«University of California Division of Agriculture and Natural Resources Committee of Experts on Dairy Manure Management September 2003 February 2004, ...»
The several variables combinations which were simulated are summarized in Table 1 for uniform irrigation and Table 2 for nonuniform irrigation. The amounts of applied N were 1.0, 1.2, and 1.4 times the N uptake for a non-stressed crop. When the N was applied with each irrigation, the amount of N applied at each irrigation was related to the potential N uptake for the succeeding period of time until the next irrigation. In some cases the N was applied with one-third at time of planting and then one-third each at 30 and 75 days after planting. The N applied in all of the above stated simulations were equally divided between ON and NH4+.
Other simulations were conducted when the applied N was entirely ON. These simulations had the total N being applied at the time of planting, and also having the N equally applied at 0, 30, and 75 days after planting. All simulations were conducted with the summer time mineralization rate one-half lifes of 90 and 280 days.
The effects of irrigation uniformity were investigated by doing simulations with irrigation uniformity CUC values of 73, 86, and 100. The N sources were equally divided between ON and NH4+ and applications were with each irrigation and also at three times during the cropping season.
E.5 Input Data for the Model The simulations were conducted for a soil bulk density of 1.40 g cm-3 and a saturated water content of 0.48 cm3 cm-3. The saturated hydraulic conductivity was chosen at 2.0 cm hr-1. The parameters used in the Hutson and Cass (1987) hydraulic function were as follows: water content at the inflection point ( i ) was 0.48 cm3 cm-3; the matric potential at the inflection point (hi ) was
-0.0028 MPa; the air entry matric potential (a) was -0.0027 MPa; and exponent (b) of the equation relating matric potential to water content was 3.8. The exponent (bhb) for the equation E-7 relating hydraulic conductivity to water content was set as 15.0. These functions are typical for a loam soil.
The lower boundary was set at 2m with 5-cm increments of soil depth for computation. The bottom boundary condition was set as free drainage. The upper boundary condition was set as flux control conditions with infiltration of irrigation according to the input rate. The bottom of the root zone was set at 1.5 m where drainage and N leaching were calculated.
The values, in units of MPa, for the threshold matric water stress (ht) was equal to -0.05 for both crops and the matric stress causing 50% growth reduction (h50) was equal to -0.14 for corn and for forage. Irrigation water, even with lagoon water mixed in, was assumed to be sufficiently low in salinity to not affect plant growth.
The initial water content distribution was established by setting the soil profile at saturation and then allowing redistribution for 14 days with free drainage as the bottom boundary condition.
This resulted in a water content of 0.34 cm3 cm-3 and the matric potential equal to -0.012 MPa at the bottom boundary and 0.32 cm3 cm-3 at the upper boundary. This soil water content profile was taken as the initial water content condition for corn in the first year, thereafter continuous simulation was conducted. The initial inorganic N distribution was 150 kg ha-1 evenly distributed over the top 20 cm and 100 kg ha-1 evenly distributed over the 20-200 cm layer. The initial water and N distribution only affected the results for the first and sometimes second year of the multiyear simulations in a manner similar to how the initial soil condition affects results in the field. The reason for running multiyear simulations was to determine the long term consequences of a management scheme and eliminate the effects of the initial conditions.
The Tp was taken as the reference ET0 times a crop coefficient. ETo values for Fresno, California, and crop coefficients for corn were taken from a report by Letey and Vaux (1985).
The crop coefficient for the forage crop was assumed to be 1.0.
E.6 Results The organic N mineralization rate is plotted as a function of day of year for five years in figure 2 when Np = 1.4, summer half-life is 280 days, and the N was applied with each irrigation. Note that steady state values are reached after about five years. The temperol rate of mineralization does not coincide with the temperol rate of N uptake (Fig. 1). Therefore the timing of leaching events will significantly affect the results. Large leaching rates at the initial and final stages of corn growth would particularly cause much N leaching that could affect crop yields as well as ground water degradation. In our simulations no large leaching events were programmed.
The relative yield (RY) of corn and forage and the amount of NO3 -N leaching (NL) are shown in Figure 3 over a 10-year period. The simulated conditions in this case were uniform application of 1.2 Np with every irrigation of 1.15 Tp. The applied N was equally divided with 50 % NH4+ E-8 and 50% ON, where the ON had 280- or 90- day summer-time half-life mineralization rate.
Maximum yields for both crops were simulated throughout. The high yields during the first one or two years can partially be attributed to the programmed high initial inorganic nitrogen content in the profile at the beginning of the simulations.
Note that the NL with a few exceptions increases with time and approaches a steady state value which is reached earlier for the 90-day than for the 280-day half-life. Eventually both reached approximately the same value. One of the important findings of these analyses is that when a given amount of organic N is applied consistently, the annual amount of mineralized N eventually equals the amount of total applied N regardless of the mineralization rate constant.
This is a fortunate circumstance because accurate information on mineralization rates is usually lacking. The mineralization rate constant determines the time period required to achieve steady state condition, but not the eventual quantitative steady state rate.
Note that after the first year essentially no NL occurs from the forage crop. (The first year results are greatly affected by the initial soil N distribution that was selected.) The low leaching under the forage crop can be attributed to two factors. The crop is grown in the winter months when the mineralization rate is very low. Therefore very little of the ON applied to the forage crop is mineralized and available for either crop uptake or leaching. The forage crop therefore was partially dependent on the mineral N remaining after the corn crop. The forage crop therefore utilized the available inorganic N sources which left little for leaching.
Much ON is mineralized between the period of forage production and maximum n uptake by the corn crop as well as after the time of maximum N uptake. These factors contribute to the significant amounts of computed NL under the corn crop.
Programming N application with time and amount consistent with crop uptake is simple when conducting computer simulations. Under a farm operation it is more common that the dairy waste only be added during some irrigations. For comparative purposes, we simulated applying the lagoon water during three irrigations for each crop, with equal amounts applied at the beginning, and approximately one third and two thirds through the crop season. All other conditions are the same as reported for Figure 3. The results of the latter simulation are illustrated in Figure 4. Note that in this case, corn yields were decreased and because of the reduced corn yields, more NL occurred. These results identify the importance of the timing as well as the total amount of N application. The decrease in corn yield is a result of having inadequate mineral N available during the relatively short time of peak N uptake requirement by the corn crop (fig. 1). Much of the applied mineral N and also the amount of mineralized ON occurred after the peak crop requirement. Note that the forage was not affected because it benefited from the carry over from the corn crop. These results illustrate one of the challenges of managing organic N such that the availability of mineral N matches the crop N requirement on a temporal basis.
E-9 The temporal effects reported above are largely associated with the mineralization of organic N.
To more completely understand the dynamics of N mineralization on crop yield and nitrate leaching, simulations were conducted for the same conditions as depicted in Figure 4 except all of the N was applied in the organic form. Results of this simulation are illustrated in Figure 5.
Note, in comparing the results depicted in figures 4 and 5 that the main difference between the presence or absence of mineral N occurs during the first few years of the simulation. After the organic N has been applied for sufficient years to achieve steady state, the response was very similar to a combination of organic and mineral N as long as the total amount of N applied was identical.
Simulations where higher and lower N applications than those depicted in figures 3, 4, and 5 were conducted. The main findings were as expected that higher N applications generally lend the higher yields and NL and the opposite for lower N applications. The temporal effects were not sensitive to total N application.
The results presented thus far are for uniform irrigation. The effect of the irrigation uniformity on results for the fifth year of simulation when steady state conditions had been approached are reported in Table 3. The simulated results are for N application with every water application and also for N application three times during each cropping season. The average water application was 1.15 Tp.
In general, RY increases with increasing irrigation uniformity. However, the difference between a CUC equal to 86 and uniform irrigation is not great. Increasing the N application rate tended to increase RY and NL.
The effect of irrigation uniformity on the amount of NL is variable. Indeed, when the N is applied three times during the cropping season, increasing irrigation uniformity resulted in a simulated increase in the amount of nitrate leached. Non-uniform irrigation results in parts of the field being "under irrigated" and other parts of the field being "over irrigated". Since the nitrogen was applied with the water, the sections of the field receiving the least amount of water also received the least amount of nitrogen. However, water rather than nitrogen was the limiting growth factor and no deep percolation of water occurred on the drier parts of the field. This would allow a small fraction of applied nitrogen to accumulate in the drier parts of the field thus leading to an overall field average reduced nitrate leaching.
Application of only organic N once at the beginning of each crop was compared to application three times during each crop growing season. The general results are as follows. During the first year, the yields for both corn and forage were higher for the one time application but in succeeding years, there was no difference between the two options. Under the steady state condition, application of 1.2 Np resulted in a relative corn yield of 0.9 and maximum yield for forages. Increasing the nitrogen application to 1.4 Np increased the relative corn yield to.97.
E-10 However, increasing the N application to 1.4 Np also increased the N leached by about 85 kg ha-1 yr-1.
E.7 Conclusions One major conclusion from this study is that when applying ON ultimately steady state conditions are achieved, whereby an equivalent amount of nitrogen applied in the organic form will be mineralized during a year. Steady state conditions are achieved more rapidly for materials with higher mineralization rates. This finding also underlines the importance that the results from short-term field experiments must be interpreted with caution. The experimental results will be very dependent upon the initial N status of the soil, mineralization rate of applied material, and whether organic N had been applied to the field several years prior to the experiment. When transitioning to an ON fertilizer source, higher amounts should be applied during the initial years and then decreasing amounts in successive years as steady state mineralization is approached.
A second conclusion is that the timing and total quantity of N application are both very important in affecting crop yield and potential N leaching. Many crops have very high N requirement over a relatively short period of time and will experience reduced growth if adequate N is not available during that period. Because mineralization of N is a continuous function, the timing of N availability with crop requirement is difficult to synchronize (Pang and Letey, 2000).
Significantly higher simulated yields were achieved when N was applied with every irrigation to meet crop demands as compared to equal applications three times during the crop season (figures 3 and 4).
Increasing irrigation uniformity resulted in increasing yield for a given N application amount.
Increasing irrigation uniformity increased, decreased, or had almost no effect on the amount of N leaching depending on the specific scenario. Because the N was applied with the water, nonuniform irrigation also caused nonuniform N application which contributed to the variable effects on N leaching.
Planting a forage crop during the winter effectively reduced the leaching of residual soil N following the corn crop. Application of ON during the winter when mineralization is slow provides very little mineral N for winter crop, but it becomes a major N source for the summer crop.
California Department of Food and Agriculture (CDFA), 2000. 1998. California Dairy Information, http://www.cdfa.ca.gov/dairy/dm98facts.htm.
Campbell, C.A., Bierderbeck, V.O., Warder, F.G., 1971. Influence of simulated fall and spring conditions on the soil system: II. Effect on soil nitrogen. Soil Sci. Soc. Am. Proc. 35, 480-483.
Frederick, L.R., 1950. The formation of nitrate from ammonium nitrogen in soils: I. Effect of temperature. Soil Sci. Soc. Am. Proc. 20, 496-500.
Hubbard, R.K., Thomas, D.L., Leonard, R.A., Batter, J.L., 1987. Surface runoff and shallow ground water quality as affected by center pivot applied dairy cattle waste. Trans. ASAE 30, 430-437.
Hutson, J.L., Cass, A., 1987. A retentivity function for use in soil-water simulation models. J.
Soil Sci. 38, 105-113.
Letey, J., Vaux, H.J., 1985. Water duties for California agriculture. A report submitted to the California State Water Resources Control Board, Sacramento, California Agreement no.