Long-term effects of crop rotation and different fertilization systems on soil fertility and productivity

Abstract

The effects of crop rotation and fertilization systems on yield and soil fertility parameters have been investigated in a long-term field trial established in southeast Norway in 1953. The results indicate the small differences between crop rotations and different fertilization systems in yield and soil fertility parameters; the decreasing trend in soil organic carbon (SOC) (from 3.8 to 3.7%) and increasing trend of N with time (from 0.32 to 0.36%) and, as a result, the decreasing trend in C/N ratio (from 12 to 10).

Keywords:

• long-term field trial
• SOC
• yield

Introduction

Although yield is often the only true measure of sustainability, because the plant itself integrates all factors including soil, climate, pests and diseases, which affect growth (Johnston, 1994), assessing the sustainability of agricultural systems unavoidably integrates monitoring soil fertility as an important quality component of agriculture (Kirchmann & Thorvaldsson, 2000). Maintaining and improving soil quality is crucial if agricultural productivity and environmental quality are to be sustained for future generations (Reeves, 1997). However, soil organic carbon (SOC) is most often reported as an attribute from long-term studies, and is selected as the most important indicator of soil quality and agronomic sustainability because of its impact on other physical, chemical and biological indicators of soil quality (Reeves, 1997). Different studies have shown that even within a crop rotation and with manure application, continuous cropping results in a decline in SOC, although the rate and magnitude of the decline is affected by cropping and tillage system, climate and soil type (Reeves, 1997; Kucharik et al., 2001; Mattsson, 2001; Albert, 2001). Therefore, the aim of a long-term field trial established in Ås, southeast Norway, in 1953 was to investigate changes in sustainability of agricultural systems under all-arable cropping compared with rotations including grassland under Norwegian conditions. In this study, we have investigated the effects of crop rotation and different fertilization systems on yield and soil fertility parameters and also their effects on changes of SOC within a period of 50 years.

Materials and methods

Experimental field

An experiment with four six-course rotations and four fertilizer treatments was established in Ås, Norway in 1953 on a clay loam soil, fluvaquentic humaquept (Soil Survey Staff, 1994). The soil contained about 25% clay, and 3.8% soil organic carbon (SOC). The detailed description of the experiment was given by Uhlen et al. (1994). The field was limed twice in the 1970s with 4 tonnes limestone per ha. Use of calcium nitrate as the main N source has also contributed to a relatively low lime requirement during the experimental period.

The experimental treatments have been subject to some changes during the 48 years; fertilizer rates have been increased in accordance with the common trends in agriculture, and also the crop sequences have been adjusted. The design includes the following crop rotations:

i.	Continuous cereal cropping (6 years)
ii.	Cereal crops (3 years)+row crops (3 years)
iii.	Cereal crops (4 years)+ley (2 years)
iv.	Cereal crops (2 years)+ley (4 years)

From 1978 the rotation plan shown in Table 1 was used.

Table 1. Rotation plan in long-term field trial

	Year
Rotation	1	2	3	4	5	6
I	Barley	Oats	Spring wheat	Oats	Barley	Spring wheat
II	Barley	Rape	Spring wheat	Oats	Potato	Rape
III	1^st year ley^a	2^nd year ley^a	Spring wheat	Oats	Barley	Spring wheat^b
IV	1^st year ley^a	2^nd year ley^a	3^rd year ley^a	4^th year ley^a	Barley	Spring wheat^b
^a Timothy and red clover (3:1).
^b Timothy and red clover seeded in spring wheat.

All crops were grown every year in two replicates, making a total of 48 rotation plots, which were located in six blocks of four plots. Each rotation plot was divided into four subplots, which were subjected to the following fertilization treatments:

a.	Low rates of NPK in mineral fertilizers (30–40 kg N, 15–25 kg P, and 30–120 kg K ha−1)
b.	Normal rates of NPK (80–120 kg N, 25–35 kg P, and 60–180 kg K ha−1)
c.	FYM (farmyard manure)+NPK to supplement treatment b.
d.	FYM+NPK to supplement to treatment b+additional NPK for grassland for rotations III and IV.

Farmyard manure (60 t ha⁻¹) was incorporated in spring by harrowing before planting, the first year in every 6-years period. After 1980 the manure was applied in the form of slurry. It contained approximately: 3.9, 1.8, 0.7, 3.4, 0.5, 1.7, 1.3, 0.6, 0.4 kg of N total, NH₄-N, total P, K, Ca, Mg, S, Na and Cl, respectively, per tonne manure. From 1992 all cereal straw has been ploughed in annually.

Soil analysis

After 31 and 48 years (in 1984 and 2001, respectively), soil samples (40) were taken from all treatments and both replications under barley and ley (0–20 cm). Soil samples were air-dried, and mildly crushed to pass a 2-mm sieve. Soil pH, content of total C and N, and extractable P, K, Ca, Mg and Na were determined.

Soil pH was measured in a soil-water suspension (1: 2.5). Total C and N content were determined using a LECO CHN 1000 instrument with IR detector, and thermal conductivity, respectively (Nelson & Sommers, 1982). In this study the total C was assumed to be organic C (SOC), because of the low pH of the soil. In the period from 1953 to 1984, soil organic C and total N were determined by a dry combustion method and Kjeldahl method, respectively (Uhlen & Tveitnes, 1995). Although there was a difference in SOC and total N contents obtained by these two methods, the effects of the different treatment combinations on C and N content in the soil were generally similar in 1984 and 2001 (Uhlen & Tveitnes, 1995).

The plant available P, K, Ca, Mg and Na were extracted with AL solution (0.1 M ammonium lactate and 0.4 M acetic acid, pH 3.75) at a soil to solution ratio of 1: 20 (w/v) (Egnér et al., 1960). The concentration of P was measured by spectrophotometer, while concentrations of K, Ca, Mg and Na were measured by atomic absorption spectrophotometry (AAS). All analyses were replicated twice.

Statistical analysis

Analysis of variance (ANOVA) was performed in a two-factorial design by SPSS statistical computer programme for soil (pH, SOC, N total, P-AL, K-AL, Ca-AL, Mg-AL and Na-AL), and the yield of barley and ley.

Results and discussion

Effects of crop rotation and fertilization on yields of barley and ley

Effects of rotation and fertilization systems on the yields of barley and ley are shown in Fig. 1. The analysis of the effects of these factors on the yield of all crops in the experiment was given by Uhlen et al. (1994).

Fig. 1. Effects of crop rotation and fertilization systems on average barley and ley DM yields (1992–2000) in long-term field experiment. Error bars show standard deviation. Within a treatment (rotation or fertilization) means followed by the same letter are not significantly different at P<0.05.

During the period from 1992 to 2000, only the small differences in yields between treatments were measured (Fig. 1). The lowest yield of barley (3.52 t ha⁻¹) was obtained in the rotation with all cereals (rotation I), the highest (4.00 t ha⁻¹) in the rotation with a higher proportion of ley (IV), while there were no significant differences between rotations II and III. Fertilization also affected the yield, but to a lesser extent. Therefore, lower yield was associated with low NPK rate (3.51 t ha⁻¹), while there were no significant differences between other treatments. In this trial, fertilization was adjusted in order to provide the same amount of nutrients with mineral fertilizer treatment (b) and FYM (c), while higher amounts of nutrients were applied only with treatment FYM+additional NPK for ley (d). The treatments with FYM gave lower yields than the standard NPK treatment, reflecting the lower percentage of N utilization from FYM compared with the standard NPK treatment. Uhlen & Tveitnes (1995) calculated utilization rate of N added in mineral fertilizers and FYM for the trial. The results demonstrated that utilization of N from FYM was only 40–45% that of mineral fertilizer, even after including all residual effects during 37 years (1954–1990). After 1980 FYM was applied in the form of slurry, the quality of which can be very variable in some years (Ekeberg & Riley, 1995), and this also can affect the yield.

Effect of rotation on yield of ley was more pronounced than effect of fertilization. Thus, higher yield was observed in the rotation with 4 years ley (6.09 t ha⁻¹) than 2 years ley (5.37 t ha⁻¹), while there were no significant differences between mineral fertilizers and FYM treatments (Fig. 1).

Effects of crop rotation and fertilization on soil fertility parameters in the long-term field experiment after 48 years

Effects of long-term crop rotation and different fertilization systems on soil properties are shown in Table 2. The fertilization systems did not affect soil pH in the topsoil (0–20 cm), but an impact of crop rotation on pH was evident. Significantly higher pH value (+0.16) was measured in the rotation with cereal cropping (rotation I) compared with other rotations (II–IV), where pH value did not differ significantly. These results can be explained by the fact that there was a lower yield (Fig. 1) and correspondingly lower uptake and removal of base cations in the rotation with cereal cropping (rotation I) than in the other rotations (II–IV). Also, since from 1992 all cereal straw has been ploughed in annually, the effects of cereals on soil pH have decreased; the content of base cations is far higher in cereal straw than grain (Uhlen & Tveitnes, 1995), and consequently the lower amount of cations is removed from the soil when cereal residues are ploughed in.

Table 2. Effects of crop rotation and fertilization systems on pH, content of SOC (%), N total (%), C/N ratio, P-AL, K-AL, Ca-AL, Mg-AL and Na-AL (mg 100⁻¹ g) in topsoil (0–20 cm) after 48 years in long-term field trial. Within a treatment (rotation or fertilization) means and standard deviation followed by the same letter in the columns are not significantly different at P<0.05

	Ph	C %	N %	C/N	P-AL mg 100^−1 g	K-AL mg 100^−1 g	Ca-AL mg 100^−1 g	Mg-AL mg 100^−1 g	Na-AL mg 100^−1 g
Crop rotations
I	5.90±0.11 a	3.25±0.39 c	0.32±0.04 c	10.33±0.18 a	9.92±1.16 a	25.32±3.79 a	175±9 b	4.78±0.83 a	1.61±0.37 ab
II	5.74±0.16 b	3.03±0.30 c	0.29±0.03 c	10.37±0.14 a	10.42±1.25 a	22.87±2.24 b	155±10 c	4.50±0.50 a	1.57±0.29 b
III	5.73±0.10 b	3.71±0.37 b	0.37±0.03 b	10.10±0.34 a	10.30±2.01 a	16.68±7.64 b	169±9 b	5.00±1.02 a	1.85±0.21 a
IV	5.75±0.11 b	3.94±0.37 a	0.39±0.03 a	10.16±0.30 a	10.73±2.66 a	11.65±3.97 c	188±4 a	5.66±1.49 a	2.12±0.26 a
Fertilization systems
a	5.71±0.15 a	3.39±0.47 b	0.33±0.05 b	10.20±0.29 a	7.94±1.37 c	14.64±6.30 a	167±16 b	4.54±1.06 b	1.81±0.28 a
b	5.74±0.12 a	3.51±0.54 b	0.34±0.06 b	10.18±0.28 a	11.15±1.44 b	18.48±6.84 a	176±19 b	4.76±1.03 b	1.82±0.34 a
c	5.77±0.10 a	3.49±0.41 b	0.34±0.04 b	10.24±0.26 a	10.66±1.19 b	17.06±8.12 a	180±18 ab	6.18±1.38 a	2.00±0.39 a
d	5.75±0.13 a	3.64±0.45 a	0.36±0.04 a	10.13±0.38 a	12.52±2.20 a	13.47±5.70 a	191±19 a	5.71±0.88 a	2.13±0.28 a

In this experiment, there was no significantly positive effect of FYM on pH value (treatments c and d). Similar effects were also noted in a long-term experiment in Bad Lauchstädt (Germany), after 96 years of applying 10 t FYM annually (Körschens & Pfefferkorn, 2001). In contrast to these results, the use of FYM (40 t ha⁻¹ once in four years) during 17 years slightly improved the pH (+0.07) in five long-term trials in the Czech Republic (Baier et al., 2001). In the same field trial, application of mineral fertilizers and FYM had a negative influence on pH (−0.13) (Baier et al., 2001), which can be explained by higher removal of the basic cations from the soil with high yield.

The soil organic C (SOC) was influenced by both rotation and fertilization (Table 2). Thus, higher SOC content was found in topsoil under rotations with ley crops compared to cereal rotations (I and II). The quantity of crop residues returned to the soil is a major factor affecting SOC. Assuming the amounts of crop residues were proportional to grain yield, the quantity of the crop residues was higher with ley rotations than that of all- cereal rotations (Uhlen, 1991). In this experiment, higher SOC content was observed in topsoil with FYM treatments (c and d) than with mineral fertilization treatments (a and b), but only FYM treatment with additional fertilization for ley (d) had a significantly higher SOC content (Table 2). Many long-term experiments have affirmed that SOC is increased by the application of both mineral fertilizer and manure (Kubat et al., 2001; Walther et al., 2001). Albert (2001) found that FYM increased SOC to a higher level as compared to mineral fertilizer. Besides supplying nutrients, FYM is also a source of organic C, and therefore adding manure is considered more effective in increasing SOC content than inorganic fertilizers (Singh et al., 1997; Uhlen, 1991). Results from an investigation by Yang et al. (2004) dealing with fractionation of organic matter in the topsoil of a field trial show that FYM not only increases SOC but also leads to a higher C concentration in humin and humic acid (HA) as compared to inorganic fertilizers. Crop rotation with ley gave a significantly higher C proportion in HA and humin, and relatively lower C proportion in fulvic acid (FA) compared with the rotation of all cereals. However, ley rotation resulted in a significantly higher black C (BC) content than that observed with cereals alone. Black carbon is considered as highly resistant to decomposition (Skjemstad et al., 1996).

In this investigation, total N content in the topsoil was affected by both rotation and fertilization (Table 2). Like SOC, a higher N content was found in topsoil under rotations with ley crops (0.37% N, rotation III; and 0.39% N, rotation IV) compared to all-cereal rotations (0.32% N, rotation I; and 0.29% N, rotation II). Since ley crops, in addition to timothy, contained clover in the first and second year, these results can partly be explained by both symbiotic and non-symbiotic N fixation (Soon & Arshad, 1996), but also by the fact that mineralization of N is retarded in grassland compared to arable land (Mattsson, 2001). Rotation with row crops+cereal crops is the most C- and N- consuming system in this experiment (Table 2). According to Uhlen (1991) and Uhlen et al. (1994), barley following row crops gave higher yields than barley in all-cereal rotations.

In this experiment, fertilizer application was adjusted in order to provide the same quantity of nutrients with mineral fertilizer treatment (b) and FYM (c), while higher quantities of nutrients were applied only with treatment FYM+additional NPK for ley (d). However, significantly higher content of N (0.36% N) was found only with treatment FYM+additional NPK (d) compared with the other treatments (a, b and c).

Effects of crop rotations and fertilization systems on the C/N ratio in this trial were not significant (Table 2), but there is an obvious trend showing that the C/N ratio is higher in rotations with cereals (10.33, rotation I; and 10.37, rotation II), compared to ley rotations (10.10, rotation III; and 10.16, rotation IV). The slightly higher C/N ratio in rotations with cereals (I and II) can be explained by ploughing-in cereal straw annually, which started in 1992. As cereal straw has a C/N ratio about 10 times that of soil organic matter (SOM), N might be taken from other sources for the straw breakdown in the soil. The results of the straw experiment at Ås (1961–84) and Øsaker (1963–84) where straw was ploughed in for more than 20 years indicates a higher C/N ratio after straw treatment (Uhlen, 1991).

The P-AL was affected by both rotation and fertilization, but effect of rotation was not significant (Table 2). Lower content of P-AL (9.9 mg P 100⁻¹ g) was observed in treatment with all-cereals (rotation I) compared to other rotations (10.3–10.7 mg P 100⁻¹ g). This decrease of P-AL can be attributed to labile P either leaching from the plough layer or being combined or fixed in less readily available forms (Blake et al., 2000). As expected, the lowest content of P-AL (7.9 mg P 100⁻¹ g) was found in the treatment with low rate of NPK (a), while the highest content (12.5 mg P 100⁻¹ g) was found in the FYM+additional NPK treatment (d). Since the same quantities of P and K were applied either in FYM+NPK (c) or in mineral fertilizers only (b), there were no differences in P-AL between treatments with a normal rate of NPK (b) and FYM (c).

Crop rotation affected K-AL content, in contrast to the fertilization treatments (Table 2). The lowest K-AL (11.6 mg K 100⁻¹ g) was observed in rotation with ley (IV), while the highest content (25.3 mg K 100⁻¹ g) was measured in the all-cereal rotation (I). It is well known that plant species differ in their nutrient requirements (Marschner, 1998). Among them, a ley is characterized not only by high K consumption but also by high ability to exploit K sources in the soil (Lanyou & Smith, 1985). The results of the soil analysis are in good agreement with the K balance calculated in 2001 for barley and ley (not shown). K balance was positive in all rotations for barley (I–IV), but it was negative for both ley rotations (III and IV).

The soil content of AL-extractable cations was influenced by both rotation and fertilization. Therefore, significantly higher Ca-AL, Mg-AL and Na-AL were measured in rotations with ley crops (III and IV) compared with cereal rotations (I and II). The content of SOC was higher in the rotation with ley crops, commonly seen as a result of a large quantity of crop residues retained in soil under ley cultivation, resulting in an increase in the cation exchange capacity (CEC) of the soil under ley cropping (Uhlen & Tveitnes, 1995). FYM application increased the content of cations (Ca, Mg and Na) in soil, partly by supplying these nutrients and partly by increasing the SOC and consequently increasing the CEC. The average amounts of Ca, Mg and Na applied per tonne of manure per year were: 1.3 kg Ca, 0.6 kg Mg and 0.4 kg Na (Uhlen et al., 1994).

Changes of soil properties during the last 17 years (after 31 and 48 years)

During the last 17 years, soil reaction (pH) decreased significantly from 5.82 to 5.74 (Fig. 2). The trial was limed twice in the 1970s, but has not been limed since. However, lime status was maintained when Ca(NO₃)₂ was used as the only N source until 1993. Therefore, pH-values decreased not only as the result of an effect of leaching of the base cations from the acid soil, but also of their removal by the growing plants.

Fig. 2. Changes of pH, content of SOC (%), N total (%), C/N ratio, P-AL, K-AL, Mg-AL, Na-AL (mg 100⁻¹ g) and Ca-AL (mg 10⁻¹ g), during the last 17 years (after 31 and 48 years) in long-term field trial. Error bars show standard error mean. Within the same indices means followed by the same letter are not significantly different at P<0.05.

The SOC and N were affected by both rotation and fertilization (Fig. 3). Thus, lower content of SOC in 2001 compared to 1984 was observed with cereal rotation I (cereal) and II (cereal+row crops), while there were no differences in SOC content between ley rotations III and IV (2- and 4-years ley rotations). However, if the mean value of SOC for the whole trial in 1984 (3.77%) is compared with 2001 (3.67%), it can be seen that the content of SOC decreased, but there were no significant differences (Fig. 2). The effect of treatments on the content of N was more pronounced than the same effect on SOC (Fig. 3). During this period, on average for the whole trial, treatments increased N total in the soil from 0.31% in 1984 to 0.36% in 2001 (Fig. 2). The highest increase in N total content was observed with the 4 years ley rotation (IV), compared with 2 years ley (III), while there were no significant differences between cereal rotations I and II.

Fig. 3. Long-term effects of crop rotation and fertilization on SOC and N total content in topsoil (0–20 cm) in long-term field trial. Error bars show standard deviation. Within a treatment (rotation or fertilization) means followed by the same letter are not significantly different at P<0.05.

As a result of the decreasing trend in SOC (from 3.85 to 3.7%) and increasing trend of N with time (from 0.31% to 0.36%), the C/N ratio decreased from 12.0 in 1984, to 10.2 in 2001 (Fig. 2).

Despite the positive balance of P in 1984 (Uhlen & Tveitnes, 1995) and also 2001 (not shown), average content of P-AL in 2001 (10.5 mg P 100⁻¹ g) was significantly lower than in 1984 (12.9 mg P 100⁻¹ g) (Fig. 2). This decrease of P-AL can be attributed to labile P being combined or fixed in less readily available forms (Blake et al., 2000). In this long-term experiment, over the experimental period, decrease of P-AL was observed in cereal rotations (−3.6 and −2.3 kg ha⁻¹ y⁻¹, rotations I and II, respectively), while in ley rotations, P-AL increased (+0.3 and +0.4 kg ha⁻¹ y⁻¹ rotation III, and +1.6 and +1.8 rotation IV). A possible explanation may be that an increase in absorptive and P combining organic surfaces by ley acts as a buffer resulting in a sustainable source of plant available P, with reduced fixation and leaching. Recent research considering three long-term field experiments at Rothamsted (UK), Bad Lauchstaedt (Germany) and Skierniewice (Poland) shows that the most efficient utilization of P was from soils treated with FYM; specifically, in soils only treated with superphosphate with low SOC content, a higher quantity of P being either leached or fixed (Blake et al., 2000).

Contrary to the negative balance in ley rotations (not shown), the content of K-AL increased during this period (Fig. 2). Clay soil contains large amounts of K, but only a small proportion, usually less than 1% of the total content, are in the soil solution (Sparks & Huang, 1981; Beringer, 1985). When exchangeable K is removed from the soil, the release of readily available non-exchangeable K may be quite rapid (Doll & Lucas, 1973). In the case when soil solution concentration is very low, it is likely that some absorbed K associated with reserve fractions will be released. Investigations by Øgaard (2000) and Øgaard et al. (2002) in 20 field experiments in Norway showed that the supply of K from the soil to grass is considerable, even at the highest level of K fertilization. The supply from easily releasable K (AL-extractable K) was reduced from year to year, whereas the supply from non-exchangeable K was increased during the period.

The contents of AL extractable Ca, Mg and Na were significantly lower in 2001 compared with 1984 (Fig. 2).

The changes of SOC content during 48 years

A high proportion of ley (4 years ley+2 years cereal crops) in rotation IV led to increase in the content of SOC (Fig. 4). The content of SOC in rotation IV was increased with 2925 kg C ha⁻¹ over 48 years, assuming a bulk density of 1.125 g cm⁻³ in the 0–20 cm layer. Average increase of the SOC content in rotation IV was about 60 kg SOC ha⁻¹ y⁻¹. In contrast, the content of SOC in the rotation with arable cropping (I) was decreased by 11,925 kg C ha⁻¹, or about 250 kg SOC ha⁻¹ y⁻¹, during the same period. A higher content of SOC in rotation IV (4 years ley+2 years cereal crops) in relation to rotation I (all cereal crops) was caused not only by the fact that amount of crop residues was higher with ley rotation than with all cereal crops rotation, but also by slower breakdown rate under the ley crops compared to arable cropping (Uhlen, 1991; Singh et al., 1997).

Fig. 4. Long-term effects of crop rotation and fertilization systems on the content of SOC in topsoil (0–20 cm) in long-term field trial.

However, the level of the SOC can be managed not only by crop rotation, but also by the use of manure and mineral fertilizers (Walther et al., 2001; Albert, 2001). Up to 1984 the use of FYM influenced the content of SOC in the soil positively (Uhlen, 1991). FYM used during 1953–1979 contained 20% DM and about 4.0 kg N t⁻¹. From 1980 slurry (liquid manure) was applied, and then no effect on SOC was observed (Fig. 4). The slurry contained on average 9% DM and 3.9 kg N t⁻¹. In 2001 only the treatment with FYM+additional fertilization for grassland gave significantly higher content of SOC compared to the other treatments (Fig. 4).

The changes of total N content during 48 years

The content of N in the topsoil in 2001 was 0.36%, while when the experiment was established in 1953, the N content was 0.31% (Fig. 5). There were significant increases of N in the rotation systems including high (1800 kg N ha⁻¹ in 48 years, or 37.5 kg N ha⁻¹ y⁻¹) and low (1463 kg N ha⁻¹ in 48 years, or 30.5 kg N ha⁻¹ y⁻¹) proportion of ley, whereas the N content in arable rotations (I and II) was almost unchanged (Fig. 5). The increase of the content of N in topsoil in the rotation with perennial grass was assumed not to be caused only by higher root biomass, but also by slower breakdown rate of crop residues and soil organic matter, and lower leaching losses compared to cereal rotations. Therefore, in field lysimeters with the same soil, leaching losses were 20 kg N ha⁻¹ y⁻¹ higher in he all cereal rotation as compared to perennial grassland in 1974–1981 (Uhlen, 1989). According to Pouder et al. (2002) mineralization rate was 100% lower in organic farming systems (with cover crops+FYM) than in conventional systems. The lower mineralization rate in the cover crop based farming system is attributed to differences in the quality of the SOM, which is linked to chemical stabilization and physical protection of the labile pool (Ladd et al., 1985). The lower mineralization rate not only corresponds with a greater accumulation of N in the system, but also includes a reduced risk for N leaching and groundwater pollution (Clark et al., 1998).

Fig. 5. Long-term effects of crop rotation and fertilization systems on the content of N total in topsoil (0–20 cm) in long-term field trial.

As expected, as an effect of high yields and crop residues, long-term fertilization with mineral fertilizers and manure affected the content of N in the topsoil (Table 2). However, a significantly greater increase in the content of N in the topsoil was obtained with FYM+additional NPK (d) for grassland compared to the other treatments (Table 2). Up to 1984 the use of manure managed the content of N in the soil. Since 1984 no positive effects of manure were measured as a result of using the manure in the form of slurry (Fig. 5).

The changes of C/N ratio during 48 years

According to Fig. 6, C/N ratio decreased from 12.1 in 1953 to 10.2 in 2001, whereas no clear differences were found between the treatments in 2001 (Table 2). However, when plant residues enter the soil, as a result of decomposition, C is released at an accelerated rate into the atmosphere. Mineralized N released from the same source in conditions of a high C/N ratio, is more likely to be immobilized in new organic matter than leached or lost as different gaseous compounds (Körschens, 1998). This suggests, therefore, that in spite of C and N produced in the breakdown process, the amount of residual organic C and N left in the soil after some decades may be somewhat different (Fig. 6). Greater accumulation of N than C after the first breakdown cycles of the organic matter led to decrease in the C/N ratio in the soil, which indicates that organic matter decomposition is reduced, resulting in a significant accumulation of stable organic C in the soil (Feng & Li, 2001). The fact that the C/N ratio of humus, the most stable form of organic C in soil, is often close to 10, as well as the C/N ratio in the ley rotations in this experiment (III and IV), can indicate that new equilibrium has been reached. A slightly higher C/N ratio in cereal rotations (I and II) can be attributed to ploughing-in cereal straw annually, which has been done since 1992.

Fig. 6. Long-term effect of crop rotations and fertilization systems on C/N ratio in topsoil (0–20 cm) in long-term field experiment.

The results show: 1) the small differences between crop rotations and different fertilization systems in yield and soil fertility parameters; 2) the decreasing trend in SOC and increasing trend of N with time; and 3) as a result, the decreasing trend in C/N ratio, which indicate that the quality of organic matter has been improved with time. Based on the above, it cannot be concluded that the sustainability of the systems applied have been depreciated during the 50 years period. Although the tendency towards declining SOC content over the years has been pronounced with all cereal rotation, the content of SOC is still satisfactory. Crop rotations with ley, as well as application of mineral fertilizers gave the highest yield and increased SOC and N content the most.

Acknowledgments

This research was supported financially by the Research Council of Norway. We are grateful for the financial support.

Additional information

Notes on contributors

Maja Cuvardic

Cuvardic, M., Tveitnes, S., Krogstad, T. and Lombnæs, P. (Faculty of Agriculture, University of Novi Sad, Trg D. Obradovica 8, SCG-21000 Novi Sad, Serbia and Montenegro and Department of Plant and Environmental Sciences, Agricultural University of Norway, P.O. Box 5003, NO-1432 Ås, Norway). Long-term effects of crop rotation and different fertilization systems on soil fertility and productivity.

Notes

Cuvardic, M., Tveitnes, S., Krogstad, T. and Lombnæs, P. (Faculty of Agriculture, University of Novi Sad, Trg D. Obradovica 8, SCG-21000 Novi Sad, Serbia and Montenegro and Department of Plant and Environmental Sciences, Agricultural University of Norway, P.O. Box 5003, NO-1432 Ås, Norway). Long-term effects of crop rotation and different fertilization systems on soil fertility and productivity.