Optimum Sustainable Nitrogen Fertilization of Cool-Season Grasses

M.H. Hall*, D.B. Beegle, R.C. Stout, and R.S. Bowersox

ABSTRACT

Cool-season grasses are an important component of farming enterprises throughout the Northeast and Midwest United States. Numerous studies have looked at the economic-optimum nitrogen application rate (cost of N vs increased yield), however, there have been limited attempts to determine level of N fertilization that is both economically and environmentally sustainable. The objectives of this study were to determine the N application rate to four cool-season grasses species that optimizes economic return and minimizes amount of N not taken up by the plant. Four N rates were applied to established stands of orchardgrass (Dactylis glamerata L.), smooth bromegrass (Bromus inermis Leyss.), tall fescue (Festuca arundinacea L.) and timothy (Phleum pratense L.) in 1999, 2000, and 2001. The grasses were harvested three and four times per year.

 

Abbreviations: DM, dry matter; NIRS, near infrared reflectance spectroscopy.

MATERIALS AND METHODS

'Pennlate' orchardgrass, 'Baylor' smooth bromegrass, 'Barcel' tall fescue, and 'Clair' timothy were established during the spring of 1998 in 2 x 6 m tilled plots of a Murrill silt loam (fine-loamy, mixed, mesic Typic Hapludult) soil at the Russell E. Larson Agric. Res. Ctr. near Rock Springs, PA. The varieties were selected because of their wide spread use in the Northeast region of the USA.  Soil pH, available P, and exchangeable K were maintained above 6.5, and 70 and 336 kg ha-1, respectively, throughout the experiment.  During the establishment year, grasses were harvested twice and received split applications of ammonium nitrate totaling 56 kg ha-1 N.

Beginning in early April of 1999, N fertilizer treatments of 0, 134, 201, 268 kg ha-1 N as ammonium nitrate were applied to each grass species (Table 1). Grass yields did not always plateau at the highest N rate in 1999 so the rates were increased to 0, 134, 268, 402 kg ha-1 N in 2000 and 2001. Grasses were harvested three or four times per year. First harvest for the three-, and four-harvests systems was made on approximately 6 June, and 20 May, respectively, and subsequent harvests taken at approximately 45, and 35 d intervals, respectively.

The experimental design was a randomized complete block in a split-plot arrangement with four replicates.  Whole plots were two harvest treatments and subplots were species x fertilizer treatments (16 total).  Weather data was collected at a weather station located within 1.0 km of the study. 

At harvest, a 1 x 5 m strip was removed from the center of each subplot with a flail type mower and a 1 kg sample of the harvested material was collected and dried in a forced-air oven at 60°C for 48 h for dry matter determination and N analyses. The samples were then ground to pass a 1-mm screen and total N content was determined by Kjeldahl procedures and NO3-N concentration was determined colorametrically after extraction with potassium cloride.  Amount of N extracted from the soil (kg ha-1) was calculated as 0.001 x [DM yield x N concentration of forage] and N recovery (g kg-1) was determined as 100 x [(N extracted in treatment - N extracted in control)/N applied] where control was the 0 kg ha-1 N application rate (Primavesi et al., 2001A).

Net revenue from each treatment was based on the DM yield and average value of forage from three and four cut systems assuming a value of $85 and $93 Mg-1 for 3x cut and 4x cut, respectively, for grass forage (Hall, 1998). A harvesting charge of $80 ha1 harvest-1 was assessed against the value of the forage. Cost for N was assessed at $0.55 kg-1. Optimum economic N fertilization rate was determined for each species and harvest system by regression of the net revenue against the N application rate. Subsequently, yield at the economically optimum N rate was determined for by solving for Y when yield was regressed against N application rate.

Soil cores were taken to a depth of 0.9 m from each plot in October following the 2001 growing season using a 50 mm diameter soil probe. Each soil core was segmented into sequential 150, 150, 300 and 300 mm segments and analyzed for total N content by xxxxx procedure.

Acceptable environmental N fertilization rate was determined for each species and harvest system by comparing soil [N] below the grass rooting zone (150 mm depth) with xxx ppm. This level was established by the Environmental Protection Agency above which there is a probable risk of ground water contamination. Nitrogen application rates x species x harvest system that exceeded soil [N] in excess of xx ppm were considered acceptable management practices.

Data were first subjected to exploratory analysis to determine if the assumptions of analysis of variance held.  Homogeneity of variance was tested using Hartley’s F-max test (Milliken and Johnson, 1984).  An additional test for homogeneity (PROC REG: SAS Institute, 1997) was conducted to determine if a common curve could be used to describe the data over all years. All regressions and statistical analyses were completed using SAS Institute (1997) software. Tukey’s multiple comparison procedure was used for mean separations. Differences reported in this paper are all at the P > 0.05 level of significance.

RESULTS AND DISCUSSION

Grass yields did not always plateau within the highest N rate in 1999 so an optimum economic yield could not be calculated. Consequently, the results from 1999 are presented only when if economic yield plateaued. Smooth bromegrass results are not presented because yields did not reach a plateau within the 402 kg ha-1 N application rate in 2000 or 2001. Economic optimum N fertilization rates for smooth bromegrass were reported by Malhi (1997) but at lower yields (4.5 Mg ha-1) than in our study (10.2 Mg ha-1).

Growing Conditions

Average monthly temperatures during the three growing seasons never exceeded 1.1 °C warmer or 2.9 °C cooler than the 10-yr average (Table 2). The cooler temperatures occurred during the 2000 growing season. Precipitation during the growing seasons was below normal all three years (Table 2).  The lowest seasonal precipitation occurred in 2001, when rainfall was 74 mm below normal with the majority of that deficit occurring in May. However, the first four months of the 1999 growing season nearly 80 mm below normal which may account for the generally lower yields than in the other years and the inability for many of the treatment to cause a yield plateau.

Optimum Economic Yield

The variance for economic return over years were homogenious but a single regression curve could not accurately represent a species over years. Consequently, individual economically optimum N rates and their regresssion equation are persented (Table 3).

Economic return increased curvalinearly for orchardgrass, tall fescue and timothy as N fertilization rate increased and was not effected by number of harvests per year (Table 3).   Dry matter yields at the optimum economic N rate were in general lower in 2001 than 2000. The amount of N to apply per Mg of forage was higher than the current recommendation, especially when four harvest are taken per year. These findings support the conclusions of Klausner et al. (1998) in that our economical optimum N rate is also approximately 25 kg N ha-1 Mg-1 forage harvested.

 

NO3-N in the Forage

NO3-N concentrations in the harvested forage in 2000 and 2001 were analyzed and are presented together because the variances of data for the two years were determined to be homogenous. There were no interactions between species, harvest or fertilization rates for NO3-N concentration. Forage NO3-N concentration increased as N fertilization rates and number of harvests per year increased (Table 4).  Klausner et al. (1998) reported similar trends in NO3-N concentrations.  In the three-harvests per year system, forage NO3-N concentration exceeded the 1000 ppm (threshold at which health of consuming animals becomes a serious concern; Emerick, 1963) only when 402 kg ha-1 N was applied on tall fescue. However, forage from the four-harvests per year system contained NO3-N levels in excess of 1000 ppm in the spring forage at both the 268 and 402 kg ha-1 N fertilization rate on all grasses.

In all harvest, timothy had lower NO3-N concentrations than either orchardgrass or tall fescue which were not different from each other (Table 4).

Certainly, caution must be exercised when feeding forages that are receiving economically optimum N rate and being harvest four times per year. This is especially true for the first two harvests of the year.

 

N Recovery

Nitrogen recovery data from 2000 and 2001 were analyzed and are presented together since there were no interactions between years and the variances of the two years were determined to be homogenous. Averaged across all grass species and N rates, N recovered was the lower in the three- than the four-harvests per year systems (Table 5). N recovery in orchardgrass and tall fescue increased as N application rate increased from 134 to 268 kg ha-1and then declined. For timothy the recovery rate was not different between the 134 to 268 kg ha-1 application rates but was lower at the 402 kg ha-1 rate. For all grass species, N recovery decreased from the 268 to the 402 kg ha-1 N application rates.

The below-ground plant structures may account for some of this “unrecovered” N, however, it is unlikely that 100% of the applied nitrogen was captured in the plant.

 

Nitrogen in the Soil

 

Conclusions

 This research indicates that current economically optimum nitrogen fertilization rates for orchardgrass, tall fescue and timothy are approximately 5 kg N too low Mg-1 of dry forage for either a three and four harvest per year system. However, at those nitrogen fertilization rates nitrate concentration of the forage from harvests one and two of the four-harvest system are generally in excess of 1000 ppm for all grass species.

Averaged across all grass species and years, nitrogen recovered in the harvested forage was 491 and 586 g kg-1 of applied nitrogen for the three and four harvest per year system, respectively. The below ground plant structures may account for some of this “unrecovered” nitrogen, it is doubtful that 100% of the applied nitrogen is captured in plant material.

Soil nitrogen concentration was ??? and ??? greater at  450 and 750 mm soil depths, respectively, at the  economically optimum nitrogen fertilization rates found in this study compared with the currently recommended economically optimum nitrogen fertilization rates. However, currently recommended economically optimum nitrogen fertilization rates had ?? and ?? greater soil nitrogen concentration at those depths than when no nitrogen was applied..

Table 1. Nitrogen application rate and timing.

Time of application                       N rate in 1999                        N rate in 2000 & 2001

                                                                                     kg ha-1

Total                                      0      134      202      268             0      134      268      403

      At greenup                       0        66        90      112             0        66      112      179

      After first harvest              0        34        56        78             0        34        78      112

      After second harvest         0        34        56        78             0        34        78      112

 

Table 2. Air temperature and rainfall at Rock Springs, Pennsylvania during the growing seasons of 1999, 2000, and 2001.

                     Average monthly air temperature                                     Rainfall

                  __________________________                  ___________________________

Month        1999      2000       2001      10-yr                   1999       2000       2001       10-yr

                                                              avg                                                                   avg

                                        °C                                                                   mm

April            9.0          8.6          9.4          8.7                    96         101           72         101

May          15.1        15.9        14.4        14.8                    36           82           36           81

June          19.2        19.7        19.3        19.5                  104         117         146           90                         

July           22.9        18.9        19.5        21.8                    49           71           84           92

Aug.          19.4        19.1        21.3        20.9                  156           99           81         111

Sept.         16.9        15.1        15.1        16.8                  122           69         106         125                             


Table 3. Economic optimum N fertilization rates per hectare (ENR/ha, regression equations), dry matter yield at ENR, and economic optimum N fertilization rates per Mg dry matter forage yield at ENR (ENR/Mg) for three forage grasses harvested three and four times per year.

                                                                                                                                     

                                                                    Three harvest year-1                                                           Four harvest year-1

                                        ___________________________________________ ________________________________________________

Species             Year         ENR                                        Yield at ENR    ENR Mg-1           ENR                   Yield at ENR                      ENR Mg-1

                                        kg ha-1                                                kg ha-1     kg Mg-1                 kg ha-1                                     kg ha-1                                        kg Mg-1

Orchardgrass    1999    145  (y = -0.1331x2 + 51.579x + 4044.2)            8725          17              192  (y = -0.0118x2 + 27.338x + 3312.1)                  8126    24

                        2000    279  (y = -0.0039x2 + 2.7282x + 185.67)       12235          23              272  (y = -0.0062x2 + 3.9231x + 168.63)              11750  23

                        2001    217  (y = -0.0054x2 + 2.8974x + 154.59)          10474          21              247  (y = -0.0065x2 + 3.7672x - 77.71)                    9169    27

 

Tall fescue        1999          ENR not achieved                                                                 251  (y = -0.0332x2 + 32.697x + 4081.1)              10196        25

                        2000    310  (y = -0.0038x2 + 2.9033x + 99.20)            13017          24              323  (y = -0.0048x2 + 3.6537x + 74.12)               11295  29

                        2001    300  (y = -0.0037x2 + 2.7719x - 57.67)                9928          30              284  (y = -0.0046x2 + 3.1597x - 151.00)                  9626    30

 

Timothy            2000    231  (y = -0.0052x2 + 2.9491x + 134.31)       10227          23              238  (y = -0.0049x2 + 2.8853x + 118.35)              10549  23

                        2001    233  (y = -0.005x2 + 2.8814x + 7.43)                   6721          35              211  (y = -0.0047x2 + 2.5325x - 78.27)                                7332    29

Values based on the DM yield and average value of forage based on a reference forage value of $85 and $93 Mg-1 from a three and four cut systems, respectively. A harvesting charge of $80 ha-1 harvest-1 was assessed against the value of the forage. Cost for N was assessed at $0.55 kg-1.

Table 4. Nitrate concentration in herbage of three cool-season forage grasses when different rates of nitrogen were applied. Values are the means of four replicates over the 2000 and 2001 growing seasons.

                                                                                                                                     

                                                   Three harvest year-1                            Four harvest year-1

                               Nitrogen      ______________________           _____________________________

Species                rate             Hvst 1     Hvst 2     Hvst 3             Hvst 1     Hvst 2     Hvst 3     Hvst 4

                            kg ha-1                                                                         ppm

Orchardgrass              0            377         393        302                  523         497        398           441

                              134            546         430        378                  690         558        448           532

                              268            784         737        761                1218       1059        872           861

                              402            877         926        956                1209       1111        913           980

                  LSD (0.05)              87         123        103                  182         160        121             93

 

Tall Fescue                 0            367         377        306                  516         510        347           392

                              134            578         446        434                  643         563        445           571

                              268            910         859        759                1154       1028        820           847

                              402          1078         982      1010                1105       1212      1031         1040

                  LSD (0.05)            139         154        177                    88         129        131           101

 

Timothy                      0            459         263        239                  603         452        281           335

                              134            590         358        308                  742         450        393           404

                              268            759         615        593                1047         774        573           603

                              402            792         678        644                1173         810        633           825

                  LSD (0.05)            146         193          94                  164         157        142           114

 

Table 5. Nitrogen recovery in the herbage of three cool-season forage grasses when different rates of nitrogen were applied. Values are the means of four replicates over the 2000 and 2001 growing seasons.

                                                                                                                                     

                                                          Harvests yr-1                                             

                               Nitrogen                ________________                                        

Species                rate                    three                 four               LSD (0.05)                          

                            kg ha-1                                   g kg-1

Orchardgrass          134                  466                  582                             

                              268                  577                  801                             

                              402                  419                  569                             

                  LSD (0.05)                    62                    70                             

 

Tall Fescue             134                  492                  536                             

                              268                  641                  736                             

                              402                  522                  544                             

                  LSD (0.05)                    44                    63                             

 

Timothy                  134                  443                  539                             

                              268                  519                  547                             

                              402                  337                  425                             

                  LSD (0.05)                    79                    72

 

All species x All N rates                  491                  586                    66

Nitrogen recovery (g kg-1) was determined as 100 x [(N extracted in treatment x N extracted in control)/N applied] where nitrogen extracted from the soil (kg ha-1) was calculated as 0.001 x [DM yield x N concentration of forage] and control was the 0 kg ha-1 N application rate.

Figure 1. Nitrogen concentration at sequencial soil depths after three cool season forage grasses were fertilized with different rates of nitrogen. Values are the means of four replicates.