DNDC simulations on Audubon, Iowa
by Chin Ng
Contents
Introduction
Description of Site
Weather Data
Crops
Tillage and Fertilization
Simulations
Results
Conclusion
References
Introduction
Denitrofication-Decomposition (DNDC) is a simulation model of soil
carbon and nitrogen biogeochemisty in agricultural ecosystems.
The model is process-oriented and including four interacting
sub-models. These sub-models are soil climate sub-model, plant growth
sub-model, denitrification sub-model, and decomposition sub-model.
The soil climate sub-model uses air temperature, precipitation data,
and soil physical properties such as pH, clay fraction, and soil
density to calculate soil temperature, moisture profiles and soil
water fluxes. Information obtained is then fed to the denitrification
sub-model, the decomposition sub-model and/or the plant growth
sub-model. The denitrification sub-model computes hourly
denitrification rates, N2O and N2 production if soil has more than
forty percent pore space filled by water. The decomposition sub-model
solves for daily decomposition, nitrification, NH3 volatilization, and
CO2 production. The plant growth sub-model calculates nitrogen uptake
by plants and plant growth, and daily root respiration.
Soil C and N dynamics are very sensitive to climate, soil properties,
and cropping practices. Therefore, to run the model, specific climate
data, soil properties, and cropping practices at the simulation site
are required. DNDC thus could be used as a tool to evaluate the
effects of various changes in climate, soil, tillage, fertilization,
and crop rotation on agricultural lands. These simulations were
performed to study responses and impact of lands in the form of
emissions.
See carbon and nitrogen cycles.
Description of Site
The simulations were performed at Audubon, which is located in
west-central Iowa. This site was chosen because model parameter
inputs such as soil information, weather data, and details about
farming could be easily available. This farm could be taken as a
representative of all farms in the area of same soil type. Audubon
is located on Marshall soil, which is one of the predominant soil
series in western Iowa. The Marshall series consists of well-drained,
moderately permeable soils on ridges and side slopes in the uplands,
and to some extent on high stream benches. These soils formed in
loess and slopes of Marshall silty clay loam are typically 2-5%. The
content of organic matter is about 2-5%. Generally, the clay
composition of these soil is about 20-35%, permeability 0.6-2.0 in/hr,
water capacity 0.18 -0.23 in/in. They belong to the Hydrologic group B
and Wind erodibility group 7. They have pH in the range of 5.6-7.3.
The densities of these soils are about 1.45 g/cm3 and have an
erosion factor of 0.32-0.43.
The climate in this site is subhumid and continental where winters are
mostly cold and summers are warm. The growing season is long enough
for the crops to grow to mature in this area. The average temperature
in the winter is about 22 degrees F. The total annual precipitation
is about 33 inches, and the average seasonal snowfall is about 34
inches. Tornadoes and severe thunderstorms accompanied by hail and
damaging winds are an occasional threat to the area particularly during
hot, humid summer.
Weather Data
Weather used in these simulations was retrieved from NCDC's "Summary
of the Day". The data was first saved in the ASCII form. A C program
was then written to convert the weather data to a desired format
acceptable by the DNDC. The final weather data must have 3 columns
where the first column is Julian day, second is daily temperature, and
third daily precipitation. The temperature unit must be in degree
Celsius and precipitation in centimeter.
Crops
Corn, soybeans, and sorghum are generally grown in the farm in this
area. The land used for corn, soybeans and sorghum is more than about
100,000 acres, 40,000 acres, and 640 acres respectively. The normal
annual yield of corn is 151.1 Bu/acre or 8008 Kg/Ha based on the year
1994. For the purpose of this study, only continuous corn was
simulated.
Tillage and Fertilization
There are total of two tills and one fertilization each year. The
field is first plowed with disk/chisel on April 4. On May 1, the
land is fertilized with 70 kg/Ha of Nitrate, 100 kg/Ha of Ammonium and
30 kg/Ha of urea. The fertilizer is applied to the surface. Corn is
planted on May 5 and harvested on October 2. 3500 gallons of liquid
hog manure/acre is applied on October 29 and plowed with moldboard plow
on the November 12. No irrigation is applied because dry-land
agriculture is normally practiced in Iowa. For the purpose of this
study, no manure is applied to the field and for the case of no till,
there is no disking but just mulched once after harvest.
Simulations
Using the data obtained above, eight simulations for five years were
performed based on the different scenario and combinations of soils,
tillage, fertilizing, manuring. The eight conditions are listed in the
table below:
Table 1: Various scenarios for the simulations
| Condition |
Till |
Soil |
Clay Frac. |
Opt. Yield, Kg/Ha |
Fertilization |
| 1 |
Regular |
Loam |
0.20 |
8,008 |
Surface |
| 2 |
No |
Loam |
0.20 |
8,008 |
Surface |
| 3 |
Regular |
Silty clay loam |
0.35 |
8,008 |
Surface |
| 4 |
No |
Silty clay loam |
0.35 |
8,008 |
Surface |
| 5 |
Regular |
Loam |
0.20 |
15,000 |
Surface |
| 6 |
Regular |
Silty clay loam |
0.20 |
15,000 |
Surface |
| 7 |
Regular |
Loam |
0.20 |
8,008 |
Surface |
| 8 |
No |
Loam |
0.20 |
8,008 |
Injection |
Results
Scenarios 1 and 2:
The first simulation (regular till) was performed on loam with annual
optimum yield of corn at 8008 kg/ha (actual annual yield based on
1994). Clay fraction in loam was fixed as 0.20. Farming schedules
were fixed as described above and simulations were performed for 5
years beginning 1990 to 1994. The second simulation (no till) was
similar to the first one, except that the soil was only mulched once
after harvest. It was found that under reduced tillage, the soil had
less carbon preserved in it than the soil under regular tillage at the
end of the fifth year (Figure 1). It was also found that the emission
as % applied carbon was higher for the case with regular till than no
till (Figure 2). This may be due to the reason that by reducing
tillage, the rates of oxidation are lowered. It has a direct effect
on decomposition occurring in the soil. Therefore, the soil
respiration due to decomposition of organic matter, for instant, CO2
emissions is lower with respect to the case of regular till.
Table 2: Comparison of SOC at the end of the year
| Year |
Regular Till
(loam) |
No till
(loam) |
Regular till
(silty clay loam) |
No till
(silty clay loam) |
| Initial value |
93191 |
93191 |
58849 |
58849 |
| End of year 1 |
93422 |
91880 |
60141 |
58910 |
| End of year 2 |
94326 |
90940 |
62287 |
59376 |
| End of year 3 |
95909 |
90498 |
64662 |
60000 |
| End of year 4 |
96422 |
89794 |
66361 |
60331 |
| End of year 5 |
96521 |
89119 |
68017 |
60608 |
*SOC in Kg C/Ha
Reduced tillage also leads to the reduction in ammonia (NH3), nitrous
oxide (N2O), and nitric oxide (NO) emissions from the soil. Since NH3
is one of the first products of decomposition of nitrogenous organic
matter, lower levels of NH3 indicate lower decomposition rates or lower
nitrification rate. The output also showed that there is greater NO
emission during the nitrification process than during denitrification.
The N2O emission from denitrification was found to be higher than the
nitrification. The denitrification process is not affected by tillage
as much as the nitrification process, so not much change is seen in the
levels of the nitrogen (N2) released into the atmosphere.
Scenario 3 and 4:
These two simulations are similar to scenario 1 and 2 except that they
were performed on silty clay loam. It has a higher clay fraction of
0.35 and is also the major soil type of this area. Higher organic
carbon was found in the soil for the case of regular till than no till.
(Figure 1). When the two scenarios are compared, it is found that the
yield of carbon in crop biomass is higher in the regular till scenario
under silty clay loam soil. A portion of this biomass is returned to
the soil as litter. In case of no till, since the biomass yield is
lower, the amount returned to the soil is also lower. The total carbon
in the roots is 4889 Kg C/Ha in the regular till case while it is 4804
Kg C/Ha for the no till case. The straw yield is 8274 and 8130
respectively. There is a reduction in the root respiration. It was
expected that the organic carbon in the soil as a result of reduced
tillage would be relatively higher, but it did not happen in this case.
The SOC depends on the balance between the residue inputs and the CO2
outputs, in this case SOC at the end of the simulation is lower than
the respective no till scenario. The total emission as percent
applied-carbon was higher in no till case (Figure 2).
The ammonia
emissions were lower in the case of no till but nitrogen emission was
higher. The emission as percent applied-nitrogen appeared to be higher
in the case of regular till (Figure 3). In the case of no till, the
organic N preserved in the soil was relatively lower. The NO emission
was found to be lower in the no till case but the N2O emissions were
generally the same. The nitrogen uptake was lower in the no till case.
It was also found that for both cases, N2O emission from
denitrification was higher than nitrification, and NO emission from
nitrification was higher than denitrification.
Scenarios 5 and 6:
In these simulations, regular till was used in both loam and silty clay
loam with the optimum yield of 15000 Kg/Ha. The SOC in the field was
found to be much more higher in the loam soil than silty clay loam
(Figure 4). The carbon output that was from the respiration in the
soil and root of silty clay loam were both found to be considerably
lower in this case. The emission of carbon was found to be
approximately the same in both loam and silty clay loam. The emission
as percent applied-carbon in the case of loam was found to be higher
than the case with lower optimum yield of 8008Kg/Ha (Figure 5). The
emission as percent applied-carbon in the case of silty clay loam was
found to be just slightly higher than the case with lower optimum
yield.
The emission as percent applied-nitrogen appeared to be lower in loam
than in silty clay loam (Figure 6). The NO and NH3 emissions in silty
clay loam were higher whereas the N2 emission and the nitrogen uptake
were much higher in loam. In both loam and silty clay loam,
NO emission from nitrification was found to be higher than
denitrification, and N2O emission from denitrification was higher than
nitrification. The emission as percent applied-nitrogen in both loam
and silty clay loam were similar to the case of lower optimum yield.
Scenarios 7 and 8:
In these simulations, regular and no till conditions were applied to
loam soil. The optimum yield used here was 8008 Kg/Ha. In scenario 7,
no fertilizer was applied to the field. The fertilizer utilized in
scenario 8 was applied by injection. It was assume to injected 10 cm
into the soil. In the case of regular till and no fertilization, the
SOC content was found to be higher than the case of no till with
fertilization (Figure 4). The respiration soil carbon emission was
higher for the case of regular till without fertilization, but the
respiration root carbon emission was higher for the case of no till
with fertilization. The emission as percent applied-carbon was lower
in the case of regular till without fertilization (Figure 5).
The emission as percent applied-nitrogen was found to be higher in the
case of no till with fertilization (Figure 6). Apparently, NO, N2O,
N2, and NH3 emission in the case of no till with fertilization were
higher than the case of regular till without fertilization. In both
cases, N2O emission from denitrification was again found to be higher
nitrification, and NO emission from nitrification was higher than that
from denitrification.
Conclusion
From the simulations performed on Audubon site, the soil carbon and
nitrogen was found sensitive to soils, optimum yield, farming practices
and weather information. By different climate change, crop grown,
tillage, soil composition, and farming practice, the changes in soil
nitrogen and carbon emissions can be predicted by the DNDC model. If
the actual values of soil nitrogen and carbon emissions data from the
field can be obtained, the results given by the DNDC model can then be
compared. The deviation between the model outputs and the actual
results can then be learned.
References
Li, C., Frolking, S., & Harriss, R. (1994). Modeling carbon
biogeochemistry in agricultural soils. Global Biogeochemical Cycles,
8, 237-254.
Lai, R., Kimble, J., Levine, E., & Stewart, B.A. (1995). Soil
Management and Greenhouse Effect: Modeling impact of agricultural
practives on soil C and N2O emissions. London: Lewis Publishers.
Collins, C.W. (1974). Atlas of Iowa. Madison, Wisconsin:
American Printing & Publishing, INC.
Li, C., Frolking, S., & Harriss, R. (1995). User's Guide for the DNDC Model.
Corn: Acreage, yield and production. Iowa Agricultural Statistics.