Adapt-N is a web-based decision support tool that provides field-specific sidedress nitrogen (N) recommendations for corn (url: http://adapt-n.eas.cornell.edu/). It was developed over the past 4 years in a collaborative effort among the Department of Crop and Soil Sciences, Department of Earth and Atmospheric Sciences, Northeast Regional Climate Center, and Center for Advanced Computing, with primary funding by the Computational Agriculture Initiative. The Adapt-N tool represents a new approach in N management by combining the use of a dynamic simulation model (Precision Nitrogen Management (PNM) model) with the most up-to-date high-resolution climate data. It was developed to improve N use efficiency in corn production. In the long term, we believe this tool can reduce excess N use and N losses to the environment while improving farm profitability.
Background
Corn (grain and silage) is the largest row crop grown in New York State (over 1 million acres in 2008; National Agricultural Statistics Service). Nitrogen management is a key component in corn production systems because of the relatively large N inputs that are used, the high cost of N fertilizer, and public concerns over reactive N in the environment. However, N is inefficiently used in corn production (Cassman et al., 2002) resulting in reduced farm profitability and potentially high environmental losses to both surface and ground water (nitrate leaching) and to the atmosphere (ammonia volatilization and denitrification). A number of studies quantified nitrate leaching potential under different crops (e.g., Logan et al., 1980; Robbins and Carter, 1980; Bergstrom, 1987; Owens, 1990; Randall et al., 1997; Mitsch et al., 2001; van Es et al., 2002). In general, they found the highest nitrate-N levels under corn, intermediate levels under less fertilized annual crops (e.g., soybeans and wheat), and lowest levels under perennial crops (e.g., alfalfa and grasses). Sogbedji et al. (2000) examined soil nitrogen dynamics using replicated plot-size lysimeters on two soil types (clay loam and loamy sand) where corn was grown on the plots under three N fertilizer levels. This study provided insights into N mass balance and the fate of fertilizer N in corn production under Northeast U.S. climate conditions. A key finding was that groundwater nitrate levels increased very little with increasing fertilizer levels until a threshold value was reached after which uptake efficiency dramatically decreases. Residual soil nitrate levels and therefore nitrate leaching are then increased. In this case, 43% of the fertilizer N above this threshold was accounted for in the groundwater (Sogbedji et al., 2000). Inefficient N management can also result in denitrification-driven losses of nitrous oxide (N2O), a potent greenhouse gas. For example, applying crop N requirements at planting can lead to high levels of denitrification, particularly under no till or minimum tillage, on poorly drained soils, and because of generally higher soil moisture levels in the early season (Tan et al., 2008).
The steep rise in the cost of N fertilizer together with increased scrutiny of off-farm N losses has focused growers on identifying more efficient, precise N management practices. Recent studies out of the CSS Department at Cornell (under the leadership of Prof. Harold van Es) and elsewhere have shown that variation in soil N associated with early season weather contributes to the well-documented variability in economic optimum corn N rates (Sogbedji et al., 2001c; Kahabka et al., 2004; Kay et al., 2006). If changes in soil N with early season weather can be estimated, N rates for corn could be adjusted to account for this variation, thus improving N management for corn production. One approach is the application of well-calibrated and tested dynamic simulation models of soil N dynamics and crop N uptake (van Alphen and Stoorvogel, 2001; Schaffer et al., 2001). Such models account for changes in soil N (sources, losses, and changes in soil N storage in the root zone) and crop N uptake. In theory, the output of these models can provide information for growers to adjust N applications to more precisely match crop N demand (Kersebaum, 1995; Smith et al., 1997; van Alphen and Stoorvogel, 2001). We have developed such a model (Precision Nitrogen Management (PNM) model; Melkonian et al., 2005) to improve N management for corn production.
The PNM model is composed of a corn growth and N uptake model (Sinclair and Muchow, 1995) linked to a soil process model (LEACHN; Hutson, 2003, Hutson and Wagenet, 1992). LEACHN simulates water and solute transport, and chemical and biological N transformations in the unsaturated soil zone (Hutson, 2003) and the corn model simulates N uptake, growth and yield of the corn crop. LEACHN has been extensively used and tested in several studies (Jabro et al., 1994; Jemison et al., 1994a, b; Lotse et al., 1992). We have calibrated and tested LEACHN for applications in the humid Northeast U.S. (Sogbedji et al., 2001a, b; Sogbedji et al., 2006). The corn model has been well tested and provides a reasonable fit to data that were collected over a range of conditions and were independent of those used in model development (Sinclair and Muchow, 1995). Parameters in the crop model have been adjusted based on corn growth and N uptake data from experiments that include a range of management practices, locations, and climate years in New York State (Cox et al., 1990a, b; Cox et al., 1993; Cox and Cherney, 2001).
The PNM model automatically links to ‘high resolution’ weather data, i.e., weather data available on a 3 mile x 3 mile grid across New York State and developed by the Northeast Regional Climate Center (Assoc. Prof. Art DeGaetano, Director) with assistance by the Cornell Center for Advanced Computing. Linkage to high resolution climate data allowed us to apply the PNM model to individual farms via Adapt-N, a web-based decision support tool that is built around the PNM model. Adapt-N requires relatively simple information related to crop, soils and N management for a given field. The inputs are processed by the PNM model and users receive a recommended sidedress N rate. Adapt-N can be used in both manured and non-manured corn productions systems, and at different years within a corn rotation. We believe that Adapt-N has the potential to increase farm profitability and reduce environmental impacts.