Segun O. Oladele , Adewole T. Adetunji

a Department of Agronomy, Adekunle Ajasin University Akungba Akoko, Ondo State, Nigeria

b Department of Crop, Soil and Pest Management, Federal University of Technology Akure, Ondo State, Nigeria

c Department of Agriculture, Cape Peninsula University of Technology, Western Cape, South Africa

Keywords:Global change Sub-saharan Africa Greenhouse gas Rainfed croplands Environment

ABSTRACT Scarce carbon dioxide (CO2) emission data, soil organic carbon (Corg) and labile Corg pool index for croplands in sub-Saharan Africa (SSA) limits our ability to define regional CO2 flux and mitigation prospects provided by biochar amendment. Hence, we conducted a two-year field scale study quantifying effects of biochar(12 t ha-1),biochar(12 t ha-1)and N fertilizer(90 kg N ha-1),N fertilizer(90 kg N ha-1) alone and control (no amendment) on Corg, Corg pools, lability index (LI) and CO2 emissions of a humid tropical alfisol under rainfed rice cultivation.Results clearly demonstrate that biochar amendment combined with N fertilizer significantly increased Corg, particulate organic carbon, and particulate organic carbon to organic carbon proportion.Furthermore,this combination decreased dissolved organic carbon,dissolved organic carbon to organic carbon proportion,easily oxidizable carbon,easily oxidizable carbon to organic carbon proportion, microbial biomass carbon, microbial biomass carbon to organic carbon proportion, LI, and CO2 emission when compared to other treatments. Our findings established relationship between labile Corg pools and CO2 emission. Overall, our study suggest that application of biochar and N fertilizer addition can decrease CO2 emission from SSA croplands under rainfed rice cultivation, increase essential labile Corg pools, protect labile Corg pools sensitive to losses and sequester Corg with positive implication for global change mitigation and soil quality improvement.

1. Introduction

Soil is widely regarded as the second largest sink of organic carbon (Corg) after the ocean (Stockmann et al., 2013; IPCC, 2014).Anthropogenic induced change in Corgstock could lead to significant contribution of CO2, a potent greenhouse gas (GHG) of global warming and atmospheric CO2accumulation.Agricultural systems in the tropics, sub-tropical and temperate regions have been reported to be a major source of GHG concentrations in the atmosphere (Hütsch, 2001). Thus, it becomes imperative to adopt or implement Corgstorage or sequestration strategies that can negate GHG emissions and sustainably mitigate climate change. In agricultural soils,Corgcontent is highly dependent on the addition and decomposition rates of exogenous organic matter (OM). The addition of stabilized exogenous OM can potentially moderate Corgsequestration or turnover rates in soils.

Biochar is a carbon-laden product derived from the thermal degradation of organic biomass in the absence or presence of little oxygen intended for soil amendment(Oladele,Adeyemo,Awodun,Ajayi, & Fasina, 2019). Its recalcitrant nature has the propensity of enhancing Corgsequestration and reducing GHG emissions (e.g.,CO2fluxes) from agricultural soils (Chavez et al., 2012; Lehmann,2007; Li et al., 2014, 2019; Stewart et al., 2013). With the foregoing, a management system promoting biochar soil technology has been proposed as a potential strategy to sequester Corg, while mitigating soil atmospheric fluxes of CO2(Liu et al.,2016).Nitrogen(N)fertilizers(such as urea)is another alternative soil amendment to increase crop yield and promote soil Corgsequestration (Malhi et al., 2011; Silveira et al., 2013) which will in turn enhance Corgpools and influences nutrients availability and turnover (Schmidt et al.,2015).However,the combined application of N fertilizer and stabilized OM has been shown to improve Corgbetter than the single addition of N fertilizer (Hai et al., 2010; Sun et al., 2015).Positive(Hyv¨onen et al.,2008;Wang et al.,2015),negative(Manna et al., 2007), and neutral effects (Chen, Liu, Deng, Yan, & Zhang,2012; Zhou et al., 2013; Sun et al., 2015) of N fertilizer addition on soil Corgsequestration have been reported. These contrasting reports may be explained by the fact that increased Corginput resulting from N fertilizer may be counteracted by soil C loss from various fractions, resulting in net negative sink of Corg(Hai et al.,2010; Fr¨oberg et al., 2013). Biochar on the other hand, owing to its stability, recalcitrant C fraction, pores and absorptive surface when combined with N fertilizer may protect Corginput in various pools thereby increasing Corgaccumulation and storage(Zimmerman et al., 2011). Conversely, Liu et al., (2016) in a metaanalysis reported a 45% increase in soil CO2fluxes when biochar was combined with N fertilizer, while Iqbal et al. (2009) in a subtropical Ultisol and Zhang et al. (2012) in a short term calcareous loam soil studies observed a total decrease in CO2flux. These contrasting results require that further insights are provided on soil Corgdynamics in biochar and N fertilizer amended soils particularly in tropical climate/Alfisols. Generally, soil Corgturnover is the interaction between exogenous OM inputs (e.g. crop residues,biochar) and decomposition of Corgwith positive, negative, or no reaction to N fertilizer additions (Huang et al., 2011). Thus, understanding the influence of biochar and N fertilizer on changes in soil Corgpools and the turnover rate of Corgis vital to revealing Corgdynamics in amended humid tropical Alfisol

Agricultural soil Corgis mainly composed of two major pools: a labile and a stabilized pool. Labile Corgare the pools with high activity and sensitive to plant interactions, microbes and highly susceptible to rapid turnover or decomposition(Chen et al.,2010).On the other hand,the stabilized pool has longer turnover period and less affected soil management practices. Labile Corgpools are further sub-divided into the following fractions; dissolved organic carbon (DOC), easily oxidizable organic carbon (EOC), microbial biomass carbon (MBC), and particulate organic carbon (POC) (Xu et al., 2011). The labile Corgpool plays an important role in soil Corgdynamics (Liang et al., 2011) and serves as useful indices of changes in soil Corgstability and nutrient turnover following soil management practices(Fierer et al.,2009).However,the influence of biochar amendments on soil labile Corgpools remains unclear and contradictory. For instance, labile Corgpools were reported to be increased in biochar amended soils than in un-amended soils(Biederman & Harpole, 2013; O’Neill, 2007; Steiner et al., 2008).While,Thies and Rillig(2008),&Lu et al.,(2014)reported negative and neutral effects respectively.The inconsistency in these findings begs the need for further studies. Furthermore, Blair et al. (1995)developed the lability index tool to quantify Corgdynamics according to changes in Corgand EOC content as affected by soil management practices. Evaluating changes in Corgcontent after biochar amendment cannot be determined without assessing labile Corgpools or lability index. This is necessary in order to quantify extent of Corgsequestration or fluxes from soils. However, scanty literature exists documenting the impact of biochar on Corglability index particularly in tropical alfisols. In addition to understanding the effects of biochar on soil Corgpools, there is a need to comprehend whether biochar amendment will enhance or reduce CO2fluxes (Luo et al., 2011). Though some laboratory studies and field experimental findings have reported increase in soil CO2fluxes following biochar amendment (Jones et al., 2011; Zheng et al.,2012), while neutral effects of biochar or decrease in soil CO2fluxes have been reported in other studies (Karhu et al., 2011; Lu et al., 2014). The underlying mechanisms behind some of these results mostly confined within temperate latitudes are yet to be fully understood while similar studies in tropical climate/soil conditions particularly in SSA are scarce. Furthermore, the dearth of Corgcontent, labile Corgand CO2flux data from biochar and N fertilizer amended SSA croplands limits current approach to quantify regional and global CO2flux and mitigation strategies.

Alfisols with a considerable level of native fertility and base saturation are the dominant soil order for crop production in southwest Nigeria (IUSS Working Group WRB, 2014). In recent years,the productivity of this soil has been declining as a result of unsustainable management practices. Continuous crop residue removal by famers for energy generation or as livestock feed andinsituburning of harvest residues/straws in preparation for next planting season has resulted in a decline of Corgpools,depletion of Corgstocks and soil degradation.Therefore,combination of biochar and inorganic fertilizers applied to croplands has been proposed to help improve soil fertility, Corgconcentration and increase crop yield(Oladele,Adeyemo,&Awodun,2019).Thus,a field study was set up in 2016 to assess the effects of biochar and N fertilizer on changes in Corgconcentration as well as its labile pools and soil CO2emissions. We hypothesized that biochar and N fertilizer addition will consistently reduce CO2flux, increase Corgconcentration and labile Corgpools on medium to long term basis.The objectives of the study were to determine the effects of biochar amendment and N fertilizer addition over a two year period on:(1)Corgconcentrations and soil CO2emissions;(2)soil labile Corgpools including POC,DOC,EOC, MBC, lability index and their relationships with Corgcontent and CO2emissions.

2. Materials and methods

2.1. Biochar and fertilizer application

The biochar used in this study was produced by slow-pyrolysis of rice husk under oxygen-limited conditions at Biochar research unit, Department of Crop Soil and Pest Management Federal University of Technology Akure, Nigeria. Pyrolysis of rice husk was performed using a biochar reactor at 350-400ᵒC; 51% of the rice husk biomass was converted to biochar. After pyrolysis, biochar samples were collected and cooled to room temperature. The moisture content of rice husk biochar was measured after drying at 105°C for 24 h and pH was determined in a 1:5 (solid/water) solution. Briefly, properties of biochar are stated-total carbon (C),Nitrogen (N), Phosphorus (P), Potassium (K), Calcium (Ca), Magnesium (Mg), Sodium (Na), Copper (Cu), Iron (Fe), Zinc (Zn), Manganese (Mn) contents of rice husk biochar were 511 g kg-1,0.3 g kg-1and 0.73 mg kg-1, 0.92 cmol kg-11.25 cmol kg-1,4.50 cmol kg-1, 0.95 cmol kg-1, 226.5 mg kg-1, 48 g kg-1,561.5 mg kg-1, 332 mg kg-1respectively, and the pH (H2O) of the biochar was 8.50. The biochar was grounded, passed through a 2 mm sieve and then thoroughly mixed to obtain a fine homogeneous material that would mix uniformly with the soil upon incorporation.

2.2. Field experiment

Experimental location and climatic conditions were earlier reported in Oladele,Adeyemo,Awodun,Ajayi,and Fasina(2019).Two year data for mean monthly ambient air temperature(°C)and total precipitation pattern (mm) are presented in Fig.1. The soil at the study site is a sandy clay loam(sand-68.8%,clay-25.1%and silt-6.1%) isohyperthermic Typic Oxic Paleustalf Alfisol. The basic physico-chemical properties of the topsoil (0-15 cm) prior to the start of the experiment are as follows:pH(H2O)-4.9,bulk density-1.63 g cm3, SOC - (3.7 g kg-1), available P- (38.06 mg kg-1),exchangeable K (0.17 cmol kg-1), total N (0.41 g kg-1). Four treatments were arranged: CT (no biochar and N fertilizer application), N (applied as urea at 90 kg ha-1), B (rice husk biochar applied at a rate of 12 t ha-1) and B + N (combined application of biochar at 12 t ha-1and N fertilizer at 90 kg ha-1).Mineral P(P2O5-30 kg ha-1as triple super phosphate and K (K2O - 30 kg ha-1as muriate of potash) fertilizer was applied equally to all treatments.Rice husk biochar was applied once, manually spread on the soil surface and incorporated into the soil by rotary hoe before fertilizer application in July 2016.Mineral P and K fertilizer was broadcasted and incorporated into all treatments before sowing, while N fertilizer was applied in two splits at designated treatments in each year respectively. Rice was sown and harvested between August and December in both years (2016 and 2017) respectively. Two weeks after seedling establishment, the rice seedlings were thinned to approximately 160,000 plant population per hectare. Three replicated plots were established for each treatment in a randomized complete block design. Each plot was 12 m2in area(3m2×4m2)and plots were separated by a 1 m2alley on each side.

Fig.1. Mean monthly ambient air temperature[°C(line graph)]and total precipitation[mm(bar graph)]in 2016 and 2017.(retrieved on January 5th,2018 from:http://climate-data.org/historical_data).

2.3. CO2 emission measurement

The CO2emissions were estimated by the widely used static chamber method.Briefly,we set up a transparent plastic cylindrical shape chamber with interior diameter and height dimensions of 15 and 25 cm, respectively. The bottomless chamber at the base was driven deep into the soil. The interior layer of the chamber was covered with aluminum foil to maximize air temperature within the chamber during gas sampling. After rice sowing, gas sampling was conducted thrice a week between 9:00 a.m. and 11:00 a.m.However, sampling time and frequency were sometimes adjusted based on precipitation pattern. CO2fluxes were determined from the slope of the mixing ratio change based on three gas samples collected at intervals of 10 min over the course of 30 min after chamber closure. Thereafter, 50 mL gas sample was collected and stored in an air vial for analysis not later than 24 h after sampling.Concentrations of CO2in gas samples were determined with a gas chromatograph equipped with a flame ionization detector (FID)(Agilent 7890A, Agilent Technologies, USA). The air temperature inside the chamber was also measured during sampling.Chambers were removed after CO2sampling terminated in mid December in both years.In the second year of study(2017),static chambers were reinstalled at the same geo-referenced sampling spot as that of 2016 in each experimental treatment.

CO2emission rates were calculated as follows:

Where,K is the CO2emission flux,p is the density of CO2at 0°C,h is the height of the closed chamber, dc/dt is the change of CO2concentration per unit in the closed chamber, and T is the temperature(°C) of the closed chamber during sampling.

Cumulative soil CO2emissions during the growing season of each year were calculated following equation derived by:

where M is the cumulative soil CO2emission,Ki+1andKiare soil CO2emission rate at the (i + 1)th and ith sampling dates, ti+1-tiis the interval days between ith and(i+1)th sampling dates,and n is the total sampling times during the growing season in a year.

2.4. Soil sampling and Corg pool analysis

Five core samples of topsoil (0-20 cm depth) from each treatment were collected and homogenized into one sample at the end of each rice growth season,sealed in nylon bags and transferred to the laboratory for analysis.Plant roots and materials were removed from soil samples, while a portion was stored at 4ᵒC for MBC determination and the remaining sample was air-dried. Soil samples were ground to pass through a 2 mm sieve prior to analysis.Subsamples were subjected to further grounding and passed through a 0.15 mm sieve for determination of Corg. The MBC was estimated from the increase in Corgin the 0.5 mol L-1potassium sulfate(K2SO4)extracts of chloroform fumigated soils compared to unfumigated extracts, and a calibration factor of 0.45 was used to calculate MBC (Vance et al., 1987). The DOC in each sample was determined as described by Zhu et al. (2015).10g of air dried soil was weighed into a 250 mL conical flask and shaken with 25 mL distilled water for 1 h at 250 rpm and then centrifuged for 10 min at 4000 rpm. The supernatant solution was filtered with a 0.45 μm filter into separate vials for analysis. The filtrate was measured using a carbon and nitrogen analyzer (Carlo Erba-1108 elemental analyzer).POC was determined according to protocols described by Camberdella and Elliott(1992).10 g of air dried soil were placed in a 250 mL conical flask, dispersed with 30 mL of Calgon (sodium hexametaphosphate) solution and consistent horizontal shaking.The entire dispersed soil sample was transferred into a 52 μm sieve and washed with distilled water to allow for separation,Corgin the suspended fraction represents POC. Thereafter, POC fraction was oven dried at 60ᵒC for 48 h, then weighed and ground to pass through a 0.05 mm sieve. EOC was determined using Blair et al.(1995) protocols. 10g of finely ground air-dried soil sample was weighed into 50 mL tubes and 25 mL 333 mM potassium permanganate(KMnO4)was added.The tubes were shaken for 1 h,and then centrifuged for 5 min.Following dilution with distilled water,supernatant solutions was measured using spectrophotometer at an absorbance of 565 nm. Soil bulk density was determined according to methods described by Page et al. (1982). Soil pH was determined using a glass electrode pH meter in 1:2.5 soil:water(w/v). Available P using the Bray-1 method (Bray & Kurtz,1945); Corgaccording to Walkley and Black(1934)method;total N content by the Kjeldahl digestion(Nelson&Sommers,1982)and exchangeable K using ammonium acetate method (Black,1965).

2.5. Lability index calculation

Based on changes in the proportion of the soil EOC,the C lability(L) was calculated according to Blair et al. (1995) using Eq. (3):

Based on changes in C lability between the control soil and treatments, the lability index (LI) was determined using Eq. (2):

2.6. Statistical analysis

Two way analysis of variance using General Linear Model(GLM)used to determine significant interactions between treatments and years of study on Corgpools and CO2efflux. One-way analysis of variance was then used to analyze differences in the tested parameters among the treatments. Tukey’s HSD test at the 5%probability level was used to test the significance of differences between treatment means.Correlations between soil Corgpools and CO2-C flux was determined by Pearson linear correlation coefficients at p ＜0.01 and p ＜0.05, respectively. All statistical analysis was carried out using MINITAB® version 17.0.

3. Results

3.1. Soil Corg

Individual factors of years of sampling and treatments significantly (p ＜0.05) affected Corgcontent (Fig. 2). Conversely, we did not detect two way interaction effects amongst both factors(Table 1).In the first year,Corgcontent significantly increased in the order of CT ＞N ＞B+N ＞B.Treatment B had the highest Corgcontent which was 60% higher when compared to CT. After two years of study, Corgcontent in treatment B + N was significantly increased by 70%,15%and 39%in contrast to CT,B and N respectively.Increase in Corgcontent followed a trend in the order of CT ＞B ＞N ＞B + N(Fig. 2).

3.2. Soil EOC

Individual factors of year of sampling and treatments significantly(p ＜0.05)affected soil EOC content in 2016 and 2017(Fig.3).However, two way interaction effects were non-significant amongst the two factors (Table 1). In the first year, EOC values ranged from 1.80 to 3.00 g kg-1, while values ranged between 1.5 and 3.5 g kg-1in the second year.Treatment N significantly had the highest EOC values after two years. High EOC values (3.00 g kg-1)recorded in N in the first year was 40%,33%and 27%higher than CT,B+N and B treatments respectively.Similar trend was repeated in the second year,treatment N significantly recorded high EOC values(3.50 g kg-1)which was 57%,29%and 34%higher than treatments CT, B + N and B, respectively(Fig. 3).

Fig.2. Influence of soil amendments on soil organic carbon.Different lowercase letters across columns indicate differences (p ＜0.05) between amendments in each year respectively,and different capital letters in a single column indicate differences of the same amendment in different years.Bars represent standard errors(n=3).CT:no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

3.3. Soil POC

The effects of different treatments on soil POC content across two years of study are shown in Fig. 4. Years of sampling had no significant (p ＜0.05) effect on POC content, while treatments showed a reverse trend; however the two factors showed no interaction effect (Table 1).POC values fluctuated between 1.5 and 3.00 g kg-1in the first and second year respectively. B + N consistently had the highest POC content(3.00 and 3.00 g kg-1)in the first and second year of study. POC values in B + N were significantly enhanced, as it recorded 50%, 33% and 16% increase over CT, N and B in 2016. Furthermore, B + N repeated similar increment in 2017 by 50%,50%and 16%in comparison to CT,N and B respectively(Fig. 4).

Table 1 Two-way ANOVA for the effects of Amendments(A),year of sampling(Y)and their interaction on CO2 emissions and soil organic carbon(Corg),easily oxidizable organic carbon(EOC),dissolved organic carbon(DOC),microbial biomass carbon(MBC)and particulate organic carbon (POC).

Fig. 3. Influence of soil amendments on soil easily oxidizable carbon. Different lowercase letters across columns indicate differences(p ＜0.05)between amendments in each year respectively, and different capital letters in a single column indicate differences of the same amendment in different years. Bars represent standard errors(n = 3). CT: no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

Fig. 4. Influence of soil amendments on particulate organic carbon. Different lowercase letters across columns indicate differences (p ＜0.05) between amendments in each year respectively, and different capital letters in a single column indicate differences of the same amendment in different years. Bars represent standard errors(n = 3). CT: no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

Fig.5. Influence of soil amendments on dissolved organic carbon. Different lowercase letters across columns indicate differences (p ＜0.05) between amendments in each year respectively,and different capital letters in a single column indicate differences of the same amendment in different years.Bars represent standard errors(n=3).CT:no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

3.4. Soil DOC

The biochar treatments affected DOC contents as shown in Fig. 5. Year of sampling and treatments significantly (p ＜0.05)affected DOC content individually, while interactions between the two factors were not significant (Table 1). DOC was observed to have a large range of values ranging from 90 to 150 mg kg-1in the first year and 85-110 mg kg-1in the second year.In comparison to CT,B+N and B,DOC content was significantly higher in treatment N(150&110 mg kg-1)in 2016 and 2017,respectively.DOC content was 40%and 26%higher in N than CT and B+N respectively in the first year.Similar trend occurred in the second year as DOC content in N increased by 23% and 14% when compared to CT and B + N,respectively.

3.5. Soil MBC

Two-way ANOVA indicated that MBC content was significantly(p ＜0.05) enhanced by treatments irrespective of the year of sampling(Table 1).MBC content was significantly increased in the order of CT ＞N ＞B ＞B+N and values ranging from 115 to 349 mg kg-1in the first year.Compared to the first year,MBC content generally decreased across all treatments in the second year. However,treatment N had the highest MBC content (255 mg kg-1), and significant increase in MBC content was in the order of CT ＞B ＞B+N＞N.Overall,MBC values in the second year fluctuated between 100.4 and 255 mg kg-1(Fig. 6).

3.6. Soil CO2-C emissions

Two-way ANOVA indicate significant interaction between amendments and years of sampling on cumulative CO2-C emissions(Table 1). Cumulative CO2-C emission was significantly (p ＜0.05)different amongst treatments after two years of study and consistently higher in treatment N (Fig. 7). In the first year, CO2-C emission significantly increased in the order: CT ＞B ＞B + N ＞N, with values ranging from 2500 to 4000 kg CO2-C ha-1.Treatment N had the highest CO2-C emission which was 38%,25%,and 13%higher in comparison to CT, B and B + N respectively. In the second year,cumulative CO2-C emission in B + N and B were markedly decreased by 40% and 7% in comparison to N and CT, respectively.On the other hand, CO2-C emissions increased in 2017 by 5% in treatment N when compared to initial values recorded in 2016.Generally,the mean cumulative CO2-C emissions after two years of sampling in treatments CT,N,B+N and B were:2590,4100,3000 and 2750 kg CO2-C ha-1, respectively.

Fig.6. Influence of soil amendments on microbial biomass carbon.Different lowercase letters across columns indicate differences (p ＜0.05) between amendments in each year respectively,and different capital letters in a single column indicate differences of the same amendment in different years.Bars represent standard errors(n=3).CT:no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

Fig. 7. Influence of soil amendments on soil CO2-C emission. Different lowercase letters across columns indicate differences(p ＜0.05)between amendments in each year respectively,and different capital letters in a single column indicate differences of the same amendment in different years.Bars represent standard errors(n=3).CT:no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

3.7. Proportion of labile Corg pools to Corg

The proportions of labile Corgand Corgpools are presented in Table 2. The mean of data collated over two years across each treatment was used in deriving these proportions. POC/Corg, was significantly(p ＜0.05)higher in CT in comparison to N,B+N and B.DOC/Corgfollowed similar trend as increase recorded were in the order:CT ＞N ＞B+N ＞B.CT and N were significantly(p ＜0.05)larger when compared to B + N and B. MBC/Corgwas significantly different across treatments. However, treatments N and B were significantly higher in comparison to CT and B + N. EOC/Corgwas significantly higher in CT and N when compared to B + N and B.

Table 2Proportions of labile organic carbon pools dominating soil organic carbon.

Fig. 8. Influence of soil amendments on organic carbon lability index. Different lowercase letters across columns indicate differences(p ＜0.05)between amendments in each year respectively, and different capital letters in a single column indicate differences of the same amendment in different years. Bars represent standard errors(n = 3). CT: no soil amendment-control, N: N fertilizer alone, B + N: combination of biochar and N fertilizer, B: sole application of biochar.

3.8. Lability index

The lability (L) and lability index (LI) of amended soils as affected by treatments are presented in Fig.8.Lability index is a tool developed to predict fertility status and quantify changes in Corgcontent in response to soil management practices. LI was consistently and significantly (p ＜0.05) higher in CT and N treatment throughout the two years of study when compared with B+N and B.LI ranged from 0.50 to 1.13 in the first year and 0.26-1.00 in the second year.After two years,LI at soil depth of 0-20 cm decreased in the following order:B+N ＞B ＞N ＞CT.CT had the highest LI,while B + N recorded the lowest LI.

3.9. Relationships between soil CO2-C emission and labile Corg pools

The Pearson correlation coefficients between the different Corgpools and soil CO2-C fluxes are presented in Table 3.After two years of study, soil cumulative CO2-C emission was highly significant(p ＜0.05) and positively correlated with Corg(r = 0.81) and MBC(r = 0.87) concentrations. Furthermore, significant positive correlation was observed between CO2-C and POC (r = 0.75), EOC(r = 0.81) and DOC (r = 0.75) pools.

4. Discussion

4.1. Effects of amendments on soil Corg

Soil Corgcontent was increased in B +N combination after two years of study in comparison to the reference (CT), N and B treatments.We attribute the higher Corgcontent in B+N amended soil to;(i)the greater quantity of C input from biochar application rate(12 t ha-1) (ii) supply of stable, labile and rhizodeposit Corginput stimulated by enhanced rice roots growth (data not presented in this study) as a result of N fertilizer addition and the indirect beneficial effect of biochar on other soil properties (e.g. aggregate size distribution, moisture content, enhanced CEC and nutrient content). Biochar has a direct effect on soil Corgincrease over time due to its intrinsic properties such as relative stability of its C content and the abundance of aromatic organic compounds which are resistant to degradation by soil microbes (Downie et al., 2009;Sohi et al.,2010).Furthermore,it’s labile C content,though readily decomposed by soil microbes coupled with N fertilizer additive effects on rice growth which in turn stimulated vigorous Corgexudates from below ground biomass of upland rice plant could have greatly contributed to the increased Corgcontent. We argue that biochar’s absorptive and protective surface could have contributed to the increase in Corgcontent by protecting available labile Corginputs from decomposition by soil microbes,counteracting possible N fertilizer priming effect on native Corgand indirect altering of microbial enzymatic activity involved in carbon decomposition process as observed in the first year of study. Similar findings as observed in our study was reported by Halvorson et al. (1999); Lu et al. (2009) & Liu et al., 2016 in a meta-analysis who all found that biochar co-applied with N fertilizer increased Corgcontent and was attributed to N fertilizer stimulatory effect which greatly enriched soil Corgpools. Nonetheless, these studies stated that increases in Corgcontent are dependent on N fertilizer source. However slight reduction in the second year for Corgvalues in treatment B could be a result of surface chemistry oxidation of incorporated biochar particles after two years, thus, creating hydroxyl and carboxylic functional groups that binds soil particles, clays minerals,Corgand micro-aggregates and creating macro-aggregates which protected soluble total Corgfrom decomposition (Jien & Wang,2013).

Table 3 Correlation coefficients between labile Corg pools and soil CO2-C emission.

4.2. Effects of amendments on soil POC

Findings from our study clearly showed that B+N combination increased POC pool and POC/Corgproportion at the soil depth of 0-20 cm when compared to other amendments. However, we admit that protocols used for POC determination in our study could plausibly cause a rise in POC levels via artificial intrusion due to fixed carbon content in biochar particles. Significant positive correlations recorded between Corgand POC suggests a strong relationship between the two components (Table 3). This implies that POC increase in B + N treatment was mediated by increase in Corgcontent.Increased POC could be attributed to the huge application rate and volume of disintegrated particles of biochar bound with soil minerals serving as sinks of labile Corg, and nutrient reservoir which are important sources of food and energy for soil microbes.Furthermore,combination of B+N contributed to POC increase via enhanced rice root and shoot growth (not reported in this study)induced by biochar’s porous medium and nitrogen availability from N fertilizer. High application rate of biochar combined with N-fertilizer will improve soil aggregates since available nitrogen can stimulate biochar’s mineralization process, thus enhancing the interaction of biochar particles with soil minerals.This will in turn induce biochar’s cementation of soil aggregates,thereby increasing POC/Corgproportion and POC. Studies conducted by Burrell et al.(2016) affirm our argument that biochar does enhance soil aggregate formation and stability which helps protects POC from quick decomposition by soil microbes.The current result is in agreement with the report of Tian et al. (2016) & Simansky et al. (2018) who observed that biochar amendment at high application rate and N fertilizer addition directly increased POC pools in long term field experiments. It is important to note that POC pools are highly sensitive to soil management practices and could be used as predictors of changes in Corgsequestration and changes in soil health.

4.3. Effect of amendments on soil DOC

DOC plays a vital role in soil Corgmineralization and nutrient cycling, reflects Corgstability (Zhao et al., 2008) and an important source of Corglosses(Gielen et al.,2011).It is active in the transport and transformation process of Corgand nutrients (Chen and Xu,2008), indicator of Corgavailability to soil microbes(Li et al., 2010,2017), highly sensitive to seasonality, soil types, amendments, fertilizer addition and land use.Findings from our study showed that B + N and B amendments markedly decreased DOC pools while N fertilizer increased DOC over two years. Though DOC was not significantly correlated with Corg, we however noticed that DOC/Corgproportion was larger in CT and N treatments indicating high tendency of Corglosses. A short term study conducted by Adams et al. (2005) found that N fertilizer addition could increase soil DOC, corroborating results from our study. Further, Bonnett et al.,(2006) also reported that DOC increases with seasonal increase in soil temperature and microbial activity.These findings support our observations in N and CT treatments. We opined that increase in DOC as corroborated by occurred due to the increase in ambient temperature and precipitation pattern recorded during the period of study - (see Fig.1). However, this observation was contrary in biochar amended treatments (B + N and B) as biochar’s intrinsic properties can reduce daytime soil temperature, fluctuations, albedo(i.e.proportion of solar radiation or light reflected by the soil back to the atmosphere) and the reflectance of the soil thereby suppressing ambient temperature influence on DOC accumulation.On the other hand,we posit that microbial decomposition-due to increase in MBC and microbial activity,and biochar’s high sorption capacity and pore sites (Pietik¨ainen et al., 2000) could be responsible for the lower DOC/Corgproportion and DOC content in B + N and B treatments. Furthermore, biochar’s functional groups and cationic elements present on its surface can form complex with DOC thereby reducing measurable soil DOC pools (Cao & Harris,2010). In addition, we refute the possibility of DOC leaching as possible pathway of decreased DOC content in B + N and B treatment due to biochar’s ability to protect Corg,provide habitat for soil biota,retention of soil moisture and nutrients which contributes to retention of DOC(Tisdall&Oades,1982).Biochar can also alter soil chemical properties by increasing pH and CEC thereby alleviating DOC leaching (Zwieten et al., 2010). Similar findings supporting decreased DOC owing to biochar amendment were reported by Eykelbosh et al.(2015)in a tropical soil under sugarcane cultivation in Brazil and Zhu et al.(2017)in a subtropical Chinese Loess Plateau.Conversely, Zhang et al., (2017) found that DOC was increased in biochar amended soils. In addition, soil pH can affect DOM mobility; however, effects are still uncertain. Nonetheless, Tipping& Woof, (1990) reported reduced adsorption capacity at high pH values with increase of DOM mobilization. Consequently, small increases in the soil pH lead to higher amounts of mobilized Corg.This study has demonstrated that biochar regardless of its combination with N fertilizer has the potential to mitigate leaching of DOC from croplands with positive implication for decreased eutrophication in surface and ground waters and protection of the environment.

4.4. Effect of amendments on soil EOC

EOC is a sensitive indicator of changes in Corgpools,soil quality and management practices that promote carbon sequestration.Our findings showed that,EOC content was consistently increased in N,B + N and B treatments when compared to the control (CT).Moreover, we observed a significant positive correlation between DOC and EOC (Table 3) which suggests that DOC is an important component of EOC. The increase in DOC may be one reason for subsequent increase of EOC content particularly in N fertilizer treatment. EOC is the most sensitive of labile Corgpools easily altered by microbial decomposition and seasonality, with implications for nutrient cycling,plant growth and soil Corgstorage(Wang et al., 2015; Weil et al., 2003). Easy access to soluble N from N fertilizer and labile nutrients from our slowly pyrolyzed biochar could have contributed to increase in root biomass and exudates of rice plants, thereby stimulating increase in EOC content. Furthermore, higher EOC/Corgproportion in treatments CT and N in comparison to biochar amended treatments(B+N&B)indicate rapid rate of Corgdecomposition and nutrient turnover. This portend serious implication for nutrient losses and environmental pollution if mineralized nutrients are not efficiently utilized by plants and an increase in CO2-C emission, a contributor to global change and ultimately deteriorating soil quality.

4.5. Effect of amendments on soil MBC

Soil MBC is an essential component of Corgplaying important role in nutrient turnover and availability and often used as indicators of changes in soil quality (Liang et al., 2012). Our study shows differences in yearly response of MBC content to amendments. In the first year, B + N and B treatments significantly increased MBC relative to CT (control) and N fertilizer treatments.Higher substrate availability provided by labile C component of biochar and additional soluble nutrient from N fertilizer induced greater initial increase of MBC concentration in B + N and B treatments. Furthermore, provision of a favorable colony and protection from predatory grazers provided by biochar enhanced rapid microbial growth, which in turn caused an increase in MBC. The decrease in MBC in the second year in biochar treatments(B+N&B)could be attributed to:(i)reduction in labile C substrates for soil microbes as most of the initial flush of labile C released upon biochar incorporation were consumed by microbial biomass. (ii) biochar’s capacity to adsorb low molecular weight OM rendering it inaccessible to soil microbes and consequently causing a decrease in MBC (Chai et al., 2018) (iii) High biochar rate used in this study(12 t ha-1) protected Corgand decreased its mineralization and decomposition resulting in low microbial activity and decreased MBC. This latter argument was supported by previous studies that have reported significant positive correlation between soil Corgand MBC(Chen et al.,2011;Zhou et al.,2014).This implies that soil Corgcontent is the primary determinant of MBC pools in the context of increase or decrease.Similar studies conducted by Li et al.(2018)&Dempster et al. (2012) reported that high rates of biochar amendment (40 kg ha-1) and (5-25 t ha-1) significantly decreased soil MBC.However,in the second year,N fertilizer had the highest MBC content and MBC/Corgproportion.We argue that higher content of MBC in N treatment in the second year could have been triggered by N fertilizer priming effect on native Corgand increased rice root exudates which caused rapid mineralization of OM thereby providing large amount of available substrates for soil microbial growth. Availability of mineralized and readily hydrolyzable Corgprovided food and energy for soil microbes which resulted into higher content of MBC. Our findings was corroborated by Li et al.(2016) who reported increase in soil MBC after N fertilizer addition but decreased microbial diversity in a Moso bamboo plantation in sub-tropical China.However,studies by Compton et al.(2004)&Johnson et al. (1998) have reported that continuous N fertilizer addition does alter soil microbial biomass and activity. Treseder(2008), in a meta-analysis, concluded that continuous N fertilizer application decreases soil MBC in the long term. It is pertinent to note that this continuous application over time can result into decreased soil C/N ratio, pH, and Corg- a source of food for soil microbes which will in turn lead to decline in MBC.

4.6. Effect of amendments on soil CO2-C emission

Soil CO2-C emissions from croplands are created by soil microbes, fauna and plant root respiration (Hanson et al., 2000). We measured soil respiration in the present study, including respiration of plant roots and soil microbes. However, we could not delineate the contribution of different sources to CO2-C emission due to limited facilities and resources. Findings from our study show significant interaction between amendments and years of sampling.This implies that increase or decrease in CO2-C emissions was dependent on amendments and year of measurement. Our analysis showed that N fertilizer treatment consistently emitted the largest CO2-C over two years. We attribute this to increase in microbial biomass (MBC) as evidently shown in our study and the activities of soil microbes which stimulated the decomposition of native Corgvia real N priming mechanism (Steinbeiss et al., 2009).Real N priming is described as the net increase or decrease in the amount of CO2-C emitted due to rapid microbial mineralization of native soil Corginduced by external N fertilizer addition(Fiorentino et al., 2019). In addition, split application of N fertilizer enhanced rice plant growth and below ground biomass which stimulated root respiration and labile C,thus providing extra Corgfor decomposition and hence an increase in CO2-C emissions. Soil moisture and temperature are two important factors modulating CO2-C emissions(Banerjee et al., 2016). Thus, ambient air temperature and precipitation during CO2-C measurement influences soil moisture and temperature, thereby determining the magnitude of CO2-C flux.Correlation between CO2-C emissions and soil temperature and moisture has been widely reported (Smith et al., 2003). We posit that increase in ambient air temperature and precipitation in 2017(Fig.1)mediated soil moisture and temperature.This consequently supported greater microbial activity and rapid decomposition of Corgand CO2-C emission particularly in N fertilizer treatment in the second year.Our finding is in contrast to studies conducted by Ding et al. (2007), Al-Kaisi et al. (2008) and who reported significant reduction in soil CO2-C emission from temperate croplands fertilized with N fertilizer (Urea) relative to tropical croplands. In the second year of study, B + N and B treatments decreased CO2-C emissions in comparison to N and CT. This suggests that initial increase in CO2-C emission observed in the first year was a transient incident which gradually diminished over time. We attribute this decrease to non-availability of labile Corgfor microbial consumption as most of the labile C has been used up in the first year and a greater dominance of fixed/recalcitrant C compounds. Similar studies by Zimmerman et al. (2011), Iqbal et al. (2009) and Zhang et al. (2012) corroborated our findings. In addition, the complexation of Corgon biochar surface and pore sites may result in cohabitation and protection of substrates, nutrients and soil microbes thus promoting enhanced microbial C-use efficiency and less CO2-C emission (Lehmann et al., 2011). CO2-C can be precipitated also onto biochar surface as carbonates via abiotic reactions owing to biochar high pH and cationic elements which consequently leads to decreased CO2-C emission (Joseph et al., 2010).However,despite decrease in CO2-C emission reported in our study we suggest caution in N fertilizer application as additions may likely induce other greenhouse gas emissions e.g. N2O emission.

4.7. Effect of amendments on soil lability index and proportion of labile Corg pool to Corg

Soil LI and related indices are used in croplands to determine effect of management practices and environmental factors on changes in Corgcontent and sequestration as well as soil fertility assessment. Documented literatures examining the impacts of applying biochar amendment and N fertilizer addition on soil LI particularly in humid tropical croplands are scarce. Blair et al.(1995) & Demisie et al. (2014) posited that a high LI rating (LI ＞1)indicates higher labile Corgpools which can be attributed to rapid Corgdecomposition and mineralization,while a lower LI rating(LI ＜1)indicates low labile Corgpools and less decomposition of organic matter and accumulation of Corg. Findings from our study clearly show that B+N treatment had the lowest LI after two years which indicate lower labile Corgpools, lesser rate of Corgdecomposition and high Corgaccumulation and sequestration.This low LI indicates enhanced Corgstability and storage in a combined biochar and N fertilizer (B + N) amended humid tropical Alfisol cropland in SSA.Furthermore, the low LI implies a slower rate of Corgturnover and nutrient cycling which consequently reduced mineralization of nutrients for rice plants from native Corg. The slow rate of Corgturnover and low LI was plausibly due to the presence of aromatic forms of C in biochar which enhanced the stability of Corgby protection both physically and chemically from microbial decomposition(Sohi,2012).On the other hand,a reverse trend was observed in the control and N fertilizer treatments as they pooled higher LI.Demisie et al. (2014) observed similar low LI values as reported in our study in a biochar amended Typic Plinthustults in China. Our study clearly shows positive and significant correlation between CO2-C emissions,Corgcontent and labile Corgpools(Table 3).These correlations imply that CO2-C emissions were mediated by Corgcontent, EOC, MBC, POC and DOC. This finding suggests that apart from soil moisture and temperature, changes in measured labile Corgpools could be used in predicting rate or magnitude of CO2-C fluxes.Correspondingly,Corgcontent was positively correlated with all labile Corgpools except DOC(Table 3).These correlations suggest massive relationship between Corgcontent and labile Corgpools and how changes in labile Corgpools could be possibly used as indicator of changes in Corgcontent.Similar findings corroborating our study were reported by Yang et al. (2012) & Zhu et al. (2015) indicating that soil Corgis an important factor influencing soil labile Corgpools and its sensitivity for use as indicators of carbon sequestration and soil quality.

5. Conclusions

Our study across two years provides data from a humid tropical alfisol cropland quantifying soil Corgcontent, CO2-C emission and lability index in response to biochar amendment and N fertilizer.We found that N fertilizer (90 kg N ha-1) applied alone in SSA cropland under rain-fed rice cultivation consistently increased CO2-C emissions and associated labile Corgpools such as DOC,EOC,MBC and LI.On the other hand,N fertilizer combination with high dose of biochar (12 t ha-1) decreased CO2-C emissions, DOC, EOC, MBC and LI,while Corgand POC were increased.Mechanisms underlying the different stimulating effects of treatments (N or B + N) on the reported results were associated with priming effects of N fertilizer(N) and protection/sorption properties of biochar (B + N). Our results contribute to rigorous documentation of soil and regionspecific emission of CO2-C in understudied SSA croplands and suggest that the Inter-governmental Panel on Climate Change(IPCC) should consider the potential synergistic effect of biochar and N fertilizer combinations in refining CO2emission factor. In addition, our result affirms the promotion of combining biochar albeit at high dosage with fertilizer-N for the potential benefit of carbon sequestration in this region. More reliablein situCO2emissions data at molecular scale using isotopic techniques are needed to cover a broader range of agroecological zones. Future studies should consider different soil types,lower doses of biochar and types,N fertilizer types and cropland management practices in SSA if we are to better define regional and global CO2emission and mitigation prospects.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

Authors acknowledge and appreciate the assistance of laboratory technicians at the Department of Crop, Soil and Pest Management, Federal University of Technology, Akure and the Department of Agronomy, University of Ibadan who helped with soil analysis. We appreciate the editor and anonymous reviewers for their shrewd comments and suggestions.