Water Quality Impacts Reduced with Adapt-N Recommendations

Aaron Ristow1, Shai Sela1, Mike Davis2, Lindsay Fennell1, Harold van Es1
Soil and Crop Sciences Section, School of Integrative Plant Science, and 2Cornell University Agricultural Experiment Station

Cornell University

Soil nitrogen (N) is both spatially and temporally variable, challenging farmers to meet optimal nitrogen (N) needs and minimize N deficiency risk. N typically is a large monetary input for corn production in part due to farmer tendency to over-apply N fertilizer and/or manure to maximize their returns to N applications in the presence of high uncertainty around the optimum N rate. This excessive N maybe be readily lost to the environment through volatilization, runoff and leaching. Not only do N losses negatively impact yield, we know a significant percentage of total N load is carried by ground water or discharged to streams, causing environmental costs. Therefore, a top priority should be the estimation of the optimum N rate that meets crop production needs while minimizing environmental impacts.

The optimum N rate depends on numerous factors including the timing and amounts of early season precipitation events, previous organic and inorganic N applications, soil organic matter, carry-over N from previous cropping seasons, soil texture, rotations, etc. There are several approaches to optimizing N rates and minimize N losses. These can be generally categorized as (i) static and (ii) adaptive. Static tools offer generalized recommendations that do not consider seasonal conditions of weather and soil/crop management, while adaptive approaches account for the variable and site-specific nature of soil N dynamics, including the effects of weather. Using data from two seasons of corn silage grown at the Cornell University research farm at Willsboro, NY, we compared the economic and environmental impacts of N rate recommendations from a conventional static approach (the Cornell Corn Nitrogen Calculator; CNC) with the adaptive Adapt-N approach (adapt-n.com).

Adapt-N and the Cornell Corn Nitrogen Calculator

The Cornell University Corn Nitrogen Calculator (CNC) is a static approach that includes a basic mass balance calculation of N demand (yield-driven crop uptake) and N supply (soil organic matter, manure, previous crops), combined with efficiency factors. The CNC approach has been the established corn N recommendation approach for several decades, and estimates can be derived from a spreadsheet downloaded from http://nmsp.cals.cornell.edu/software/calculators.html.

Adapt-N is a dynamic simulation tool that combines soil, crop and management information with weather data to estimate optimum N application rates for corn. Originally developed at Cornell University, the tool has been licensed for commercial use and is currently calibrated for use on about 95% of the US corn production area. When using the tool to inform in-season N application rates, early season weather effects and site-specific attainable yield can be incorporated into the recommendation, allowing N management precision to be improved.

The Adapt-N tool was compared to CNC recommendations in a spatially-balanced complete block design (4 replications) on two paired experimental sites for the 2014 and 2015 growing seasons. In each trial, the treatments were defined by the total amount of N applied, where the rates were:

(i)     the total N rate based on Adapt-N recommendations (including a 15 lbs/ac starter) for the date of sidedress, and

(ii)    the total recommended rate of the Cornell Corn Nitrogen Calculator (including a 15 lbs/ac starter), using realistic yield goals (rather than the database yield goals, which would have underestimated real yields for these sites).

The treatments were implemented on 16 plots, each on a Cosad loamy fine sand and a Muskellunge clay loam, in continuous corn (silage), under no-till and plow-till management. Drainage water samples were collected from the lysimeters at key time points in the spring (April 7th and April 23rd) and fall (October 1st, October 29th, and December 3rd). The lysimeters include drainage lines routed to a utility hole to allow for drain water samples to be collected. Nitrate (NO3) and Nitrite (NO2) concentration was quantified from the samples to allow us to assess differences in water quality in Adapt-N vs CNC plots. In this article, we will refer to NO3+NO2 concentrations simply as NO3 or “nitrate”, as the NO2 fraction is typically very small.

At the end of the 2014 and 2015 seasons, we measured corn yields and calculated associated partial profit differences for the two treatments. Corn yields were assessed by representative sampling (four 15 ft long row sections per plot). Partial profit differences between the Adapt-N and CNC practices were estimated using prices of $0.50/lb N and $50/T silage.


Yield and Profit: The measured agronomic and leaching losses of the two recommendation approaches are presented in Table 1. Adapt-N recommended N rates were substantially lower than the CNC rates with an average reduction of 55 lbs/ac (183 vs 126 lbs/ac), while the average yields did not differ significantly (13.0 vs 13.1 T/ac; p=0.74). Reducing N rates without compromising yields resulted in $34/ac higher partial profit from the Adapt-N treatment. The economic and agronomic benefits of Adapt-N are similar to those from a larger study conducted in IA and NY using data from 113 on-farm trials (Sela et al., 2016).

Screen Shot 2016-04-05 at 11.26.41 AM

Lysimeter measured nitrate concentrations: In addition to the economic benefits, substantial environmental advantages were found with Adapt-N. When both seasons and soil textures were combined, the average NO3 concentration from the grab samples collected from the lysimeters indicated significantly lower water quality impacts under Adapt-N management vs CNC (11.0 and 15.3 mg/L, respectively; p<0.01). On average there was a 28% reduction in NO3 concentration from the Adapt-N treatments. When analyzing the clay loam and loamy sand plots separately but still combining the two seasons, NO3 concentration was significantly higher in the CNC loamy sand treatments (20.1 vs 13.7 for Adapt-N; p<0.01) and they trended toward higher concentrations in the clay loam treatments (10.0 vs 8.0 for Adapt-N; p=0.09).

Figure 1. Total Applied N recommended from two tools (Adapt-N and CNC) compared with measured NO3 leaching concentrations over two seasons from two soil textures. In general the Adapt-N recommended lower N applications resulted in lower average NO3 concentrations, and the loamy sand showed greater leaching losses with increasing N rates than the clay loam.
Figure 1. Total Applied N recommended from two tools (Adapt-N and CNC) compared with measured NO3 leaching concentrations over two seasons from two soil textures. In general the Adapt-N recommended lower N applications resulted in lower average NO3 concentrations, and the loamy sand showed greater leaching losses with increasing N rates than the clay loam.

Figure 1 shows nitrate concentrations for each drain water sample. Generally, there was a large range of losses throughout the year, but they trended up with more applied N. As could be expected, we saw that the loamy sand plots had higher losses, regardless of treatment, due to the lower water holding capacity of the coarse textured soil. Similarly, NO3 concentrations from the clay loam plots were less responsive to the amount of applied N compared to the sandy plots, but there were still substantial losses, especially at the higher rates. We conclude that the lower applied N rate in the Adapt-N treatments resulted in an overall lower concentration of NO3 in leachate from the lysimeters.


This study proves both economic and environmental gains from using Adapt-N’s adaptive approach to estimating in-season N rates across two distinct soil types in Northern New York. In all, the Adapt-N recommended rates were lower than the CNC rates but maintained the same yield and showed greater profits. Overall, the use of Adapt-N can significantly contribute to nitrogen reduction goals by reducing overall inputs, minimizing environmental losses, and improving farmer profits.


This work was supported by funding from the USDA-NRCS, New York Farm Viability Institute, USDA-NIFA, and USDA-Sustainable Agriculture Research and Extension, and the Northern New York Agricultural Development Program.


L. Fennell, S. Sela, A. Ristow, H. van Es, S. Gomes. 2015. Comparing Static and Adaptive N Rate Tools for Corn Production. What’s Cropping Up? 25:5

L. Fennell, S. Sela, A. Ristow, B. Moebius-Clune, D. Moebius-Clune, B. Schindelbeck, H. van Es, S. Gomes. 2015. Adapt-N Recommendations Reduce Environmental Losses. What’s Cropping Up? 25:5

Sogbedji, J.M., H.M. van Es, J.J. Melkonian, and R.R. Schindelbeck. 2006. Evaluation of the PNM Model for Simulating Drain Flow Nitrate-N Concentration Under Manure-Fertilized Maize. Plant Soil 282(1-2): 343–360

Sela. S, H.M. van Es, B.N. Moebius-Clune, R. Marjerison, J.J. Melkonian, D. Moebius-Clune, R. Schindelbeck, and S. Gomes. 2015. Adapt-N Outperforms Grower-Selected Nitrogen Rates in Northeast and Midwest USA Strip Trials. Agronomy Journal (accepted for publ.)


Print Friendly, PDF & Email