Nitrogen fertilization of winter wheat in spring in Finland : comparison of calcium ammonium nitrate and calcium nitrate , and the effect of split application

The effectiveness of calcium nitrate (CN) and calcium ammonium nitrate (CAN) as N sources for winter wheat in spring was studied in seven field trials in Southern Finland, in the province of Uusimaa. The experiments were carried out on clay soils in 1986—1989. Nitrogen dressings of 120—150 kg ha-1 were broadcasted either as a single application or split into two applications. The two applications were in most cases 30—40 and 90 —110 kg ha-'. The single or the first split application was given in the beginning of the growing season and the latter application at the end of tillering. The grain yields obtained by a single N application of 140—150 kg ha-1 ranged from 2 610 to 7 550 kg ha~' and the protein content of grains from 9.4 to 14.5 %. The nitrogen form of the fertilizer had no significant effect on the grain yield and on the protein content of grains. Splitting the nitrogen increased the grain yield by 400 kg ha~' and the protein content of grains by 0.8 percentage points. Index words: Calcium ammonium nitrate, calcium nitrate, fertilization, nitrogen, split application, winter wheat


Introduction
Calcium ammonium nitrate (CAN) is the most commonly used straight nitrogen fertil- izer in Finland.An amount of 79 000 tons of CAN was sold in Finland during the period July 1, 1988 to June 30, 1989 (Anon. 1989).Calcium nitrate (CN) is today of minor im- portance, 3 000 tons being used annually.The nitrogen of CN is in nitrate form, whereas CAN contains nitrogen in nitrate and in am- monium form in equal proportions.
There are only few papers comparing the effect of CN and CAN on the grain yield of winter cereals in Finland or other Scandina- vian countries.Jaakkola (1978) compared five forms of fertilizer N in a five-year field experiment on winter wheat, winter rye and spring wheat; there was no difference in the effect of CN and CAN on the grain yield.The effect of CAN and CN on the grain yield of winter wheat has been compared also in Sweden in a series of 97 trials; CN was found to be a more effective N source than CAN in the eastern and in the western part of Sweden, but not in the southern part (Jonsson 1971).The rate of nitrogen application in both trials (Jaakkola 1978, Jonsson 1971) was lower (75 and 60 Nkg ha -1 , respectively) than used nowadays.
The N fertilizer dressing of winter cereals in spring is traditionally given in one application in Finland.Splitting the fertilizer dress- ing into two applications is a common prac- tice in Sweden, where the climatic conditions and the length of the growing period are equal to those in Finland.A low application of N is given in the beginning of the growing sea- son and the rest of the nitrogen at the tillering stage, if growth is exuberant.If the vegeta- tion is scanty, the higher dose is given first.
The aim of this study was to compare CN and CAN in spring fertilization of winter wheat.The effect of a split fertilizer application was also investigated.

Materials and methods
The seven field experiments reported in this paper were carried out in 1986-1989 in Southern Finland, in the province of Uusimaa.The experimental crop in all trials was the winter wheat cultivar "Aura".The quantity of seeds sown was 220-250 kg ha -1 .Plant-available nutrient resources of the soils were determined before starting the trials by acid ammonium acetate extraction (Vuorinen and Mäkitie 1955).At the time of sowing in September the experimental fields were given a NPK fertilizer application according to their fertility.The N application in autumn ranged from 48 to 110 kg ha" 1 .Details of the ex- perimental fields and their autumn fertiliza- tion are given in Table 1.
In spring the experimental fields were given a N dressing as calcium nitrate (15.5 % N) or calcium ammonium nitrate (27.5 °/o N).The nitrogen dressing was broadcasted on the field either as one application or split into two ap- plications.The first application was given at the beginning of the growing period in late April or in early May.The second application was given at the end of the tillering stage in late May.The rates of fertilizer dressings (N kg ha -1 ) were as follows: The field of trial 1 was divided into four blocks, each comprising six plots of 60 m 2 .
Trial 3 consisted of two blocks and trials 2, 4,5, 6 and 7 of three blocks, each comprising 8 plots of 30 m 2.
After harvest the grain yields were weighed, the moisture content was determined and the grain yields per hectare were calculated at a moisture level of 15 Vo.The replicate samples were combined for protein analysis, and the analysis was carried out in duplicate.The grain yield results were tested statistically using the analysis of variance, which was made separately for each trial.The paired t-test was applied for the data of all trials to test the ef- fect of the N form of the fertilizer, split ap- plication and rate of N dressing on the pro- tein content of grains.

Results and discussion
The average yields of winter wheat when using CAN and CN were 5 480 kg ha -' and 5 380 kg ha -1 , respectively (Table 3).The low yields of 1987 and 1988, ranging from 2 430 to 3 590 kg ha ', were mainly due to the low average temperature in 1987 and the dryness in 1988.The Summers of 1986 and 1989 were beneficial for winter wheat and the yields were relatively high, ranging from 4 630 to 8 210 kg ha '.The low protein contents were typical of high grain yields and vice ver- sa.The protein content of grains ranged from 9.4 to 15.1 % (Table 2).
Raising the rate of N application from 120 to 140 kg ha -1 did not have any significant effect on the quantity of grain yields, although it decreased the average yield by 100 kg ha -1 .Increasing the N application by 20 kg ha -1 elevated the average protein content of grains by 0.5 percentage points (t = 4.6o***).The autumn fertilization showed no effect in any trial.
The grain yields obtained with CAN were in five trials slightly higher than those ob- tained with CN, but the difference was signifi- cant only in trial 7 (F= 10,27**).The type of Table 2. Grain yield (kg ha -') and protein content (%) of grains of winter wheat fertilized with CAN and CN.N fertilizer did not affect the protein content of grains.The average protein content ob- tained with CN and CAN was 12.4 %.Jaak-  kola's (1978) results of a five-year field trial also indicated that there was no significant difference in the size of grain yield or in the protein content of grains when winter wheat and winter rye were fertilized with CN or CAN.According to Esala (1989), CN and  CAN were also equal N sources as the second N dressing for spring wheat.Because of the fast-acting nitrate (Tisdale et al. 1985), CN is claimed to be a useful fertilizer for winter cereals in spring.The present results do not, however, distinguish between the effectiveness of CN and CAN.Splitting the fertilizer dressing into two ap-plications increased the grain yields by 2-25 % (Fig. 1); the difference was 400 kg ha -1 on average.The split application affected similarly the yields obtained with CN and CAN.The increase of grain yield was statisti- cally significant in trial 2 (F= 15.85**), trial 5 (F = 11.93**) and trial 7 (F = 84.25***).Split application of N increased the protein content of grains in all trials except trial 3 (Fig. 2).The maximum increase of 2.3 percentage points was obtained in trial 7. The increase of the protein content was 0.8 percentage points on average (t = 3.71**), which is noteworthy as the protein content of winter wheat is low, as compared to spring wheat.
Based on three field experiments, Varis  and Juuti (1975) reported that splitting a N dressing of 100 kg ha -1 into two applications in spring increased both the quantity and pro- tein content of grain yield of winter wheat.
The present results, which are based on more extensive material, were in agreement with their observations.Inversely, Raininko (1966)  did not obtain higher yields by giving an ad- ditional N dressing of 25 or 50 kg ha -1 to wheat at the flowering stage.The grain yield of winter rye increased in a Danish field trial when a N application of 200 or 300 kg ha -' was split in spring, while splitting a N application of 100 kg ha -1 had a negative effect (Olsen 1986).Indeed, Varis and Juuti (1975)   suggested that splitting the N application of winter cereals is beneficial only if relatively high amounts of nitrogen are used.A large amount of N in the beginning of the growing season increases the vegetative growth too much, which can disturb later growth by causing lodging, by shadowing the lower leaves and by spending the plant-available nu- trient resources of the soil in too early a growth stage.These negative effects can be prevented by splitting the N dressing, which guarantees a more steady availability of N during the growing season.
All trials reported in this paper were car- ried out on clay soils.According to the results of Danish field experiments, winter wheat can benefit even more from splitting the N dress-  ing when grown on coarse soil, because nitro- gen is more readily leached from sandy soils than from clay soils (Olsen and Larsen  1984).Furthermore, Gregersen and Hej- lesen (1985) found that higher winter wheat yields can be obtained on sandy soils if the nitrogen dressing is split into three applications instead of two.
According to the present results, a N dress- ing of 120 kg ha -1 can be considered suffi- cient for winter wheat.Raising the amount of N by 20 kg ha -1 did not further increase the grain yield, even though the protein content increased by 0.5 percentage points.Splitting the N application into two parts was a more effective way of increasing the size and the protein content of the yield than was the use of an increased single application.There was no difference in the size of the yield and the protein content of grains whether CN or CAN was used.However, the use of CAN is more economical, because the N content of CN (15.5 %) is considerably lower than that of CAN (27.5 %), resulting in higher transpor- tation, storage and broadcasting costs when using CN, as compared to CAN.Further- more, CN is extremely hygroscopic and this property makes CN a less attractive fertilizer (Tisdale et al. 1985).

Fig. 1 .
Fig. 1.Grain yields of winter wheat fertilized by single and split N dressing in seven (1 -7) field trials.

Fig. 2 .
Fig. 2. Protein content of grains of winter wheat fertilized by single and split N dressing in seven (1 -7) field trials.

Table 1 .
Location, soil type and autumn fertilization of the experimental fields.