Same Farms, Different Answers: When Nitrogen Pollution Estimates Don’t Add Up

To reduce nitrogen pollution stemming from synthetic fertilizer and livestock waste, California's Central Valley relies on five datasets to estimate surplus nitrogen. However, most parts of the United States use just one or two datasets — the U.S. Department of Agriculture Agricultural survey and census — to estimate nitrogen surplus. 

Using the Central Valley as a “living laboratory,” a University of California, Davis, research team coordinated by postdoctoral researcher Baibaswata Bhaduri compared these five different datasets and found that each produced significantly different results. The range of uncertainty was so wide, it could reshape which crops and regions receive priority for groundwater protections. 

Their study, published this week in the journal Nature Sustainability, offers ranges for surplus nitrogen in everything from tomatoes and nectarines to corn and grapes. Growers worldwide can look to these estimates for guidance in the absence of local data. 

“Our paper confirms the approach that California nitrogen policy is pursuing,” said corresponding author Thomas Harter, a Cooperative Extension specialist and professor in the UC Davis Land, Air and Water Resources department. His research in 2012 called attention to the problem of nitrogen groundwater pollution and helped set the stage for California’s nitrogen management strategies. 

“We’re saying, be cautious,” said Harter. “Inevitably, between the various existing data sets, there can be large differences with respect to surplus that could rearrange the shape of your map and where you want to prioritize your conservation efforts. We must take into account the uncertainty in current data about fertilizer use and harvest. Improving data accuracy by engaging with farmers is critical.”

In addition to USDA’s survey and census data, the study compared the following datasets from the Central Valley: grower-reported records on the total nitrogen applied and crop yields, now annually reported by each farm under California’s recent nitrogen tracking policy; UC Davis Cost and Return Studies, which provide expert estimates of crop-specific fertilizer rates and yields; and the California Department of Food and Agriculture’s annual crop-specific acreage and harvest records, which have been collected and reported each year by county agencies since the early 1900s.

The same farm or crop could appear to need less mitigation under one dataset and more under another, the report said.

“The Central Valley is not a rare place with a rare problem,” the authors wrote in the study. “It is one of the few places with enough data to see a problem that single‑dataset assessments carry everywhere, unseen.” 

“Disagreement between datasets is not just a limitation; it reveals where our knowledge is weakest,” said Bhaduri. “Understanding those differences is essential to improving nitrogen assessments and ensuring that groundwater protection decisions are scientifically defensible and fair to the farmers and communities affected.”

Additional coauthors include Kenneth Miller of Formation Environmental LLC, and Giorgos Kourakos of UC Davis.

The study was funded by the California State Water Resources Control Board and the USDA National Institute of Food and Agriculture.