Effect of different types of organic fertilizers on the chemical properties and enzymatic activities of an Oxisol under intensive cultivation of vegetables for 4 years

Abstract

The effects of different types of organic fertilizers on the chemical and enzymatic properties of an Oxisol were studied after being fertilized for four consecutive years (26 crops) in a greenhouse under intensive cultivation of vegetables. Seven treatments, consisting of five types of organic fertilizer treatments, one “sequential application” (SA) treatment, and a chemical fertilizer treated plot were compared. The organic fertilizers used were dairy cattle dung compost (DCDC), hog dung compost (HDC), chicken dung compost (CDC), pea residue compost (PRC) and soybean meal (SBM). After 4 years of cultivation, the soils were analyzed for their chemical properties and enzymatic activities. The microbial carbon (C) and nitrogen (N), basal respiration and nitrification rate were also measured. The results showed that the SBM significantly lowered the soil pH, and that the HDC and DCDC raised the soil pH. The SBM and CDC resulted in the lowest soil electrical conductivity. The SBM had no significant effect on soil organic C and total N contents when compared with the CF plot. However, the DCDC resulted in the highest contents of soil organic C and total N. The organic fertilizers applied did not significantly affect the soil available copper, zinc, cadmium, lead and nickel. The effects of the different organic fertilizers on soil enzymatic activities depended on the types of organic fertilizers applied. The SBM and CDC often resulted in a lower microbial C (or N) and respiration rate, while in contrast DCDC and PRC resulted in high measurements. Most of the measured soil enzymatic activities in the SBM treatment, except for acid phosphatase, were the lowest. Differing contents of different heavy metals in the organic fertilizers resulted in different Mehlich III extractable heavy metal contents in the soils. From the point of view of the soil chemical and enzymatic properties, SBM is not an appropriate organic fertilizer for continuous application to an Oxisol.

Key words:

• microbial biomass
• organic fertilizer
• soil enzyme activity

INTRODUCTION

Organic matter is the principal indigenous source for soil available nitrogen (N), insofar that it contains as much as 65% of the total soil phosphorus (P), and provides significant amounts of sulfur (S) and other nutrients essential for plant growth (Bauer and Black 1994). Adequate nutrients, good structure and abundant biological life are important factors of a productive soil. There is a consensus that soil organic matter has a significant role to play in the sustainability of farming systems (Swift and Woomer 1993) and that it is an important indicator of soil quality and productivity (Larson and Pierce 1994). Under tropical and subtropical climatic conditions, high cultivation frequency and a low input of organic matter the organic matter contents in the farmland soils of Taiwan are generally low, and it is common for the soil organic matter content to be lower than 20 g kg⁻¹ (Huang 1994). Therefore, in recent years the application of organic fertilizers has received great attention from researchers investigating the sustainability and productivity of agricultural soils.

Animal dung composts, such as cattle dung compost, hog dung compost and chicken dung compost, which come from the livestock industry, are the principle organic sources being applied to increase the fertility and organic matter content of agricultural soils in Taiwan. Meanwhile, other agricultural wastes, such as soybean meal (SBM) and pea residue compost (PRC), are also used in large quantities as fertilizers. Soybean meal is what remains after grinding up and extracting the oil from soybeans. The meal is rich in protein, which is why it is usually sold as feed for pigs, cows and chickens. Soybean meal is also often applied directly to the soil as a fertilizer because it provides N for plant growth faster than other organic fertilizers. A large quantity of so-called pea seedlings are produced and sold as a vegetable in Taiwan. The pea seeds are germinated on a bed of rice hulls. Pea seedling shoots are cut at the bed surface after they reach approximately 15 cm in height. Then the residue together with the rice hulls is composted, resulting in pea residue compost.

The biological properties of soils usually respond more rapidly to changing soil conditions than either the chemical or physical properties (Kowaljow and Mazzarino 2007; Melero et al. 2007; Powlson 1994; Tejada et al. 2006). Soil enzymatic activities have been used as indicators of soil fertility because they are reflections of the effects of cultivation, soil properties and pedological amendments (Avidano et al. 2005; Bastida et al. 2006; Ceccanti et al. 1993; Corstanje et al. 2007). Furthermore, soil enzyme activities and microbial biomass, especially in combination with an activity parameter such as CO₂ production rate, are very sensitive to both natural and anthropogenic disturbances and show a quick response to the induced changes (Dick 1992; Hernández et al. 2007; Nourbakhsh 2007; Renella et al. 2007).

As there are threats of heavy rain and typhoons in the summer months, the cultivation of vegetables in greenhouses is a common practice in Taiwan, especially in organic farming. Organic farming has become very popular and the acreage under cultivation is steadily increasing. For vegetable cultivation, six to ten crops in a year, depending on the type of crop, are common. Vegetable crops require an adequate and continuous supply of N for proper growth and maximal high quality yields. Therefore, a high rate of N fertilizer is applied in vegetable cultivation in Taiwan (Lian et al. 1996).

Little research has been carried out regarding the effects on soil chemical and enzymatic activities when different types of organic materials are continuously applied in large quantities under intensive cultivation of vegetables in a greenhouse. Therefore, the aim of this study was to examine the effects of different types of organic fertilizers on the chemical properties and enzymatic activities of an Oxisol after the fertilizers had been applied to the soil for 4 years in a greenhouse under intensive cultivation of different vegetables.

MATERIALS AND METHODS

An experiment was conducted in a greenhouse in Taoyuan District Agricultural Research and Extension Station, located in Taoyuan in northern Taiwan. The mean annual precipitation of 1,500 mm per year has a distinct seasonal pattern, with 65% of it falling from May to September and 16% from October to January. The mean daily temperature is 22.3°C and the highest monthly temperature is in July (32.7°C) and the lowest monthly temperature is in January (12.7°C). The experiment was carried out between April 2000 and July 2004 (a total of 26 crops of vegetables were cultivated). The soil used was Typic Halpludox, clayey, mixed, allic and thermic (0–15 cm). Some selected characteristics of the soil before the experiment began (0–15 cm depth) are shown in Table 1 and Table 2. There were a total of seven plots. Six of these plots received treatment as follows. Five plots were treated continuously with one of five types of commercial organic fertilizer: dairy cattle dung compost (DCDC), hog dung compost (HDC), chicken dung compost (CDC), pea residue compost (PRC) and soybean meal (SBM). The sixth plot received each of the above mentioned five organic fertilizers sequentially (SA), and the seventh plot received a chemical fertilizer treatment, and is referred to as CF. For the SA treatment the organic fertilizers were applied in the following order for each crop: DCDC, PRC, HDC, SBM and CDC and then repeated. The locations of the individual research plots and their respective treatments remained unchanged over the 4-year period. However, chemical fertilizers were applied to the CF plots from the 24th crop (January 2004) onwards and no fertilizer was applied before the 24th crop. The chemical fertilizers used were urea for N, superphosphate for P and KCl for K. The total amounts of chemical fertilizers applied in the three crops on the basis of N, P₂O₅ and K₂O were 475, 240 and 320 kg ha⁻¹, respectively. Dairy cattle dung compost is composed of dairy cattle dung and sawdust, which is the waste of mushroom cultivation; CDC is composted chicken dung and rice hulls; PRC is composted pea seedling residues and rice hulls; and HDC is composted hog dung and sawdust. Some selected chemical properties of the organic fertilizers used are shown in Table 3. The 24th, 25th and 26th crops planted were spinach (Spinacia oleracea L.) (180 kg N ha⁻¹) (from 5 January 2004 to 5 March 2004), lettuce (Lactuca scariola L. var. sativa Bisch.) (145 kg N ha⁻¹) (from 10 March 2004 to 14 May 2004), and water convolvulus (Ipomoea aquatica Forssk.) (150 kg N ha⁻¹) (from 20 May 2004 to 2 July 2004), respectively. Fertilizers were applied to each crop based on the recommendation rate for that crop. Only N was considered in the application of the organic fertilizer. No residual effect of the fertilizer was considered. Dairy cattle dung compost, HDC and PRC were applied based on the assumption that 50% of the N will become available to the plants of each crop, while for SBM and CDC that number was 80%. Therefore, in the past 4 years the total amount of N applied through the SA treatment was 7,991 kg ha⁻¹, through the DCDC, HDC and PRC treatments it was 9,360 kg ha⁻¹, and through the SBM and CDC it was 5,850 kg ha⁻¹. All the fertilizers were applied to the soil surface and incorporated by means of rototilling before transplanting the seedlings.

Table 1 Some selected chemical properties of the original soil and the soils after harvesting the final three crops

Treatment		pH (1:1)	EC (saturated)(dS m^−1)	Total C	Total N	Mehlich III extractable
						P	K	Ca	Mg
				(g kg^−1)		(mg kg^−1)
Original soil		6.7	0.2	14.0	4.42	93	127	3054	207
	CF	5.19 c	3.33 b	13.0 e	4.43 d	328 cd	265 bc	753 d	239 e
	SBM	4.30 e	1.16 d	14.8 e	2.31 e	234 e	200 e	43 f	48 f
After 24th crop	PRC	4.83 d	2.10b cd	57.4 b	8.13 b	367 bc	235 bc	384 e	430 c
	DCDC	4.96 cd	4.88 a	81.3 a	12.25 a	408 b	720 b	1278 b	667 a
	HDC	6.13 a	2.68 bc	42.0 c	6.43 c	291 de	346 de	2033 a	436 c
	CDC	6.79 a	1.79 cd	30.7 cd	4.75 d	549 a	691 a	1847 a	506 b
	SA	5.18 c	2.66 bc	40.6 d	6.46 c	346 bcd	372 bcd	948 c	332 d
	CF	5.55 d	1.05 de	11.3 f	3.79 e	282 b	292 c	983 e	181 c
	SBM	4.58 d	0.73 e	14.4 f	2.34 f	198 c	141 d	53 g	40 f
After 25th crop	PRC	4.70 cd	1.85 bc	60.9 b	7.94 b	268 b	317 bc	779 f	381 e
	DCDC	4.89 c	4.43 a	89.0 a	12.27 a	272 b	580 a	1662 c	592 a
	HDC	6.67 a	2.08 b	46.6 c	6.43 c	222 c	360 bc	3014 a	379 e
	CDC	6.90 a	1.28 d	31.6 de	5.07 d	459 a	545 a	2078 b	448 b
	SA	5.62 b	1.51 cd	40.8 cd	6.44 c	261 b	365 b	1270 d	293 d
	CF	5.40 e	2.32 c	12.3 d	4.12 f	309 cd	281 c	920 d	211 e
	SBM	4.15 f	1.74 c	14.9 d	2.32 g	258 de	172d	65 e	89 f
After 26th crop	PRC	4.96 cd	2.85 b	57.6 b	8.04 b	333 bc	539 a	1025 d	462 b
	DCDC	4.73 d	4.06 a	81.4 a	9.35 a	382 b	433 b	1509 c	183 a
	HDC	6.83 a	2.27 bc	50.3 b	5.37 d	237 e	261 cd	4463 a	396 c
	CDC	6.56 b	2.12 c	31.6 c	4.53 e	553 a	625 a	2177 b	473 ab
	SA	5.17 c	2.99 b	49.1 b	5.86 c	319 c	342 bc	1443 c	325 d
Within columns of the same crop, means followed by the same letter are not significantly different (P = 0.05) using a Duncan's multiple range test. CDC, chicken dung compost; CF, chemical fertilizer; DCDC, dairy cattle dung compost; HDC, hog dung compost; PRC, pea residue compost; SA, sequential application; SBM, soybean meal.

For each crop, uniform seedlings were transplanted to each plot. No pesticides were applied. After harvesting the plots were immediately rototilled to a depth of 15 cm.

All treatments were replicated four times in a randomized complete block design using 1.1 m × 5.7 m individual plots. Soil samples were taken after the harvesting of the last three crops. Composite soil samples that were composed of three randomly extracted soil cores for each plot were collected from the surface layer (0–15 cm). Field-moist soil samples were divided into two subsamples. One soil subsample was sieved to pass through a 2 mm mesh and was then stored at 2°C for biochemical analysis and a second soil subsample was allowed to air-dry at room temperature. The air-dried soil samples were ground to pass through a 2 mm screen for chemical analysis.

The chemical properties, microbial biomass and basal respiration of soils were analyzed using the method of a previous study (Chang et al. 2007). The levels of eight enzymatic activities were measured in the soils: protease (enzyme commission [EC] 3.4.2.21-24) (Ladd and Bulter 1972), urease (EC 3.5.1.5) (Kandeler and Gerber 1988), alkaline phosphatase (EC 3.1.3.2) (Tabatabai and Bremner 1969), acid phosphatase (EC 3.1.3.2) (Tabatabai and Bremner 1969), arylsulphatase (EC 3.1.6.1) (Tabatabai and Bremner 1970), β-glucosidase (EC 3.2.1.21) (Tabatabai 1982), dehydrogenase activities (EC 1.1.) (Thalmann 1968), and nitrification rate (Berg and Rosswall 1985). Statistical analysis was carried out using multiple anovas (Duncan's multiple range test at a 0.95 level of probability) to determine significant differences between the treatments.

Table 2 Some selected heavy metal contents of the original soil and the soil after harvesting the final three crops

Treatment		Mehlich III extractable (mg kg^−1)
		Cu	Zn	Cd	Pb	Cr	Ni
Original soil		2.34	15.6	0.13	4.85	0.37	2.13
	CF	2.55 a	28.9 a	0.14 b	5.11 a	0.28 a	2.31 a
	SBM	1.95 b	11.0 e	0.06 d	1.43 b	UD	1.75 ab
After 24th crop	PRC	0.08 f	21.8 d	0.12 c	0.28 ef	0.55 a	1.55 ab
	DCDC	0.13 f	23.3 c	0.16 b	0.15 f	UD	1.43 b
	HDC	1.20 c	24.1 c	0.20 a	0.65 c	UD	1.83 ab
	CDC	1.00 d	26.8 b	0.22 a	0.33 de	UD	1.68 ab
	SA	0.58 e	23.3 c	0.15 b	0.43 d	0.28 a	2.33 a
	CF	3.14 a	26.4 bc	0.13 b	5.02 a	0.32 b	2.22 a
	SBM	2.23 b	10.8 e	0.06 c	1.80 b	0.02 c	1.30 bc
After 25th crop	PRC	0.20 d	23.4 d	0.08 c	0.38 d	0.11 c	0.98 c
	DCDC	0.30 d	25.5 c	0.15 b	0.35 d	0.13 c	1.32 bc
	HDC	1.60 c	27.5 b	0.20 a	1.10 c	1.12 a	2.00 a
	CDC	1.30 c	29.8 a	0.21 a	0.50 e	0.50 ab	1.48 b
	SA	0.58 d	25.9 c	0.17 ab	1.18 c	0.53 ab	1.25 bc
	CF	2.89 a	27.7 cd	0.18 a	4.65 a	0.25 c	2.18 a
	SBM	0.95 b	12.0 d	0.03 c	1.53 b	1.95 ab	1.22 b
After 26th crop	PRC	UD	30.5 c	0.08 bc	0.13 c	1.60 b	0.95 b
	DCDC	UD	34.8 c	0.12 b	0.48 c	0.45 c	1.30 b
	HDC	0.65 c	45.5 b	0.21 a	1.15 b	2.38 a	2.03 a
	CDC	0.38 d	57.0 a	0.20 a	0.30 c	2.00 ab	1.35 b
	SA	0.18 e	35.8 c	0.08 bc	0.23 c	2.70 a	1.20 b
Within columns of the same crop, means followed by the same letter are not significantly different (P = 0.05) using a Duncan's multiple range test. CDC, chicken dung compost; CF, chemical fertilizer; DCDC, dairy cattle dung compost; HDC, hog dung compost; PRC, pea residue compost; SA, sequential application; SBM, soybean meal, UD, under the detection limit.

RESULTS AND DISCUSSION

Chemical properties of the soils after the planting of vegetables

Table 3 shows that the chemical properties of the selected organic fertilizers used in the experiment differed depending on the type of compost. All organic fertilizers except SBM could be considered to be matured compost according to their C/N ratios (Bernal et al. 1998; Kato et al. 2005; Zmora-Nahum et al. 2005). There was no significant difference in organic matter content and pH between the different organic fertilizers. However, the total N content of SBM was higher and the C/N ratio of SBM was lower than those of the other organic fertilizers. The electrical conductivity (EC) of CDC was significantly higher than that of the other organic fertilizers and this can be attributed to the fact that the chicken dung decomposed rapidly during composting, resulting in a higher content of soluble salts. The CDC also has higher contents of P, K, Ca and Mg, probably because most of the P, K, Ca and Mg in the feed of the chickens were excreted. On the contrary, the PRC contained less P, K and Ca. The higher contents of Zn and Cu in HDC and CDC resulted from the zinc oxide (ZnO), zinc sulfate and cupric sulfate that are added to the feed of hogs and chickens as growth promoting substances (Bou et al. 2005; Namkung 2006; Zacharias et al. 2003; Zhang and Guo 2007). The organic fertilizers also differed in their contents of cadmium (Cd), nickel (Ni), chromium (Cr) and lead (Pb); however, the levels were acceptable for fertilizers.

Zaller and Koepke (2004) showed that the soil pH, C/N ratio, available P and available K contents were all significantly increased by farmyard manure (FYM) application. Table 1 shows that no significant change in soil pH was observed by applying CDC and HDC for four consecutive years when compared with the initial soil pH (6.7). This result could be attributed to the high amount of bases also applied with the composts. Similar results have been obtained in other studies (Hue 1992; Materrechera and Mkhabela 2002; Whalen et al. 2000). Owing to an assumption of differing mineralization rates for different organic fertilizers, the total amount of N applied with DCDC, HDC and PRC was higher than that with SBM, CDC and SA. However, SBM application resulted in the lowest soil pH level (Table 1). Soybean meal is rich in protein with a low C/N ratio and, therefore, its mineralization rate in the soils is greater than that of the other organic fertilizers (Van Kessel et al. 2000). The mineralization of organic N and the subsequent nitrification resulted in acidifying the soil (Graham and Haynes 2005). The pH of the CF soil was lowered, compared with its initial pH, even when applying urea for three crops (475 kg N ha⁻¹ in total). This can be attributed to the acidification effect of urea during transformation in the soil (Kemmitt et al. 2006). Owing to the fact that the equivalent of 1,137.5 kg N ha⁻¹ of SBM has also been applied in the SA treatment, the pH of the SA plot also declined compared with its initial pH.

Table 3 Some selected characteristics (on a dry-weight basis) of the organic fertilizers used in the experiment

Parameter	Soybean meal	Pea residue compost	Dairy cattle dung compost	Hog dung compost	Chicken dung compost
Organic matter (g kg^−1)	700 a	690 a	670 a	600 a	620 a
Total N (g kg^−1)	52.0 a	27.9 b	29.5 b	26.2 b	30.2 b
C/N	6.0 b	11.0 a	10.2 a	10.2a	9.1 ab
pH (1:5)	6.10 a	6.40 a	6.45 a	7.30 a	6.50 a
EC (dS m^−1) (1:5)	4.75 b	3.83 b	7.79 ab	3.84 b	13.08 a
Total P (g kg^−1)	9.1 b	6.7 b	10.8 ab	10.7 ab	17.1 a
Total K (g kg^−1)	21.3 b	10.3 b	19.7 b	14.4 b	37.0 a
Total Ca (g kg^−1)	3.0 b	6.5 b	13.1 b	44.5 a	46.5 a
Total Mg (g kg^−1)	2.8 b	3.2 b	5.5 b	4.5 b	8.4 a
Total Cu (mg kg^−1)	11.6 c	17.3 c	29.8 c	101.8 b	222.5 a
Total Zn (mg kg^−1)	36.7 c	78.7 c	136.8 b	228.6 a	265.2 a
Total Mn (mg kg^−1)	28.3 d	178 c	372 b	507 a	516 a
Total Cd (mg kg)	0.7 c	0.7 c	1.5 bc	3.1 a	2.5 ab
Total Ni (mg kg)	25.9 b	34.5 b	42.0 b	89.6 a	89.7 a
Total Cr (mg kg)	0.3 b	3.3 b	7.3 b	91.0 a	8.9 b
Total Pb (mg kg)	3.9 b	7.3 b	14.0 b	36.7 a	16.2 b
Within rows, means followed by the same letter are not significantly different (P = 0.05) using a Duncan's multiple range test. EC, electrical conductivity.

The effects on soil EC were different according to the different organic fertilizers that were applied. The application of DCDC resulted in the highest EC level of the soil and the lowest EC level was obtained when applying SBM. The study was carried out in a greenhouse; therefore, leaching loss of soluble salts was not an issue. High EC, high decomposition rate and high application rate of the compost could be the main reasons resulting in high EC of the soil. Even if the EC of CDC is higher than that of DCDC, the higher total amount of DCDC applied (317,280 kg ha⁻¹) over the 4 years resulted in the higher EC of the soil compared with CDC. The application of organic fertilizers excluding SBM for four consecutive years resulted in a significant increase in soil organic matter and total N contents. There was no significant effect on soil organic C content with the application of SBM compared with organic C (14.0 g kg⁻¹) of the original soil.

Applying fertilizer resulted in a significant increase in Mehlich III extractable P and K contents compared with those of the original soil. These results indicated that the amount of nutrients applied by the fertilizers was higher than those removed by the vegetables. Similar results were reported by Carpenter et al. (1998), Aulakh and Pasricha (1999) and Van Den Bossche et al. (2005). The fact that little leaching occurs under greenhouse conditions is another reason for the rapid accumulation of nutrients in the soil (Chen et al. 2002). The Mehlich III extractable Ca of all treatments was lower than that of the original soil (Table 1). This could be attributed to the formation of low solubility Ca–P compound through application of high amounts of P-containing compost. The application of SBM resulted in a significant lowering of Mehlich III extractable Ca and Mg compared with the other treatments. This result could be because of the fact that reduced amounts of Ca and Mg were applied with the SBM.

The application of organic fertilizers with different heavy metal contents resulted in significant differences in Mehlich III extractable Cu, Zn, Cd, Pb, Cr and Ni of the soils (Table 2). The application of organic fertilizers resulted in decreases in Mehlich III extractable Cu and Pb compared with the CF and original soil, probably because of the formation of insoluble Cu–organic and Pb–organic complexes (Covelo et al. 2007; Egiarte et al. 2006; Karaca et al. 2006; Kumpiene et al. 2007). The higher Zn contents in the CF treatment compared with the original soil could be attributed to its lower pH levels. However, no significant difference in the concentrations of the heavy metals of vegetables was observed (data not shown) probably because the contents of the metals in the soil were not high enough.

Microbial biomass, respiration rate and nitrification rate

Microbial biomass measurements have been used to give an early indication of the changes in the organic matter content of a soil as a result of variation in soil management (Garcia-Gil et al. 2000; Hargreaves et al. 2003; Insam et al. 1989; de Vries et al. 2007; Zhang et al. 2004). Leita et al. (1999) reported that the soil microbial biomass, which represents approximately 1–4% of total soil organic C, is a more sensitive indicator of changing soil conditions than direct analysis of the organic C content. Fauci and Dick (1994) showed that the highest microbial activity in soils usually occurs shortly after the application of organic matter. The effect of the application of different organic fertilizers on microbial biomass C and N was significantly different (Fig. 1). Incorporation of PRC and DCDC to soil for four consecutive years raised the microbial C and N significantly. Taking temporal variations in microbial C and N into consideration, the application of DCDC resulted in the highest microbial C and N among the different treatments. This is probably because of the material being obtained by processes involving a high degree of microbial activity, and once these processes have terminated part of the microorganisms are retained in the compost. In addition, the DCDC-treated soil also had the highest content of total C (Table 1), which acts as an energy source for the authochthonous microorganisms (Perucci 1992). Soil treated with SBM had the lowest microbial C and N. Even if SBM treatment resulted in the lowest microbial biomass, this effect could be masked by applying other organic fertilizers before or after the application of SBM. The SBM was applied at the planting of the 24th crop, however, there was a significant difference in soil microbial biomass C between the SA and SBM treatments after harvesting of the 24th and 25th crops (Fig. 1).

Basal respiration rate can be another useful parameter in measuring the biological activity of a soil (Dinesh et al. 2004; Ros et al. 2006; Vanhala et al. 2005). The application of organic fertilizers (except for SBM) resulted in significant differences in the basal respiration rates of the soils studied, and on average the PRC and DCDC treatments had the highest respiration rates (Fig. 1). Basal respiration is considered to reflect the availability of C for microbial maintenance and is a measure of the basic turnover rates in soil (Dinesh et al. 2004; Insam et al. 1991; Vanhala et al. 2005). There was a consistent relationship between microbial C and N and respiration rate, and a significant linear correlation was observed between microbial C and microbial N with the respiration rates (Table 4). Dinesh et al. (2004) reported a similar result. The low respiration rate of the soil after harvesting of the vegetables in the SBM treatment means that the SBM decomposed rapidly and that the microbial activity declined shortly after the SBM was applied to the soil. Respiration rates were also positively significantly correlated with EC, and the organic matter contents, total N, Mehlich III extractable K and Mg of soil (Table 4).

Figure 1  Effect of the treatments on the microbial biomass content, respiration and nitrification over 4 years of cultivation. Bars with the same letter are not significantly different at the 0.95 level of probability, according to a Duncan's Multiple Range Test. CDC, chicken dung compost; CF, chemical fertilizer; DCDC, dairy cattle dung compost; HDC, hog dung compost; PRC, pea residue compost; SA, sequential application; SBM, soybean meal.

Table 4 Linear correlation coefficients between selected chemical properties and the enzymatic activities of the soils

	Mic-C	Mic-N	pH	EC	OM	TN	EXT-P	EXT-K	EXT-Ca	EXT-Mg
Mic-C	–	0.8923**	–0.1342	0.7411**	0.9007**	0.9548**	0.1015	0.4709*	0.0995	0.7120**
Mic-N	0.8923**	–	–0.0490	0.6672**	0.8205**	0.8901**	0.1425	0.5839**	0.2047	0.7273**
CO2-resp	0.5847*	0.5317*	0.1597	0.4454*	0.7555**	0.6952**	0.1418	0.5375*	0.4277	0.6279**
dHase	0.2280	0.4539*	0.7916**	–0.0144	0.1196	0.2042	0.2427	0.4150	0.5940**	0.5941**
Gluase	0.9194**	0.8311**	–0.1470	0.5346*	0.8055**	0.8720**	0.0781	0.4156	–0.0245	0.6773**
Protease	0.8503**	0.8552**	0.2285	0.6408**	0.7587**	0.8392**	0.2054	0.6105*	0.3592	0.7743**
Urease	0.2347	0.3320	0.8353**	0.2480	0.3150	0.3095	0.5411*	0.6288**	0.8066**	0.7001**
Nitrifi	–0.2674	–0.0141	0.8874**	–0.1997	–0.1495	–0.1783	0.1543	0.4284	0.6202**	0.3480
aSase	0.0943	0.2544	0.5552**	0.2364	0.3338	0.2202	0.1889	0.3517	0.7139**	0.3750
aPase	0.7886**	0.6155**	–0.5878**	0.6091**	0.7758**	0.7816**	–0.1480	0.1364	–0.3036	0.2662
Pase	0.2744	0.4612*	0.8299**	0.2304	0.3391	0.3191	0.0616	0.4800*	0.8976**	0.7273**
Mic-C, microbial C; Mic-N, microbial N; CO2-resp, respiration rate; dHase, dehydrogenase; Gluase, β-glucosidase; Nitrifi, nitrification; aSase, arylsulphatase; aPase, acid phosphatase; Pase, alkaline phosphatase; OM, organic matter content; TN, total N content, EXT-P, Mehlich III extractable P. *P < 0.05;
**P < 0.01.

Nitrification is a key process in agricultural and natural ecosystems and plays an important role in the global N cycle (Innerebner et al. 2006). Different organic fertilizers affected the rates of nitrification differently (Fig. 1). On average, the highest nitrification rate was obtained with CDC followed by HDC. This means that the decomposition rates of organic N in CDC-treated and HDC-treated soils were higher than the rates of the other organic matter treated soils. Owing to the fact that chemical N fertilizer (urea) was applied to the CF plot at the last three crop cultivations, the nitrification rates were induced to a considerably high level compared with the SBM and PRC treatments. The lower nitrification rate can also be attributed to the lower pH of the soils because low pH is the most likely factor restricting its activity (Egiarte 2005; Graham and Haynes 2005). A significant linear correlation was observed between nitrification rates and soil pH values (Table 4). Nitrification rates were also positively significantly correlated with the organic matter contents, total N, Mehlich III extractable K and Mg of the soil (Table 4).

Soil enzyme activities

Dehydrogenase activity in soil is considered to be one of the best indicators of overall microbial activity (Masciandaro et al. 2001) because dehydrogenase occurs only within living cells, unlike other enzymes, which can occur in extracellular states (Cooper and Warman 1997). Marinar et al. (2000) reported that a higher level of dehydrogenase activity was observed in soil treated with vermicompost and manure. Figure 2 shows that the dehydrogenase activities of soils were different based on the types of organic fertilizers they were treated with. Figure 2 also shows that there was a temporal variation in dehydrogenase activities. However, the dehydrogenase activities of soils treated with HDC and CDC were higher than those treated with the other fertilizers. The SBM-treated soil had the lowest dehydrogenase activity, which could be attributed to the low soil pH. Berg and Rosswall (1985) also showed that the dehydrogenase activities were very low in acid soils (pH < 5). There is a positively significant correlation between soil pH values and dehydrogenase activities (Table 4). Table 4 also shows that there were significant positive correlations between dehydrogenase activities and available Ca and Mg of soil.

β-glucosidase, which is an enzyme involved in the C cycle, has been widely used in the evaluation of soil quality in soils subjected to different management procedures (Gil-Sotres et al. 2005). β-glucosidase activities differed depending on the types of organic fertilizers added (Fig. 2). Treatment with SBM resulted in the lowest β-glucosidase activity. The β-glucosidase activities when receiving PRC or DCDC treatments were the highest among the group. The existence of a highly significant linear positive correlation between organic C of the soil and β-glucosidase activities (Table 4) indicates that organic C is an important substrate of β-glucosidase (Dinesh et al. 2004). There are also significant positive correlations between β-glucosidase and soil microbial C and N, EC, total N and Mehlich III extractable Mg contents (Table 4).

Figure 2  Effect of the treatments on soil enzymatic activities over 4 years of cultivation. Bars with the same letter are not significantly different at the 0.95 level of probability, according to a Duncan's Multiple Range Test. CDC, chicken dung compost; CF, chemical fertilizer; DCDC, dairy cattle dung compost; HDC, hog dung compost; PRC, pea residue compost; SA, sequential application; SBM, soybean meal.

Nitrogen mineralization is an important reaction in soils because it is related to plant growth by supplying sufficient amounts of N. Figure 2 shows that the different organic fertilizers affect the protease activities of the soils differently. Treating a soil with SBM resulted in the lowest protease activity of the soil, even lower than that of CF treatment. However, soil amended with DCDC stimulated protease activity and showed levels significantly higher than that of the other treatments. Loll and Bollag (1983) also showed that soil amended with organic compounds, such as straw, would increase the protease activity. Okur et al. (2006) showed that the activity of protease decreased by 26% after solarization. However, in this study there was a significantly linear correlation between protease activity and EC (Table 4). This means that the contents of soluble salts in the soils studied did not have hazardous effects on protease activity. The protease activities were also significantly positively correlated with microbial C and N, contents of organic matter, total N and Mehlich III extractable K and Mg of the soil. There was temporal variation in the protease activities of the soil, and PRC, HDC and CDC had similar effects on soil protease activity irrespective of the sampling time.

Pascual et al. (1999) and Chakarabarti et al. (2000) observed that the activity of urease, the enzyme that catalyzes the hydrolysis of urea and which has been widely used in the evaluation of changes in soil quality as a result of soil management (Díaz-Marcote and Polo 1995; Gil-Sotres et al. 2005), increased when organic matter was added to the soils. García et al. (1994) and Nogales and Benitez (2007) reported that the application of organic matter increased the supply of nutrients to soil microorganisms and increased enzymatic activities, including urease activity. The increase in soil organic matter content resulting from the application of compost, in addition to the incorporation of stable enzymes contained in the compost (Díaz-Marcote and Polo 1995), also favors the formation of complexes with free enzymes and, as a result, soil enzyme activities increase. Figure 2 shows that the effects of organic matter on the activity of urease differ with the type of organic fertilizer. This result is different from what García et al. (1994) reported. However, Tejada et al. (2006) also found that cotton gin crushed compost and poultry manure affected the soil urease activity differently. Application of SBM resulted in the lowest urease activity of the soil, even lower than that of the CF treatment (Fig. 2). In general, HDC and CDC resulted in a higher urease activity. McCarty et al. (1994) reported that the application of different types of liming materials to soil resulted in soil pH and urease activity differences among treatments. They suggested that the influence of these materials on urease activity in soil largely results from their ability to influence soil pH. McCarty et al. (1994) also showed that urease activity could increase when the soil reaction was neutral and activity was inhibited at pH levels below 5.0. In this study, there was a significant linear positive correlation between the soil pH values and urease activities (Table 4). This study also shows that there was no correlation between soil organic C contents and urease activities. However, in the study of Dinesh et al. (2004) a significantly positive correlation was observed between soil organic C contents and urease activities. Table 4 shows that there are positive significant correlations between urease activities and Mehlich III extractable K, Ca and Mg of the soil (Table 4).

Arylsulfatase is believed to be partly responsible for S cycling in soils. Arylsulfatase activity (Fig. 2) was on average the lowest when receiving SBM treatment and the highest when receiving HDC. Deng and Tabatabai (1996, 1997) and Dinesh et al. (2004) reported that arylsulfatase activity was highly correlated with soil C content. No such relationship has been observed in our study. Although the microorganisms are an important source of this enzyme (Elsgaard et al. 2002) there was no correlation between microbial biomass C (or N) and arylsulfatase activities. However, there were significant linear positive correlations between arylsulfatase activities and soil pH and Mehlich III extractable Ca values (Table 4), suggesting that arylsulfatase activity was dependent on soil reaction.

The activities of soil acid and alkaline phosphatases were affected by different organic fertilizer treatments (Fig. 2). Dick (1992) reported that soil acid phosphatase activities increased with the addition of organic materials. However, our results show that the treatments of HDC and CDC lowered the phosphatase activity compared with that of the CF plots (Fig. 2). There was no significant difference in acid phosphatase activity between the CF and SBM treatments. In general, DCDC-treated and PRC-treated soils had the highest acid phosphatase activity, however, the highest alkaline phosphatase activity occurred with HDC treatment. Although the SBM applied in the SA treatment at the 24th crop lowered the alkaline phosphatase activity, applying CDC before the 25th crop alleviated this effect. Owing to acid in reaction of the soils, the acid phosphatase activity is much higher than that of the alkaline phosphatase for each corresponding treatment (Fig. 2). The same result was reported by Renella et al. (2006). There was a positive significant linear correlation between alkaline phosphatase activities and soil pH, however, a negative correlation exists between acid phosphatase activities and soil pH values (Table 4). Alkaline phosphatase activities were also significantly positively correlated with Mehlich III extractable K, Ca and Mg of soil; however, no such correlation existed between acid phosphatase activities and Mehlich III extractable K, Ca, and Mg (Table 4).

Table 5 Linear correlation coefficients between the extractable heavy metal concentrations and the enzymatic activities of the soils

	Cu	Zn	Cd	Pb	Cr	Ni
Mic-C	–0.6181**	0.0374	0.1196	–0.4538*	–0.2526	–0.2784
Mic-N	–0.5287**	–0.0265	0.2735	–0.5251*	–0.3327	–0.1622
CO2-resp	–0.7271**	0.4583*	0.1166	–0.5914**	0.3962	–0.5973**
dHase	–0.0173	0.2179	0.7805**	–0.2042	–0.1838	0.2685
Gluase	–0.6277**	–0.0335	0.0436	–0.4839*	–0.2460	–0.3245
Protease	–0.4604*	0.1024	0.4801*	–0.4205	–0.3043	–0.0623
Urease	–0.2651	0.8064**	0.8087**	–0.2598	0.2816	0.1044
Nitrifi	0.1200	0.4563*	0.7410**	–0.0883	0.1204	0.2352
aSase	–0.3447	0.6307**	0.3695	–0.3365	0.5451*	–0.0469
aPase	–0.5485**	–0.2435	–0.3603	–0.2934	–0.2052	–0.3629
Pase	–0.2337	0.5209*	0.8315**	–0.3460	0.1221	0.1852
Mic-C, microbial C; Mic-N, microbial N; CO2-resp, respiration rate; dHase, dehydrogenase; Gluase, β-glucosidase; Nitrifi, nitrification; aSase, arylsulphatase; aPase, acid phosphatase; Pase, alkaline phosphatase. *P < 0.05;
**P < 0.01.

Heavy metal contents and enzymatic activities of the soil

Different heavy metals affected the enzymatic activities of the soils differently (Table 5). Higher Cu concentration resulted in lowering the microbial C and N, respiration rate, β-glucosidase, protease and acid phosphatase activities. Respiration rate, urease, nitrification, arylsulfatase and alkaline phosphatase activities were significantly negatively correlated with extractable Zn concentrations. The current extractable Cr and Ni concentrations had no significant effect on most enzymatic activities studied except that Cr on arylsulfatase and Ni on respiration rates could be attributed to the low Cr and Ni levels. However, there are significant positive correlations between Cd levels and the activities of hehydrogenase, protease, urease, alkaline phosphatase and nitrification.

Conclusions

Owing to the low buffering capacity of the Oxisol the pH values of the soil were affected significantly by the organic fertilizers applied. Low Ca content of the organic fertilizers resulted in lowering the soil pH values and resulted in significant effects on the soil enzymatic activities studied, except β-glucosidase and protease. With a high N content and low C/N ratio of the SBM, SBM is not an appropriate organic fertilizer for continuous application to an Oxisol. The effects of different organic fertilizers with different contents of heavy metals on the contents of soil-extractable heavy metals were significant. The contents of Cu and Zn were not high enough, however, despite this, their effects on some soil enzymatic activities were obvious.