Biochar and DMPP regulated root-zone CO 2 , N 2 O and 1 CH 4 emissions of soil-ridged/substrate-embedded 2 cultivation sweet pepper in Chinese solar greenhouse

: Northern China is a major production area for off-season vegetables in Chinese solar 10 greenhouse. Usually, greenhouse gas emission flux and coefficient in Chinese solar greenhouse are 11 higher than those in the open field. The reason for this phenomenon is heavy nitrogen (N) 12 fertilization (esp. chemical N and organic manure N) and frequent irrigation during year-round 13 cultivation. A novel substrate cultivation method for vegetable production in Chinese solar 14 greenhouse, called soil-ridged/substrate-embedded cultivation (SSC), was put forward to reduce 15 environmental pollution and increase use efficiency of nutrients. To clarify the characteristics of SSC 16 root-zone greenhouse gas emissions, and the regulation effects of biochar and DMPP addition, five 17 treatments were designed in Chinese solar greenhouse under the same nitrogen application level, 18 including soil-ridge cultivation (SC, as a control), SSC (peat: vermiculite: perlite (v/v=2:1:1), 19 SSC-B50% (biochar: vermiculite: perlite,v/v=2:1:1), SSC-B25% (biochar: peat: vermiculite: perlite, 20 v/v=1:1:1:1), and SSC-DMPP (SSC supplemented with 1% (w/w) DMPP of N fertilizer). Results 21 showed that SSC improved fruit yield of sweet pepper of by 10.99% compared to SC. SSC-B50% and 22 SSC-DMPP significantly improved sweet pepper growth compared to SSC. Moreover, SSC-DMPP 23 increased sweet pepper yield by 10.30% compared to SSC treatment, while SSC-B50% and SSC-B25% 24 treatments lowered the yield by 47.1% and 13.7% separately. Five treatments presented various 25 root-zone temperature features. Also, substrate pH of SC, SSC-B50%, and SSC-B25% is alkaline, 26 while SSC and SSC-DMPP treatments is acidic. Besides, the Global Warming Potential was 27 significantly mitigated in the SSC cultivation compared with the SC. Similarly, the greenhouse gas 28 intensity decreased from 0.074 to 0.038 kg CO 2 -eq kg -1 yield. Compared with the SSC treatment, 29 cumulative N 2 O emissions were significantly reduced in the SSC-DMPP treatment. The greenhouse 30 gas intensity also decreased from 0.038 to 0.033 kg CO 2 -eq kg -1 yield. Thus, we concluded that SSC 31 was a promising method characterized with reduced greenhouse gas emissions and increased fruit 32 yield. Application of DMPP in SSC cultivation significantly reduced N 2 O emissions. We recommend 33 SSC method use in Chinese solar greenhouse with DMPP addition in substrate to optimize 34 greenhouse gas mission. 35


Introduction 38
North China is a major production area for off-season vegetables using Chinese solar 39 greenhouses (CSG). However, a series of agricultural resource and environmental problems, such as 40 soil salinization, underground water pollution, soil continuous cultivation obstacles [1] , and greenhouse 41 gas (GHG) emissions etc. have occurred after the long-term soil cultivation in CSG. Recent data 42 indicated that greenhouse vegetable production is one of the main sources of agricultural N2O 43 emissions [2,3] . In particular, vegetable production in CSG resulted in a higher level of GHG emissions 44 (particularly N2O) because of the overuse of N fertilizers, frequent irrigation, and intensive soil 45 cultivation year-round. This issue is increasingly becoming too important to ignore [4] . Thus, it is vital 46 to explore some simple and effective ways to reduce agricultural resource and environment problems 47 in CSG in North China to enable clean and sustainable greenhouse vegetable production. 48 Soilless culture constituted the main cultivation system primarily used in horticultural facilities in 49 Europe and North America [5] . Studies have shown that soilless cultivation is the most effective way to 50 overcome a series of problems arisen from soil cultivation, such as soil salinization and continuous 51 cropping obstacles [6] . Furthermore, recent reports have also showed that soilless culture can decrease 52 GHG emission, such as CO2, N2O, and CH4 [7][8][9] . Llorach-Massana et al. [7] estimated an emissions factor 53 (EF) of 0.0072-0.0085 kg N2O -1 per kg N -1 for substrate cultivation lettuce. Similarly, Daum and 54 Schenk [8] experimentally assessed an EF of 0.004-0.016 kg N2O -1 per kg N -1 for cucumbers grown in 55 rockwool substrate. These studies had proven that soilless cultivation can produce an EF value that is 56 50% percent less than that accepted by the IPCC (2006) [10] . Furthermore, hydroponics that contain 57 rockwool reduced CO2 emissions by reducing the number of rhizosphere microorganisms [9] . However, 58 the cultivation substrates usually perform poor in temperature stability, leading to large fluctuation in 59 diurnal root-zone temperature of vegetables. Frequent occurrence of low-and high-temperature 60 stresses in the CSG impacted the performance of substrate cultivation [11] . To overcome this issue, Fu et 61 al. [12] invented a novel substrate-cultivation method, called as soil-ridged/substrate-embedded 62 cultivation (SSC), to substitute for soil cultivation. SSC combines the best features of the root-zone 63 temperature buffer capacity of soil cultivation and the high-yield performance of soilless cultivation 64 [13] . Moreover, drip irrigation techniques and custom-made grooves with inserted plastic film were 65 used to accurately control the amount and fates of water and fertilizer, preventing the downward 66 leaching of nutrients. However, compared with soil cultivation, substrate cultivation has different 67 properties and microbial populations in root zone [14] . Accordingly, the root-zone GHG emission 68 characteristics of the SSC root zone may differ from those of soil cultivation. Clarifying the 69 characteristics of GHG emissions of the SSC root zone, and developing effective mitigation pathway is 70 important basis for the evaluation and large-sacle application of the SSC method. In addition, there is 71 no research report on the effects of 3,4-dimethylpyrazole phosphate (DMPP) and biochar on GHG 72 emissions of soilless crops in CSG, particularly SSC method. 73 In the intensive plant cultivation system of the CSG, the use of biochar as a growth substrate may 74 have the benefit to replace non-renewable media, such as peat [15] . Furthermore, it may have the 75 potential to reduce GHG emissions [16] . Biochar is a carbon-rich product produced by the pyrolysis of 76 biomass with little or no oxygen [17] . Biochar improved plant growth had reported by changing the soil 77 properties, such as pH, water retention, and porosity, and promoting microbial activity [18,19] . 78 Numerous studies have highlighted the impacts of biochar on soil environment, crop yields and GHG 79 emissions [20][21][22] , but there are few studies on the use of biochar as a soilless crop cultivation substrate. 80 Awad et al. [23] indicated that using a combination of rice husk biochar and perlite as a matrix increased 81 the growth of leafy plants by approximately two-fold compared with the use of perlite alone. 82 However, Daniele et al. [24] found that biochar only increased the biomass of green tomato, but not the 83 tomato yield, and most of its quality indicators. Furthermore, 550℃ maple bark biochar had the 84 highest inhibitory effect on CO2 emissions, and reduced them by 50%. However, at 700℃, pine chip 85 biochar stimulated CO2 emission and increased its cumulative emissions [16] . Differences between 86 reported GHG emission results may be due to differences in the physical and chemical properties of 87 biochar, depending on the feedstock and the pyrolysis temperature used in the production process [16] . 88 3,4-Dimethylpyrazole phosphate(DMPP) is one of the most effective nitrification inhibitors (NIs) 89 [25] . It has been reported that DMPP has several obvious advantages over other currently used NIs. 90 This compound is highly effective at inhibiting soil nitrification, reducing N2O emissions, increasing N 91 fertilizer efficiency and crop yields [25,26] . Previous studies have shown that the application of DMPP in 92 intensive vegetable fields significantly reduced the cumulative soil N2O emissions by 75% [27][28][29][30] . In 93 addition, this inhibitor significantly increased the yield of corn, wheat, carrots, and lettuce [31,32] . 94 However, to our knowledge, the growth of soilless crops supplied with nitrogen fertilizer combined 95 with DMPP has not yet been reported. To maximize the effective use of N fertilizer, and to minimize 96 the negative environment impacts of greenhouse vegetable production, further research is urgently 97 needed to determine use strategies of biochar and DMPP for SSC method management. 98 In this study, an experiment was conducted in CSG to investigate the characteristics of SSC 99 root-zone GHG emissions, and the regulation effects of biochar and DMPP addition. The purpose of 100 our research was (i) to quantify the characteristics of CO2, N2O and CH4 emissions in the SSC root 101 zone; and (ii) to evaluate the effects of sweet pepper yield and the GHG emissions from SSC in the 102 root zone following the addition of DMPP and biochar. Finally, effective cultivation management 103 strategies were put forward. 104 Our hypothesis were: a) SSC can mitigate GHG emissions in the root zone compared with SC; b) 105 SSC with DMPP can further mitigate GHG emissions in the root zone; c)the ability of biochar to 106 mitigate greenhouse gas emissions in the root zone will vary according to the proportion of biochar in 107 the mixed matrix. 108 Construct an SSC ridge through the following steps. SSC ridge structure and seedlings plant were 132 described by Fu et al. [13] as follows: specially made grooves of wire-mesh within a placed plastic film 133 all-embracing substrate were inserted trimly on the ground with the north-south direction and 134

Materials and methods
subsequently was filled with homogeneous peat matrix, perlite, and vermiculite along with a volume 135 ratio of 2:1:1. Moreover, in order to enhance the capacity of the buffer attributed to the root-zone 136 temperature, the soil was stacked up and piled with considering both sides of the ridge related to the 137 particular grooves of wire-mesh. Ultimately, the ridge of SSC was covered via plastic mulch, and also 138 drip irrigation was implemented to provide water as well as fertilizing through nutrient solution. The 139 characterization of wire-mesh groove in a U-shape is as follows: 10 cm width, 300 cm length, and 10 140 cm height. Further, the ridges distance, namely, 40, 20, and 10 cm were considered for bottom width, 141 top width, and height, respectively, that was declared owing to the production requisiteness. The 142 specification of SC was similar to the regular SSC ridges. All procedures were performed by three 143 replicates. The ridge length in the direction of the north to south remained 3.0 m, and ten sweet 144 peppers are planted. 145 The implemented nutrient solution had the following composition (mmol·L -1 ): 0.77 K2SO4，0.1 146 KCl, 0.5 KH2PO4，0.65 MgSO4·7H2O，1×10 -3 H3BO3，1×10 -3 MnSO4·4H2O，1×10 -4 CuSO4·5H2O，1× 147 10 -3 ZnSO4·7H2O，5×10 -6 (NH4)6MO7·4H2O，0.1 EDTA-Fe，3.39 Ca(NO3)2 ·4H2O [34] . Except for the 148 SSC-DMPP treatment, the other four treatments are provided by a unified nutrient solution tank.

149
SSC-DMPP treatment added DMPP with a nitrogen application rate of 1% and irrigated the plants 150 together with the nutrient solution. The EC and pH of the nutrition materials have remained at 2.0 151 dS·m -1 and 6.5, respectively, via modifying the solution existed in the tank of nutrient solutions every 152 10 d.

153
Before transplanting seedlings, samples were collected in the soil or substrate of the cultivation 154 area, mixed and stored at -20°C for analysis of their physical and chemical properties. Total N, 155 available P, K contents of experimental soil, and substrate before cultivation were measured using the 156 procedure detailed by Lu et al. [35] , and the results are presented in Table 1. 157

163
When all considered procedures were fulfilled successfully, a drip irrigation system was 164 implemented. The cultivation process was divided into three major steps, including vegetative 165 growth, flowering, and ripening processes as the first to third steps. Afterward, the achieved seedlings 166 (30 d after transplantation) were watered by using automatic systems via a nutrient solution at 10:00 167 within the vegetative growth, and also the volume of irrigation related to each seedling was around 20 168 ml at a precise time. At the second step (flowering), the sweet peppers were all irrigated by taking 169 advantage of nutrient solution at 10:00, while the volume of irrigation for each plant was about 40 ml 170 at a time. At the third step (ripening), planted sweet peppers were irrigated through nutrient solutions 171 at 10:00, and 13:00, respectively, and also the volume of daily irrigation for each plant was around 80 172 ml. The irrigation amount was assigned according to the growth of sweet pepper. 173

Growth parameters measurement 174
Leaf chlorophyll content, plant height, and stem diameter were evaluated via chlorophyll meter 175 (SPAD-502, Konica Minolta Sensing Co., Osaka, Japan), ruler, as well as vernier calliper at 30 d, 60 d, 176 and 90 d after transplantation. The fruits of ripen sweet pepper were harvested on Jan. 15, 2020, Jan. 177 30, Feb. 15, Feb. 29, and Mar. 15. Five plants were randomly selected from each treatment, and the 178 yield of pepper was measured by accumulative balance to get the yield of the plant. Total output 179 calculations were based on planting density and area. After the sweet pepper fruits were harvested on 180 March 15th, the plants were subjected to a seedling pulling treatment. The five plants tested were 181 washed thoroughly with faucet water, rinsed well with deionized (DI) water, and next separated into 182 stems and roots. Fresh masses of roots and shoots were measured. Plant organs were dried for at least 183 48 h at 105°C in a ventilated oven to determine their dry weights [36] . 184

Root-zone temperature, substrate EC and pH determination 185
The root-zone temperature was recorded every 10 min by the CR1000 data collector and T-type 186 thermocouple wire produced in the United States. Root-zone temperature measurement was performed 187 in the vertical profile at a depth of 5 cm. According to the different growth stages of sweet peppers, we 188 continuously selected Nov. 7-11, 2019, Dec. 7-11, and Jan. 7-11, 2020 for data analysis. 189 Soil and substrate samples were collected before seedling transplantation and after harvest, and 190 fresh samples were dried. Unless otherwise specified EC and pH measurements were carried out in 191 triplicate. EC and pH were assessed as defined by Lu et al. [35] via adding the 10 g of soil and substrate in 192 25 mL DI water. Then, after 1.5 h of stirring, pH (PHS-2F, Shanghai INESA Scientific Instrument Co., 193 Ltd., China) and EC (DDSJ-308F, Shanghai INESA Scientific Instrument Co., Ltd., China) were 194 evaluated on the passing of filtration using a filter paper with VWR grade 413 (5µm). 195

Gas sampling and determination 196
Greenhouse gas(GHG) fluxes were evaluated via the method of the static chamber [37] . Before gas 197 extraction, One PVC collar was inserted in per plot, and the top side of the collar possessed a groove, 198 loaded with water 5 cm high in order to seal the flange of a 10cm × 10cm × 40cm gas-sampling chamber.

199
A fan was fixed on the top of the box to mix the gas, and the gas sampling tube and temperature probe 200 were respectively placed inside the box. Briefly, gas samples were collected between 10:00 and 11:00 in 201 the early hours of each sampling day. The utilized lid apparatus in the top-side has a gas sampling 202 three-way valve connected to the gas sample bag, five 100 mL headspace samples were taken using a 203 150 mL gas sample bag at 0, 10, 20, 30, and 40 min after the shutting of the chamber. Then, a 100ml 204 gas-tight syringe was used to transfer 40 mL of gas in the gas sample bag to a special vacuum glass 205 bottle, which was instantly taken to the laboratory for further investigation. The concentration of 206 existed gas in the specimen was studied comprehensively in 24 h after sampling procedure, an Agilent 207 7890A gas chromatograph provided with a flame ionization detector (FID) for CO2 and CH4 detection 208 as well as an electron capture detector (ECD) in order to detect N2O detection [38] . The carrier gas in 209 ECD and FID were Ar-CH4 and High-purity N2, respectively. The gas specimen was separated via 210 columns made of stainless steel and packed with Porapak Q (80/100 mesh. The oven temperature was 211 adjusted at 55℃, and the temperature of the FID and ECD was fixed at 330℃ and 200℃, respectively. 212 The gas chromatography configurations defined by Xiang et al. (2015) [38] were implemented for the 213 analysis of gas concentration. GHG fluxes were assessed via utilizing the linear increases within gas 214 concentration with time. Specimen sets were denied unless they created a linear regression value of 215 R 2 ≥0.90 [39] . 216

Data analysis and statistics 217
The flux of Gas emissions (mg·m −2 ·h −1 ) was measured using the following equation [38] : Where F is indicated in mg·m −2 ·h −1 ; ρ states the gas density at the standard state, V(m 3 ) represents 220 the chamber volume, A (m 2 ) states the area of the pot, dc/dt is the modification of gas concentration 221 with time, and T (°C) is the average temperature inside the considered chamber within the sampling 222 period, P0(mm•Hg) represents standard atmospheric pressure, P/P0≈1. 223 The seasonal cumulative gas emissions (E, kg N2O-N ha -1 ) were calculated regarding to the 224 equation characterized utilizing the Eq. (2) [38] : In this equation, Fi and Fi+1 state the N2O (or CH4, CO2) efflux at the ith and (i+1)th assessment time 228 (mg N/C m -2 h -1 ), respectively, ti+1-ti is the interval between the ith and (i+1)th assessment time (d), and n 229 is the total assessment time.

230
In order to realize the climatic effects of the vegetable fields systems under the addition of DMPP 231 and biochar, we present GWP to determine greenhouse impacts applying the equation altered from 232 Zhang et al. [37] . 233 GWP(kg 2 − equivalent ha −1 ) = 25 × GWP ( 4 ) + 298 × GWP ( 2 ) + GWP ( 2 ) 234 (3) 235 The greenhouse gas intensity (GHGI) is the other concept that make a connection between 236 biochar and DMPP to GWP, which was measures via dividing GWP by the yield from vegetable 237 [40]. 238 Statistical analyses were carried out utilizing SPSS 25  height was always highest in SSC-DMPP treatment, and it was significantly higher than the other four 255 treatments in the first two stages. In the ripening stage, the SSC-DMPP treatment changed from 256 vegetative growth to reproductive growth, and its growth rate was relatively lagging, while the growth 257 rate of SC treatment in the flowering to ripening stages accelerated, and its plant height in the ripening 258 stage was not statistically different from SSC-DMPP treatment. SSC-B50% and SSC-B25% of biochar 259 treatments grew slowly in the growing season of sweet pepper, and was consistently significantly 260 lower than SC treatment. 261 It can be seen from Fig. 1(B) that the stem diameter of sweet peppers at different growth stages 262 that have been treated with SSC-DMPP to be the highest, and SSC-B50% is the lowest, and there are 263 significant differences between the two treatments (P < 0.05).

264
SPAD value indirectly evaluates the chlorophyll content of sweet pepper leaves. It can be seen 265 from Fig. 1(C) that the SPAD value of each treatment did not show an increasing trend in the first two 266 stages, only the chlorophyll content of the sweet pepper leaves increased during the ripening stage. In 267 addition, the chlorophyll content of sweet peppers in different growth stages has been the highest in 268 SSC-DMPP treatment, and was significantly higher than the other four treatments in the vegetative 269 growth period. Compared with SC treatment, SSC treatment had no marked difference in chlorophyll 270 content during the whole growth stage. The two treatments of SSC-B50% and SSC-B25% significantly 271 reduced the chlorophyll content.

Root-zone temperature at different growth stages of sweet pepper 292
As shown in Table 3-5, with the continuous growth of sweet pepper, the various indexes of 293 root-zone temperature in each treatment gradually decreased. The root-zone temperature reached the 294 highest on November 7-11, and the lowest on January 7-11. Whether it is the root-zone temperature 295 during the day or night, the difference between every two months reaches about 5°C. The highest 296 root-zone temperature varies between November and December. The lowest root-zone temperature in 297 November is 10°C higher than that in December. The lowest temperature in the root-zone gradually 298 decreases to about 3°C every two months. In the same month, the differences of various indicators 299 between different treatments were all between 0-3°C. 300

Substrate electric conductivity and pH 307
As shown in Fig. 2, SC, SSC-B50%, and SSC-B25% treatments are an alkaline cultivation media, 308 while SSC and SSC-DMPP treatments are an acidic cultivation media. After the sweet pepper was 309 harvested, the pH value of SC, SSC and SSC-DMPP treatment increased, while the pH value of 310 SSC-B50% and SSC-B25% treatment decreased. Different from pH value, the (electric conductivity)EC 311 value of substrate cultivation treatment was significantly higher than that of soil (P<0.05), and the EC 312 background value of SSC-B50% treatment was the highest. After the sweet pepper is harvested, the EC 313 value of SSC-B25% treatment is the highest. Besides, the EC value of SSC-DMPP treatment was 314 significantly lower than that of SSC treatment(P < 0.05).

Root-zone CO2, N2O and CH4 emissions at three growth stages 321
Among the three stages, the CO2 emission in vegetative growth period was the highest, and all 322 treatments showed a trend of gradual decline with the delay of growth period, as shown in Fig. 3A. 323 The CO2 emissions of different treatments in the same growth stage were significantly different (P < 324 0.05, Fig. 3B). In the first four weeks after transplantation (growth stage I), SSC-B50% treatment had 325 the highest CO2 emissions (approximately 427 kg CO2-C ha -1 ), which was significantly higher than 326 other treatments. SSC treatment has the lowest CO2 emissions (approximately 229 kg CO2-C ha -1 ), 327 which is not significantly different from SSC-B25% and SSC-DMPP treatments. In the latter two stages, 328 the CO2 emissions of SC treatment are the highest and are significantly higher than the other several 329 treatments. The CO2 emissions of SSC, SSC-B25%, and SSC-DMPP treatment are about half that of SC 330 treatment.

331
In contrast, the N2O emissions of each treatments are significantly different at the same stage (P 332 <0.05, Fig. 3C), but there are differences from the CO2 emissions model. Unlike the CO2 emissions, the 333 SC treatment was significantly higher than the other four treatments throughout the growth stage and 334 showed a gradual upward trend as the growth period delayed (Fig. 3C). However, the N2O emissions 335 of SSC, SSC-B50%, SSC-B25%, and SSC-DMPP treatments remained low, and the four treatments had 336 the highest N2O emissions in the first 4 weeks after transplantation (growth stage I). In the four weeks 337 after transplantation, compared with SSC treatment (approximately 0.125 kg N2O-N ha -1 ), SSC-B50%, 338 SSC-B25%, and SSC-DMPP treatments all significantly reduced N2O emissions, and there was no 339 significant differences in N2O emissions during the entire growth period among the three 340 treatments (Fig. 3D). 341 The CH4 emissions fluctuated greatly during the cultivation period (Fig. 3E), while the CH4 342 emissions between the treatments were significantly different at the same stage (P < 0.05). In the first 343 two growth stages, except for the SC treatment, the CH4 emissions of the other four treatments did not 344 exceed 0.2 kg CH4-C ha -1 . But in the ripening period, except for the SSC treatment, the CH4 emissions 345 of the other four treatments all exceeded 1.5 kg CH4-C ha -1 (Fig. 3F).  The cumulative N2O emissions in the control was significantly higher than those in the SSC, 354 SSC-B50%, SSC-B25%, and SSC-DMPP treatments, which was up to 0.79 kg N2O-N ha -1 (     SSC is a novel substrate cultivation method for fruit and vegetable production in CSG with 375 significant practical advantages that can solve a series of environmental problems caused by the 376 production of fruits and vegetables [13] . Compared with SC treatment, that of SSC improved the yield of 377 peppers by 10.99%, although it did not facilitate the growth of plants. The reason for this result could 378 be that the SSC model takes full advantage of the high yield of soilless culture and is more conducive 379 to promoting the reproductive growth of crops [41] . In addition, as a root-limited cultivation and plastic 380 film barrier, SSC can effectively retain nutrients in the zone of root growth, which is convenient for 381 sweet peppers. However, soil cultivation cannot maintain nutrients in the root zone, which then easily 382 diffuse into the surrounding areas, resulting in insufficient nutrients in the root zone, thus, weakening 383 its reproductive growth [13] .

384
In this study, biochar was tested as a cultivation substrate. Compared with the SSC treatment, the 385 two treatments with the addition of biochar resulted in poorer growth during the three stages of 386 growth of sweet pepper (Fig.1). In fact, the low temperature pyrolysis method for making biochar may 387 have created toxic compounds that inhibit plant growth [42] . In addition, biochar with highly variable 388 amounts of volatile matter will also inhibit plant growth in the short term owing to the immobilization 389 of nitrogen [42] . However, the biomass accumulation of the SSC-B50% treatment increased significantly 390 by 7.60% compared with the SSC control. In terms of biomass accumulation, the reason why 391 biochar-containing substrates perform better may be owing to the interaction between biochar and 392 plant nutrition, which is based on what was proposed in previous studies with vegetable crops [15] . It is 393 worth noting that the higher accumulation of biomass that was found in sweet peppers grown on 394 biochar did not lead to enhancements in yield. In fact, compared with treatment with peat, the plants 395 grown in biochar media exhibited different trends of the distribution of organs, which aids in the 396 development of accumulation of plant biomass instead of reproductive organs [15] . Furthermore, the 397 allelopathic effect of biochar-derived hydrocarbons or toxic levels of heavy metals can also have 398 deleterious effects on yields [42] . There are fewer studies on the use of biochar as a soilless culture, and 399 more reports explain the mechanism for the positive impact on crop yields [23,43] . Therefore, insights 400 into the causes of negative effects of biochar on yield are necessary.

401
Applications with DMPP fertilizer promoted the growth of sweet peppers (Fig. 1). In many 402 experiments with vegetables, when fertilizer is applied in concert with DMPP, NH4 + -N is supplied to 403 crops at a higher rate over a longer time. Plants will absorb a large amount of NH4 + -N, which will 404 reduce the concentration of NO3 - [32] . In addition, DMPP-induced partial NH4 + -N may decrease the pH 405 and increase microbial activity in the rhizosphere [32] . This lower pH should improve the ability of 406 plants to absorb other nutrients, particularly micronutrients and lay a foundation for crops to absorb 407 the nutrients needed to grow and improve their quality [32] . This could be confirmed in this experiment 408 (Fig. 2). More importantly, the application of DMPP not only increased the quality of the plants' dry 409 matter but also significantly increased the yield of sweet peppers by 10.30% compared with the SSC 410 treatment. The results of applying DMPP to promote vegetable yield have been reported in previous 411 studies [32] . This could be because the application of DMPP reduces the leaching of nitrate nitrogen, 412 maintains the efficiency of fertilizer of the nutrient solution, and ensures that there is sufficient fertility 413 in the substrate during the latter period of nutrient absorption of sweet peppers, thereby increasing 414 their yield [31,32] . Therefore, fertilizers combined with DMPP may improve the yields of agricultural 415 crops and promote the green, clean and sustainable development of agriculture. 416

Effects of adding biochar and DMPP into SSC substrate on root-zone GHG emissions 417
An increasing number of studies have shown that GHG emissions from horticultural products are 418 mostly because of the cultivation procedure, rather than the industrial products of materials attained 419 from cultivation: viz., biocides, fertilizer, electricity, etc. [7,9,44] . Our research concentrated on the 420 cultivation procedure and compared CH4, CO2, and N2O emissions from soil cultivation and substrate 421 cultivation. The results show that the rhizosphere regions GHG emissions of SSC treatment are 422 significantly reduced compared to SC treatment (Fig. 3). The total GHG emissions of SC and SSC are 423 3387 and 1899 kg CO2-eq ha -1 , respectively. This explains that the SSC procedure indicated a positive 424 impact on global warming. In fact, the soil includes a huge pool of carbon and abundant 16S rRNA 425 genes. Therefore, a greater amount of CH4 and CO2 emissions from soil could be easily described as a 426 great number of microorganisms and also resulted in a high rate of microbial soil respiration [9,45] . 427 Interestingly, the N2O emissions in the soil come from nitrification and denitrification reactions. 428 However, the inert conditions of substrates used in soilless culture and their poor water retention 429 capacity inhibit the growth of microbial populations, making denitrification may be the main 430 mechanism [7,46] . But the denitrification route is further limited by the high porosity of the substrate, 431 which leads to the aerobic environment [46] . Therefore, soilless cultivation may alleviate N2O emissions. 432 Generally, peats in enhanced decomposition release less amount of CO2 and are rich in recalcitrant 433 C [47] . In the present study, both SSC-B50% and SSC-B25%, which were replaced by peat increased the 434 cumulative emissions of CO2 and CH4 (P < 0.05). Although we did not observe the total C content in 435 the mixed matrix, the cumulative CO2 emissions of SSC-B50% treatment was significantly higher than 436 that of SSC-B25% treatment. The increase in CO2 emissions caused by the addition of biochar may be 437 because of: (i) abiotic release of inorganic C [48]; (ii) microbial decomposition of more unstable biochar 438 components [49] . Moreover, there was evidence that a strong correlation was seen between the 439 cumulative emissions of CO2 and CH4, merely anaerobic methanogenic archaea utilize CO2 as an 440 energy resource, releasing CH4 [16] . Especially when the rhizosphere regions were at a neutral to 441 alkaline pH, the addition of biochar will enhance CH4 emissions [50] . In contrast, the addition of biochar 442 mitigating N2O emissions ( Table 2). The mitigation of N2O by biochar addition was related to several 443 potential mechanisms: (i) Inhibit denitrification by improving rhizosphere regions aeration; (ii) 444 Increase pH to promote complete denitrification; (iii) decrease in soil N availability for microbial 445 activities through immobilization or sorption; (iv) induce the toxic impacts on microorganisms 446 involved in cycle of N [51] . Overall, Biochar completely replaces peat and may lead to increased 447 emissions of CO2 and CH4, but the partial addition of biochar into the peat matrix seems to be a 448 fascinating practice for developing clean and green production. Future investigation should survey the 449 impact of the increment of various rates of biochar to peat matrix on yields of myriad CSG crops as 450 well as GHG emissions during production. 451 Notably, DMPP treatment with 1% pure nitrogen reduced potential N2O emissions by a further 452 52.2% compared to SSC treatment (Table 6). This was also found in our previous experiments. This 453 could be elucidated through the following description: (i)DMPP straightly repress N2O emissions from 454 nitrification-mediated pathways via hindering the metabolic activity and growth of AOB and feasibly 455 AOA [52] ; (ii)The increasing of DMPP significantly improved the affluence of both nosZI-N2O and 456 nosZII-N2O reducers, and the ability of N2O reduction to N2 was enhanced, thus reducing N2O 457 emissions. The impact of DMPP on CO2 and CH4 emissions is still controversial [53] . Although, in this 458 experiment, the addition of DMPP reduced the cumulative CO2 emissions, and it promoted CH4 459 emissions. DMPP can reduce the release of CO2, which may be due to the reduction of organic carbon 460 mineralization and carbon decomposition in the rhizosphere [54] . For the increase of CH4 emissions, we 461 cannot exclude the increase of rhizosphere microorganisms or N application [9] . In fact, the specific 462 micro-mechanism of how DMPP affects CO2 and CH4 is still unclear, which needs further study and 463 discussion. 464

Effects of the root-zone environment on N2O emissions and plant growth 465
N2O emissions from the root zone are not only affected by N fertilization but also by factors, such 466 as temperature, pH and EC [8,9] . Although our experiments showed that the temperature of root zone 467 gradually decreased with the growth period, the N2O emissions did not decrease as a result. 468 Interestingly, the N2O emissions of soil cultivation increased in the latter two periods compared with 469 the first period. The reasons for this phenomenon could include the following: (i) the root zone 470 temperature is maintained at 10~30℃, which is conducive to nitrification and denitrification to 471 produce N2O, and (ii) we provided abundant NO3 -substrates for the production of N2O by 472 denitrification. Thus, this promotes the production of N2O emissions. However, the role of 473 temperature in the nitrification and denitrification process in the root zone is more complicated. 474 Future research should comprehensively consider the interactive effects of temperature and other 475 control factors. 476 The optimal pH in the root zone of most substrates ranges from 5.5 to 6.5. Since the nutrient 477 solution of each plant in the root zone is limited, the risk of exceeding or falling below this value will 478 increase. The growth of most plants is restricted when they are exposed to external pH levels > 7 or < 5 479 [55] . This experiment showed that alkaline biochar as a cultivation substrate has a negative impact on 480 the growth of sweet pepper plants. In addition, pH is an important and complex influencing factor in 481 the process of producing N2O emissions in the root zone. It affects N2O emissions by directly or 482 indirectly affecting the activities of microorganisms involved in the nitrogen conversion process and 483 the activities of enzymes at different stages of action [56] . Our results show that soil cultivation provides 484 a more suitable pH value for nitrification and denitrification and promotes the emissions of N2O. The 485 EC is considered to be one of the most important characteristics in soilless culture [57] . Typically, the EC 486 value is used to evaluate the growth and yield of plants. If the EC is too low, the supply of some 487 nutrients to the crop may be inadequate. Similarly, when the EC is too high, the plants are exposed to 488 salinity [58] . In summary, it is very important to determine the substrate ratio suitable for the growth of 489 sweet pepper, because it not only affects the yield but also affects the production of GHG emissions. 490 The ratio of biochar to peat merits further testing and exploration. 491

Conclusions 492
In this study, the SSC method significantly increased the fruit yield of sweet pepper by 10.5% 493 compared to SC. SSC-B50% and SSC-DMPP plants had a significant increase in root and shoot fresh 494 weight compared with SSC treatment. SSC-DMPP treatment increased the yield of sweet pepper per 495 plant by 10.7%, but SSC-B50% and SSC-B 25% treatments reduced by 45.2% and 13.1%, respectively. 496 Furthermore, SSC significantly reduced the cumulative emissions of CO2, N2O, and CH4 compared to 497 SC. The SSC-DMPP treatment significantly reduced the cumulative N2O emissions by reducing the 498 N2O emissions during the vegetative growth period and fruiting period of the sweet pepper, 499 compared with the SSC treatment. However, the SSC-B50% treatment increased the CO2 and CH4 500 emissions, thereby significantly improved GWP. This study shows that SSC reduce N2O emissions 501 when compared to conventional crops, making soilless crops an attractive practice for reducing GHG 502 emissions. It should be emphasized that the results of this study are based on the parameters 503 described here, such as substrate type, fertilizer and irrigation systems, but it does not provide a single 504 GHG emission result for all soilless crops. We hope that the results of this study will help to provide a 505 new way of thinking about greenhouse vegetable cultivation.

509
Declaration of competing interest: The authors declare that they have no known competing financial interests or 510 personal relationships that could have appeared to influence the work reported in this paper.