Effect of different fertilization modes on soil organic carbon sequestration in acid soils

ABSTRACT

A meta-analysis of 297 treatment data from the Vezaiciai Branch of the Lithuanian Research Centre for Agriculture and Forestry long-term field experiment published from 2006 to 2015 was used to characterize the changes in SOC under different fertilization treatments and residue management practices in Lithuania’s acid soil. A meta-analysis was performed to quantify the relative annual change (RAC) of SOC content and the average RAC rate of SOC under four fertilization modes (farmyard manure (FYM) (40 t ha⁻¹)); alternative organic fertilizers (in the manure background (40 t ha⁻¹)); FYM (60 t ha⁻¹); alternative organic fertilizers (in the manure background (60 t ha⁻¹)) in two soil backgrounds (naturally acid and limed soil). The average RAC under four fertilization modes was 1.46 g kg⁻¹ yr⁻¹, indicating that long-term fertilization had considerable SOC sequestration potential. Incorporation of alternative organic fertilizers in unlimed soil showed negative effects (−0.39 and −0.66 g kg⁻¹ yr⁻¹) in the observed long-term experiment. The RAC in the limed soil with incorporated organic fertilizers (FYM and alternative organic fertilizers), compared to the control, and varied from 0.25 g kg⁻¹ yr⁻¹ in the treatment with incorporated alternative organic fertilizers (in the manure background (40 t ha⁻¹)) to 0.71 g kg⁻¹ yr⁻¹ in the soil with FYM (60 t ha⁻¹). In this study, the average RAC rate of SOC under organic fertilization treatments in limed soil (5.07–6.54%) was longer than organic fertilization in unlimed soil (2.11–3.49%), which might be attributed to the application of organic manure that would result in a slow release of fertilizer efficiency. Our results indicate that the application of manure (40 or 60 t ha⁻¹) showed the greatest potential for C sequestration in agricultural soil and produced the longest SOC sequestration duration.

KEYWORDS:

• Acid soils
• soil organic carbon
• carbon sequestration
• long-term fertilization
• meta-analysis

Introduction

Globally, soil organic matter (SOM) contains more than 2–3 times as much carbon as either the atmosphere or terrestrial vegetation, and is central to sustainable grain production (Schmidt et al. 2011). Soil organic carbon (SOC) determines the functioning of large parts of biogeochemical interfaces in the soil. SOC is an important source of plant nutrients, which stabilizes soil structure and plays a central role in soil surface-atmosphere exchange of greenhouse gases. Because SOM can be associated with different soil chemical, physical and biological processes, it has been widely considered as one of the best soil quality indicators (Grandy and Robertson 2006; Liaudanskiene et al. 2013).

In arable soil, the preservation or improvement of soil quality and productivity is of major importance. Fertilizer application has been widely used as a common management practice to increase soil carbon (C) sequestration and SOC level. The application of organic manures is considered to be an important strategy for the build-up of carbon stock in arid soil. The issue is directly related to maintain SOM quantity, which is a critical component of soil productivity (Powlson et al. 2012). SOC and dissolved organic carbon (DOC) are important indexes for SOM. DOC is a key component of the active SOM pool and serves as a source and sink of soil nutrients (Xu et al. 2012) and/or as an ecological marker to understand soil fertility (Ge et al. 2010). Fertilization could affect the content of DOC, leading to alterations in the formation of complexes between organic ligands and metals, and the SOM sequestration (Yu et al. 2012; Wen et al. 2014). On the other hand, increased decomposition of stabilized material induced by addition of fresh organic material triggering microbial activity can result in higher C losses from soil (Schlüter et al. 2011; Kirchmann et al. 2013). These indices are influenced by agricultural practices (e.g. cultivation, fertilization and irrigation), and fertilization is one of the most important practices affecting SOC turnover, which in turn affects the soil fertility, temperature, moisture and C/nitrogen (N) ratio (Pan et al. 2009; Schmidt et al. 2011).

Over the past few decades the need has risen for comprehensive studies on the soil properties and environmental factors controlling DOC quantity and quality. However, there is limited information available about the effects of different fertilization on the changes of SOC, and the data on the dynamics of SOC in soils are often contradictory, especially in arable soils. Meta-analysis is an effective statistical method to quantitatively summarize the results from a large number of studies and allows general conclusions to be drawn at regional scales (Guo and Gifford 2002).

The aim of this research is to characterize the changes in SOC under different fertilization treatments and residue management practices in Lithuania’s acid soil.

Materials and methods

Study area

The study site is a long-term field experimental established at Vezaiciai Branch of Lithuanian Research Centre for Agriculture and Forestry in 1959. This site is located in a transitional climate (between maritime and continental) region in the western part of Lithuania. The region’s mean annual amount of precipitation is more than 800 mm, the average annual air temperature is 6.7°C. The soil of the experimental site is Bathygleyic Dystric Glossic Retisol (WRB 2014) with a moraine loam texture with clay-sized particles content of 13–15%.

Data sources

Field experiment with a period of 57 years that recorded the response of SOC to changes from control treatment (CK) to fertilization treatments in acid fields in Western Lithuania were selected as the criteria for the meta-analysis in this study. The data included the initial and final content of SOC in the control and different fertilization modes. Farmyard manure (FYM) (40 and 60 t ha⁻¹) and alternative organic fertilizers (wheat straw, rape residues, roots, stubble, perennial grasses) were applied on two soil backgrounds – acid and limed. The experimental design included the following treatments: (1) unlimed soil; (2) unlimed soil + FYM (40 t ha⁻¹); (3) unlimed soil + alternative organic fertilizers (in the manure background (40 t ha⁻¹); (4) unlimed soil + FYM (60 t ha⁻¹); (5) unlimed soil + alternative organic fertilizers (in the manure background (60 t ha⁻¹); (6) limed soil (1.0 rate in 5 years); (7) limed soil + FYM (40 t ha⁻¹); (8) limed soil + alternative organic fertilizers (in the manure background (40 t ha⁻¹); (9) limed soil + FYM (60 t ha⁻¹) and (10) limed soil + alternative organic fertilizers (in the manure background (60 t ha⁻¹). The soil was limed at 1.0 rate to maintain optimal soil pH_KCl (5.8–6.0). The amount of manure applied per rotation amounted to 40 or 60 t ha⁻¹. Alternative organic fertilizers were applied every year. The mineral fertilization in both acid and limed soil was the same: N₆₀P₆₀K₆₀ for cereals, N₆₀P₆₀K₆₀ for barley, N₃₀P₆₀K₆₀ for lupine + oats mixture, N₆₀P₉₀K₁₂₀ for rape, P₆₀K₉₀ for perennial grasses.

A resampling based on 297 samples was used to generate the mean effect size. Each variable included in the meta-analysis was assumed to be a random sample of a relevant distribution of effects, and the combined effect estimates the mean effect in this distribution. If the 95% confidence intervals did not overlap zero, the treatment land use transition was regarded as having significantly different SOC content than the control land use. The meta-analysis was weighted in that each study-wise effect size was weighted by the inverse of its variance.

Methods of analyses

Soil chemical analyses were carried out at the Chemical Research Laboratory of Institute of Agriculture, Lithuanian Research Centre for Agriculture and Forestry. Soil reaction was determined in 1M KCl (soil-solution ratio 1:2.5) using an IONLAB pH meter, mobile aluminium by Sokolov method. Soil total nitrogen (N_tot) was determined by Kjeldahl method, and soil mobile phosphorus (P₂O₅) and potassium (K₂O) content – by Egner-Riehm-Domingo (A–L) method. SOC content was determined by photometric procedure at the wavelength of 590 nm using the UV–VIS spectrophotometer Cary 50 (Varian).

Data analysis

The annual change of SOC content under different fertilization treatments during the experiments was calculated as follows:(1) where AC is the annual change of SOC content (g kg⁻¹y⁻¹); SOC₀ and SOC_t are the SOC contents of the initial and final year during the experiments, respectively; and t is the period of each experiment.

In order to determine the actual effects of different fertilization modes on SOC sequestration, the AC under no fertilizer (control (CK)) was deducted from that under each fertilization treatment (TR) during the experimental period, resulting in a relative annual change (RAC) of SOC content, and was calculated as follows:(2)

The formula to obtain average RAC rate of SOC under different fertilization modes was used as follows:(3) where Y is the average RAC rate of SOC content (%) after fertilization treatment relative to CK.

Statistical analysis

Statistical computations were performed using the statistical software package One-way analysis of variance (ANOVA). ANOVA was used to analyse the differences in the tested parameters among the treatments. The least significant difference method (LSD) at the 5% probability level was used to test the significance of differences between treatment means.

Results and discussion

In the past decades raised the need for comprehensive studies on the soil properties and environmental factors controlling SOC quantity and quality. SOM accumulation is a slow process and considerably slower than the rate of decline.

Fortunately, accumulation can be enhanced by positive land management techniques, such as green manures and applications of FYM. The application of organic amendments to soil has beneficial effects, mainly because such amendments supply organic matter and other nutritive elements to the soil–plant system (Purakayastha et al. 2008; Powlson et al. 2012).

There was found a positive statistically significant effect of fertilization on SOC amount in the soil (Figure 1). SOC amount was 1.44% for the non-fertilized treatment and in fertilized treatments it varied from 1.58 to 1.75%. SOC content in the fertilized plots was thus approximately by 0.1–0.3 percentage points higher compared to the unfertilized plots. In the limed soil (1.0 rate to maintain optimal soil pH), the content of SOC increased by 0.12 percentage points compared with the CK. The highest amount of SOC (1.67% and 1.75%) was obtained in the limed soil applied with FYM at both manuring rates (40 and 60 t ha⁻¹). These trends could attribute to greater C inputs through the input of manure and root biomass due to better crop growth. Long-term applications of animal manure plus mineral fertilizers (liming) increase SOC in two ways: by adding organic matter content and by increasing organic matter in crop residues due to higher crop yields (Kundu et al. 2007). Similar results have been reported from other long-term (10–100 years) fertilizations in cropland soils (Blair et al. 2006; Kundu et al. 2007; Purakayastha et al. 2008; Giacometti et al. 2013). The fertilization with alternative organic fertilizers generally showed relatively less effect on SOC accumulation. The SOC content in these treatments was approximately by 0.04–0.08 percentage points lower compared to the treatment where FYM was incorporated. The lack of response of SOC accumulation might be due to a lower C input, to low decomposition of added straw or to an induced SOC decomposition through the straw return (Ju et al. 2009).

Figure 1. The effect of organic fertilizers on SOC amount (%) in the soil, which has been treated as follows: (1) unlimed and limed (1.0 rate in 5 years) soil; (2) FYM (40 t ha⁻¹); (3) alternative organic fertilizers (in the manure background (40 t ha⁻¹); (4) FYM (60 t ha⁻¹); (5) alternative organic fertilizers (in the manure background (60 t ha⁻¹). *Differences significant at 95% probability level.

The assessment of soil chemical composition and the investigation of condition of acid and fertilized areas are very important. Application of manure and alternative organic fertilizers not only added the valuable nutrients but also improves soil physical environment for crops to better take up water and nutrients from the soil (Giacometti et al. 2013). The main soil chemical properties influencing crop productivity and soil fertility are soil acidity and the amount of mobile aluminium. The greatest soil acidity neutralizing effect resulted from the combination of manuring and liming, when after incorporation FYM and alternative organic fertilizers in limed soil, pH increased by 2.00 and 1.95 units, respectively (Table 1).

Table 1. The effect of manuring and liming combination on topsoil chemical properties, 2011–2013.

Treatments	pH KCl	Mobile Al (mg kg^−1)	Total N (g kg^−1)	Mobile P2O5 (g kg^−1)	Mobile K2O (g kg^−1)
(1) Unlimed	4.1	88.4	1.21	152.8	106.5
(2) Unlimed +  FYM (40 t ha^−1)	4.3	43.6*	1.26	180.6*	118.9*
(3) Unlimed + alternative organic fertilizers (in 40 t ha^−1 FYM background)	4.2	27.5*	1.13*	155.2	114.0*
(4) Unlimed +  FYM (60 t ha^−1)	4.4*	6.59*	1.43*	202.0*	140.1*
(5) Unlimed + alternative organic fertilizers (in 60 t ha^−1 FYM background)	4.3	49.6*	1.33*	165.2*	143.7*
(6) Limed (1.0 rate in 5 years)	5.8*	0.00*	1.27*	143.8*	105.2
(7) Limed +  FYM (40 t ha^−1)	6.1*	0.00*	1.28*	192.0*	126.0*
(8) Limed + alternative organic fertilizers (in 40 t ha^−1 FYM background)	5.9*	0.00*	1.17*	146.0	116.0*
(9) Limed +  FYM (60 t ha^−1)	6.1*	0.00*	1.43*	220.5*	151.3*
(10) Limed + alternative organic fertilizers (in 60 t ha^−1 FYM background)	5.9*	0.00*	1.28*	161.7*	123.3*

*Differences significant at 95% probability level.

The pH is an important chemical factor for the solubility and production of DOC and the relationship between pH and DOC is generally thought to be a complex one, partly because of the influence on charge density of the humic compounds, partly because of stimulation of the microbial activity. In our study we found that the amount of DOC depended on soil pH (Figure 2). The lowest amounts of DOC (0.15–0.16 g kg ⁻¹) were determined in the soil, which pH varied from 4.0 to 2.0. The amount of DOC in soil increased with increasing soil pH. The highest amount of DOC was determined at the soil where pH varied from 5.9 to 6.1. Similar consistent patterns on relationships between soil pH and DOC concentrations have been identified in others studies. This relationship could be attributed to differences in decomposition rates (higher at elevated pH), differences in DOC sorption to the soil complexes and complex formation with aluminium, and differences in DOC quality (phenol content lower at high pH and therefore more readily decomposable DOC) (Kemmitt et al. 2006; Löfgren and Zetterberg 2011).

Figure 2. Changes of DOC content (g kg⁻¹) in soil at different pH levels. External graph axis displays different pH levels, internal graph axis – DOC content (g kg⁻¹) in soil.

There is also a possibility that not only pH but also the N status influences DOC amount in soil, for either biological or physicochemical reasons. Furthermore, N limited sites may respond differently to sites where N availability is already high. Possible effects as a result of altered enzyme activities include however both changes in production of DOC and changes in mineralization of DOC (Zak et al., 2011). In the present study, the DOC content in soil was increased by increasing soil N. The limed and FYM-applied soil had a higher nitrogen (1.43 g kg⁻¹) content compared to the other treatments (Table 1). In this treatment we also determined the highest DOC amount. It is possible that increase of total N content favour the microbial degradation of both DOC and solid organic matter, the latter resulting in the production of DOM (Filep and Rékási 2011).

We conducted a meta-analysis to quantify the RAC of SOC content and the average RAC rate of SOC under four fertilization modes and to integrate the results of previous studies that examined changes in SOC due to fertilization. The average of RAC of SOC content (RAC) under four fertilization modes was 1.46 g kg⁻¹ yr⁻¹, indicating that long-term fertilization had considerable SOC sequestration potential (Figure 3). Incorporation of alternative organic fertilizers in unlimed soil showed negative effects (−0.39 and −0.66 g kg⁻¹ yr⁻¹) in the observed long-term experiment. The RAC in the limed soil with incorporated organic fertilizers (FYM and alternative organic fertilizers) varied from 0.25 g kg⁻¹ yr⁻¹ in the treatment with incorporated alternative organic fertilizers (in the manure background (40 t ha⁻¹)) to 0.71 g kg⁻¹ yr⁻¹ in the soil with FYM (60 t ha⁻¹), which was similar to the result determined by Zhu et al. (2012). It is worth to note that a higher C addition with manure produced significant increase of sequestrated C. An explanation would be that the addition of FYM with mineral fertilizers could promote the formation of micro-aggregates in macro-aggregates, leading to more particulate organic matter fixation in a newly formed micro-aggregate. Moreover, compared to the large aggregates, soil micro-aggregates have a lower turnover rate and higher stability (Tian et al. 2015).

Figure 3. The effect of different fertilization modes on the RAC of SOC content (g kg⁻¹ yr⁻¹) in soil which has been treated as follows: (1) control; (2) FYM (40 t ha⁻¹); (3) alternative organic fertilizers (in the manure background (40 t ha⁻¹); (4) FYM (60 t ha⁻¹); (5) alternative organic fertilizers (in the manure background (60 t ha⁻¹).

SOC sequestration is not infinite but will reach saturation after decades with the implementation of various fertilization modes. Carbon sequestration duration is synonymous with the time it takes for the soil C to reach a stable level. In this study, the average RAC rate of SOC (Y) under organic fertilization treatments in limed soil (5.07–6.54%) was longer than organic fertilization in unlimed soil (2.11–3.49%), which might be attributed to the application of organic manure that would result in a slow release of fertilizer efficiency (Figure 4). Our results indicate that the application of manure (40 or 60 t ha⁻¹) have further potentials to sequestrate considerable C, suggesting the significance of more organic C input from manure could build-up SOC pool in our study soils. These consistencies indicated that the average RAC rate of SOC summarized in this research could be used as an index to assess the SOC sequestration potential of the four fertilization modes in acid soils in Lithuania. The results of this research provide us with the clear information on the characteristics on SOC change, which could be very useful to guide climate policy and cropland fertilization strategies.

Figure 4. The effect of different fertilization modes on the average RAC rate of SOC (%) in soil which has been treated as follows: (1) FYM (40 t ha⁻¹); (2) alternative organic fertilizers (in the manure background (40 t ha⁻¹); (3) FYM (60 t ha⁻¹); (4) alternative organic fertilizers (in the manure background (60 t ha⁻¹).

Disclosure statement

No potential conflict of interest was reported by the author(s).