By Dave Mengel, Ray Asebedo
Soil testing for nitrate-N in the fall for making nitrogen (N) recommendations on winter wheat is a valuable practice, particularly when using 24-inch profile sampling. Unfortunately, few farmers utilize this tool, and its value has been questioned in some areas due to the potential for overwinter N loss.
However, with the exception of sands, N losses over winter in Kansas are normally quite low due to our low rainfall in December, January, and February.
To evaluate the relationship between wheat yield and fall soil nitrate-N — and to determine if it is still a viable practice to utilize in N management of wheat — we summarized data from 26 different N management experiments conducted across Kansas from 2007 through 2014. Most were from 2010 through 2013.
The driving force behind this study is the growing interest in improving N management in winter wheat production. Recent efforts have been focused on improving nitrogen use efficiency (NUE), or the portion of the fertilizer N we apply that is used by the plant. This has resulted in the creation of N fertilizer products designed to reduce N loss, optical sensors that can evaluate wheat’s N status, and changes in methods and timing of N applications. With so many new practices incorporated into N management systems, older practices are starting to be considered dated and discarded.
Taking fall soil profile-N samples has been a recommended practice for making an N recommendation for winter wheat for many years. However, due to the mobility of nitrate-N in the soil, soil test values observed in the fall may be completely different than values observed in the spring, particularly on soils prone to leaching. Because many producers wait until spring greenup to make their N application, does soil sampling in the fall for nitrate-N really provide useful information for N management in wheat? That’s a legitimate question.
The objective of our study was to evaluate the relationship between N fertilizer response by wheat and fall soil nitrate-N and determine if it is still a viable practice to utilize in N management of wheat.
Data were drawn from 26 dryland wheat experiments conducted in 2007 through 2014 throughout Kansas in cooperation with producers and Kansas State University experiment stations. Locations included Manhattan, Tribune, Partridge, Johnson, Randolph, Rossville, Ottawa, Sterling, Pittsburg, Silver Lake, Solomon and Gypsum.
Soil samples to a depth of 24 inches were taken prior to planting and fertilization. Samples from 0 to 6 inches were analyzed for soil organic matter, phosphorus, potassium, pH, and zinc. Soil profile 0- to 24-inch samples were analyzed for nitrate-N, chloride, and sulfate. Fertilizer needs other than N were applied in the fall at or near seeding.
1) Analysis of yields taken from plots that received no N fertilizer shows a strong positive relationship with fall soil profile nitrate-N (Figure 1). Wheat yields increased rapidly as soil N levels increased to about 80 pounds soil N per acre, and then leveled off.
Figure 1. Relationship between fall soil profile nitrate-N level and wheat yield with no N fertilizer applied
2) We then converted check plot yields to a relative yield, or percentage of the maximum fertilized yield obtained at each location (Figure 2). The results reveal not only the yield of the check plot, but also the N responsiveness of the site. This shows that at low soil nitrate levels, sites respond well to applied fertilizer. When fall soil profile nitrate-N levels are greater than 80 to 100 pounds per acre, relative yield is approaching 100%, and it is unlikely the site will respond to additional fertilizer N applied in the spring.
Figure 2. Relationship between fall soil profile nitrate-N and Relative Yield, or percent check plot yield of the maximum obtained with fertilizer at each site.
3) A third way to show this relationship between fall soil nitrate and N response is to calculate the Delta Yield, or the increase in yield obtained from the addition of fertilizer at each site. This is a good measure of N responsiveness of an individual research site. The relationship between fall profile N level and Delta Yield is shown in Figure 3. It is clear from this graph that at low soil nitrate levels in the profile, sites respond well to applied nitrogen fertilizer. However, as the profile N level increases beyond 75 to 80 pounds N per acre, little or no N fertilizer response was found.
Figure 3. Increase in yield due to N fertilization, Delta Yield, as a function of soil N level
4) A commonly used way to measure the efficiency of N use is to determine the amount of N fertilizer required to produce one additional bushel of yield. This relationship is shown in Figure 4.
Figure 4. Pounds of N fertilizer required per bushel of yield increase at different levels of N responsiveness, or Delta Yield.
On highly N-responsive sites, those with a large Delta Yield, the amount of N required to increase yield by one bushel is relatively low, near the 2.4 pounds N per bushel used in the K-State fertilizer recommendations. However, as the yield response decreases, the amount of N required to obtain that response increases dramatically. This relationship provides a good explanation of why fertilizer recommendations are generally made not to obtain the maximum yield, but rather the economic optimum yield.
The efficiency of squeezing out those last one or two bushels is just too low. The cost of the added fertilizer will exceed the value of the extra grain produced. A number of additional conditions such as drought, disease, and poor root growth can influence this relationship. Many of the new technologies being developed to enhance N management and NUE, should help reduce the pounds of N fertilizer required to obtain a bushel of N response.
Wheat yield with no N fertilizer applied was compared with fall nitrate-N levels and a strong relationship was established. Although new practices have been developed to improve N management in winter wheat, soil sampling in the fall for nitrate-N remains an important practice to manage N efficiently and can result in considerable savings for producers.
When soil sampling for N is not done, the K-State fertilizer recommendation formula defaults to a standard value of 30 pounds per acre available N. In this particular dataset, the average profile N level was 39 pounds N per acre. However the N level at individual sites ranged from 11 to 197 pounds N per acre. Most recommendation systems default to a standardized set of N recommendations based on yield goal and/or the cost of N. Without sampling for N or using some alternative method of measuring the soil’s ability to supply N to a crop, such as crop sensing, the recommendations made for N will be inaccurate, resulting in a reduction in yield or profit per acre and increased environmental impact.
Due to the drought of the past three years, there have been many situations where large amounts of N have been present in the soil at planting of wheat or summer crops such as corn or grain sorghum. Early samples requesting soil N tests from western Kansas coming to the lab are already showing high soil N levels from some areas. Failure to account for that valuable resource can result in excess foliage, increased plant disease, inefficient use of soil water, and reduced yield.
Soil sampling in fall for nitrate-N can have a significant impact on N recommendations for winter wheat, thus improving N management, and is still strongly recommended.