Revisiting nitrogen fertilizer recommendations for Saskatchewan: Are we measuring the right soil nitrogen pool?
Date: August 2022
Term: 3 years
Status: Completed
Researcher(s): Richard Farrell, University of Saskatchewan; Fran Walley, University of Saskatchewan; Garry Hnatowich, ICDC; Vince Zhang, University of Saskatchewan
SaskCanola Investment: $65,525
Total Project Cost: $262,102
Funding Partners: WGRF, SaskWheat, AWC
Grower Benefits
Microwave-assisted citrate extraction (MACE) provides a viable and more rapid alternative to autoclaved citrate extraction (ACE) for soil protein determinations.
Soil protein pool extracted using the ACE/MACE methodology is not an appropriate measure of the N that potentially becomes biologically available during the season.
Trends suggested that the largest response in canola yield occurred with the first increment of applied N with further increases occurring at the 1x soil test recommendation rate.
Statistically significant yield increases often were generally not observed above the 1x rate, suggesting that current soil test recommendations provide an acceptable measure of biologically available N, with the understanding that the vagaries of weather strongly influence N responses.
Soil tests identified sites less likely to respond significantly to fertilizer N due to relatively high levels of initial soil inorganic N.
Project Summary
With nitrogen (N) being so important to crop production, lack of appropriate soil N tests have been criticized as being a limitation to developing sound fertilizer N recommendations. Soil testing is meant to measure how much N is in the soil that will supply the crop during the growing season.
However, what is typically measured is the soil inorganic N (a product of mineralization), and then predictions are used to estimate how much additional N will be released depending on the soil zone. Although measuring mineralization during a growing season could determine actual nitrogen release, it is a matter of too little too late so an estimate of potentially mineralizable nitrogen is required. Research has been done in the past but no test that has been developed has been widely adopted. There needs to be an affordable, easily reproducible, and chemically defensible test that measures the substrate for N mineralization (soil protein), and not the product (inorganic N). Soil proteins are the key substrate that contributes to the plant available N pool which contains both organic monomers, which are not measured by typical soil N tests, and inorganic N. The soil N tests currently used in Saskatchewan usually involve the use of a chemical extractant that determines the amount of plant available nitrate N in the soil or short-term incubation release.This provides only a snapshot of the amount of inorganic N (nitrate and/or ammonium) either in the inorganic N pool or entering this pool. Mineralizable N and proteins could be a better measure to use. Researchers have found a soil protein pool that is believed to represent potentially mineralizable N that can be rapidly tested in a lab. The standard method for testing soil protein is referred to as Autoclaved Citrate Extractable (ACE) protein, however this method is not regularly used for commercial soil tests as it is time consuming. Most labs are equipped with microwave systems that can provide the high temperature and pressure needed for soil protein extraction. These researchers wanted to determine whether microwave-assisted extraction of soil protein yields the same results as the standard ACE method, optimize the method, and demonstrate that this soil protein pool is directly related to potentially mineralizable N and can provide a basis for improved fertilizer recommendations.
Researchers started by assessing the ACE soil protein concentration in 55 different soils in Saskatchewan. Microwave-assisted cirate extractable (MACE) soil protein was determined on a sub-set of these samples and compared to ACE protein levels. Soil protein concentration was not affected by extraction method. Moreover, MACE soil protein extraction can be performed in half the time of the ACE method, without affecting protein recovery. The 55 soils collected showed varied levels of ACE protein N and total N across and within the major soil zones. There were strong correlations between soil protein N, total soil N and soil organic C, but only a weak correlation between soil protein N and inorganic N.
Fertilizer N response studies also were conducted for two years, at seven of the Agri-ARM sites across the province: Conservation Learning Centre (CLC) at Prince Albert, Northeast Agriculture Research Foundation (NARF) at Melfort, Irrigation Crop Diversification Corporation (ICDC) at Outlook, Indian Head Agricultural Research Foundation (IHARF) at Indian Head, Wheatland Conservation Area (WCA) at Swift Current, Western Applied Research Corporation (WARC) at Scott, and East Central Research Foundation (ECRF) at Yorkton. Crop management operations for wheat and canola were site specific and followed a standard protocol. There were five treatments replicated four times. Treatments consisted of a non-fertilized control (0x) and N applications of 0.5x, 1x, 1.5x and 2x the soil test recommendation. At harvest, above ground biomass and yields were determined. In the second year of the trial, soil samples were collected and sent for analysis. In this year, mid-season biomass samples were also collected, and following harvest, total biomass and yields were taken. Following harvest, soil samples were again collected from the 0x, 1x and 2x treatments and sent for analysis. They were tested for various indices of soil N availability, including ACE protein and potentially mineralizable N. In 2019, the largest response in canola yield occurred with 0.5x the soil test recommendation at the WARC site. There was no significant yield increase at N rates greater than the soil test recommendation. In 2020 yields varied more than in 2019. Nitrogen rates had no significant effect on canola yield at NARF, CLC or WCA sites. The strongest yield responses in both 2019 and 2020 were at WARC and IHARF.
This research has found that soil protein pool extracted using the ACE/MACE methods are not an appropriate measure of the N that potentially becomes available during the growing season. Researchers concluded that the ACE/MACE methods of soil protein are not suitable as a seasonal N availability index. Since statistically significant yield increases often were not observed above the 1x rate, this suggests that current soil test recommendations provide an acceptable measure of available N, with the understanding that weather strongly influences N responses. Researchers have also indicated that more research is needed to develop a rapid chemical test to determine biologicially available soil N.
Other Acknowledgements of this Research Project
Some collaboration with Kate Congreves - CARP ADF 2017.151
All field operations, and sampling in 2020, were carried out by staff at the Agri-ARM sites whose cooperation during the pandemic is greatfully acknowledged.