Paddy rice yield and greenhouse gas emissions: Any trade-off due to co-application of biochar and nitrogen fertilizer? A systematic review

Abstract

Combined application of biochar and nitrogen (N) fertilizer could offer opportunities to increase rice yield and reduce methane emissions from paddy fields. However, this strategy may increase nitrous oxide (N₂O) emissions, hence its interactive effects on GHG emissions, global warming potential (GWP) and GHG intensity (GHGI) remained poorly understood. We conducted a systematic review to i) evaluate the overall effects of combined application of biochar and N fertilizer rates on GHGs emissions, GWP, rice yield, and GHGI, ii) determine the quantities of biochar and N-fertilizer application that increase rice yield and reduce GHGs emissions and GHGI, and iii) examine the effects of biochar and different types of nitrogen fertilizers on rice yield, GHGs, GWP, and GHGI using data from 45 research articles and 183 paired observations. The extracted data were grouped based on biochar and N rates used by researchers as well as N fertiliser types. Accordingly, biochar rates were grouped into low (≤9 tons/ha), medium (>9 and ≤ 20 ton/ha) and high (>20 tons/ha), while N rates were grouped into three categories: low (≤140 kg N/ha), medium (>140 and ≤ 240 kg N/ha), and high (>240 kg N/ha). For fertiliser types, N rates were grouped as: low (≤150 kg N/ha), medium (>150 and ≤250 kg N/ha), and high (>250 kg N/ha) and N types into: urea, NPK, NPK plus urea (NPK_urea) and NPK plus (NH₄)₂SO₄ (NPK_(NH₄)₂SO₄). Results showed that biochar and N fertiliser significantly affected GHGs emissions, GWP, GHGI and rice yield. Compared to control (i.e., sole N application), co-application of high biochar and medium N rates significantly decreased CH₄ emission (82 %) while low biochar with low N rates enhanced CH₄ emission (114 %). In contrast, high biochar combined with low N decreased N₂O emission by 91 % whereas medium biochar and high N rates resulted in 82 % increase in N₂O emission relative to control. The highest GWP and GHGI were observed under co-application of medium biochar and low N rates. Highest rice yield was observed under low biochar rate and high N rate. Regardless of N fertiliser type and biochar rates, increasing N rates increased rice yield and N₂O emissions. The highest GWP and GHGI were recorded under sole NPK application. Combination of low biochar and medium N produced low GHGs emissions, high grain yield, and the lowest GHGI, and could be recommended to smallholder farmers to increase rice yield and reduce greenhouse gas emissions from paddy rice field. Further studies should be conducted to evaluate the effects of biochar properties on soil characteristics and greenhouse gas emissions.

1. Introduction

Globally, rice is a strategic crop for combatting hunger and food insecurity [1]. More than half of the global population depend on rice for meeting their food needs [[2], [3], [4]]. Consequently, rice has become one of the fastest growing crops in the world in terms of production area [5]. However, rice production depends on large nitrogen (N) fertilizer and water inputs for optimal yield. More than 24 % of the global N consumption is linked to rice production [6], with less than 40 % N use efficiency [7]. Particularly, in high yielding rice varieties, maximum grain yield is achieved by reliance on chemical fertilizer [1]. Their growths and developments are hinged on the presence of N [[8], [9], [10]], without N, yield is significantly reduced [[11], [12], [13]], thus leading to excessive N application which pollutes the environment [14,15], causes disruption in the N cycle, and lowers N use efficiency by plants [16]. It is estimated that between 50 and 70 % of the added N is lost to the environment [17] where they are transformed into different greenhouse gases (GHGs), air pollutants such as ammonia (NH₃) or N monoxide (NO), or are leached through nitrates (NO₃⁻), thereby making rice an important contributor to the current anthropogenic global warming [[18], [19], [20]].

Paddy rice production is a major source of methane (CH₄) [[21], [22], [23]] and accounts for about 11 % of the global CH₄ emission [24]. The high moisture [25] and N demand by rice [1,8,9,11,12] provide perfect conditions for the anaerobic production of CH₄ and nitrous oxide (N₂O) [20,26]. The flooded condition and N fertilization associated with rice production provide favourable environment and substantial ammonium (NH₄⁺) and nitrate (NO₃⁻) substrates for the microbial production of N₂O through nitrification and denitrification [25,27]; two processes that account for 39 % and 60 % of the global N₂O emission from the soil [28]. Aside from N₂O, the resulting lack of oxygen and other oxidant prevalent in paddy fields allows methanogenic archaea to decompose soil organic matter to release CH₄ [[29], [30], [31]]. From the foregoing, it is evident that alternative approaches to rice cultivation with concomitant advantages of raising crop yield and reducing emission footprint from paddy rice production are urgently needed. The situation is more compelling considering the growing demands for food especially rice by the teeming global population [32].

Different strategies have been deployed in reducing GHGs emissions from rice fields. The most popular are water management [[33], [34], [35], [36]] and use of nitrification inhibitors like biochar [37]. However, results from researchers have shown mixed effects from the application of water management and biochar. For instance, changing irrigation regimes from continuous flooding to alternate wetting and drying results in either yield increase [38], or trade-off between CH₄ and N₂O [23,33,39]. While biochar application leads to either reduction of CH₄ emission and increase in carbon dioxide (CO₂) and N₂O emission or vice versa or other forms of unwanted effects such as yield decreases [[40], [41], [42], [43], [44], [45]] or no effect [46]. Some studies showed that little effect of biochar application on rice yield was due to the inability of biochar to supply sufficient plant nutrients especially at low biochar rate, and other studies reported that high rate of biochar application leads to nutrient immobilization [44] and toxicity to plants due to the presence of toxic polycyclic aromatic hydrocarbons [40]. To overcome these challenges, the combined application of biochar and N fertilizer was advocated by several authors [[47], [48], [49], [50]], and has shown good results in both rice yield [51] and GHGs reduction [52] albeit with some trade-offs for CH₄ and N₂O emissions, suggesting that optimal combination rate of these two amendments needs to be explored. Therefore, using the currently available data, this review i) assessed the overall effects of the combined application of biochar and N-fertilizer on GHGs emissions, global warming potential (GWP), rice yield, and yield-scaled GHG intensity (GHGI), ii) determined the quantity of biochar and N-fertilizer combinations that increases crop yield and reduces GHGs emissions and GHGI, and iii) determined the effect of different N types when combined with biochar, on rice yield, GHGs and GHGI.

2. Materials and method

2.1. Approach for data collection

A mixed method involving literature search, systematic and bibliographic review of literatures as well as assessment and analysis of published literature was adopted for retrieving articles for this review [53].

2.2. Systematic literature search

In principle, our literature collection approach as shown (Fig. 1) followed the guidelines for the preferred reporting items for systematic reviews and meta-analysis (PRISMA) [54]. The process involved an exhaustive systematic literature search of Web of Science, Science Direct and Google Scholar using the following keywords: “biochar and N-fertilization” OR “biochar and nitrogen fertilizer” OR “biochar and N fertilizer” OR “biochar and urea” OR “biochar and fertilizer” OR “biochar and inorganic amendment” OR “char and N-fertilization” OR “char and fertilizer” OR “char and urea” OR “char and inorganic fertilizer” OR “charcoal and inorganic fertilizer” OR “charcoal and nitrogen fertilizer” OR “charcoal and urea” OR “charcoal and N fertilizer” OR “charcoal and N-fertilization” OR “charcoal and fertilizer” AND “GHGs” OR “N2O” OR “nitrous oxide” OR “CO2” OR “carbon dioxide” OR “greenhouse gas emission” OR “methane” OR “CH4” AND “rice yield” in the title, keywords, abstract and in the papers body.

Fig. 1.

PRISMA diagram of article retrieval.

In addition to the systematic literature search, and to ensure that we captured all the relevant studies, bibliographies of the selected articles were also screened. The search returned a total of 4138 articles which were subjected to double entry check. After checking for double entry, a total of 3992 articles were retained for further screening of eligibility.

2.3. Eligibility criteria and screening

To avoid any form of bias and to restrict articles to only those relevant to our review, inclusion criteria were set. For a study to be included, it must have fulfilled the following criteria:

• i.must be an original research article conducted on rice.
• ii.must have applied both biochar and N fertilizer (either as treatment or soil amendment) with a fertilized or non-amended control.
• iii.must have measured and reported any of the two main GHGs (N₂O and CH₄) as well as rice yield or provided other proxy data such as GHGI or GWP through which these variables can be extracted with relative accuracy.

Using the “abstract screener” function of metagear library in R, all the titles and abstracts of the retrieved articles were screened. Where the information contained in the titles and abstracts were insufficient for deciding on inclusion or exclusion, such articles were downloaded and the whole texts screened to determine their eligibility.

Furthermore, to ensure uniform comparisons, in a multi-season or year experiment, where biochar and N fertilizer were applied only in the first season or year, only the results obtained from the first season or year were considered. On the other hand, if an experiment lasted for more than a year or season with the same rate of biochar and N fertilizer application as the preceding seasons or years, and results reported on yearly or seasonal basis, the average of such results was calculated and used. However, if varying biochar and N fertilizer rates were applied in the succeeding years or seasons, each of the seasons or years was treated as a separate experiment. Similarly, where a study lasted for more than one year and the researcher failed to indicate whether biochar (treatment) was applied in the succeeding years, only the results of the first season trial was extracted and used. Also, in multi-water management or location studies, each water management or location constituted an independent experiment.

At the end of the screening, a total of 45 articles passed the eligibility criteria (Table 1). For each of these studies, the reported means of CH₄, N₂O, rice yield, GWP, and GHGI were extracted either directly from tables or graphs using WebPlotDigitizer 4.5. Also, biochar and N fertilizer application rates, as well information on N fertilizer sources were extracted.

Table 1.

Summary of the retrieved articles used for this systematic review.

S/N	Country	Biochar feedstock	Soil type	Biochar pH	Biochar N (g/kg)	Biochar OC (g/kg)	Soil pH	Soil N (g/kg)	Soil OC (g/kg)	Reference
1	Bangladesh	Bagasse	Silt loam	9.5	6.5	597	6.2	–	23.2	[55]
2	China	poplar saw dust	–	–	–	–	6.42	1.72	16.98	[56]
3	China	Wheat straw	Hydragric anthrosol/Halic acrisol	9.51	13.3	–	6.38	1.56	22.8	[57]
				9.51	13.3	–	5.05	1.9	18.1
4	China	Wheat straw	Stagnic anthrosols	10.5	11.29	513.17	5.6	–	20.26	[58]
5	China	wheat straw	littoral clay salt soil	9.4	5.6	467	8.5	0.9	11.5	[39]
6	China	wheat straw	Irragric Anthrosols	9.4	5.9	467	5.7	1.32	14.6	[50]
7	China	wheat straw	Anthrosol	10.3	10.7	490	7.35	2.1	20.3	[59]
8	China	Wheat	–	10.48	10.7	513.17	5.3	1.82	20.26	[60]
				10.48	10.7	513.17	5.64	1.45	14.62
9	China	Maize straw	Histosols	9.8	20	624.2	6.8	1.87	15.39	[40]
				10.4	16.4	655.5
10	South Korea	Sesame	–	10.1	10.9	710.7	4.95	0.58	2.02	[61]
11	South Korea	Bamboo	–	9.71	12.8	764	4.95	0.58	2.03	[48]
12	South Korea	Rice husk	–	–	5.7	454	5.4	–	12.79	[62]
13	Japan	poultry excreta	Alluvial	10	26.7	284.5	5.24	1.4	14.5	[63]
14	China	wheat straw	–	–	–	–	5.99	1.81	20.11	[46]
							6.21	1.79	18.76
							4.89	1.59	17.7
15	China	wheat straw	Hydragric Anthrosol	10.4	6	470	6	1.8	20.1	[64]
16	China	Bamboo	Loamy paddy	9.84	4.1	709	6.03	0.8	6.1	[65]
17	China	Maize straw	Sandy Loam		0.11	25.3		0.1	20.2	[66]
18	China	Rape straw	Orthic Entisol/Regosol	8.9		625.8	6.3	0.1	11.1	[41]
19	China	Pig manure compost/Maize straw/peanut husk/municipal waste	Entic Hydragric Anthrosol	6.84	16.8	508	5.82	3.08	33	[67]
				9.74	11.08	520.7
				9.16	10.2	341
				7.88	10.5	295.3
20	Thailand	Mangrove	Vertisol	7.8	2.8	585	7	0.6	17.72	[68]
21	China	Rice straw	–	9.1	8.1	670.7	6.7	1.2	11	[69]
22	Thailand	Eucalypt wood	Loamy soil	7.98	5.4	614.3	5	0.8	7.1	[43]
23	Thailand	Mangrove tree	Clay loam	5.8	2.9	282.3	7.5	0.6	0.53	[52]
24	China	Rice husk	Orthic Anthrosols	9.1	6.2	465.4	7.8	0.42	4.8	[70]
25	China	Rice husk	Gleyi-Stagnic Anthrosol	10.9	0.81	238	7	2.83	26.6	[71]
26	China	rice straw/waste wood	Silt loam	–	–	–	6.5	1.93	18.16	[72]
27	China	Wheat straw	hydroagric Stagnic Anthrosol	10.4	5.9	467	6.5	1.8	24	[73]
28	Thailand	Mangrove	Alfisols	7.94	2.4	659	6.17	0.45	4.3	[74]
29	Cambodia	Rice straw	Sandy	–	15.1	503.4	4.79	0.53	5.25	[38]
30	Vietnam	Rice straw	–	10.1	1.9	209	5.7	1.63	12.55	[75]
31	China	Wheat straw	Latosolic soil	8.1	3.29	605	7	1.87	31.2	[76]
32	Japan	Rice husk	Haplic andosol	8.4	4.01	427	5.8	4.2	58.7	[77]
33	Japan	Rice husk	Haplic andosol	7.07	3.15	361.7	6.63	2.7	32.3	[78]
34	China	Rice straw/Bamboo	–	10.2	10.7	478	5.2	3.2	20.6	[79]
				9.81	5.9	815
35	China	Rice straw	Stagnic anthrosol	9.8	5.8	500	5.31	1.98	18.4	[49]
36	China	Wheat straw	Stagnic Anthrosol	10.4	5.9	467	6.5	1.8	23.5	[80]
37	China	Rice straw	–	10.58	10.8	472.1	5.39	2.8	49.3	[81]
38	China	Rice straw	Stagnic Anthrosol	10.4	5.9	467	6.5	1.8	23.5	[82]
39	China	Rice straw	Anthrosols	9.16	13.3	620	6	1.79	16.6	[51]
40	China	Rice straw	Hydragric anthrosol	10.1	7.5	426	7.4	1.03	21.9	[83]
41	China	wheat straw	–	8.91	14.15	–	6.38	1.56	22.8	[84]
				9.63	4.72	–
				4.64	12.26	–
				3.74	4.5	–
42	Vietnam	Rice straw	Alluvial	–	12.1	353.2	6.21	10.3	12.6	[44]
43	Malaysia	Rice straw	Dystric Fluvisol	10.08	–	–	5.9	6	240.9	[85]
44	China	wheat straw	Stagnic Anthrosol	9.1	5.8	418.3	5.1	2.1	18.9	[86]
45	China	Wheat straw	Stannic Anthrosol	9.3	5.8	418.3	5.1	1.62	17.5	[87]

Alternate Wetting and Drying = AWD, Continuous Flooding = CF, Intermittent Irrigation = IT, mgt = Management.

2.4. Data standardization and grouping

To ensure uniform comparison, all the extracted data were standardized to the same metric units across all the studies for the same variables. Consequently, CH₄, N₂O, rice yield, biochar and N rates were converted to kg-C/ha, kg-N/ha, ton/ha, ton/ha, and kg N/ha, respectively. Similarly, GWP of CH₄ and N₂O were calculated and standardized for studies that reported both CH₄ and N₂O. A global warming potential of 27 and 273 for CH₄ and N₂O, respectively, relative to the CO₂ warming potential over a 100-year timescale [88] were used for the standardization. The resulting GWPs (Mg CO₂ equivalent) were used to calculate GHGI (Mg/ton). Additionally, when pH was reported in KCl, it was converted to pH_[H2O] with the formula pH_[H2O1:5] = −1.95 + log₁₀(pH_{[KCl 1:2.5]})* 11.58 [89].

After data harmonization and standardization, the data were categorized into groups based on the rates of biochar, and the rates of N co-applied by different researchers. Similarly, based on N rates and N fertilizer types applied, another categorization was also made to assess the role of N fertilizer types and rates on the variables of interest as stated in the objectives of the review. But to determine the number of groups and to ensure even representation or spread of data in each group that will enable statistical comparison, all the extracted data on biochar, N rates, and N fertilizer types and rates of N fertiliser types were displayed on histograms. Thereafter, the frequencies of the rates of N, biochar, and N types applied were used to categorize biochar, and N rates data into three groups each while the treatment without biochar but N only irrespective of N fertilizer type was used as control (CN). The choice of N fertilizer as the control instead of no amendment or biochar only treatment is because we wanted to mimic the most wide-spread nutrient management practice by farmers. Also [90], have demonstrated through a meta-analysis that regardless of whether in comparison to no amendment control or fertilized control that biochar application alone is inefficient in increasing crop yield. Hence, for this review, we adopted a fertilized treatment with no biochar as our control. The biochar rate groups (derived from biochar and N fertilizer co-application treatments) included: low biochar (biochar application rate of ≤9 tons/ha; LB), medium biochar (biochar application rate between 9 and ≤20 ton/ha; MB), high biochar (biochar application rate of >20 tons/ha; HB), while N fertilizer rates (derived from biochar and N fertilizer co-application treatments) were grouped as: low N fertilizer (N fertilizer application rate of ≤140 kg N/ha; LN), medium N fertilizer (N fertilizer rate between 140 and ≤240 kg N/ha; MN), high N fertilizer (N fertilizer application rate of >240 kg N/ha; HN). The combination of biochar and N fertilizer groups resulted in a total of nine treatment groups (LBLN, LBMN, LBHN, MBLN, MBMN, MBHN, HBLN, HBMN, HBHN), and a total of ten groups when the CN is included. CN consisted of 62, HBHN 14, HBLN 10, HBMN 14, LBHN 10, LBLN 19, LBMN 10, MBHN 18, MBLN 8, and MBMN 18 data points.

Another group was also formed based on N fertilizer types, and N fertilizer rates combined with biochar. As stated above, the extracted data of yield and GHGs emissions from N fertiliser types and rates of N fertiliser types were displayed on histogram in order to ensure uniform distribution of data in each group, and thereafter categorized into groups. A total of three classes of N fertilizer rates were obtained: ≤150 kg N/ha as low rate, >150 and ≤ 250 kg N/ha as medium rate and >250 kg N/ha as high rate. Similarly, for N fertilizer types and the variables, N₂O, GWP, and GHGI, a total of three groups were formed (urea, NPK, NPK_urea) because under NPK_NH4SO4 [i.e., NPK_(NH₄)₂SO₄], N₂O was not reported and as such it was not possible to calculate GWP and GHGI. But for the variables; rice yield, and CH₄, a total of four groups were realized (urea, NPK, NPK_urea, NPK_NH4SO4[i.e., NPK_(NH₄)₂SO₄]). For each of the groups, urea was used as the control.

In the studies that make up urea, and NPK groups, only one of these N fertilizer types was used by the researchers while in NPK_urea, and NPK_NH4SO4 [i.e., NPK_(NH₄)₂SO₄] groups, researchers applied NPK and urea, and NPK and (NH₄)₂SO₄ respectively at different stages of the experiment. Overall, urea group consisted of 82 datasets, NPK_urea 16, NPK_NH4SO4 [i.e., NPK_(NH₄)₂SO₄] 5, and NPK 16. Please note that, under this classification, biochar rate was not considered. We therefore suggest that future research should consider biochar rates under N fertilizer types.

2.5. Data analysis

Data analyses for this review were performed using R statistical software. All the groups obtained after data standardization and grouping were subjected to one way analysis of variance and means separated by TukeyHSD test (p < 0.05). Results were presented in bar plots with different letters indicating statistical significance. The essence of the statistical tests was to understand the effects of biochar and N fertilizer application rates, N fertiliser types, and rates of N fertiliser types on rice yield, N₂O, CH₄, GWP and GHGI, and to determine the optimal rates of biochar and N to increase rice yield and reduce GHG emissions.

Additionally, rice yield, CH₄ and N₂O emissions were also regressed on the most widely reported soil and biochar properties (soil organic carbon, soil total N, soil pH, biochar organic carbon, biochar total N and biochar pH) using pairwise regression approach. The failure of many of the researchers to report climate data made it impossible to assess the impact of climate on the variables of interest. The Akaike Information Criterion (AIC) was used to determine the best model for the prediction of our dependent variables (rice yield, CH₄ and N₂O emissions) [91,92]. The model with the lowest AIC value was retained.

3. Results

In terms of geographical spread, the studies used for this review were conducted in eight countries (Supplementary Fig. S1). All the countries are in Asia. Most of the studies were carried out in China using urea as N fertilizer source and biochar derived from cereal waste (Table 1).

3.1. Effect of nitrogen (N) fertilizer types and rates on methane (CH₄) and nitrous oxide (N₂O) emissions

Greenhouse gas (GHG) evolution in rice field varied significantly among the different N fertilizer types and rates (Fig. 2). At low (≤150 kg N/ha), and high N rates (>250 kg N/ha), the application of NPK with biochar consistently produced the highest N₂O emission compared to other forms of N fertilizer. Specifically, NPK (364 %) and NPK_urea (318 %) caused more than double fold increase in N₂O emission under low N rate relative to urea. However, increasing N rate caused a deviation in the pattern of N₂O emission under NPK fertilizer. At medium N rate (150 < N ≤ 250 kg N/ha), the use of NPK as N source resulted in about 100 % decrease in N₂O emission while NPK-urea induced N₂O emission by about 56 % compared to urea. Similar to the pattern observed under low N rate, the addition of NPK, or NPK_urea with biochar caused a rise in N₂O emission by 189 % and 51 %, respectively, compared to urea under high N rate (>250 kg N/ha). It appears that irrespective of N fertiliser types, N₂O emission increased as N rate increased except for NPK under medium N rate (Fig. 2A).

Fig. 2.

Effect of biochar application with nitrogen fertilizer types and rates on N₂O (A) and CH₄ (B) emissions, GWP (C) and GHGI (D). Bars with different letters are statistically different from each other (p < 0.05). The error bars represent the standard deviation of the mean.

On the other hand, a contrasting pattern was observed for CH₄ emission under different N fertilizer types and rates. Under low N rate, combined application of biochar with either NPK or NPK_urea decreased CH₄ emission by 23 % and 70 %, respectively compared to the use of urea. However, these inhibitory effects waned at medium N rate (150 < N ≤ 250 kg N/ha), resulting in 444 % (NPK) and 33 % (NPK_urea) increase in CH₄ emission relative to urea. Furthermore, at high N rate (>250 kg N/ha), NPK_urea promoted CH₄ emission more than any other N source, followed by NPK_NH4SO4 [i.e., (NH₄)₂SO₄]. Both N sources (NPK_urea and NPK_NH4SO4 [i.e., (NH₄)₂SO₄]) increased CH₄ emission by 63 % and 48 %, respectively, compared to urea. Contrastingly, under the same N rate (high N rate), NPK application sharply decreased CH₄ emission (31 %) relative to urea. More so, urea application also indicated a decreasing pattern in CH₄ emission as N rate increased (Fig. 2B). At low rate, urea strongly stimulated CH₄ emission but the stimulatory effect was less evident at higher N rates particularly at medium N rate (150 ≤ N250 kgN/ha). Medium rate of urea application significantly inhibited CH₄ emission. Except for NPK_urea which clearly showed an increasing pattern in CH₄ emission as N rate increased, no defined pattern of relation could be established between N fertilizer types and N rates. Different N fertilizer types appear to have different optimal rate for decreasing CH₄ emission. For instance, NPK at high rate (>250 kg N/ha) inhibited CH₄ emission, while NPK_urea and urea, were both more efficient in decreasing CH₄ emission at medium rates (150 < N ≤ 250 kg N/ha). No pattern of effect on CH₄ emission could be established for NPK_NH4SO4 [i.e., (NH₄)₂SO₄] since it was only applied at high rate.

Global warming potential (GWP) varied significantly among the various N fertilizer types. NPK application increased GWP (122 %) compared to urea while the effect of NPK_urea on GWP relative to urea was moderately lower (51 %) (Fig. 2C). The high N₂O and CH₄ emissions associated with NPK application are responsible for its high GWP value. On the other hand, the greenhouse gas intensity (GHGI) from the various N sources varied non-significantly across the N fertilizer types (Fig. 2D). The use of NPK as N source resulted in highest GHGI compared to other N fertilizer types as was the case in GWP driven majorly by high GHG emissions.

3.2. Rice yield under different N fertilizer types and rates applied with biochar

Different N fertilizer types and rates combined with biochar significantly influenced paddy rice yield (Fig. 3). Overall, the calculated rice yield varied approximately from 5 ton/ha to 7 ton/ha under low N rate, 7 ton/ha to 11 ton/ha under medium N rate and 9 ton/ha to 15 ton/ha under high N rate across N fertilizer types regardless of biochar combination rates. For all the N fertilizer types, increase in N rate resulted in increase in rice yield. However, the magnitude of the increase varied according to N fertilizer types. Considering individual N fertilizer types applied with biochar, NPK_urea as N source was least responsive to increasing N rate while NPK was most responsive. Increasing N rate from (≤150 kg N/ha) to between >150 and < 250 kg N/ha almost did not result in any change in rice yield for NPK_urea as N source. However, increasing N rate from 150 < N < 250 kg N/ha to >250 kg N/ha caused the highest increase in rice yield especially under NPK as N source. In addition, the application of the mixture of NPK and (NH₄)₂SO₄ with biochar similarly enhanced rice yield relative to the application of biochar with either urea or a mixture of NPK and urea. Under combined application of biochar and low N rate (≤150), NPK as N source decreased rice yield by 27 % and 24 % relative to urea and NPK_Urea respectively. But under medium N rate, combination of biochar with NPK enhanced rice yield by 26 % relative to urea while NPK_urea as N source applied with biochar caused a yield decrease of 20 % in comparison with urea applied with biochar. Furthermore, the use of high rate of (NH4)₂SO₄ (>250 kg N/ha) with biochar induced a 41 % increase in rice yield compared to applying high urea (>250 kg N/ha) with biochar. Similar increase in rice grain yield (61 %) under high NPK rate (>250 kg N/ha) applied with biochar relative to combined application of high urea rate (>250 kg N/ha) with biochar was also observed. However, high rate of NPK_urea (>250 kg N/ha) applied with biochar decreased rice grain yield (8 % decrease) compared to high urea rate applied with biochar (Fig. 3).

Fig. 3.

Rice yield under different nitrogen fertilizer types and rates applied with biochar. Bars with different letters are statistically different from each other (p < 0.05). The error bars represent the standard deviation of the mean.

3.3. CH₄ and N₂O responses to rates of biochar and N fertilizer combination

The quantity of biochar combined with a given quantity of N fertilizer played a significant role in CH₄ emission (Fig. 4). Regardless of N fertiliser types, combination of low biochar (LB) and low N (LN) stimulated largest CH₄ emission (274 kg-C/ha) while the lowest mean CH₄ emission (24 kg-C/ha) was observed under high biochar (HB) and medium N (MN) group. Increasing biochar rate under low N rate resulted in significant decrease in CH₄ emission (Fig. 4). Similar effect was also observed when increasing biochar rate under medium N rates. But under high N rates, increase in biochar resulted in no definite pattern in CH₄ emission. Although, CH₄ emission decreased and increased at high biochar rate under moderate and high N rates, respectively. Contrarily, a linear decrease in CH₄ emission was observed under low biochar rate with increasing N rates. This pattern was retained as biochar rate was increased to medium and N rates gradually increased. Under high biochar rate with increasing N rates, no pattern could be established, but high N rate with high N increased CH₄ emission (Fig. 4). Also, no linear relationship between rates of biochar and N on CH₄ emissions was observed. Instead, a decrease in CH₄ emission was observed under medium biochar (MB) and MN rates while LBLN, as well as HB and high N (HN) combinations enhanced CH₄ emission. Nevertheless, the increase in CH₄ emission under HBHN was still smaller compared to LBLN.

Fig. 4.

CH₄ emissions under different levels of nitrogen fertilizer (N) applied with different biochar rates (B). CN (sole N; average, 195 kg N/ha), LBLN (B ≤ 9 t/ha + N ≤ 140 kg N/ha), LBMN (B ≤ 9 t/ha+140 < N ≤ 240 kg N/ha), LBHN (B ≤ 9 t/ha + N > 240 kg N/ha), MBLN (9 < B ≤ 20 t/ha + N ≤ 140 kg N/ha), MBMN (9 < B ≤ 20 t/ha+140 < N ≤ 240 kg N/ha), MBHN (9 < B ≤ 20 t/ha + N > 240 kg N/ha), HBLN (B > 20 t/ha + N > 240 kg N/ha). Bars with different letters are statistically different from each other (p < 0.05). Error bars represent the standard deviation of the means.

According to Fig. 4, LBLN, MBLN, HBHN, and LBMN rates all enhanced CH₄ emission compared to the control (CN). Respectively, under these biochar and N combination rates, CH₄ emission increased by 114 %, 50 %, 38 % and 8 % compared to control N (CN). On the other hand, decreases in CH₄ emissions were found under HBMN (82 %), HBLN (53 %), MBHN (35 %), LBHN (6 %), and MBMN (5 %) compared to CN (Fig. 4). The result indicated that rates of biochar and N affect CH₄ emissions.

Fig. 5 shows the mean cumulative N₂O emission as affected by different rates of biochar and N fertilizer combinations. From the figure, rates of biochar and N strongly controlled the mean cumulative N₂O emission from rice field. Cumulative N₂O emission ranged from 0.14 kg-N/ha observed under HBLN rates to 2.74 kg-N/ha found under MBHN rate. The result (Fig. 5) showed that increasing N rate under low biochar application enhanced N₂O emission. Similar progressive N₂O enhancement can also be observed under medium and high biochar rates as N rate increased, indicating that at constant biochar rate, progressive increase in N rate results in gradual increase in N₂O emissions. On the other hand, gradually increasing biochar rates under low N level resulted in decrease in N₂O emission (Fig. 5), indicating the effectiveness of biochar in decreasing N₂O emissions at low N application rate (≤140 kg N/ha). Nevertheless, as N rate was increased to medium and biochar rates progressively increased, only high biochar rate was effective in decreasing N₂O emission. Under high N rate, no definite trend could be established as biochar rate was gradually increased, however, the mitigatory influence of biochar at high rate on N₂O emission was retained.

Fig. 5.

Effects of biochar (B) and nitrogen fertilizer (N) rates on N₂O emission. CN (sole N; average, 195 kg N/ha), LBLN (B ≤ 9 t/ha + N ≤ 140 kg N/ha), LBMN (B ≤ 9 t/ha+140 < N ≤ 240 kg N/ha), LBHN (B ≤ 9 t/ha + N > 240 kg N/ha), MBLN (9 < B ≤ 20 t/ha + N ≤ 140 kg N/ha), MBMN (9 < B ≤ 20 t/ha+140 < N ≤ 240 kg N/ha), MBHN (9 < B ≤ 20 t/ha + N > 240 kg N/ha), HBLN (B > 20 t/ha + N > 240 kg N/ha). Bars with different letters are statistically different from each other (p < 0.05). Error bars represent the standard deviation of the mean.

Irrespective of biochar rate, high N rate significantly promoted N₂O emission while high biochar reduced N₂O emission (p < 0.05). The efficiency of high biochar in decreasing N₂O emission increased as N rate decreased (Fig. 5). Relative to control, LBLN inhibited N₂O emission by 16 %, LBMN by 55 %, MBLN by 67 %, and MBMN by 28 %. The highest decrease in N₂O emission (91 %) compared to control was recorded when high biochar was applied with low N (HBLN). On the other hand, LBHN, MBHN, HBHN all enhanced N₂O emission by 81 %, 82 % and 49 % respectively compared to the CN.

3.4. Effect of biochar and N fertilizer rates on rice yield, GWP and GHGI

Rice yield responded significantly to varying rates of biochar and N fertilizer combinations (Fig. 6). Higher N rate was most effective in increasing rice yield especially under low biochar rate. Maximum rice yield was observed under combined application of LBHN, while low biochar combined with low N fertilizer (LBLN) resulted in the lowest rice yield. Under low and medium N rates, increasing biochar rate promoted rice yield. However, at high N rate, progressive increase in biochar rates resulted in decrease in rice yield. Conversely, increasing N availability by increasing N rate in the presence of low biochar rate induced a progressive increase in rice yield. The positive effect of increase in N on rice yield continued even at medium biochar rate but faded under high biochar rate. Therefore, in order to avoid nutrient immobilization associated with biochar use, biochar should be applied at rates lower than 20 ton/ha.

Fig. 6.

Rice yield response to biochar (B) and nitrogen (N) fertilization rates. CN (sole N; average, 195 kg N/ha), LBLN (B ≤ 9 t/ha + N ≤ 140 kg N/ha), LBMN (B ≤ 9 t/ha+140 < N ≤ 240 kg N/ha), LBHN (B ≤ 9 t/ha + N > 240 kg N/ha), MBLN (9 < B ≤ 20 t/ha + N ≤ 140 kg N/ha), MBMN (9 < B ≤ 20 t/ha+140 < N ≤ 240 kg N/ha), MBHN (9 < B ≤ 20 t/ha + N > 240 kg N/ha), HBLN (B > 20 t/ha + N > 240 kg N/ha). Bars with different letters are statistically different from each other (p < 0.05). The error bars represent the standard deviation of the mean.

Regardless of N fertilizer type and biochar rate, high N ensured high rice yield (Fig. 6). Consequently, relative to CN, LBHN promoted rice yield by 45 %. Whereas retaining high N rate but gradually increasing biochar rates enhanced yield by 30 % and 11 % under MBHN and HBHN respectively, compared to LBHN. This result indicated that increasing biochar rates under high N condition does not favour rice yield.

The calculated GWP and GHGI (indicator for GWP per rice yield) for different biochar and N fertilizer rates are shown in Fig. 7. Biochar and N rate significantly affected GWP (Fig. 7A) and GHGI (Fig. 7B). GWP and GHGI were generally similar for most of the treatments, however, highest and lowest GWP were recorded under MBLN rate, and HBMN rate, respectively. The general trend reveals that low N is the major contributor to GWP especially under low biochar rate while the combination of HBHN was also critical in enhancing GWP. GHGI followed similar pattern as GWP. The variations in GHGI were significant with medium rate of biochar combined with low rate of N (MBLN), and high rate of biochar combined with high rate of N (HBHN) generating the highest GHGI. The most promising biochar and N fertilizer combination for reducing GHGI is the combination of low biochar and medium N rates (LBMN) since it led to the lowest GHGI value (0.22 Mg CO₂ eq-ha⁻¹ ton⁻¹). Moreover, Fig. 7A showed that low N also plays a key role in enhancing GHGI, probably due to its inability to enhance rice yield. It is important to state that we have not considered the effect of water management in arriving at the overall effect of N fertilizer types as well as different rates of biochar and N fertilizer combinations on GHGs emissions and rice yield. Water regimes and climatic conditions can severely interfere with rice yield and GHGs emissions. As such caution must be taken in interpreting and applying the results and recommendations of our finding.

Fig. 7.

Effect of combined biochar (B) and nitrogen fertilizer (N) rates on GWP (A) and GHGI (B). CN (sole N; average, 195 kg N/ha), LBLN (B ≤ 9 t/ha + N ≤ 140 kg N/ha), LBMN (B ≤ 9 t/ha+140 < N ≤ 240 kg N/ha), LBHN (B ≤ 9 t/ha + N > 240 kg N/ha), MBLN (9 < B ≤ 20 t/ha + N ≤ 140 kg N/ha), MBMN (9 < B ≤ 20 t/ha+140 < N ≤ 240 kg N/ha), MBHN (9 < B ≤ 20 t/ha + N > 240 kg N/ha), HBLN (B > 20 t/ha + N > 240 kg N/ha). Bars with different letters are statistically different from each other (p < 0.05). The error bars represent the standard deviation of the mean.

3.5. Effect of some selected soil and biochar properties on rice yield, CH₄ and N₂O emissions

The most widely reported initial soil and biochar properties were selected to assist in determining some of the biochar and soil properties that regulate rice yield, CH₄ and N₂O emissions. It is important to note that failure of most of the researchers to report the soil properties at the end of their studies prevented the assessment of the after-effect of biochar and N fertilizer application on soil properties, GHGs emissions and rice yield. Hence the initial soil properties and biochar were used to assess the relationship of the effects of biochar and N application on the variables of interests. Thus, pairwise regression and Akaike Information Criterion (AIC) approach was used to determine the most important soil and biochar properties that control CH₄ and N₂O emissions as well as rice yield based on the initial soil and biochar properties. Although the p values of some of these properties were less than the set alpha level (p < 0.05) (Table 2), they were however retained as their inclusion optimised the model. The influences of these soil properties are presented in Table 2. According to the results (Table 2), rice yield were mainly regulated by initial soil total N, pH, biochar total N, and total soil organic carbon. Increase in soil total N and soil pH promoted rice yield, while initial soil organic carbon and biochar total N were less dominant in their influence on rice yield. On the other hand, the amount of CH₄ emitted from rice field was a function of initial soil pH, N content of applied biochar, biochar pH, soil organic carbon content, carbon content of biochar and soil total N with a regression coefficient (R²) of 30 %. From the model (Table 2), high initial soil pH and total organic carbon, and applying biochar of high N content will enhance CH₄ emission, whereas high biochar pH, soil total N, and the application of biochar with high carbon content will all result in decrease in CH₄ emission. Similarly, N₂O emission was also driven by initial soil and biochar pH, soil total N and soil organic carbon, with a weak positive correlation (R² = 29 %). The model suggests that positive relationship exists between N₂O emission, soil organic carbon and biochar total N while increase in initial soil total N and biochar pH could result in decrease in N₂O emission.

Table 2.

Response of rice yield, CH₄ and N₂O on some selected initial soil and biochar properties using a multilinear regression model.

Variables	Estimates	95 % confidence interval		p-value
		Lower limit	Upper limit
Rice yield				3.6e-06
Intercept	−0.33	−4.20	3.52	0.86
Soil pH	1.00	0.43	1.56	0.00
Soil_TN	0.85	−0.32	2.04	0.15
Soil_OC	0.03	−0.05	0.11	0.46
B_TN	0.08	−0.01	0.18	0.07
CH4emission				2.9e-06
Intercept	278.15	64.34	491.97	0.01
Soil pH	13.54	−6.62	33.71	0.19
Soil_TN	−52.48	−92.29	12.68	0.01
Soil_OC	4.42	1.78	7.07	0.001
pH biochar	−26.22	−39.60	12.85	0.00
B_TN	6.69	2.93	10.46	0.00
B_OC	−0.12	−0.26	0.01	0.08
N2O emission				3.6e-06
Intercept	3.97	1.68	6.26	0.00
B_TN	0.06	0.01	0.10	0.02
Soil_TN	−1.72	−2.43	−1.02	4.8e-06
Soil_OC	0.08	0.02	0.13	0.01
pH biochar	−0.19	−0.41	−0.03	0.10

pH_soil = soil pH, B_TN = biochar total nitrogen (g/kg), pH_biochar = biochar pH, soil_OC = soil organic carbon (g/kg), B_OC = biochar organic carbon (g/kg), and soil_TN = soil total nitrogen.

4. Discussion

4.1. Influence of initial soil and biochar properties on rice yield, methane (CH₄), nitrous oxide (N₂O) emission, global warming potential (GWP) and greenhouse gas intensity (GHGI) under combined biochar and nitrogen (N) fertilizer application

Soil physicochemical properties are important indicators of the general soil health and productivity and are strongly affected by managements. These factors affect both crop yield and greenhouse gases (GHGs) emissions and include soil pH, aggregate stability, and soil organic matter [93]. Table 2 indicated that the initial state of soil as well as biochar properties have varying impacts on GHGs emissions and rice yield. Specifically, our regression analysis indicated a positive relationship (R² = 22 %) between rice yield and initial soil pH, soil total N, soil organic carbon (C), and biochar total N, and that soil pH and biochar N content are the key properties that regulate rice yield. Soil pH controls the solubility and bioavailability of many chemicals in the soil and as such regulates environmental pollution and nutrients intake by plants [94]. Low soil pH can cause toxicity to the microbial population in the soil [60] and thus affect nutrient cycling within the soil eco-system. Low pH can also result in nutrient immobilization and availability of some heavy metals such as manganese and boron which are prevalent in rice fields and toxic to plant [95]. The application of biochar either singly or with N raises soil pH and improves nutrient availability in the soil. Moreover, some studies have attributed the improved crop yield observed under biochar and N especially in the tropics to the fertilization effect of biochar and N on the soil [96,97].

From Tables 2 and it is obvious that biochar and N application can significantly impact emission of CH₄ and N₂O, GWP and GHGI. According to [98], soil can be both source and sink of CO₂, CH₄ and N₂O gases depending on its management. Documented evidences suggest that management practices such as biochar addition improves soil functions [[99], [100], [101]]. Although biochar influence depends on biochar and soil properties [75,77]. Increased GWP due to increase in biochar pH and decrease in soil pH under co-application of biochar and N has been reported [96]. Biochar through its synergistic effect on added N enhances cation availability [69] and alters soil nutrient dynamics and biological activities [102]. Particularly, under anaerobic condition as obtained in rice field, presence of labile carbon in biochar can serve as source of substrates for the activities of methanogens and hence promote the production of CH₄ [55,73] and N₂O [39,103]. [49] attributed the decrease in CH₄ emission following biochar application to biochar induced increase in soil pH which led to increase in the population of methanotrophs. Methanogenic bacteria which break down organic matter to produce GHGs are known to be neutrophilic and as such are negatively impacted by large increase in soil pH [60]. Similarly, the change in pH from the liming effect of biochar facilitates the transfer of electrons to denitrifying bacteria and induces N₂O emissions [50]. Also, the decomposition of biochar by these organisms adds carbon [51,104] and a net negative charge to the soil [105]. Increase in negative charge affect soil's capacity to act as an electron acceptor and thus impact emission of both CH₄ [72] and N₂O [50] and potentially GWP. Moreover, the combined application of biochar with N has been noted to enhance soil content of both dissolved organic and inorganic carbon necessary for biological activities within the soil [102] which can impact gas production.

4.2. Biochar and N fertilizer rates regulate CH₄ and N₂O emissions, rice yield, GWP and GHGI

Nutrient management is a key strategy that influence gas evolution from soil and crop yield. Fig. 4 revealed that low biochar rates induced high CH₄ emission while high biochar rate decreased CH₄ emission. Decrease in CH₄ emission following biochar application has been reported [86]. Biochar interferes with soil aeration [62], carbon cycle [60], soil porosity [62] as well as soil pH [47]. The magnitude of such interference is linearly related to biochar application rate [72] and depends on soil properties [47]. [69] found that the addition of biochar at 29.6 ton/ha with N fertiliser increased soil C and C/N ratio, and soil porosity compared to treatment without biochar [69]. Similarly, a meta-analysis showed that biochar application resulted in 11.5–11.9 % increase in soil pH compared to untreated plots [103] and about 50 % increase in soil porosity with or without N fertilizer [40]. Such changes favour the growth and activities of methanotrophs [49] and CH₄ oxidation.

As demonstrated by [60], biochar application caused a significant reduction in CH₄ emission which was attributed to increase in the population of methanotrophs in response to biochar-induced increase in soil pH. Methanotrophs have higher sensitivity to rising soil pH than methanogens [50]. Ideally, methanogenic activities are maximum within the pH range of 6.0–7.5 [49,60,68,106]; a pH value obtainable under low biochar application. This partly explains the high CH₄ observed in this study at low biochar rate. Moreover, CH₄ production in paddy field only starts when iron (III) oxide (Fe³⁺) is reduced to iron (II) oxide (Fe²⁺) which occurs around pH neutral point [107] and is favoured by low biochar rate. Reduction of Fe³⁺ to Fe²⁺ decouples organic matter from Fe³⁺ which stimulates methanogenic activities and increases CH₄ emission [72]. Also, low biochar application with N promote CH₄ emissions by enhancing N utilization and plant growth [40] and soil redox potential [97] which is crucial in regulating CH₄ emissions [55,62]. It therefore implies that in the absence of enough biochar, microbial oxidation of CH₄ could be hampered as was the case in our analysis. Nevertheless, some studies have also indicated that the application of biochar can result in decrease in CH₄ emission. But decreases in CH₄ emissions are more frequently reported at high biochar rates [49,69,87]. Besides shifting soil pH to strongly alkaline (pH > 10), at medium to high rates, biochar through its large surface area adsorbs indigenous carbon and restricts their availability to microbial decomposition [43,49]. Medium to high biochar application in paddy fields also promotes oxygen availability within the root zones and oxidation of CH₄ [52,62] as well as increases the microbial abundance of particulate monooxygenase gene (pmoA) known to enhance oxidation of CH₄ [56] and hence leads to decrease in CH₄ emission.

On the other hand, low N application enhanced CH₄ emission (Fig. 4). [108] reported that low N rate (<100 kg N/ha) stimulates the emission of CH₄ while high N rates (>200 kg N/ha) results in CH₄ inhibition. N at 70 kg N/ha applied with biochar induced 10 % higher CH₄ emission relative to biochar only treatment [22]. Low N rate (≤150) enhances CH₄ emissions by limiting N supply to methanotrophs. Nitrogen is an essential nutrient for methanotrophs and therefore N limitation may inhibit CH₄ oxidation in soils, and increase CH₄ emissions [22,109,110]. Furthermore, low N rate also tends to moderate the soil pH within the neutral point which enhances the activities of methanogens and production of CH₄ [49,67]. Although, CH₄ emission is controlled by the interaction of different factors, the role of low N in moderating soil pH within the neutral range may have contributed significantly to the observed increase in CH₄ emission under the combination of biochar with low N.

In contrast, high N rate inhibits CH₄ emission [111] by decreasing soil pH with a resultant toxicity effect on methanogen population [60]. Different studies have reported [54,60,112] N induced toxicity to methanogenic activities at high N rate as the leading cause of CH₄ inhibition in fertilized paddy fields. High N application also leads to the production of lower root biomass and activities within the rhizospheres which decreases carbon substrates necessary for CH₄ production [113]. Increasing N application especially at panicle initiation stage enhanced the production of organic acid within the rhizosphere which decreased CH₄ emission [113]. However [41,97,114], have reported enhancing effect of high N on CH₄ emissions especially when N is applied with biochar. The likely explanation for this is the inhibition of CH₄ oxidation by NH₄⁺ [111]. At high NH₄⁺, methanotrophs use oxygen as an electron acceptor, and NH₄⁺ for energy derivation instead of CH₄ which leads to increase in CH₄ emissions [111]. N application at over 500 kg N/ha in combination with biochar inhibits the growth of methanotrophs and enhances CH₄ emission [41]. Also, the interactive effect of biochar and N fertiliser can also alter the diversity of soil microbes which can affect CH₄ pattern [115].

Apart from the increases observed under low biochar and low (LBLN) rates as well as high biochar and high N (HBHN) rate, large part of the groups agreed with previous researchers who reported increased CH₄ uptakes after biochar applications with N [79,81,116]. It is therefore evidently clear that proper management is crucial to the mitigation of GHGs emissions from soils under combined biochar and N fertiliser.

Contrary to CH₄ emission pattern, N₂O emission increased with N rate irrespective of biochar rate (Fig. 5). The presence of high N substrate under high N rate stimulated N₂O emission. In soils, substrate availability is the main factor that determines whether N₂O emission will be enhanced or suppressed. Therefore, the application of fertilizer promoted N₂O emission [20,64] by providing substrates for microbial nitrification and denitrification [48]. However, the addition of biochar especially at high rate, disrupted the pattern of N₂O emission under low N application. Some of the reasons are the changes in soil aeration and oxygen availability [62], porosity [40] and C/N ratio following biochar application [117]. Application of biochar changes the soil C/N and causes a net N immobilization which hinders N₂O production [97]. C/N ratio >35 is linked with N immobilization [118,119]. Specifically, application of N rate>15 kg N/ha in combination with biochar leads to inhibition of N₂O emission, and rate of inhibition increases with N rates until about 500 kg N/ha after which additional N enhances N₂O emission [120]. Large biochar surface area also causes adsorption [22,52,56,121,122] of NH₄⁺ which leads to reduced substrate availability for the production of N₂O [123]. About 67 % reduction in N₂O was reported by Ref. [122] following the addition of biochar to a vegetable soil which can be linked to the complete reduction of N₂O to N₂ during denitrification by the presence of easily decomposable carbon from biochar [63]. Reduction in N₂O following biochar application has also been attributed to biochar-induced changes in soil pH which favours the growth of N₂O reductase that are inhibited under acidic condition [63,64,124]. Also, biochar related processes such as improvement in soil moisture content inhibits nitrification process and alters N₂O production [22,23].

From Fig. 6, combined application of biochar and N significantly increased rice yield especially at low biochar and high N (LBHN) rate. High yield under LBHN may be related to the availability of N for rice utilization. N application is closely associated with increased grain yield [54,125,126]. Depending on biochar rate, biochar inclusion with N reduces N loss from soil and hence increases crop yield [120]. Various researchers have similarly reported crop yield increases after biochar application [90,122,127]. Biochar through its effects on soil chemical, biological and physical properties can elicit improved crop performances. The main pathway is by directly supplying nutrients [106] such as Ca, K, P, Mg, S, and Si to plants [104,128], improving nutrient adsorption for plants uptake [40,56,106], and increase in soil biological diversity [64]. However, high biochar rates, causes N immobilization through negative priming [125]. According to Ref. [90] application of biochar at more than 10 ton/ha can potentially lead to decrease in plant biomass probably due to nutrient immobilization. Further negative impact of biochar is observed if rate reaches 100 g kg^-1 of soil [129]. Biochar-induced N immobilization and the consequent reduction in bioavailability of essential nutrients through their sorption on biochar has been reported [130] as the leading cause of yield reduction under biochar and is related to pyrolysis temperature [40].

Highest GWP was found under medium biochar and low N (MBLN) while lowest GWP was observed under low biochar and medium (LBMN) rate irrespective of N source. Yield and GHG emissions regulate GWP and GHGI [131] as can be seen with low biochar and low N (LBLN) as well as MBLN combinations (Fig. 6). Similarly, despite increasing yield, the high emission associated with high biochar and high N (HBHN) resulted in high GHGI.

4.3. Responses of CH₄ and N₂O emission, rice yield, GWP and GHGI to N fertiliser types and rates of N fertiliser types

Nitrogen is the most limiting nutrient for crop production [126,132]. To overcome this challenge, N fertilizer is used [40,57] particularly in rice fields. These fertilizers are derived from different N sources (e.g., NPK, urea, ammonium sulphate, ammonium nitrate etc.) and behave differently in the soil [133]. As such, they have varying impacts on crop yield and GHGs emissions. As indicated in Fig. 3, different N fertilizer types influenced rice yield differently under the same N rates. Altering the rates of N applied with biochar also produced varying effects on N₂O and CH₄ emissions under different N fertilizer types. Urea and NPK increased N₂O emission mostly at medium N rate probably due to increased N substrate availability (Fig. 2A). While increase in micro-organism diversity may have been responsible for the enhanced CH₄ emission observed under both NPK and urea at low and medium N rates. Nitrogen rates generally produced inconsistent pattern on both CH₄ and N₂O emissions under different N fertiliser types. Only NPK_urea showed a consistent increase in CH₄ emission as N rate increased. The varied responses of both CH₄ and N₂O to N rates and types is a clear indication that N fertiliser types impact gas emissions from paddy fields differently [132]. However, the nature of this impact and the mechanism involved are still uncertain [11,54]. Varying sensitivity to the products of the hydrolyzation of N fertilizer sources by soil micro-organisms might be responsible for the variation in the observed N₂O and CH₄ emission as N rate changed. For instance, by-products of sulphate and nitrate-based N fertilizer are toxic which may affect nitrification-denitrification process by micro-organisms [132] and eventually N₂O production and emission. These N fertilizers also appear more effective in reducing N-induced CH₄ emission through competition for electrons under anaerobic condition [54,132]. Contrarily, ammonium-based N fertilizer enhances CH₄ emission through the inhibition of CH₄ oxidation by NH₄⁺ [123]. The process involves blocking the systems of methanotrophic enzymes, which causes the inhibition of CH₄ oxidation. Other mechanisms such as the stimulation of CH₄ within the rhizosphere by urea and other N types have also been reported [56]. Increased growth and root respirations, within the rhizosphere depletes the oxygen needed for CH₄ oxidation [72] and thus increases CH₄ emission. It also appears that while (NH₄)₂SO₄ inhibits CH₄ emission, it promotes N₂O production [134] compared to other N fertilizer types [135]. Similarly, the use of urea also promotes N₂O emission [136]. N₂O is produced from nitrification-denitrification process which depends on weather conditions, therefore, varying N₂O production from N types are site and weather dependent [137]. It is important to note that, other elements like phosphorus and potassium were applied to all the experiments, although not as treatments. The varied time and rates of application of these elements may have also influenced the dynamics of CH₄ and N₂O emissions.

Aside GHGs emissions, N rates and N fertiliser types also significantly influenced rice yield (Fig. 3). For all the N fertiliser types, increasing N rate resulted in higher yield. This is due to increased availability of N for plant utilization. NPK and (NH₄)₂SO₄ produced the highest rice yield. The variation in rice yield under the same N rate but not N fertiliser types is an indication that N types affect rice yield.

Combining N fertilizers especially NH₄⁺ and NO₃⁻ based fertilizers have been reported to promote crop yields [133]. However, their effects differ depending on the environment. Compared to NO₃⁻ fertilizers, NH₄⁺ based fertilizer gives better yield under anaerobic conditions [138]. Hydrolysis of NH₄⁺ [e.g. (NH₄)₂SO₄] and NO₃⁻ based fertilizers results in the production of NH₄⁺ and NO₃⁻ respectively. Whereas NO₃⁻ are easily leached out, NH₄⁺ tends to be fixed in the soil [138] for plants uptake. Our result (Fig. 3) shows that the co-application of biochar with NPK resulted in highest rice grain yield followed by combined biochar and (NH₄)₂SO₄ mixed with NPK [90]. has observed that N fertilizer type applied with biochar significantly influences the effect of the co-application to crop yield. High yield from NPK may be linked to increased availability of other nutrients such as phosphorus and potassium contained in the fertilizer. On the other hand, the improved rice yield observed under (NH₄)₂SO₄ fertilizer maybe linked to the toxicity of SO₄²⁻ to soil micro-organisms which may have reduced the microbial transformation of NH₄⁺ and thus made more nutrients available to the plants. Additionally, the sulphur content of (NH₄)₂SO₄ may have also served as an additional source of nutrients for plants’ uptake. Our finding agrees with the result of [133] who reported an improved maize yield under (NH₄)₂SO₄ fertilizer than urea. Similarly, results from [139] showed a 22 % increase in rice yield under (NH₄)₂SO₄ fertilizer compared to urea. While low sugarcane yield from urea amended plots have also been reported by [140] which suggests that urea is less efficient in increasing crop yield possibility due to high loss of N though ammonia volatilization [141,142] and N₂O emissions [135].

Highest GWP was recorded under NPK application due to its low mitigatory role on CH₄ and N₂O emissions (Fig. 2C). Despite its high rice yield, the high N₂O and CH₄ emitted under NPK ensured that GHGI under NPK N type remained the highest. The combined application of organic materials such as biochar and inorganic fertilizer triggers the evolution of CH₄ and N₂O emission which can cause more than 80 % increase in GWP [143]. It is therefore important that caution should be taken in the choice of N fertilizer type to be applied with biochar. Our results suggest that high rate of urea applied with biochar can enhance rice yield and mitigate GHG emissions from rice field.

5. Conclusion

In conclusion, combined application of biochar and N fertilizer is beneficial to both crop yield and GHGs mitigation. However, the benefits depend on both biochar and N rates as well as N source and rate. Low biochar and high N (LBHN) stimulated high N₂O emission, although it increased rice yield and decreased CH₄ emission. Application of low biochar and medium N rates (LBMN) reduced GWP and GHGI while achieving high rice yield.

At high N rate, sole NPK application with biochar, or combined application of NPK and (NH₄)₂SO₄ with biochar were most efficient in increasing rice yield. But due to lack of data, we could not assess the effect of (NH₄)₂SO₄ applied with biochar on N₂O emission, as such, the result should be taken with caution. On the other hand, medium rate of NPK applied with biochar, significantly reduced N₂O but promoted CH₄ emissions. However, the application of medium rate of NPK and biochar appeared most profitable for both rice yield and environmental protection since its stimulatory effect on CH₄ is compensated for by increased rice yield and decreased N₂O. Biochar pH, biochar total N, biochar carbon, soil pH, soil carbon, and soil total N were the most important biochar and initial soil properties regulating rice yield, N₂O, and CH₄ emissions under the biochar and N fertilizer application.

6. Recommendations

• 1.From our review, we found that most researches have been focussed on the effects of nitrogen (N) fertilizer rates on rice yield and greenhouse gases (GHGs) emissions. However, considering the variation in hydrolysation process of various N fertilizer types and the peculiarity of paddy fields, we recommend that future researches on biochar and N application should focus more on the role of N fertilizer types on GHGs emissions.
• 2.Also, given the strong influence of environmental factors on GHGs emissions and the scope of our current study, we recommend that future review should integrate environmental factors in analysing the factors driving GHGs emission from rice field under biochar and N fertilizer application. The integration of environmental factors will help to better isolate under what environmental conditions the recommended biochar and N rate is most suitable.
• 3.There is also an urgent need for more GHGs studies in Africa as the literature clearly shows dearth of research on the subject area in the region.
• 4.Since the current study reviewed only CH₄ emission from NPK and (NH₄)₂SO₄ application with biochar, and given the contradictory pattern of CH₄ and N₂O emissions, our finding should be taken with caution and future research should focus on the simultaneous emission of CH₄ and N₂O from this management practice to make a more informed recommendation.

Data availability

Data will be made available on request.

Additional information

No additional information is available for this paper.

CRediT authorship contribution statement

Maduabuchi P. Iboko: Writing – review & editing, Writing – original draft, Methodology, Funding acquisition, Formal analysis, Data curation, Conceptualization. Elliott R. Dossou-Yovo: Writing – review & editing, Supervision, Methodology, Data curation, Conceptualization. Sunday E. Obalum: Writing – review & editing, Supervision, Methodology, Conceptualization. Chidozie J. Oraegbunam: Writing – review & editing, Resources, Data curation. Siméon Diedhiou: Writing – review & editing, Formal analysis, Data curation. Christian Brümmer: Writing – review & editing, Supervision, Data curation, Conceptualization. Niaba Témé: Writing – review & editing, Supervision, Data curation, Conceptualization.

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.

Appendix A. Supplementary data

The following is the supplementary data to this article: