| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Dep. de Ciencias y Recursos Agrícolas y Forestales, Univ. of Córdoba, Campus de Rabanales, Edificio C-4 "Celestino Mutis," Ctra. Madrid km 396, 14071 Córdoba, Spain
b Dep. de Ciencias Agroforestales, Univ. of Huelva, Spain
c Dep. de Producción Vegetal y Tecnología Agraria, Univ. of Castilla La Mancha, Spain
* Corresponding author (cr1lobel{at}uco.es)
Received for publication January 12, 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Abbreviations: HI, harvest index • NUE, nitrogen use efficiency • NF, nitrogen-15 fertilizer • NR, labeled-fertilizer nitrogen recovery • NUpE, nitrogen uptake efficiency • NUtE, nitrogen utilization efficiency
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Proper N application timing and rates are critical for meeting plant needs and improving NUE. In addition, the growth stage of plants at the time of fertilizer application also determines NUE, with significant genotypic variations (Ashraf and Azam, 1998). Limaux et al. (1999) report that the timing of fertilizer N applications has a significant effect on the uptake of fertilizer N by the crop and the resulting partitioning of added N between soil and plant. Nitrogen applications split between fall and spring have been found to increase yield, NUE, and N uptake efficiency (NUpE) compared with fall application in hard red winter wheat under temperate conditions (Mahler et al., 1994; Sowers et al., 1994).
The efficiency of N fertilizer use when applied as a topdressing in wheat is influenced by timing, fertilizer rate, and rainfall. Maximum efficiency should be achieved with the latest possible application, as long as the growing plant is still capable of swift N uptake; this would avoid unnecessary vegetative growth and the risk of lodging and also reduce N loss through leaching, denitrification, volatilization, and runoff since an active root system ensures uptake of the N fertilizer applied (Alcoz et al., 1993). Both Alcoz et al. (1993) and Stockdale et al. (1997) reported increased wheat yield when N fertilizer was applied between the end of tillering and formation of the first node (Stages 4 to 6 on the Feekes scale; Large, 1954) compared with application at planting or during heading (Feekes Stage 10). Mossedaq and Smith (1994) suggest that N applications should occur immediately before the period of peak N demand (i.e., the onset of stem elongation) and speculate that this will result in minimizing N losses from leaching.
Fertilizer experiments using 15N for winter wheat in the temperate climate of northwestern Europe confirmed that crop N fertilizer uptake was significantly lower after early applications (e.g., planting and tillering) than after applications later in the crop growth cycle (e.g., shooting) (Recous and Machet, 1998; Tran and Tremblay, 2000). Field trials in central Europe have recorded an average 50 to 60% recovery of N fertilizer applied to winter wheat (grain and straw) (Powlson et al., 1992; Blankenau et al., 2002; Macdonald et al., 2002). Wuest and Cassman (1992) report that the recovery of N applied at planting ranged from 30 to 55% while that of N applied at anthesis ranged from 55 to 80%. Sowers et al. (1994) found that more 15N-labeled fertilizer was recovered with split N application than with fall-applied N. These results suggest that spring application of N as a topdressing may improve fertilizer N recovery and NUE over presowing applications in dryland winter wheat.
In dry regions, much of the fertilizer N not absorbed by the crop in years with inadequate rainfall may remain in the soil as nitrate available to subsequent crops. The fate of this residual nitrate will depend on the incidence of waterlogging, which causes denitrification (Craswell and Godwin, 1984). In the Mediterranean climate, periodical soil water shortages have a considerable impact on fertilizer efficiency. Generally speaking, the efficiency of fertilizer N in Mediterranean climates is lower than that observed in temperate areas. Garabet et al. (1998) indicate that for wheat grown under rainfed Mediterranean conditions, recovery is <50% of the applied fertilizer N. However, the influence of N fertilization timing on grain yield is less well documented for fall-sown spring wheat in Mediterranean environments although Mossedaq and Smith (1994) suggested that timing of application be considered as a critical factor in N fertilizer management.
Traditionally, considerably less N fertilizer has been applied to durum wheat than to bread wheat in the Mediterranean region due to lower grain yield and greater height of the cultivars, which are susceptible to lodging. Over the last 25 yr, however, new cultivars selected by the CIMMYT that are earlier, lowstanding, and disease resistant and have greater yield potential have been introduced in this region. The use of these cultivars has prompted an increase in fertilizer N inputs. Currently, the N fertilizer rate applied to new durum cultivars sown on rainfed Vertisols in subhumid areas or irrigated land in the Mediterranean region is broadly similar to that applied to bread wheat (López-Bellido, 1992).
Numerous published studies have addressed optimal N fertilization rates and timing, generally in the context of intensive wheat management. However, research has been focused mainly on bread wheat, notwithstanding the importance of durum wheat. Few field studies have dealt with the behavior and response of durum cultivars to N fertilization. The aim of the present study was to determine the effects of different fertilizer N timings on grain yield, N uptake, NUE, and recovery of 15N-labeled fertilizer by durum wheat in a field experiment under rainfed Mediterranean conditions.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
At harvest, a 0.5-m2 (0.18 by 3 m) portion at the center of each wheat plot was sampled. From this sample, the biomass, harvest index (HI), heads per square meter, seeds per head, thousand-seed weight, and N uptake were measured. Dry matter and seed weight were determined by drying the sampled plants at 80°C to constant weight. Nitrogen content of straw and grain was determined using the Dumas combustion method (Leco FP-428 analyzer). Nitrogen uptake was calculated by multiplying dry weight by N concentration in straw and grain.
The following N efficiency parameters were calculated for each treatment: NUE (kg kg–1) as the ratio of grain yield to N supply, where N supply is the sum of soil NO3––N at planting, mineralized N, and N fertilizer; NUpE (kg kg–1) as the ratio of total plant N uptake to N supply; N utilization efficiency (NUtE; kg kg–1) as the ratio of grain yield to total plant N uptake; N harvest index (NHI; kg kg–1) as the ratio of N in grain to total plant N uptake; and N apparent recovery fraction (NRF; kg kg–1) as the ratio of N uptake at Nx – N uptake at N0 to applied N at Nx. Nitrogen efficiency terminology follows Moll et al. (1982), Pierce and Rice (1988), Huggins and Pan (1993), Sowers et al. (1994), and Delogu et al. (1998).
Labeled Nitrogen Experiment
Microplots (1 by 2 m) were established within the main experiment area to monitor uptake of 15N-labeled fertilizer. Microplots were arranged in a randomized complete block with four replications of six treatments. All microplots received 150 kg N ha–1, with the following application timings: (i) 100% fall (15N labeled), (ii) 50% fall (15N labeled)–50% topdressing (TP), (iii) 50% fall–50% TP (15N labeled), (iv) one-third fall (15N labeled)–two-thirds TP, (v) one-third fall–two-thirds TP (15N labeled), and (vi) 100% TP (15N labeled). The data from treatments (ii) and (iii), and (iv) and (v), were combined to determine the total contribution of fall-applied and topdressed N fertilizer to plant N in this application. Fertilizer solutions were formulated with urea 46% and urea 15N enriched (2.5 atom % excess 15N) for fall application and ammonium nitrate 27% and ammonium nitrate 15N enriched (2.5 atom % excess 15N) for topdressing application. Fall applications were made immediately after planting, and topdressings were applied at the start of stem elongation. The treatments were applied to the soil surface of the microplot area in 4 L of distilled water per microplot, using a hand sprayer.
At maturity, a 0.5-m2 (0.18 by 3 m) area of plants was harvested from each microplot, threshed, dried, and ground. All samples were analyzed for 15N content with a IRMS Delta Plus XL mass spectrometer (Thermo Electron Corp., San Jose, CA, USA).
Labeled-fertilizer N recovery (NR) in the plant on an area basis and percentage basis was calculated as follows, after Hauck and Bremner (1976): NR =Nt x (c – b)/(a – b) and percentage NR = (NR/f) x 100, where Nt = total plant N at maturity in kg ha–1, a = atom % 15N in the fertilizer, b = atom % 15N in the unfertilized plant, c = atom % 15N in the fertilized plant, and f = fertilizer rate in kg N ha–1.
The NF was calculated as: NF = NR/Nt x 100.
Crop Management
Experiments were performed on a farm representative of local farming conditions using a different plot every year, always following a cotton crop. The N fertilizer treatments listed above were applied as urea and incorporated. The topdressing treatments were applied as ammonium nitrate. Every year wheat plots were also supplied with P fertilizer before sowing at a rate of 65 kg P ha–1; the fertilizer was incorporated into the soil. Soil available K was adequate.
Wheat was planted in December in all years at a seeding rate of 150 kg ha–1 in 18 cm wide rows. The cultivar used was D. Pedro, a short-cycle durum wheat of medium height. The cultivar displays high productivity and very good pasta quality. Wheat was harvested early in June each year, using a 1.5 m wide Nurserymaster Elite plot combine (30 m2 plot–1).
Meteorological data were obtained from the weather station situated at Córdoba airport, 2 km from the farm.
Statistical Analyses
Experimental design was a randomized complete block with four replications. Annual data for each parameter over the whole 3-yr period were subjected to analysis of variance (ANOVA), using a year-combined randomized complete block design according to McIntosh (1983). Treatment means were compared using Fisher's protected least significant difference (LSD) test at P 
0.05. The LSDs for different main effect and interaction comparisons were calculated using the appropriate standard error terms following Gómez and Gómez (1984). The Statistix v. 7.0 (Analytical Software, 2000) package was used for this purpose.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Rainfall distribution also differed between years (Fig. 1). In 1999–2000, a mean 39% of total rainfall was recorded in the fall, before wheat planting; rainfall was scarce in winter, 18%, while spring accounted for 27% of total annual precipitation. By contrast, in 2000–2001 rainfall was low in the fall, with a mean value of 18%, and abundant in winter with a mean value of 48% of total annual rainfall; mean spring rainfall was 31%. In 2001–2002, the rainfall distribution pattern was more balanced, with a mean 27% in fall, 27% in winter, and 33% in spring.
Yield and Yield Components
The response of wheat grain yield to N fertilizer timing varied depending on the season (Table 2). The highest yield occurred in 1999–2000, the driest of 3 yr (Fig. 1), while the lowest yield occurred in 2000–2001, the wettest year. Total dry matter was similar in 1999–2000 and 2000–2001, and significantly higher in 2001–2002. Harvest index values were significantly higher in the year with the lowest grain yield and biomass (2000–2001) (Table 2). This year recorded the smallest number of heads m–2 and the highest number of seeds head–1 (Table 2). Since rainfall was excessive in the fall–winter period of 2000–2001, this probably resulted in poor establishment of the crop and poor tillering due to waterlogging.
|
The lack of any clear response by durum wheat to N fertilizer timing reflects the effect of annual variations in the amount and distribution of rainfall on the uptake and efficiency of fertilizer N at various stages in wheat, and the resulting variations in soil residual N content before planting, as noted earlier. The year x treatment interaction was therefore significant for grain yield, total dry matter, and grain protein content (Table 2). As Fig. 2 shows, the effect of N fertilizer timing on grain yield was more marked in 1999–2001 than in the other 2 yr; topdressing exerted a clearer influence on grain yield. Data for 1999–2001 were characterized by better rainfall distribution, greater N uptake, and greater yield, as observed earlier.
|
|
Soil mineralized N content also varied from year to year: 39, 43, and 76 kg ha–1 in 1999–2000, 2000–2001, and 2001–2002, respectively. Yearly variations in rainfall distribution over the growing season mentioned earlier may have prompted differences in the mineralization rate, linked both to favorable soil moisture levels during fall months and an excess of soil water during the winter, given the strongly clayey nature of Vertisols, as indicated by Stockdale et al. (1997).
Based on soil NO3––N content at planting and the levels of soil mineralized N, a considerable amount of N was available to wheat in all three study years: 156, 163, and 181 kg ha–1 in 1999–2000, 2000–2001, and 2001–2002, respectively. Large amounts of native N may account for the weak and sometimes inconsistent response of biomass and grain yield to variations in the timing of 150 kg ha–1 of N fertilizer. Craswell and Godwin (1984) suggest that considerable quantities of fertilizer N are accumulated as nitrate in the soil profile in dryland regions, particularly when high rates of N are applied, since much of the fertilizer N not absorbed by the crop in years with inadequate rainfall may remain in the soil, and thus available to subsequent crops. For irrigated crops in the Mediterranean region, Abad et al. (2004) report soil NO3––N levels ranging from 77 to 157 kg ha–1, and assert that rates of 100 kg N ha–1 may be sufficient to obtain high grain yields and good durum wheat quality.
Nitrogen Efficiency
Nitrogen use efficiency was not significantly affected by the various N fertilizer timing treatments (Table 3). However, the control treatment (zero N) gave a NUE index of almost double the mean value for timing treatments using 150 kg N ha–1 (Table 3). This increase is far greater than that reported by Sowers et al. (1994), and agrees with the results obtained for bread wheat by López-Bellido and López-Bellido (2001) and López-Bellido et al. (2005), under similar Mediterranean conditions. This confirms the major contribution of high levels of residual soil N to grain yield, due to the accumulation of N fertilizer over time, as indicated earlier.
|
Nitrogen utilization efficiency was also influenced by year, though the behavior of this index in response to variations in timing of N fertilizer was erratic (Table 3). Like NUpE, the highest value for NUtE was recorded with the zero N treatment, a finding also reported by Delogu et al. (1998) and López-Bellido et al. (2005) for bread wheat.
Nitrogen harvest index was significantly affected by year; the highest values were recorded in 2000–2001, the year with the lowest biomass and grain yield (Table 3). Nitrogen fertilizer timing also had an appreciable effect on HI (Table 3), values generally being significantly higher when N was applied at stem elongation; however, similar values were recorded for some treatments involving a final application at tillering (Table 3). Ehdaie and Waines (2001) report very similar values for NHI in durum wheat, ranging between 63 and 71%, depending on sowing date.
Finally, the apparent N recovery fraction (NRF) showed—depending on the study year—a direct correlation with grain yield (Table 3). Although it was significantly affected by N fertilizer timing, the effect of the various treatments was neither clear nor consistent (Table 3). However, mean values were higher than those reported by Ichir et al. (2003a) for rainfed Mediterranean conditions (33.1%). The year x N timing interaction was highly significant for all efficiency indices (Table 3).
Recovery of Labeled Nitrogen Fertilizer
A significant difference in recovery of labeled N fertilizer was observed between fall applications (before planting) and topdressing at stem elongation (Fig. 4
). The percentage recovery of NR was 19.4, 11.2, and 7.6% for rates of 150, 75, and 50 kg N ha–1, respectively, applied at planting (Fig. 4). Application at stem elongation yielded NR rates of 49.6, 38.2, and 35.4% for application rates of 150, 100, and 75 kg N ha–1, respectively (Fig. 4). Mean NR for fall applications was 12.7% compared with 41.1% for applications as topdressing (at stem elongation) (i.e., more than a threefold difference). Ichir et al. (2003b) also obtained in durum wheat a recovery rate of 56% with 15N at tillering compared with 28% with 15N just after planting. A number of studies addressing recovery of 15N labeled fertilizer in temperate climates report that late N application in wheat improves the uptake and efficiency of fertilizer N compared with early application before fall planting (Wuest and Cassman, 1992; Sowers et al., 1994; Tran and Tremblay, 2000; Melaj et al., 2003).
|
The higher efficiency of high N applications shown in Fig. 2 may be due to greater concentration of fertilizer in the root area at higher 15N rates, thus stimulating development of a larger and more effective root system for the recovery of soil N, rather than to the effect of mineralization–immobilization turnover suggested by Strong (1995), due to which losses of labeled N through immobilization would be proportionately greater at lower 15N fertilizer rates. The low soil C/N ratio (Table 1) and the habitual existence of large amounts of available N in the system suggest that differences in efficiency at different labeled N rates may be attributable to the former hypothesis.
Native soil N contribution to total plant N was substantial over the 3-yr study as a whole. Soil N comprised 80.4% of total plant N for 150 kg N ha–1 all-fall application. The soil N contribution to total N uptake in the split N treatments (planting + stem elongation) decreased on average to 57.6%. Soil N uptake accounted for 56.3% total N uptake by the whole plant for 150 kg N ha–1 all-topdressing application (stem elongation). The large contribution of soil N to plant N accumulation is related to the high residual levels of preplanting inorganic N and to mineralization, estimated to provide (on average over the 3 yr) roughly the same amount of N as the 150 kg N ha–1 rate applied as fertilizer.
Nitrogen fertilizer recovery varied from among years, although differences were significant only for grain 15N recovery (Table 4). There was also a yearly variation in N fertilizer recovered by the grain as a proportion of total plant recovery. The highest recovery occurred in 2000–2001 (83.8%), the year of lowest biomass and grain yield, while recovery for the other 2 yr was significantly lower: 64.2 and 59.6% in 1999–2000 and 2001–2002, respectively (Table 4). The amount of N accumulated in the grain as a proportion of total N recovery by the whole plant was similar for N-labeled and standard N fertilizers (mean 69.4 and 70.6%, respectively).
|
Most of the increase in aboveground N derived from fertilizer was accounted for in the grain. Split application and application at stem elongation resulted in a higher percentage of grain N being derived from fertilizer, with the remainder from soil N (Table 4). However, in the all-fall application, the percentage of grain N derived from fertilizer was much lower, 46% of the mean recorded both for split treatments and for stem elongation application (Table 4).
Native N was the largest component of grain N for all treatments, and particularly when N fertilizer was applied only in the fall (Table 4). This underlines the need to take into account residual N and N mineralization when predicting crop N fertilizer requirements.
Estimates of the amount of fertilizer N recovered by wheat using the difference method (NRF) were higher than those obtained using the 15N recovery method. The NRF was 44%, compared with 42.1% for labeled N recovery (Tables 3 and 4). The difference was more marked when N fertilizer was applied only at planting. Strong (1995) reports similar differences between methods, attributing them to the effects of "added N interaction" and "mineralization–immobilization turnover," which lead to overestimation and underestimation, respectively, of fertilizer efficiencies.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
Nitrogen fertilizer recovery in durum wheat increased when fertilizer was applied as a topdressing before stem elongation, whereas application of N in the fall, before planting, yielded poor recovery.
Use of 15N labeled fertilizer suggests that in rainfed Mediterranean Vertisols, fertilizer N should be applied mainly as a topdressing, between tillering and stem elongation. This might even allow total N fertilizer rates to be reduced to <150 kg N ha–1, given the importance of soil residual and mineralization when predicting crop requirements.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. L. Mulvaney, S. A. Khan, and T. R. Ellsworth Synthetic Nitrogen Fertilizers Deplete Soil Nitrogen: A Global Dilemma for Sustainable Cereal Production J. Environ. Qual., October 29, 2009; 38(6): 2295 - 2314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Takahashi, M. R. Anwar, and S. G. de Vera Effects of Compost and Nitrogen Fertilizer on Wheat Nitrogen Use in Japanese Soils Agron. J., June 26, 2007; 99(4): 1151 - 1157. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||
