Toward solving the nitrogen mineralization puzzle
Approaches to quantifying nitrogen (N) mineralization from soil organic matter under field conditions have been put forward with varying degrees of success. Estimates have been made from crop N uptake and yield data. Gilmour (2021) detailed a computer model that predicts N mineralization from soil organic matter based on weather (temperature, rainfall and pan evaporation) for a specific location with minimal soil data (total N, bulk density, nitrogen use efficiency, and soil depth). This article provides background information and describes that computer model, NminSOM2.1, in detail.
Approaches to quantifying nitrogen (N) mineralization from soil organic matter under field conditions have been put forward with varying degrees of success. Estimates have been made from crop N uptake and yield data. A successful and easy-to-use example is the N mineralization mapping tool for the Midwest U.S. published by Iowa State University (https://crops.extension.iastate.edu/facts/soil-n-mineralization) based on years of field research. Attempts at computer modeling have been modestly successful. Benbi and Richter (2002) found that models were limited as inputs were not typical soil properties. Morris et al. (2017) concluded that model inputs were too detailed and that real-time weather was not possible.
Laboratory analyses like potentially mineralizable N (PMN) have been proposed to mirror N mineralization in soil health studies, and there are correlations among the proposed laboratory methods (Cappellazzi & Morgan, 2021). However, no combination of laboratory methods predicts actual N mineralization, and studies have found little correlation between these laboratory tests and crop yield or commercial N fertilizer recommendations (Clark et al., 2019; Sullivan et al., 2020). Moebius-Clune et al. (2016) wrote “Although in-season soil N testing is available …, employing models that account for the impact of weather on fertilizer needs …, along with soil testing, is likely the future of nitrogen management.” Sullivan et al. (2020) recommended that laboratory tests be verified in the field where no N had been applied using soil buried in bags that could be recovered periodically or by measuring N uptake by crops. They also wrote “In the future, soil N min tests may find semiquantitative applications within computer simulation models.”
Gilmour (2021) detailed a computer model that predicts N mineralization from soil organic matter based on weather (temperature, rainfall, and pan evaporation) for a specific location with minimal soil data (total N, bulk density, nitrogen use efficiency, and soil depth). This article provides background information and describes that computer model, NminSOM2.1, in detail.
Selecting a Soil Depth to Sample
A study conducted at Purdue University where tillage treatments had been maintained for 28 years and total N and bulk density (BD) determined at various depths illustrates the importance of soil depth (Gál et al., 2007). Total N decreased with depth in no-till plots over all depths while total N in moldboard plow plots decreased after 12 inches as shown in Figure 1.
Bulk density for moldboard plow plots had much lower BD than no-till plots in the first 12 inches (Figure 2). This BD difference remained to 39 inches. Bulk density in no-till increased with depth in five of six depths.Total N concentration times BD (plus conversion factors) determines the amount of total N in pounds of N per acre at each depth from the soil surface shown in Figure 3.
For the 0-to-6-inch depth, no-till and moldboard plow soils contained 29 and 22% of the total N to 39 inches, respectively. For the 0-to-12-inch depth, no-till and moldboard plow soils contained 55 and 46% of the total N to 39 inches, respectively. It is important to note that total N (g N/kg) in Figure 1 declined with depth while total N (lb N/acre) in Figure 3 increased with depth. It is the latter measure of total N that determines amount of N mineralization.
Researchers have used the 0-to-12-inch depth increment for estimating N mineralization. That depth increment will be used in NminSOM2.1 predictions that follow.
NminSOM2.1 Climate Data
The climate portion of NminSOM2.1 was first used in the prediction of organic soil amendment decomposition by Gilmour (1998). NminSOM2.1 uses monthly data to decrease temperature, rainfall, and pan evaporation variability that occurs in daily data. Calculations on a daily time step using monthly data allow seasonal estimation of N mineralization.
Monthly temperature (Tair,˚F) and rainfall (inches) can be found at www.usclimatedata.com and other websites. Monthly pan evaporation (Ep, inches) is not readily available on the internet for most locations. To simplify Ep estimation, the American Rose Trials for Sustainability climate zones (Figure 4, M. Schwartz, personal communication) were used to group locations in the U.S. where Ep vs. Tair relationships might be similar. Linear regressions of observed Ep vs. Tair for each location within a climate zone provided a database to develop new linear models for an entire climate zone. In this way, linear Ep models included the impact of other atmospheric factors such as relative humidity, wind speed, and solar radiation. Equations for each of the eight climate zones were statistically significant (Table 1). Equation 1 presents the relationship where m was the slope and b was the intercept. Units are given in Table 1.
Table 1. Number of data points (n), Equation 1 slope (m), and intercept (b) and statistical results for climate zones in the lower 48 states of the U.S.
| Climate zone1 | n | m | b | R2 | RMSE |
|---|---|---|---|---|---|
| inch/˚F | inch | inch | |||
| Marine West Coast | 60 | 0.185 | −6.5 | .88 | 0.7 |
| Mediterranean | 84 | 0.270 | −10.6 | .80 | 1.5 |
| Semiarid Steppe | 281 | 0.253 | −8.1 | .83 | 1.5 |
| Midlatitude Desert | 125 | 0.282 | −10.1 | .85 | 1.8 |
| Humid Continental (cool summer) | 111 | 0.179 | −6.0 | .80 | 1.1 |
| Humid Continental (warm summer) | 342 | 0.153 | −4.8 | .84 | 0.9 |
| Humid Subtropical | 564 | 0.162 | −5.3 | .82 | 1.0 |
| Tropical Wet/Dry Season | 12 | 0.258 | −12.6 | .99 | <0.1 |
NminSOM2.1 Water Balance
A monthly water balance is constructed for 36 inches of soil. Each month, soil water is initially set to 55% water-holding capacity (WHC). Rainfall and irrigation depths are added, and evapotranspiration (ET) depth equal to 0.8 × Ep is subtracted from initial soil water. The 0.8 factor is the typical ratio of ET to Ep and can be changed to fit specific situations.
If rainfall and/or irrigation cause initial WHC to exceed 55%, runoff/or drainage occurs until 55% WHC is reached. Soil is not allowed to get drier than two-thirds of the permanent wilting point. The relationship between percent WHC and inches of water in the 36 inches of soil is determined for one of three BD ranges: >1.49 g/cm3, 1.49 to 1.21 g/cm3, and < 1.21 g/cm3 corresponding to the location. This relationship is used to determine percent WHC each month.
NminSOM2.1 Nitrogen Mineralization Kinetics
NminSOM2.1 is a first-order kinetic model where the value of the first-order rate constant (0.00021/d) is the same for soils where total N is less than 3.0 g N/kg soil (0.3%). Above 3.0 g N/kg soil, the rate constant declines as the total N (g N/kg) increases because that organic matter is more stable (Gilmour, 2021). The first-order rate constant is corrected for temperature and percent WHC effects using an equation from Gilmour (2021). The first-order rate constant is zero below 33 ˚F, and no N mineralization occurs.
NminSOM2.1 Input and Output (West Lafayette, IN)
In this example, BD was reported (Gál et al., 2007). If BD is not known, it is estimated from BD vs. soil organic C data of Ruehlmann and Körschens (2009) where soil organic C is converted to soil organic N assuming SOM has a C:N ratio of 12. That relationship is shown in Figure 5.
Monthly maximum and minimum Tair are entered in Figure 6. NminSOM2.1 calculates the means. Monthly rainfall and irrigation are also entered in Figure 6. NminSOM2.1 calculates ET using the appropriate climate zone equation in Table 1 and the ET/Ep factor for that month.
The first part of Figure 7 shows the remaining input data and the model output for moldboard plow plots in the Gál et al. (2007) study. Inputs are sample ID, total N, soil depth, and start month where Tair is a minimum of 49 ˚F. Climate zone, location, and BD are from Figure 6. Cumulative and monthly N mineralization predictions are shown in the output portion of Figure 7. Annual cumulative N mineralization was 3.2% of total N, which is within the range (2 to 4%) reported by Sawyer et al. (2006) for the Midwest U.S. Corn Belt.
Traditionally, nitrogen use efficiency (NUE) is the N in the fertilized crop minus the N in the unfertilized crop divided by fertilizer N applied. Here, NUE of mineralized N equals N in the unfertilized crop divided by seasonal N mineralization from NminSOM2.1. Monthly values multiplied by NUE are estimates of the commercial N fertilizer equivalency of mineralized N.
The seasonal value of cumulative N mineralization in the second part of Figure 7 is obtained by entering the start date and end date of the N uptake portion of the cropping season. The value for the 104-day growing season in West Lafayette, IN (101 lb N/ac) in Figure 7 was 13 lb N/ac larger than the estimate (about 88 lb N/ac) using maps from Iowa State University (https://crops.extension.iastate.edu/facts/soil-n-mineralization). This difference might be attributed to mineralized N absorbed by the corn root system. Ordóñez et al. (2020) reported 24 lb N/ac for corn root N uptake in the Midwest U.S. for a yield of 189 bu/ac for fertilized corn. About half that amount would be expected for corn receiving no commercial fertilizer.
Any of the other input variables can be changed to see their impact on N mineralization. An example is irrigation. When 4.5 inches of irrigation are applied in July and August at the West Lafayette site, yearly cumulative N mineralization increases to 230 lb N/ac, and seasonal cumulative N mineralization increases from 101 lb N/ac to 117 lb N/ac (data not shown).
Another example is entering real-time temperature and rainfall data, so the output represents real-time conditions.
NminSOM2.1 has been evaluated for corn production in Iowa (Gilmour, 2021). Fifty-eight percent of the variability in yield for unfertilized corn could be accounted for by the model. While this is a promising result, future evaluation of NminSOM2.1 for a wide variety of crop production systems in a variety of locations is warranted. Toward that end, it is the author’s intention to make NminSOM2.1 available to CCAs and others on the internet (http://nmin.us/). A last point: it would be better to use N uptake than yield to evaluate NminSOM2.1 as suggested by Sullivan et al. (2020). Many factors contribute to yield that make it less desirable as an evaluative tool.
Nitrogen Mineralization and Soil Health
The Iowa State University mapping tool presents low, medium, and high ranges of daily N mineralization that would be ideal as input data for a cumulative normal distribution graph described by Moebius-Clune et al. (2016) where soil health score is plotted versus N mineralization. Potentially, the output of NminSOM2.1 could also be used to develop soil health scores vs. N mineralization from soil organic matter.
References
Benbi, D.K., & Richter, J. (2002). A critical review of some approaches to modelling nitrogen mineralization. Biology and Fertility of Soils, 35, 168–183.
Cappellazzi, S., & Morgan, C. (2021). Assessing soil health: Soil nitrogen cycling. Crops & Soils, 54(1), 26–30.
Clark, J.D., Fernández, F.G., Veum, K.S., Camberato, J.J., Carter, P.R., … & Shanahan, J.F. (2019). Predicting economic optimal nitrogen rate with the anaerobic potentially mineralizable nitrogen test. Agronomy Journal, 111, 3329–3338.
Gál, A., Vyn, T.J., Michéla, E., Kladivko, E.J., & McFee, W.W. (2007). Soil carbon and nitrogen accumulation with long-term no-till versus moldboard plowing overestimated with tilled zone sampling depths. Soil and Tillage Research, 96, 42–51.
Gilmour, J.T. (1998). Carbon and nitrogen mineralization during co-utilization of biosolids and composts. In S. Brown, J.S. Angle, and L. Jacobs (Eds.) Beneficial co-utilization of agricultural, municipal and industrial by-products. (pp. 89–112). Kluwer Academic Publishers.
Gilmour, J.T. (2021). Predicting soil organic matter nitrogen mineralization. Soil Science Society of America Journal, 85, 353–360.
Moebius-Clune, B.N., Moebius-Clune, D.J., Gugino, B.K., Idowu, O.J., Schindelbeck, R.R., … & Abawi, G.S. (2016). Comprehensive assessment of soil health—The Cornell framework ( Edition 3.2). Cornell University.
Morris, T.F., Murrell, T.S., Beegle, D.B., Camberato, J.J., & Ferguson, R.B. (2017). Strengths and limitations of nitrogen rate recommendations for corn and opportunities for improvement. Agronomy Journal, 110, 1–37.
Ordóñez, R.A., Archontoulis, S.V., Martinez-Feria, R., Hatfield, J.L., Wright, E.E., & Castellano, M.J. (2020). Root to shoot and carbon to nitrogen ratios of maize and soybean crops in the US Midwest. European Journal of Agronomy, 120, 1–11.
Ruehlmann, J., & Körschens, M. (2009). Calculating the effect of organic matter concentrations on soil bulk density. Soil Science Society of America Journal, 73, 876–885.
Sawyer, J., Nafziger, E., Randall, G., Bundy, L., Rehm, G., & Joern, B. (2006). Concepts and rationale for regional nitrogen rate guidelines for corn (PM 2015). Iowa State University Extension.
Sullivan, D.M., Moore, A.D., Verhoeven, E., & Brewer, L.J. (2020). Baseline soil nitrogen mineralization: Measurement and Interpretation (EM 9281). Oregon State University Extension Service.
Text © . The authors. CC BY-NC-ND 4.0. Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.
