Evaluation of nitrogen status of agricultural soils in Java, Indonesia

Abstract

To evaluate the content of nitrogen (N) fractions of agricultural soils in Java, Indonesia, in relation to soil type and land use, 46 surface soil samples, 23 from paddy and 23 from upland, were collected throughout Java to include various types of soils. Soil N was separated into four fractions according to form and availability: inorganic extractable nitrogen (Iex-N), fixed ammonium nitrogen (Ifix-N), organic mineralizable nitrogen (Omin-N) and organic stable nitrogen (Osta-N). The total-N content was determined by the dry combustion method. The Iex-N content was determined by extraction with a 2 mol L⁻¹ potassium chloride (KCl) solution and the Ifix-N content by extraction with an hydrofluoric and hydrochloric acid (HF-HCl) solution after removal of organic-N. The Omin-N content was evaluated as the potentially mineralizable N based on a long-term incubation method. The Osta-N content was calculated as the difference between the contents of total-N and the three other fractions. The total-N content was 2.06 g kg⁻¹ on average. The contents of Iex-N, Ifix-N, Omin-N and Osta-N were 25.8, 99.1, 103 and 1,832 mg kg⁻¹, respectively, and corresponded to 1.3, 4.8, 5.0 and 88.9% of the total-N. Hence, available (Iex-N and Omin-N) and stable (Ifix-N and Osta-N) fractions accounted for 6.3% and 93.7% of the total-N, respectively. Correlation analysis indicated that the contents of total-N and Osta-N had positive correlation with (Alo + 1/2Feo) as an index of amorphous minerals (p < 0.01), suggesting strong influence of volcanic materials for the accumulation of organic matter in Java soils. The content of Ifix-N had a positive correlation with nonexchangeable potassium (K) content (p < 0.01), suggesting the contribution of 2:1 clay minerals which can fix both ammonium (NH₄⁺) and K⁺ in their interlayer sites. On the contrary, Omin-N did not have any significant correlation with soil properties, implying the importance of management for the improvement of the available N level in soils, rather than intrinsic soil properties. Soil N status further showed strong topographical trends depending on the elevation where soil developed. The contents of total N, Iex-N, Ifix-N, Omin-N and Osta-N in Java soils were on average 80, 69, 90, 65 and 80% of those in Japanese soils, respectively, suggesting that the soil N level in Java was lower than that in Japan, probably due to accelerated decomposition of organic matter, especially degradable fractions, reflecting high temperature, but that the level was relatively high for tropical soils due to the effect of volcanic materials. In conclusion, these results should be taken into account for the sustainable management of soil N in agricultural fields in Java, Indonesia.

Key words:

• Availability
• Java
• nitrogen
• soil type
• tropical soil

INTRODUCTION

Nitrogen (N) is one of the most important essential nutrients for plants and N content in the soil significantly affects plant growth and crop yields (Brady and Weil, 2002). The total N content in soils is reported, on average, to be 2.0 g kg⁻¹ with a range of 0.2–5.0 g kg⁻¹ (Bowen, 1979), or 1.63 g kg⁻¹ with a range of 0.15–2.18 g kg⁻¹ (Batjes, 1996). It is well known that soil N content in the tropics is generally much lower than that in temperate areas, as a result of the accelerated decomposition of organic matter at higher temperature in the tropics. For example, Kawaguchi and Kyuma (1977) reported that the total N content of 410 soils from tropical Asia was on average 0.13 g kg⁻¹. Soil N status in the tropics is, therefore, regarded to be one of the greatest determining factors of crop production.

Java in Indonesia, located from about 6–8° in the southern latitude, is one of the main food production areas in Indonesia and nourishes more than 134 million people in 1.32 × 10⁵ km² with a population density of more than 1000 people km⁻². Information on the N status of agricultural soils in Java is therefore important to understand, as is the relatively high soil fertility level of Java which enables it to nourish so many people. So far, the total N content has been reported by several authors (Tan, 1968; Kawaguchi and Kyuma, 1977; Darmawan et al. 2006) and the content of mineralizable N in soils has been also reported (Manguiat et al. 1996; Kadono et al. 2009). However, a comprehensive evaluation of soil N fractions has not been carried out, in spite of its potential importance for the establishment of rational soil management for sustainable and productive agriculture.

Soil N has been fractionated into labile (active) and stable fractions (Motavalli and McConnell, 1998; Westerhof et al. 1998) or into various forms such as amino acid-N, amino sugar-N, ammonia-N, acid-insoluble-N and hydrolysable unknown N (Stevenson, 1996). In order to evaluate soil N status comprehensively in terms of form and availability, soil N can be also fractionated into inorganic labile-N, inorganic stable-N, organic labile-N and organic stable-N (Sano et al. 2004; Yanai et al. 2011). The objectives of this research were, therefore, (1) to evaluate N fractions of agricultural soils in Java in terms of form and availability, (2) to elucidate the relationship between N fractions and selected physicochemical properties of the soils, (3) to examine soil N fractions in relation to soil type, land use and topography and (4) to compare soil N fractions in Java with those in Japan for the elucidation of the characteristics of N status in the tropics relative to those in temperate areas.

MATERIALS AND METHODS

Soil samples

Soil sampling was carried out from February to March 2012 in Java. According to Köppen climate classification, climates of the western, central and eastern part of Java are classified as tropical rain forest climate (Af), tropical monsoon climate (Am) and tropical savannah climate (Aw), respectively. For example, the mean annual temperature and annual precipitation of Jakarta located in western Java is 26.5°C and 1821 mm, respectively.

Forty-six soil samples were collected from the surface layer (0–15 cm) of agricultural fields all over Java (Fig. 1). Sampling sites ranged from 06°16′25.0′′ to 08°06′27.2′′ in the south latitude and from 107°17′08.7′′ to 112°32′15.0′′ in the east longitude. Elevation of sampling sites ranged from 16 to 1664 m above sea level. In terms of land use, 23 samples were collected from paddy fields and 23 samples were from upland fields. These sampling numbers roughly corresponded to the relative abundance of paddy fields (3.24 × 10⁴ km²: 47%) and upland fields (3.70 × 10⁴ km²: 53%) in Java island, according to statistical data on land area by utilization in Indonesia (2005). Soil samples were collected to include various types of soils, i.e. soils investigated were classified into six soil types: Andosols, Acrisols, Fluvisols, Luvisols, Nitisols and Vertisols, according to IUSS Working Group WRB (2006). Numbers of samples for each soil group and land use are shown in Table 1. Crops cultivated were variable in the upland fields and the timings of the soil samplings were also variable in relation to the growth stage of crops in both paddy and upland fields.

Figure 1 Location of the sampling sites in Java, Indonesia.

Table 1 Number of soil samples in relation to soil group and land use

Soil group	Paddy	Upland	Total
Andosols	2	8	10
Acrisols	3	5	8
Fluvisols	8	1	9
Luvisols	6	4	10
Nitisols	2	2	4
Vertisols	2	3	5
Total	23	23	46

Aliquots of moist soil collected were immediately sieved with 4-mm mesh and used for the determination of the contents of organic mineralizable-N. The remaining moist soil samples were air-dried, sieved with 2-mm mesh, and used for the determination of the contents of total-N, inorganic extractable-N, fixed ammonium nitrogen (NH₄⁺-N) and other physicochemical properties.

Analytical methods

Evaluation of soil nitrogen fractions

Soil N was fractionated into four fractions according to form and availability as follows: inorganic extractable nitrogen (Iex-N), fixed ammonium nitrogen (Ifix-N), organic mineralizable nitrogen (Omin-N) and organic stable nitrogen (Osta-N), as described in detail by Sano et al. 2004). Among these fractions, Iex-N and Omin-N are relatively available, and Ifix-N and Osta-N are relatively stable.

1. Total-N content was determined by the dry combustion method (Sumigraph NC analyzer NC-800, Sumika Chem. Anal. Service, Osaka, Japan) using finely ground (< 0.25 mm) air-dried soil samples.

2. Iex-N content was determined as described by Mulvaney (1996). In this method, NH₄⁺-N and nitrate nitrogen (NO₃^–-N) were extracted from air-dried soil with a 2 mol L⁻¹ potassium chloride (KCl) solution at a soil:solution ratio of 1:5. The amount of NH₄⁺-N was determined by the indophenol blue method. The NO₃^–-N was reduced to nitrite nitrogen (NO₂^–-N) by passage through a column of copperized cadmium and the amount of NO₂^–-N was determined by the modified Griess-Ilosvay method.

3. Ifix-N content was determined as described by Silva and Bremner (1966). Zero point five grams of finely ground (< 0.25 mm) air-dried soil was treated with an alkaline potassium hypobromite (KOBr) solution to decompose organic fractions and washed with 30 mL of 0.25 mol L⁻¹ KCl solution three times to remove exchangeable NH₄⁺-N and organic N compounds. The residue was then digested with 10 mL of 5 mol L⁻¹ hydrofluoric acid (HF) - 1 mol L⁻¹ hydrochloric acid (HCl) solution with 24 h of reciprocal shaking, and the amount of extracted NH₄⁺-N was determined by the steam-distillation method.

4. Omin-N content was determined using a long-term aerobic incubation (Sanford and Smith, 1972; Saito and Ishii, 1987; Sano et al. 2004). Fresh soil samples, 8 g on a dry weight basis, were incubated in plastic bottles at 30°C for 2, 4, 6, 10, 15 and 20 weeks until the N mineralization pattern was found to reach a plateau, keeping the soil moisture at 50–60% of the maximum water-holding capacity. The amount of N mineralized during this incubation period was determined using the same method as that for the determination of Iex-N content. In the present study, the content of potentially mineralizable N was determined on the assumption that the rate of N mineralization was proportional to the amount of mineralizable N present, according to the first order kinetic model:

   (1)

   where N_t is the amount of N mineralized at time t, N₀ is the amount of N mineralized after an infinite time (potentially mineralizable N or Omin-N) and k is a mineralization rate constant.

5. The content of Osta-N was calculated by subtracting the sum of the contents of Iex-N, Ifix-N and Omin-N from that of total-N.

Evaluation of selected physicochemical properties of the soils

To investigate the relationships with the soil N fractions, selected physicochemical properties of the soils were analyzed. Total carbon (C) content was determined by the dry combustion method (Sumigraph NC analyzer NC-800, Sumika Chem. Anal. Service). The content of oxalate extractable aluminum (Alo) and iron (Feo) were determined by inductively coupled plasma-atomic emissions spectroscopy (SPS1500, SEIKO, Chiba, Japan) after extraction with an acid ammonium oxalate solution for 4 h in darkness (Blakemore et al. 1987). For the determination of non-exchangeable K, 2.5 g of finely ground (< 0.25 mm) air-dried soil sample was extracted with boiling 1 mol L⁻¹ nitric acid for 15 min, and the concentration of K was determined by atomic absorption spectrophotometry (AA-6200; Shimadzu, Kyoto, Japan) (Helmke and Sparks, 1996). The clay, silt and sand contents were determined by the pipette method after organic matter removal by oxidation with hydrogen peroxide and ultrasonic dispersion. The pH of the soil samples was determined electrochemically at a soil:water ratio of 1:5 (HORIBA F-23, Kyoto, Japan).

Statistical analysis

Two-way t-tests and analysis of variance (ANOVA) followed by a least significant difference (LSD) test were applied. Correlation analysis and stepwise multiple regression analysis were also applied. SYSTAT 13 was used for the analysis.

RESULTS AND DISCUSSION

Descriptive statistics of the contents of N fractions in agricultural soils in Java

Descriptive statistics of the content of total-N, Iex-N, Ifix-N, Omin-N and Osta-N are indicated in Table 2. The total-N content was 2.06 g kg⁻¹ on average and ranged from 0.76 to 6.79 g kg⁻¹ with a coefficient of variation of 58.3%. The content of Iex-N was 25.8 mg kg⁻¹ on average and corresponded to 1.3% of the total N content. This fraction, most labile and most affected by fertilizer application, was smallest among all the fractions. The content of Ifix-N was 99.1 mg kg⁻¹ on average and corresponded to 4.8% of the total-N content. It ranged most considerably with a coefficient of variation of 142%, probably reflecting parent materials, weathering degree, clay content and clay minerals. The content of Omin-N was 103 mg kg⁻¹ on average and corresponded to 5.0% of the total-N content. The Omin-N had a wide range from 5.3 to 368 mg kg⁻¹ with a coefficient of variation of 66.2%. As available fraction, Omin-N, the organic fraction, was about 4 times larger than Iex-N, the inorganic fraction. The content of Osta-N was 1.83 g kg⁻¹ on the average and corresponded to as much as 88.9% of the total-N content as the dominant N fraction in most of the soil samples. In summary, the available fraction (Iex-N and Omin-N) amounted to 128 mg kg⁻¹ and corresponded to 6.3% of the total-N, whereas the stable fraction (Ifix-N and Osta-N) amounted to 1.92 g kg⁻¹ and corresponded to 93.7% of the total-N.

Table 2 Descriptive statistics of the contents of Total-N, Iex-N, Ifix-N, Omin-N and Osta-N of the soils

	Total-N g kg^−1	Iex-N mg kg^−1	Ifix-N mg kg^−1	Omin-N mg kg^−1	Osta-N g kg^−1
Arithmetic mean	2.06	25.8	99.1	103	1.83
Geometric mean	1.81	22.0	–	81.1	1.57
Median	1.84	20.0	36.7	87.6	1.67
Maximum	6.79	83.1	580	368	6.68
Minimum	0.76	9.9	0.0	5.3	0.58
Coefficients of variation (%)	58.3	63.5	142	66.2	64.4
Ratio to total N (%)	100	1.3	4.8	5.0	88.9

Iex-N: inorganic extractable N, Ifix-N: fixed ammonium N, Omin-N: organic mineralizable N, Osta-N: organic stable N.

It should be noted that the arithmetic mean of the total-N content of 2.06 g kg⁻¹ was relatively higher than the reported value of 1.3 g kg⁻¹ for 410 tropical paddy soils from nine countries in Asia, i.e., Bangladesh, Burma, Cambodia, India, Indonesia, West Malaysia, Philippines, Sri Lanka and Thailand (Kawaguchi and Kyuma, 1977). The mean total-N content was also higher than the reported values for total-N content in paddy soils in Java, i.e., 1.2 g kg⁻¹ (Kawaguchi and Kyuma, 1977, n = 44) and about 1.6 g kg⁻¹ (Darmawan et al. 2006, n = 46), possibly because of accelerated application of organic matter and/or inorganic N fertilizer.

Relationship with selected physicochemical properties of the soils

Table 3 shows description statistics of the soil physicochemical properties. Arithmetic means of the total C content, Alo + ½ Feo and nonexchangeable K were 21.6, 25.2 and 338 mg kg⁻¹, respectively. Considerably high content of Alo + ½ Feo, an index of amorphous mineral, indicated that soils in Java are strongly influenced by volcanic materials, as shown in the case of the latest eruption of Mt. Merapi in 2010 (Anda and Sarwani, 2012). Considerably high Alo + ½ Feo values were observed for four soil samples which were collected from sites with altitudes above 551 m (all Andosols), which was consistent with the fact that Andosols are generally observed in areas with relatively high altitude under cool climate in combination with strong effect of volcanic materials. Arithmetic means of the sand, silt and clay content were 180, 327 and 493 g kg⁻¹, respectively. The arithmetic mean of the pH was 6.58, which was relatively high compared with reported values for tropical soils (Kawaguchi and Kyuma, 1977), implying artificial reclamation of the acidic nature of the soils. From these data, the strong influence of volcanic materials was confirmed for the agricultural soils in Java.

Table 3 Descriptive statistics of selected physicochemical properties of the soils

	Total-C g kg^−1	Alo + 1/2Feo g kg^−1	Nonex-K mg kg^−1	Sand g kg^−1	Silt g kg^−1	Clay g kg^−1	pH
Arithmetic mean	21.6	25.2	338	180	327	493	6.64
Median	17.8	11.4	257	87	325	512	6.69
Maximum	79.4	258.2	1368	704	548	807	8.55
Minimum	7.6	1.4	70.4	16	139	68	4.99
CV (%)	68.5	194	76.9	112	33.7	42.9	15.1

Nonex-K: nonexchangeable K.

The correlation coefficients between soil N fractions and selected physicochemical properties of the soils are indicated in Table 4. The contents of total-N and Osta-N correlated positively with total C content, Alo + ½ Feo and silt content (p < 0.01), and negatively with pH (p < 0.01). Positive correlation with total C content and Alo + ½ Feo as an index of amorphous minerals indicates a strong influence of volcanic materials on the accumulation of organic matter in Java soils. The content of Iex-N correlated positively with total C content (p < 0.05) and negatively with pH (p < 0.01). The content of Ifix-N correlated positively with nonexchangeable K content (p < 0.01) and clay content (p < 0.05), and negatively with sand content (p < 0.01), suggesting the importance of 2:1 clay minerals which can fix both NH₄⁺ and K⁺ in their interlayer sites (Fig. 2). On the contrary, Omin-N did not have any significant correlation with soil properties investigated. The fact that Alo + ½ Feo did not show significant positive correlation with Omin-N even though it showed significant positive correlation with Total-N would indicate that amorphous minerals derived from volcanic materials may inhibit the mineralization of accumulated organic N compounds due to strong binding of the materials. This result may also indicate the importance of soil management practices for the improvement of available N level in soils rather than intrinsic soil properties (Yanai et al. 2011).

Figure 2 Relationship between non-exchangeable potassium (K) and fixed ammonium-nitrogen (Ifix-N).

Table 4 Correlation coefficients between nitrogen (N) fractions and selected physicochemical properties of the soils

	Total-N	Iex-N	Ifix-N	Omin-N	Osta-N
Total-C	0.96**	0.33*	−0.12	0.21	0.95**
Alo + 1/2Feo	0.82**	0.20	−0.20	0.11	0.83**
Non-exchangable K	−0.08	−0.06	0.92**	−0.07	−0.18
sand	−0.13	−0.02	−0.34**	−0.16	−0.09
silt	0.41**	0.03	0.05	−0.17	0.41**
clay	−0.09	0.01	0.29*	0.24	−0.13
pH	−0.57**	−0.40**	0.19	−0.16	−0.58**

** and * indicate statistical significance at the level of p < 0.01 and p < 0.05, respectively.

Iex-N: inorganic extractable N, Ifix-N: fixed ammonium N, Omin-N: organic mineralizable N, Osta-N: organic stable N.

Soil N fractions in relation to soil type

The contents and percentages of the soil N fractions in relation to soil type are shown in Table 5. The total N content of Andosols (3.38 g kg⁻¹) was significantly higher than that of other soils except for Acrisols (p < 0.05). This would be due to high amorphous minerals as indicated by high Alo + 1/2Feo value in Andosols (78.6 g kg⁻¹). The content of Iex-N of Acrisols (42.9 mg kg⁻¹) was significantly higher than that of other soils except Andosols (p < 0.05) but there were no significant differences among the percentages of Iex-N. The content and percentage of Ifix-N of Fluvisols were significantly higher than those of the other soils (p < 0.05). This may be due to relatively high NH₄⁺ concentration in Fluvisols reflecting the submerged condition as well as the dominance of 2:1 type clay (mainly smectite) in the paddy. The contents of Omin-N of Andosols (128 mg kg⁻¹) and Vertisols (125 mg kg⁻¹) tended to be higher than those of the others. This result was in accordance with the general understanding that Andosols and Vertisols are relatively fertile in Java. The percentage of Omin-N of Vertisols tended to be higher than those of the others, even though the difference was statistically insignificant. This would suggest that organic N compounds in Vertisols are relatively more vulnerable to mineralization by soil microorganisms, possibly due to the relatively wet conditions in the rainy season reflecting a high amount of swelling of 2:1 clay minerals. The contents and percentages of Osta-N showed similar tendencies to those of Total N. Judging from the fact that Acrisols are mainly observed in the relatively humid west Java and Luvisols, Nitisols and Vertisols are mainly observed in the relatively dry central and east Java, these trends mentioned above can be regard as the reflection of climatic conditions. From these results, it was concluded that soil type affected soil N status considerably in the tropical area with the effect of volcanic materials as in Java, Indonesia.

Table 5 Arithmetic mean of the contents and percentages of the soil nitrogen (N) fractions in relation to soil type

		Total-N	Iex-N		Ifix-N		Omin-N		Osta-N
Soil type	No.	g kg^−1	mg kg^−1	%	mg kg^−1	%	mg kg^−1	%	g kg^−1	%
Andosols	10	3.38a	27.9ab	0.8a	43.4b	1.3b	128a	3.8a	3.18a	94.1a
Acrisols	8	2.26ab	42.9a	1.9a	37.2b	1.7b	109a	4.8a	2.07ab	91.6a
Fluvisols	9	1.78b	21.2b	1.2a	295a	17a	103a	5.8a	1.36b	76.5b
Luvisols	10	1.42b	20.0b	1.4a	51.2b	3.6b	81.0a	5.7a	1.27b	89.3a
Nitisols	4	1.55b	19.9b	1.3a	80.7b	5.2b	56.3a	3.6a	1.40b	89.9a
Vertisols	5	1.30b	18.9b	1.5a	68.0b	5.3b	125a	9.6a	1.08b	83.7ab
Average	46	2.06	25.8	1.3	99.1	4.8	103	5.0	1.83	88.9

Means followed by the same letter are not significantly different at the p < 0.05 level.

Iex-N: inorganic extractable N, Ifix-N: fixed ammonium N, Omin-N: organic mineralizable N, Osta-N: organic stable N.

Soil N fractions in relation to land use

The contents and percentages of the soil N fractions in relation to soil land use are shown in Table 6. The total N content of upland soils (2.29 g kg⁻¹) was higher than that of paddy soils (1.83 g kg⁻¹), even though the difference was not significant. The percentage of Osta-N and the contents of Total N and Osta-N were higher in upland soils even though only the percentage of Osta-N showed a significant difference (p < 0.01). The content and percentage of Iex-N were also higher in upland soils (p < 0.05). On the contrary, the content and percentage of Ifix-N were higher in paddy soils (p < 0.01). For Omin-N, the content was higher in upland soils and the percentage was slightly higher in paddy soils, even though they were not statistically significant. This contrasting result would be due to the accumulation of higher total organic matter in the upland reflecting a stronger effect of volcanic materials, and the accumulation of relatively degradable organic matter in the paddy under submerged condition. In addition, the high amount of organic matter and inorganic fertilizer application to cash crops, especially in relatively cool areas at high elevations, may have increased soil N contents in upland soils. These tendencies mentioned above would be more strongly affected by the difference of the soil types used for upland and paddy—i.e., Andosols were the dominant soil type in the upland and Fluvisols were the dominant soil type in the paddy (Table 1)—rather than land use per se. It was concluded from these results that land use in combination with soil type affected soil N fractions considerably, reflecting the intrinsic chemical structures and reactivity of each fraction.

Table 6 Arithmetic mean of the contents and percentages of the soil nitrogen (N) fractions in relation to land use

		Total-N	Iex-N		Ifix-N		Omin-N		Osta-N
Land use	No.	g kg^−1	mg kg^−1	%	mg kg^−1	%	mg kg^−1	%	g kg^−1	%
Paddy	23	1.83	20.8	1.1	172	9.4	95.9	5.2	1.55	84.3
Upland	23	2.29	30.8	1.3	26.3	1.2	110	4.8	2.12	92.7
Significance		ns	*	*	**	**	ns	ns	ns	**

**, *, and ns indicate statistically significant at the level of p < 0.01 and p < 0.05 and not significant, respectively.

Iex-N: inorganic extractable N, Ifix-N: fixed ammonium N, Omin-N: organic mineralizable N, Osta-N: organic stable N.

Soil N fractions in relation to topography

The contents and percentages of the soil N fractions in relation to elevation are shown in Table 7. It is clear that total N content and Osta-N content were significantly higher for the samples with high elevations, whereas Ifix-N content was significantly higher for the samples with low elevations. Iex-N and Omin-N contents did not show significant trends, even though Omin-N content tended to be higher for the samples with high elevations. These results strongly suggest that soil N status showed strong topographical trends depending on the elevation where soil developed, in combination with those of climatic conditions, soil type and land use.

Table 7 Arithmetic mean of the contents and percentages of the soil nitrogen (N) fractions in relation to elevation

		Total-N	Iex-N		Ifix-N		Omin-N		Osta-N
Elevation(m)	No.	g kg^−1	mg kg^−1	%	mg kg^−1	%	mg kg^−1	%	g kg^−1	%
1000∼	3	5.46a	40.4a	0.8a	17.8b	0.3b	117a	2.3a	5.29a	96.6a
500∼1000	5	2.71b	25.7a	1.1a	19.5b	0.8b	110a	4.3a	2.55b	93.9a
200∼500	6	1.17c	16.7a	1.5a	22.9b	2.0b	126a	10.5a	1.00c	86.0a
100∼200	9	2.16bc	35.4a	1.7a	24.7b	1.7b	120a	6.1a	1.98bc	90.6a
50∼100	5	1.42bc	31.3a	2.0a	54.4ab	4.5ab	94.3a	6.6a	1.24c	86.9a
0∼50	18	1.74bc	20.1a	1.2a	210a	11.9a	85.3a	4.7a	1.42c	82.2a
Average	46	2.06	25.8	1.3	99.1	4.8	103	5.0	1.83	88.9

Means followed by the same letter are not significantly different at the p < 0.05 level.

Iex-N: inorganic extractable N, Ifix-N: fixed ammonium N, Omin-N: organic mineralizable N, Osta-N: organic stable N.

Comparison of soil N status of agricultural soils between Java and Japan

N fractions of agricultural soils in Java, Indonesia, were compared with those in Japan, as shown in Table 8. This comparison was carried out because both regions are affected by volcanic activities in the circum-Pacific orogenic zone and they have basically different climatic conditions: a tropical climate in Java and a temperate climate in Japan.

Table 8 Comparison of the contents of soil nitrogen (N) fractions between Java and Japan

		Total-N	Iex-N	Ifix-N	Omin-N	Osta-N
	No.	g kg^−1	mg kg^−1	mg kg^−1	mg kg^−1	g kg^−1
Japan^1	147	2.59	37.2	124	159	2.27
Java	46	2.06	25.8	99	103	1.83
Significance		*	**	ns	**	*
Java/Japan (%)		80	69	90	65	80

¹Sano et al. (2004).

Iex-N: inorganic extractable N, Ifix-N: fixed ammonium N, Omin-N: organic mineralizable N, Osta-N: organic stable N.

Table 8 clearly indicates that the N contents of all the fractions except Ifix-N in Java soils were statistically lower than those in Japanese soils. Significantly lower Omin-N, Total-N and Osta-N contents in Java can be explained because soil organic matter tends to be decomposed more rapidly in Java due to higher temperature and precipitation reflecting tropical climate conditions, and significantly lower Iex-N in Java may be because the intensity of inorganic fertilizer application tended to be lower in Java. The insignificant difference between the contents of Ifix of Java and Japan may indicate that clay content and type of clay minerals of the soils were more influential factors than the climatic conditions. Their relative ratios were, however, variable depending on the fractions. The content of Total-N in Java soil was on average 80% of that in Japanese soils. The contents of available N fractions, i.e., Iex-N and Omin-N, in Java soils were on average 69 and 65% of those in Japanese soils, whereas those of stable N fractions, i.e., Ifix-N and Osta-N, in Java soils were 90 and 80% of those in Japanese soils. This result directly indicates that relative percentages of available N fractions in Java soils were lower than those in Japanese soils, probably due to accelerated decomposition of organic matter, especially relatively degradable available fractions, in tropical soils. This would be one of the reasons why N deficiency of plants is so prevalent and hence soil N level often becomes the limiting factor for the growth and yield of agricultural crops in the tropics (Yanai et al. 2007, 2010), even though application of inorganic N fertilizers and organic matter should also be taken into account. In conclusion, the N status of agricultural soils in Java was considerably lower than that of soils in Japan, reflecting their environmental conditions.

CONCLUSION

The soil N status of the agricultural soils in Java was evaluated in terms of form and availability. The contents of total-N and Osta-N were mainly affected by the amount of amorphous minerals, whereas the content of Ifix-N was mainly affected by the amount of 2:1 clay minerals. The contents of total N, Iex-N, Ifix-N, Omin-N and Osta-N in Java soils were on average 80, 69, 90, 65 and 80% of those in Japanese soils, probably due to accelerated organic matter decomposition reflecting high temperature, but the levels were relatively high in tropical soils due to the effect of volcanic materials. In conclusion, these results should be taken into account for the rational management of soil N in agricultural fields in Java, Indonesia.

ACKNOWLEDGMENTS

We are grateful to Dr. Tetsuhiro Watanabe and Ms. Iva Dewi Lestaringingsih, Kyoto University, for their assistance during a part of the soil sampling, and Prof. Spiandi Sabiham, Bogor Agricultural University, for his arrangement to carry out this collaborative research between Indonesia and Japan. This study was partly supported by a Grant-in-Aid for Scientific Research (C) from the Ministry of Education, Science, Sports and Culture, Japan (No. 22580070).