that amount necessary to oxidize planktonic biomass upon its death. He put these extraordinary relationships together in a diagrammatic model that integrated these relationships (Fig. 1), illustrating their relevance to the entire ocean, and in fact, for the entire planet. The upper part of the diagram poses that seawater concentrations of ionic N and P are incorporated through photosynthetically-driven biosynthesis of plankton biomass in the photic zone, and consequently mineralized by decomposition upon biomass consumption or death in a fast, local cycle. Biological cycling sets the concentration of P and N; there is, on average, no excess of either. Dissolved O2, CO2, and N2 are in exchange equilibrium with their concentrations in the atmosphere. Ostensibly, if P becomes less abundant, ionic N concentrations, largely NO3, decrease in parallel through denitrification. If P becomes more abundant, as through eutrophication, ionic N indirectly increases via biological N-fixation and subsequent mineralization. To the that extent biomass sinks or is mixed by downwelling into the hypolimnion, P, N, and organic-C are eventually mineralized in deep waters, and then returned by upwelling in other parts of the ocean. Exchange between surface and deep waters involves much longer turnover times of thousands of years, compared with cycling in surface waters.

Figure 1.

The central diagram and caption of Redfield's 1958 paper. Note that the equation for sulfate reduction is misleading. Oxygen is not a direct product of the process. Published with permission of American Scientist.

If O2 concentrations of the atmosphere were to decline, a higher rate of undecomposed biomass is exported to bottom waters leading to more O2-consuming respiration. Larger areas of anaerobic bottom waters lead to greater P solubility that, with upwelling, restores lost P from the photic zone, increasing photosynthesis, and restoring higher O2 concentrations in the photic zone and the atmosphere. Inorganic carbon is always available in excess because of the large concentrations of carbonate-bicarbonate ions. Carbonate and bicarbonate ions are, in turn, in equilibrium with carbonate sediments (not shown in Fig. 1). For a more complete discussion of the place of Redfield's conceptualization in modern understanding of ocean biogeochemistry, see Schlesinger and Bernhardt (2013).

Redfield's conceptualization has many implications and uses. It is a central process in the assessment of varying turnover rates of the oceanic conveyor system (Broecker 1991), the contrasting time scales of different biogeochemical processes, evaluation of land sources of nutrients, and incorporation of other biogenic elements into the system such as Si, Fe, and other transitional metals (Schlesinger and Bernhardt 2013). Perhaps most important, Redfield's conceptualization incorporates feedbacks that help understand homeostatic behavior of Earth's biogeophysical system. Redfield's paper put muscle on the bones of earlier concepts about biological influences of Earth's environment (Henderson 1913). Fifty-six years later, Redfield's conceptualization and his stoichiometric ratios continue to be important parts of understanding oceanic and global biogeochemistry, and are useful hypothetical tools for understanding terrestrial systems as well (Vitousek et al. 1988, Cleveland and Liptzin 2007). An April 2014 query of Redfield and ratio on Google Scholar indicated 2307 citations.

Returning to my own narrative, Redfield's global conceptualization vastly expanded my view of biogeochemistry. It became a framework for my research (Reiners 1992), but even more so for teaching. It was through teaching that I came to incorporate Redfield's ratios for protoplasmic biomass with sequestration of elements into structural mass such as wood, shells, and bones. Aided by more recently available data on biological tissues (Bowen 1979), I arranged this information in an evolutionary context and presented it as a deductive system of axioms (well-established facts) and theorems (deductions from the facts) for ecological stoichiometry (Reiners 1986).

This paper attracted immediate attention (215 citations April 2014), but more importantly, later became a stimulus for the rise of something even bigger. Ecologists associated with James Elser at Arizona State University and Robert Sterner at the University of Minnesota, taking advantage of newer understandings of genetics and biochemistry, applied stoichiometric ideas to food chains and tropic structures in aquatic systems. They discovered that protoplasmic Redfield ratios were not fixed but could vary in ways important to all levels of ecological focus, from cellular metabolism to ecosystem structure and function. Sterner and Elser (2002) summarized this exciting work in their remarkable book, Ecological Stoichiometry: The Biology Of Elements from Molecules to the Biosphere,. The work of Sterner, Elser, and their colleagues launched an exciting subdiscipline termed “ecological stoichiometry” (for a review see Martinez del Rio 2003). In April, 2014, the combination of ecological” and “stoichiometry” elicited 582,000 references in a Google Search.

What are some of the generalizations one can deduce from this paper trail? For me there are five. First, in ecology one cannot take too large a view of the problem one is addressing. Second, it is useful to step out of one's science into others to gain useful new ways of addressing questions. Collaboration with others outside ones field facilitates this complementarity. Third, teaching provides a useful forum for developing one's ideas. Fourth, there is no literature that is too old to have no value for current issues. And fifth, one must take time to read to be a thoughtful, creative scholar.

I gratefully acknowledge William H. Schlesinger's significant improvements to this manuscript.