Stanford Research Communication Program
  Home   Researchers Professionals  About
Archive by Major Area


Social Science

Natural Science

Archive by Year

Fall 1999 - Spring 2000

Fall 2000 - Summer 2001

Fall 2001 - Spring 2002

Fall 2002 - Summer 2003




Agriculture and Pollution in the Developing World:
Understanding the Link Between Fertilizer Use, Greenhouse Gases, and Coastal Change in Sonora, Mexico

John Harrison
Geological and Environmental Sciences Department
Stanford University
June 2001

Half the nitrogen fertilizer ever produced on Earth was used during the last fifteen years. This increase in the use of nitrogen fertilizer has led to massive increases in agricultural yield (the amount of food grown per unit area) and has, in fact, allowed humans to largely avoid the food shortages historically predicted to accompany our recent population boom. In this sense, nitrogen fertilizer has been an enormous boon to humans. However, the recent increase in nitrogen use may have serious potential drawbacks as well, such as coastal pollution and the increased production of greenhouse gases, leading to global climate change. My research takes as an example the drainage system of the Yaqui Valley in Sonora, Mexico and investigates the link between agricultural fertilizer use (the largest human source of reactive nitrogen) and its environmental impact.

One of the biggest changes that humans have wrought upon the Earth over the last 50 years is our profound alteration of the global nitrogen cycle. Reactive nitrogen is essential to life, constituting an important building block for DNA and proteins. However, reactive nitrogen has historically been scarce, frequently constituting a limiting resource to living beings and systems. This scarcity is a bit odd considering that dinitrogen gas (N2) comprises 78 percent of the atmosphere. However, it is explained by the fact that N2 is chemically inert and must be altered (or fixed) in order to be useful to living beings.

Through activities such as the production of fertilizer (the largest single contributor), the burning of fossil fuel, and the planting of nitrogen-fixing crops, humans have more than doubled the rate at which this reactive nitrogen enters the living world. When farmers apply fertilizer to fields, half (or more) of that fertilizer generally does not stay on the field to nourish crops, but rather is carried away in water and air to adjacent ecosystems where it can fundamentally change the way those ecosystems function.

I am particularly interested in the water-borne portion of escaping nitrogen fertilizer, which can flow downstream to coastal ecosystems and lead to the suffocation of fish and invertebrates, the loss of biodiversity, the creation of toxic and overabundant algae blooms, and sedimentation. This nitrogen can also be transformed while en route to the coast into inert N2 via a microbial process called denitrification. In denitrification, microbes essentially eat reactive nitrogen and digest it into more chemically reduced forms using the energy that is produced in this process to live, grow, and reproduce. The principal end-product of denitrification is N2, and since, as stated above, N2 is a harmless inert gas, when fertilizer is transformed into it, it is effectively "cleaned" out of a system.

Some denitrification, however, results in the emission of another gas, nitrous oxide (also known as laughing gas). Nitrous oxide traps light energy and re-radiates it as heat. This, together with its rapidly increasing atmospheric concentrations, makes nitrous oxide an important greenhouse gas. We are not sure where excess nitrous oxide comes from, but we believe that agricultural runoff is a major source, and that nitrous oxide production in this runoff is regulated, at least in part, by the amount of nitrogen fertilizer applied to upstream fields. Very few scientists, however, have studied nitrous oxide emissions from agricultural drainage waters, and none have looked at this issue outside of the temperate zone, where the greatest increases in fertilizer use are currently occurring. This lack of study has meant that our estimates of nitrous oxide production have been rather uncertain, and that our understanding of the mechanisms controlling nitrous oxide production has been rather rudimentary.

In my research, I address the following three questions: 1) What factors control the amount of runoff fertilizer converted into greenhouse gases, including nitrous oxide? 2) What controls the amount of nitrogen fertilizer runoff that makes it to the coast? and 3) How important is agricultural runoff likely to be as a global source of nitrous oxide? To address these questions, I am monitoring nitrogen dynamics and nitrous oxide production in a relatively simple model system; the drainage system of the Yaqui Valley in Sonora, Mexico. This drainage system consists of a network of several drainage canals carrying agricultural runoff untreated from 225,000 hectares (about 2,250 square kilometers) of farm fields to the sea. In these canals, the factors that are likely to affect nitrous oxide production and denitrification include nitrogen availability, organic matter availability, oxygen concentration, and temperature. These factors vary with space and time, and I have been using this variability to better understand how each factor correlates with in-stream nitrogen transformations (e.g. the amount of nitrous oxide produced and the amount of reactive nitrogen converted to inert N2). To test the relationships I have observed via correlation analyses, I have done a number of experiments wherein I altered nitrogen availability, organic matter availability, oxygen concentration, and temperature under controlled laboratory conditions in order to examine how nitrous oxide and denitrification respond.

I have found that rates of nitrous oxide production are extremely high in this system compared to other previously studied systems (20 percent higher than the highest previously published values). However, due to the small surface area of agricultural drainages in the Yaqui Valley, it appears drainage waters contribute less nitrous oxide to the atmosphere than other scientists working in other systems have estimated. I am also finding that this type of system is likely to function quite differently from other previously studied systems. Whereas it is commonly assumed that the rate of nitrous oxide production is controlled by the dissolved concentration of reactive nitrogen, I am finding that the availability of another chemical group, organic carbon, is more likely to control nitrous oxide production in the Yaqui Valley.

My next steps are to further improve my estimates of nitrous oxide production and to begin to develop a computer model that quantifies my current hypotheses and uses them to make some predictions about system response to different environmental stimuli. It is my hope that such a model will constitute a first step toward predicting and controlling nitrous oxide emissions and the transfer of agricultural fertilizer pollution from fields to the sea.