Crop Nitrogen Uptake

[sumanager56]

• [ ] Summary of N controlling parameters
• [ ] N uptake and leaching response to parameters
• [ ] Corn Yield response

[sumanager56]

@julieshortridge @ryestewart As I was looking through the crop yield and nitrogen response to our calibrated hydrological parameters, the N uptake and yield values were very low as before. Also, the LAI values always stayed below 1 which means that the corn is not growing at all (Figure 1). The corn was not taking Nitrogen from the soil and the yield was very low (mentioned on figure caption for each plot below)

Figure 1. Plot with latest hydrological parameters, default crop parameters (except temperature sum required to reach anthesis and maturity) and using 180 kg/ha of mineral fertilizer on April 20, 2021.

Total N uptake = 33 kg/ha. Total NO3 leaching = 73 kg/ha. Total NO3_denitr = 0.15kg/ha

There is an option in the .crp input file within SWAP to control the fraction of N uptake by biological fixation. I tried changing this from 0 to 0.1 and the result changed significantly (Figure 2). Yield, LAI values are all looking good. NFIXF = 0.1 ! fraction of crop nitrogen uptake by biological fixation [-]

Figure 2. Plot with biological fixation turned on, everything same as in Figure 1.

Total N uptake = 28 kg/ha. Total NO3 leaching = 73 kg/ha. Total NO3_denitr = 0.16kg/ha

One thing that I changed from the CN module example was that I didn't simulate heat transport in all of my simulations (just turned off this option so far). Keeping everything same (no biological fixation), and just including heat flow calculation (Figure 3) in the simulation apparently seem to have solved most of the issues. I doubt if the corn yield (both potential and actual) are close to the expected yield range. The yield, N uptake, leaching, and denitrification rate have all increased significantly (Figure 4)

Figure 3. Heat flow section in swap input file. All are default values except TIMREF

Figure 4. Plot with just the heat flow calculation turned on.

Total N uptake = 114 kg/ha. Total NO3 leaching = 288 kg/ha. Total NO3_denitr = 155kg/ha

The denitrification rate can be reduced to 18kg/ha by reducing a denitrification coefficient (RateConDenitr_ref) from its default value, 0.01, to a new value, 0.001. However, those ultimately contribute to No3 leaching rather than N uptake (N uptake increased from 114 to 123 kg/ha, whereas leaching increased from 288 to 340 kg/ha).

I also ran the CN module example and extracted just the first year 1974 result from their 8 years of corn simulation (Figure 5). The only difference with ours is they have multiple application of manures rather than single and they have calibrated their parameters focusing more on dry matter yields rather than the grain yield.

Figure 5. Plot for CN_module example. Yield shown in the plot is grain yield. Crop period is April 19 to October 8.

Total N uptake = 129 kg/ha. Total NO3 leaching = 131 kg/ha. Total NO3_denitr = 0.0kg/ha

[sumanager56]

Updated the ".sme" file that contains the dates, source, and rate of fertilizers/manures. I initially had series of application dates starting from 2003 to 2022 (Figure 1) and ignored them since I was simulating only the year 2021. When I deleted all other events and ran the model with only one line shown in Figure 2, the corn yields, leaching, and N uptake have been improved (Figure 3).

Figure 1. Old .sme input file with series of soil management events (MatNum 10 is Mineral N fertilizer with NH4NFrac=0.5 and NO3NFrac=0.5)

Figure 2. Updated .sme file showing soil management events.

Figure 3. Updated plot after correcting .sme input file.

Total N uptake = 162 kg/ha. Total NO3 leaching = 142 kg/ha. Total NO3_denitr = 194kg/ha

[ryestewart]

Hi Suman,

Do you know why the N values are so different between scenarios? There are 100's of kg/ha differences in the total N mass for different scenarios. You are simulating some kind of N fertilization rate, correct? Is there a biological N component as well (fixation or mineralization?) Not sure from a mass balance if these numbers make sense, but I would expect the total N to be similar. Maybe there is a N storage term that is not being shown that accounts for the differences...?

Thanks, Ryan

Ryan D. Stewart, Ph.D. Associate Professor School of Plant and Environmental Sciences Virginia Tech, 241 Smyth Hall, Blacksburg, VA 24061 Phone: (540) 231-0253; Email: @. @.> Zoom: https://virginiatech.zoom.us/my/ryanstewart

On Thu, Feb 9, 2023 at 6:11 PM Suman Budhathoki @.***> wrote:

Updated the ".sme" file that contains the dates, source, and rate of fertilizers/manures. I initially had series of application dates starting from 2003 to 2022 (Figure 1) and ignored them since I was simulating only the year 2021. When I deleted all other events and ran the model with only one line shown in Figure 2, the corn yields, leaching, and N uptake have been improved (Figure 3). Figure 1. Old .sme input file with series of soil management events (MatNum 10 is Mineral N fertilizer with NH4NFrac=0.5 and NO3NFrac=0.5)

[image: Sme_old] https://user-images.githubusercontent.com/99036349/217959389-9ce375d1-b69c-49dc-a651-938ee916bd7e.PNG Figure 2. Updated .sme file showing soil management events.

[image: sme_new] https://user-images.githubusercontent.com/99036349/217957439-42fa3f54-256d-4b3e-9c97-0df1901306c8.PNG Figure 3. Updated plot after correcting .sme input file.

Total N uptake = 162 kg/ha. Total NO3 leaching = 142 kg/ha. Total NO3_denitr = 194kg/ha [image: Corrected_sme_dates] https://user-images.githubusercontent.com/99036349/217957608-76495e64-782c-485e-b8c5-098549a1b1a5.png

— Reply to this email directly, view it on GitHub https://github.com/sumanager56/SWAP-WOFOST/issues/11#issuecomment-1424963322, or unsubscribe https://github.com/notifications/unsubscribe-auth/A3TXL5UBT4L2QW2TQEDHAJDWWV2TZANCNFSM6AAAAAAUUL4EZE . You are receiving this because you were mentioned.Message ID: @.***>

[julieshortridge]

Yes, I would echo Ryan's comment - a complete mass balance of soil N for each simulation would be helpful. I'd imagine this would include the following, all in kg/ha (@ryestewart please advise if I'm missing something):

• mineral fertilizer application (+)
• biological fixation (0 in corn?) (+)
• nitrogen uptake (-)
• nitrogen leaching (-)
• denitrification (-)
• change in storage (could be + or -)

I think you are correct to remove the prior years of applications if we're doing a single-year simulation - I'm not sure why that would increase the N uptake/leaching/denitrification so much (if anything I'd expect the opposite), but in any case there's not a reason they should be included.

On your N uptake plot, could you put root water uptake on the same axes? I imagine they track with each other but it would be helpful to see. Could you also add precip events to the nitrate leaching plot?

Also at some point we'll need to look into why the potential yield is so low. That should be up closer to 200 bu/acre. I'm going to find the agronomic yield/N uptake data from 2020-2021 so you have that to compare, let me find it.

[sumanager56]

Thanks @julieshortridge , and @ryestewart. The mineral fertilizer application is 180kg/ha and there are other N components that I didn't include before. Here's an example of N balance (NH4-N and NO3-N) which constitutes all the inputs and outputs (Figure 1 and Figure 2, respectively for NH4-N and NO3-N). These numbers are highly variable depending on the temperature sum that I use in the model and different Nitrification and Denitrification coefficients. There are three different temperature sum inputs required for our crop (Temperature sum to emergence, temperature sum to flowering, and temperature sum to maturity). Our potential yield can also be increased by just changing the temperature sum requirement until flowering. For example, if temperature sum to flowering is simulated too late, biomass yield will higher compromising the grain yield and vice versa.

<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

### Figure.1 NH4-N balance NH4-N |   |   |   -- | -- | -- | -- Input |   | Output |   Ammendments | 90 | Volatilisation | 22.5 Crop Residues | 0 | Plant uptake | 5.81 Mineralisation | 660 | Nitrification | 751 Influx top | 29.6 | Outflux | 4.61 Influx lat | 0 |   |   Influx bot | 0 |   |   Total | 779.6 | Total | 783.92   |   | Diff.Stor | -4.32

<html xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:x="urn:schemas-microsoft-com:office:excel" xmlns="http://www.w3.org/TR/REC-html40">

### Figure.2 NO3-N balance NO3-N |   |   |   -- | -- | -- | -- Input |   | Output |   Ammendments | 90 |   |   Crop Residues | 0 | Plant uptake | 153 Nitrification | 751 | Denitrification | 207 Influx top | 29.6 | Outflux | 281 Influx lat | 0 |   |   Influx bot | 0 |   |   Total | 870.6 | Total | 641   |   | Diff.Stor | 229.6

The temperature sum seems to be highly dependent on the maize variety used and the climate. The temperature sum for sowing to emergence + emergence to flowering + flowering to maturity that I used before was 70+800+1700. For example when I change this sum to 70+700+400, the crop maturates to full grain ripeness very early and the potential yield in this case turns out to be 150 bu/acre (was around 100 bu/acre before).

[julieshortridge]

NOTE: REPOSTED WITH UPDATED SUMMARY YIELD AND N UPTAKE DATA FOR IRRIGATED TREATMENT

Thanks Suman. Regarding the temperature sum, have you found any literature values for these that are representative of the mid-Atlantic?

On the N balance information, are NO3 and NH4 the only two components? What are the units here? I'll admit this is a bit out of my area of expertise. @ryestewart what do you make of those tables?

I'm attaching a table summarizing the trial results from 2021 in the non-irrigated, 160 lb/acre N corn. It also includes the 0 lb/acre N, non-irrigated corn as a comparison. Yield ranges from 131 to 202 bu/acre, and N uptake ranged from 151 to 210 lb/acre (there is one 431 lb/acre N uptake value but Hunter was skeptical of that one and thought it could be an error). In the 0 N treatment, N uptake ranged from 40-50 lb/acre, with one anomalous high value of 102 lb/acre.

We can't use these results for a PEST-style calibration, but at least we should aim to have similar values for yield and N uptake.

CornSummary.xlsx

[sumanager56]

Thanks, Dr. Shortridge. I did find some literature values for the growing degree days (E.g., https://sites.udel.edu/agronomy/2020/09/23/2019-2020-corn-growing-degree-days/) and these seem reasonable as compared to the default values that I was using before. Also, when I quickly reviewed the GDD from Lal's email, it seems that the GDD falls quite in the same range as in the literature values. NO3-N and NH4-N are the only inorganic form of Nitrogen that is considered in the plant uptake here and the units are in Kg/ha. I am assuming the plant dry weight lb/ac is the dry matter yield which would be helpful when trying to adjust the crop parameters to get the ratio between grain yield and dry matter yield accurate. It's very easy to change the parameters and get some reasonable values, but I think I need to have a set of criteria built and focus on that.