Pine forests in the southeastern United States are more productive than ever, and fertilizers can take some of the credit. But not all fertilizer goes toward plant growth. Some of it runs off into rivers and streams, where it can degrade water quality. A number of water quality models are available to predict fertilizer runoff. A new study by U.S. Forest Service scientists, researchers at other Federal agencies, university partners, and private industries and research institutions reviews five of the most commonly used models.
“We synthesized existing knowledge on the applications, limitations, uncertainties, and other characteristics of the models,” says Devendra Amatya, research hydrologist at the Forest Service Southern Research Station Center for Forested Wetlands Research and lead author of the study. The findings were recently published in the Transactions of the American Society of Agricultural and Biological Engineers.
Nitrogen is critical for plant growth; however, although nitrogen gas makes up almost 80 percent of our atmosphere, plants are unable to use nitrogen gas for growth. Bacteria were once the only living things that could convert the gas into forms that plants could use, but in the early 20th century, scientists developed an industrial chemical process to convert nitrogen gas into forms that could be added to fertilizer. As of 2000, about a billion tons of reactive nitrogen were released each year from nitrogen fertilizer spread on farmlands around the world. The amount of fertilizers used in forests is dwarfed by the amounts used in agriculture and urban areas, but land managers and scientists want to know where the nitrogen in forest fertilizers goes and how runoff containing it could affect rivers and streams.
Nitrogen takes a winding path through the environment, cycling through plants, animals, bacteria, soils, and water. At each stage, new complications arise. “The nitrogen cycle in forests, soils, air, and water bodies is very complex,” says Amatya. Complicating factors include the area’s hydrology and water management, topography, vegetation type, soils, management practices, disturbance history, and others. Each of the five models the researchers studied – APEX, MIKESHE-DNDC, DRAINMOD-FOREST, REMM, and SWAT – approach these factors in different ways and vary widely in their design, scope and potential applications.
“The strengths and limitations of each model reflect the original purposes they were designed for,” according to the article. “REMM, for example, was developed to model small areas – the riparian buffer zone of fields. While this is important, models looking at water quality generally need to work at scales of at least 20 square miles.”
DRAINMOD-FOREST was also designed to be used at field scales but only for poorly drained lowlands, and APEX was developed for small upland farms or small watersheds. The other two models, MIKESHE-DNDC and SWAT apply to larger scales, but SWAT is primarily for uplands and cannot simulate tree growth. Of the five models, MIKESHE-DNDC is the only one capable of describing both nitrogen transformation and hydrology for both uplands and lowlands, but it does not model nitrogen movement and transport through the soil.
“Independently, none of these models are optimal,” says Amatya. “No single model is capable of describing the fate of nitrogen fertilizer applied to forest stands at both small and large scales, uplands and lowlands.” The models are continually being updated and improved, but because of their different designs and intents, it can be difficult to make direct comparisons between them. Amatya and his colleagues recommend creating a common database so that all the models can run simulations with the same information. The database should represent approximately 20 square miles, and have long-term data on hydrology, water quality, and forest growth. “The Forest Service, the universities, other Federal agencies, and private industries have this information,” says Amatya. “We should organize it and make it available so that these models can be tested and compared.”
For more information, email Devendra Amatya at email@example.com