Composting modifies the patterns of incorporation of OC and N from plant residues into soil aggregates

1 Composting modifies the patterns of incorporation of OC and N from 1 plant residues into soil aggregates 2 Remigio Paradelo1, Thomas Lerch2, Sabine Houot3, Marie-France Dignac3 3 1. Departamento de Edafoloxía e Química Agrícola, Universidade de Santiago de 4 Compostela, Santiago de Compostela, Spain 5 2. iEES-Paris, UMR Université Créteil, UPMC, CNRS, IRD, INRA, Bondy, France 6 3. UMR Ecosys, INRA AgroParisTech Paris Saclay University, 78850 Thiverval-7 Grignon, France 8 9 10 Abstract 11 In order to understand if and how composting modifies the dynamics of incorporation of 12 organic carbon and nitrogen to soil aggregates during residue decomposition, we studied 13 samples of a soil amended with fresh or composted plant residues in the laboratory. 14 Samples from the surface horizon of a Luvisol (representative of large agricultural regions 15 in northern Europe) were amended with fresh or composted 13C-15N-labelled wheat, corn 16 and rapeseed residues and incubated in the laboratory during three years. Aggregates of 17 samples taken at the initial time, after 45 days and after three years of incubation were 18 fractionated by size; OC and N concentrations plus isotopic signatures of C and N were 19 analysed in each aggregate fraction. Both fresh and composted residues addition 20 increased aggregate stability at 45 days, but the effect almost disappeared after three years. 21 The addition of the residues increased OC and N contents with respect to the control soils 22 after three years. This additional OC and N coming from residues were initially 23 incorporated preferentially into 0.05-0.2-mm aggregates, as indicated by isotopic 24 signature analyses, and later redistributed homogeneously among all size fractions. After 25 2 three years of incubation, 11% of plant residue OC remained in soil when these were 26 added fresh, and 22% if they were previously composted, whereas similar N percentages 27 (26-27%) remained for fresh or composted residues. Overall, a higher amount of inputs 28 from plant residues remained in the soil in the long term when they were previously 29 composted, with respect to non-composted residues. 30 Keywords: Soil organic matter; 13C; 15N; Isotopic fractionation; Soil aggregation; 31 Compost. 32 33 Introduction 34 The biological interest of organic matter in ecosystems relies to a great extent on its 35 decomposition that releases the nutrients and energy that are necessary for plants and 36 microorganisms (Janzen, 2006), but also in its potential to offset current greenhouse gas 37 emissions (Paustian et al., 2016) and limit climate change (Smith, 2016). Incorporation 38 of organic carbon (OC) and nitrogen (N) in soil aggregates during plant residue 39 decomposition is a key edaphic process with consequences for the global biogeochemical 40 cycles of these elements. This process is responsible to a great extent for the development 41 of soil structure, which is one of the specific features of soils, as well as for the protection 42 and accumulation of soil organic matter (SOM). Aggregate formation and the C cycle are 43 related through SOM protection and occlusion, as physicochemical protection offered by 44 aggregate structures and mineral surfaces is critical for building and maintaining soil OC 45 and N stocks (O’Brien and Jastrow, 2013). 46 The canonical theory of soil aggregation, by Tisdall and Oades (1982), states that 47 aggregates are formed when input of plant residues to soil stimulates microbial growth 48 and activity, which generates transient binding agents mostly involving polysaccharides 49 in the first few weeks. In the modification of the model by Oades (1984), aggregation 50 3 proceeds firstly by the formation of macroaggregates (>250 µm) due to mineral 51 associations with particulate organic matter (POM) via temporary binding agents. 52 Afterwards, microaggregates (<250 µm) are formed once the decomposition of POM 53 within macroaggregates by microorganisms produces exopolysaccharides and other 54 metabolites which act as persistent binding agents. In this sense, microaggregates are 55 increasingly considered as key constituents for soil structure, properties and functions 56 (Six et al., 2004; Totsche et al., 2018). 57 Thus, aggregate formation and OM protection are intimately related to the dynamics 58 of organic matter (plant residues, organic amendments) in soil. The amount of fresh 59 organic matter and its degree of decomposition are controlling the processes involved in 60 the formation and stabilization of aggregates (Haynes and Beare, 1996). In this sense, the 61 nature and composition of residues entering the soil are key factors that influence these 62 processes. The conceptual model by Monnier (1965), reviewed later by Abiven et al. 63 (2009), explains the differences in the increase of aggregate stability as a consequence of 64 differences in the composition of the residues. Some products, such as green manure, 65 produce intense effects on aggregate stability within a month, while less readily 66 decomposable materials, such as mature manure, have small initial effects but relatively 67 large effects at the annual scale. Monnier proposed that the short-term effects of organic 68 matter on aggregate stability are due to the turnover of microbial products and cells, while 69 the long-term effects are due to compounds in a more advanced stage of decomposition. 70 In this sense, composting plant residues before adding them to soil may affect their 71 dynamics of decomposition and stabilisation (Lerch et al., 2019). Composting is a process 72 for stabilization of organic wastes, where OM is progressively processed by 73 microorganisms and converted into “stable” or mature compost (Zucconi and de Bertoldi, 74 1987). Obviously, the nature of the process implies a transformation of OM and the decay 75 4 in soil is different depending on whether residues are composted or not, and even for 76 composted materials, depending on the degree of maturation (Bernal et al., 1998a; 77 Francou et al., 2005). Aggregate stability in the long term will depend on the existence of 78 permanent (non-transient) aggregation agents or binding compounds: their production is 79 supposed to be enhanced by compost application (Whalen et al., 2003; Annabi et al., 80 2007) and might be different with the application of composted and non-composted plant 81 residues (Annabi et al., 2007). 82 In view of the close linkage between aggregate structure and OM turnover, aggregate 83 fractionation techniques are regularly used to study the carbon sequestration process 84 (Blanco-Canqui and Lal, 2004; Six et al., 2004; O’Brien and Jastrow, 2013; Guan et al., 85 2015). Stable isotope analysis (13C and to a lesser degree 15N) is an increasingly used tool 86 for estimating SOM dynamics (Balesdent and Mariotti, 1996; Ehleringer et al., 2000) and 87 in association with aggregate fractionation can help determine the precise location of C 88 and N from plant residues, their evolution and turnover rate in soils (Angers et al., 1997). 89 In order to study the processes of C and N incorporation into soil and of stable 90 aggregate formation, as well as whether differences exist in these processes due to 91 composting of plant residues before soil application, we have separated and characterized 92 aggregates in samples from a three-year incubation experiment of soil with composted 93 and non-composted isotopically-labelled plant residues (Lerch et al., 2019). We have 94 measured aggregate stability at two moments of incubation: 45 days (short-medium term) 95 and three years (long term). The use of isotopically labelled residues (13C and 15N) 96 allowed looking for differences in the incorporation of OM into aggregates. Our 97 objectives were to better understand the effect of composting on the dynamics of C and 98 N in soils and to investigate the potential relationships with aggregate stability. 99 100 5 Materials and methods 101 Plant material 102 Plant residues were a mixture of wheat straw, corn stalks and leaves, and rapeseed leaves. 103 All these residues were enriched highly in 13C, some were also enriched in 15N. These 104 highly enriched plants were mixed with grass clippings and wheat straw with natural 105 isotopic contents. Boxes containing 13C and 15N enriched plant mixtures were placed in 106 pilot composting devices where natural abundance plant residues were composted during 107 12 weeks with forced aeration (Doublet et al., 2011). Mass loss during composting was 108 80% dry weight, 82% for C, 73% for N. 109 The OC contents for fresh and composted residues were 39% and 35%, respectively, 110 with C/N ratio of 11.3 and 7.7, respectively. The 13C signatures were 271‰ and 263‰ 111 and the 15N signatures were 1064‰ and 1473‰, respectively, for composted and fresh 112 materials. The relative amount of non-cellulosic sugar measured by the acetate alditol 113 method (Rumpel and Dignac, 2006) was 36 and 28 µg g-1 C for fresh and composted plant 114 residues, whereas the amounts of lignin were 15 and 25 µg g-1 C for fresh and composted 115 plant residues, respectively. Further details about the composition of plant residues, the 116 composting process and analysis of the materials can be found in Lerch et al. (2019). 117 118 Soil incubation set up 119 The soil is a silt loam (7% sand, 76% silt, 17% clay) Haplic Luvisol (IUSS Working 120 Group WRB, 2014), with 8.9 g kg-1 of total organic carbon, 1.1 g kg-1 total N, and a pH 121 of 6.9, obtained from the surface layer (0-28 cm) of an experimental plot located at 122 Feucherolles (France). The bulk isotopic compositions were –24.8±0.2‰ for δ13C and 123 5.9±0.2‰ for δ15N. Soil samples (125 g dry weight equivalent) were introduced in serum 124 bottles with Teflon® rubber stoppers crimped on with aluminium seals and amended 125 6 either with 360 mg of composted plant residues or 330 mg of non-composted plant 126 residues, corresponding to a C addition of 11.5% of the soil OC. Control samples did not 127 receive any amendment. Every sample moisture was adjusted to pF 2 (approximately 0.25 128 g H2O g-1) and incubated at 20°C in the dark for 3 years. A whole description of the 129 incubation set up and analyses performed during the incubation can be found in Lerch et 130 al. (2019). 131 132 Aggregate fractionation and analysis 133 Samples from the incubation were selected for analysis at three times: initial, 45 days, 134 three years, in order to follow the effect of the amendments in the medium and long-term. 135 Three replicates for each incubation duration were used. The study of aggregate 136 composition was performed following the same procedure as in Paradelo et al. (2016). 137 Water-stable aggregates were fractionated according to sizes using the slow wetting test 138 under controlled tension (Le Bissonnais, 1996), assumed to mime soil rehydration by 139 gentle rain in field conditions. It is less destructive than fast wetting and enhances better 140 discrimination between little stable and unstable soils which is the case for the agricultural 141 silty soils developed on loess which are typical of the region (Jamagne, 2011). Briefly, 142 five grams of calibrated air-dried aggregates (3.15–5 mm) were placed on a filter paper 143 on a tension table and maintained at a matrix potential of -0.3 kPa for 60 min. It is 144 important to say that the C and N composition of the calibrated aggregates is not different 145 from that of the bulk soil (Lerch et al. 2019). Once moisture equilibrated, aggregates were 146 transferred on a 0.05-mm sieve immersed in ethanol for the determination of aggregate 147 size distribution. Ethanol was used for wet sieving because it reduces additional 148 breakdown due to its low surface tension. The 0.05-mm sieve was gently moved five 149 times with a ‘Hénin’ apparatus (Feodoroff, 1958; Hénin et al., 1958) to separate 150 7 aggregates <0.05 mm from those >0.05 mm. Then the >0.05-mm fraction was collected, 151 oven-dried and gently dry-sieved by hand using a column series of six sieves: 2, 1, 0.5, 152 0.2, 0.1 and 0.05-mm mesh. Mass percentages of aggregate-size fractions were then 153 calculated. The mass of the <0.05-mm fraction was evaluated as the difference between 154 the initial mass and the sum of the six other fractions. The aggregate mean weight 155 diameter (MWD) for each sample was also calculated. Given that all samples come from 156 the same soil, sand corrections were not performed. Three replicate fractionations were 157 performed for each treatment. 158 The aggregate fractions were milled (<0.2 mm), and OC and N contents were 159 measured by dry combustion in a CHN autoanalyser (Carlo Erba NA 1500). Carbon and 160 nitrogen isotopic contents were measured by EA-IRMS (NA-1500, Carlo-Erba) and 161 expressed with δ notation on a per mil basis (‰), as follows: 162 163 𝛿(‰)=𝑅!− 𝑅!"# 𝑅!"# ·1000 164 165 where Rs is the 13C/12C or 15N/14N isotope ratios of the sample and Rstd is the 13C/12C ratio 166 of the VPDB standard (Coplen, 1995) or 15N/14N ratio of atmospheric N2 (Mariotti, 1983). 167 Precision of measurements was 0.1% for δ13C and 0.2% for δ15N. 168 In the amended soils, the contribution of labelled plant residues to aggregate OC and 169 N was calculated as follows: 170 171 𝐶$=100 ·𝛿%&𝐶'( − 𝛿%&𝐶)( 𝛿%&𝐶$− 𝛿%&𝐶)( 172 173 𝑁$=100 ·𝛿%*𝑁'( − 𝛿%*𝑁)( 𝛿%*𝑁$− 𝛿%*𝑁)( 174 8 175 where CR and NR are the percentages of soil OC or N coming from added plant residues, 176 AS is the amended soil and CS is the control soil. 177 178 Statistics 179 Linear regression analysis and mixed-model ANOVA were performed in order to test the 180 effect of aggregate size, time and treatment on aggregate composition (p<0.05). The 181 homogeneity of variance was tested using the Levene test, and the normality of residuals 182 was also checked applying the Shapiro-Wilk test to the residuals. All statistical tests were 183 conducted using the R statistical package for MacOSX (R Core Team, 2018). 184 185 Results 186 Table 1 shows the bulk composition of the 3.15-5 mm aggregates used in the aggregate 187 stability test. After 45 days, the OC content of the aggregates of soil amended with non-188 composted residues was slightly higher than that of the control soil. After 3 years, 189 aggregates of soils amended with plant residues, composted or not were richer in OC than 190 those of the control soil. OC concentrations decreased with time in all treatments, but 191 decline was less marked in the amended soils: between 45 days and 3 years, 19% of initial 192 OC was lost from the control soil, 16% from the soil amended with plant residues, and 193 7% from the compost-amended soil. Nitrogen contents were similar in the 3.15-5 mm 194 aggregates of all treatments at all analysed times. The C/N ratios thus decreased with time 195 due to differential consumption of C and N by microorganisms. Isotopic 13C and 15N 196 signatures became less negative with either residue or compost inputs, and in all cases, 197 values decreased with time. After 45 days, 4.5 and 6.7% of the aggregate bulk OC and 198 7.7 and 8.6% of the N originated from the non-composted and composted amendment, 199 9 respectively. After 3 years, these proportions were 2.7 and 5.2% (OC) and 8.2 and 10% 200 (N) for plant residue and compost addition, respectively. 201 Table 2 shows the size distribution of stable aggregates after the stability test. The 202 trends for the control and amended soils were very similar. Large aggregate fractions (> 203 0.5 mm) contributed more to the total aggregates at 45 days than after three years for all 204 treatments. Soils are unstable to very unstable according to the classification of Le 205 Bissonnais and Le Souder (1995), a typical feature of agricultural soils developed on loess 206 at the Paris Basin. Aggregate stability, indicated by the values of MWD, decreased with 207 the incubation time. 208 For the control soil, the contribution of 0.5-2 mm aggregates increased at the expense 209 of <0.5 mm fractions between 0 and 45 days of incubation, but this effect was no longer 210 observed in the long term. This short-term effect was likely due to an increase in the 211 microbial activity as a consequence of the better cultivation conditions at the start of the 212 laboratory incubations. After 45 days, the non-aggregated soil (<0.05 mm) decreased in 213 all treatments, from 21% in the control soil at t=0 to 12.1 % in the control and 9.8-14.8% 214 in the amended soils. This fraction increased again after three years. The addition of both 215 amendments increased soil aggregate stability (MWD increasing from 0.58 in the control 216 to 0.73-0.79 in the amended soils) after 45 days, without any significant difference 217 between soil amended with composted and non-composted residues. However, this 218 aggregating effect of amendments did not last, and at the end of the incubation (three 219 years), the aggregate stability of the amended soils was not significantly different from 220 that of the control soil. After three years, the contribution of the 0.5-1 mm size fraction 221 of the soil amended with non-composted plant residues was higher compared to control 222 and soil amended with compost, at the expense of 0.2-0.5 mm and 0.1-0.2 mm size 223 fractions. 224 16 addition of plant residues produced an enduring increase of SOM content (OC and N), as 374 well as a transient increase of aggregate stability. Exogenous OM was incorporated to 375 soil in a non-homogeneous manner that is related to the three-dimensional structure of 376 soil: this is, during amendment decomposition, C and N incorporate differently to 377 different size aggregates. In the short term, they were mainly stored in aggregates larger 378 than 0.5 mm (macroaggregates) but after three years of incubation a shift happened and 379 C and N were transferred to smaller aggregates. Therefore, on the long term, OM was 380 transferred from larger aggregates to smaller aggregates. Microaggregates were more 381 enriched with amendment C and N with the composted residues compared to the non-382 composted plant residues. This might improve soil functioning by limiting losses of C 383 and nutrients through leaching or erosion. Overall, the fact that more C and N from initial 384 plant residues are transferred to the soil when they are composted than when they were 385 not and that these C and N were preferentially transferred in the smaller fractions of the 386 soil, suggests that composting is a better option than using fresh materials from the point 387 of view of C sequestration and nutrient preservation in soils. 388 389 Acknowledgements 390 Dr. R. Paradelo thanks the Spanish Ministry of Economy and Competitiveness 391 (MINECO) for his Ramón y Cajal fellowship (RYC-2016-19286). 392 393 References 394 Abiven, S., Menasseri, S., Chenu, C., 2009. The effects of organic inputs over time on 395 soil aggregate stability – A literature analysis. Soil Biol. Biochem. 41, 1-12. 396 17 Angers, D.A., Recous, S., Aita, C., 1997. Fate of carbon and nitrogen in water-stable 397 aggregates during decomposition of 13C15N-labelled wheat straw in situ. Eur. J. Soil 398 Sci. 48, 295-300 399 Annabi, M., Houot, S., Francou, C., Poitrenaud, M., Bissonnais, Y.L., 2007. Soil 400 aggregate stability improvement with urban composts of different maturities. Soil Sci. 401 Soc. Am. J. 71, 413-423. 402 Baveye, P.C., Berthelin, J., Tessier, D., Lemaire, G., 2018. The “4 per 1000” initiative: 403 A credibility issue for the soil science community? Geoderma 309, 118-123. 404 Bernal, M.P., Navarro, A.F., Sanchez-Monedero, M.A., Roig, A., Cegarra, J., 1998a. 405 Influence of sewage sludge compost stability and maturity on carbon and nitrogen 406 mineralization in soil. Soil. Biol. Biochem. 30, 305–313. 407 Bernal, M.P., Sánchez-Monedero, M.A., Paredes, C., Roig, A. 1998b. Carbon 408 mineralization from organic wastes at different composting stages during their 409 incubation with soil. Agric. Ecosyst. Environ. 69, 175-189. 410 Le Bissonnais, Y., 1996. Aggregate stability and measurement of soil crustability and 411 erodibility. I. Theory and methodology. Eur. J. Soil Sci. 47, 425–437. 412 Le Bissonnais Y., Le Souder C., 1995. Measurement of the structural stability of soils for 413 the evaluation of their susceptibility to surface crusting and erosion. Etud. Gest. Sol. 414 2, 43-56. 415 Coplen, T.B., 1995. New IUPAC guidelines for the reporting of stable hydrogen, carbon, 416 and oxygen isotope-ratio data. J. Res. Natl. Inst. Stand. Technol. 100, 285. 417 Darboux, F., Algayer, B., 2014. Croûtes de battance, ruissellement, érosion hydrique. In: 418 Baize, D., Duval O., Richard, G. (Eds), Les sols et leurs structures. Observations à 419 différentes échelles. Éditions Quæ, Versailles, pp. 141-153. 420 18 Dignac, M.F., Bahri, H., Rumpel, C., Rasse, D.P., Bardoux, G., Balesdent, J., Girardin, 421 C., Chenu, C., Mariotti, A., 2005. Carbon-13 natural abundance as a tool to study the 422 dynamics of lignin monomers in soil: an appraisal at the Closeaux experimental field 423 (France). Geoderma 128, 3-17. 424 Dignac, M.-F., Derrien, D., Barre, P., Barot, S., Cécillon, L., Chenu, C., Chevallier, T., 425 Freschet, G., Garnier, P., Guenet, B., Hedde, M., Klumpp, K., Lashermes, G., Maron, 426 P.-A., Nunan, N., Roumet, C., Basile-Doelsch, I., 2017. Increasing soil carbon storage: 427 mechanisms, effects of agricultural practices and proxies. A review. Agron. Sustain. 428 Dev. 37, 14. 429 Féodoroff, A., 1960. Evaluation de la stabilité structurale d'un sol (indice S). Nouvelles 430 normes d'emploi pour l'appareil à tamiser. Ann. Agron. 11, 651-659. 431 Francou, C., Poitrenaud, M., Houot, S., 2005. Stabilization of organic matter during 432 composting: influence of process and feedstocks. Compost Sci. Util. 13, 72–83 433 Goebel, M.-O., Bachmann, J., Reichstein, M., Janssens, I.A., Guggenberger, G., 2011. 434 Soil water repellency and its implications for organic matter decomposition—is there 435 a link to extreme climatic events? Global Change Biol. 17, 2640–2656. 436 Guan, S., Dou, S., Chen, G., Wang, G., Zhuang, J., 2015. Isotopic characterization of 437 sequestration and transformation of plant residue carbon in relation to soil aggregation 438 dynamics. Appl. Soil Ecol. 96, 18-24. 439 Haynes, R.J., Beare, M.H., 1996. Aggregation and Organic Matter Storage in 440 Mesothermal, Humid Soils. In: Carter, M.R. and Stewart, B.A. (Eds.), Structure and 441 Soil Organic Matter Storage in Agricultural Soils, CRC Press, Boca Raton, pp. 213-442 262. 443 Hénin, S., Monnier, G., Combeau, A., 1958. Méthode pour l’etude de la stabilité 444 structurale des sols. Ann. Agron. 9, 73–92. 445 19 IUSS Working Group WRB, 2014. World Reference Base for Soil Resources 2014. 446 International soil classification system for naming soils and creating legends for soil 447 maps. World Soil Resources Reports No. 106. FAO, Rome. 448 Jamagne, M., 2011. Les paysages pédologiques de France. Editions Quae, Paris. 449 Janzen, H.H., 2006. The soil carbon dilemma: shall we hoard it or use it? Soil Biol. 450 Biochem. 38, 419–424. 451 Le Guillou, C., Angers, D.A., Leterme, P., Menasseri-Aubry, S., 2011. Differential and 452 successive effects of residue quality and soil mineral N on water-stable aggregation 453 during crop residue decomposition. Soil Biol. Biochem. 43, 1955-1960. 454 Lerch, T.Z., Nunan, N., Dignac, M.-F., Chenu, C., Mariotti., A., 2011. Variations in 455 microbial isotopic fractionation during soil organic matter decomposition. 456 Biogeochemistry 106, 5–21. 457 Lerch, T.Z., Dignac, M.-F., Thevenot, M., Houot, S., 2019. Composting plant residues 458 changes their mineralisation, priming effect and soil carbon budget. Soil Biol. 459 Biochem. 136, 107525. 460 Mariotti, A., 1983. Atmospheric nitrogen is a reliable standard for natural N-15 461 abundance measurements. Nature 303, 685–687. 462 Méndez-Millán, M., Nguyen Tu, T.T., Balesdent, J., Derenne, S., Derrien, D., Egasse, C., 463 Thongo M’Bou, A., Zeller, B., Hatté, C., 2014. Compound-specific 13C and 14C 464 measurements improve the understanding of soil organic matter dynamics. 465 Biogeochem. 118, 205–223. 466 Monnier, G., 1965. Action des matières organiques sur la stabilité structurale des sols. 467 Ann. Agron. 16, 327-400. 468 Oades, J.M., 1984. Soil organic matter and structural stability: mechanisms and 469 implications for management. Plant Soil 76, 319-337. 470 20 O'Brien, S.L., Jastrow, J.D., 2013. Physical and chemical protection in hierarchical soil 471 aggregates regulates soil carbon and nitrogen recovery in restored perennial grasslands. 472 Soil Biol. Biochem. 61, 1-13. 473 Paradelo, R., Van Oort, F., Barré, P., Billiou, D., Chenu, C., 2016. Soil organic matter 474 stabilization at the pluri-decadal scale: insight from bare fallow soils with contrasting 475 physicochemical properties and macrostructures. Geoderma 275, 48-54. 476 Pascual, J.A., García, C., Hernandez, T., 1999. Comparison of fresh and composted 477 organic waste in their efficacy for the improvement of arid soil quality. Bioresource 478 Technol. 68, 255-264. 479 Paustian, K., Lehmann, J., Ogle, S., Reay, D., Robertson, G.P., Smith, P., 2016. Climate-480 smart soils. Nature 532, 49–57. 481 Puget, P., Chenu, C., Balesdent, J., 1995. Total and young organic matter distributions in 482 aggregates of silty cultivated soils. Eur. J. Soil Sci. 46, 449-459. 483 Puget, P., Chenu, C., Balesdent, J., 2000. Dynamics of soil organic matter associated with 484 particle-size fractions of water-stable aggregates. Eur. J. Soil Sci. 51, 595-605. 485 R Core Team, 2018. R: A language and environment for statistical computing. R 486 Foundation for Statistical Computing, Vienna, Austria. http://www.R-project.org/. 487 Rumpel, C., Dignac, M.F., 2006. Gas chromatographic analysis of monosaccharides in a 488 forest soil profile: Analysis by gas chromatography after trifluoroacetic acid 489 hydrolysis and reduction–acetylation. Soil Biol. Biochem. 38, 1478-1481. 490 Six, J., Bossuyt, H., Degryze, S., Denef, K., 2004. A history of research on the link 491 between (micro)aggregates, soil biota, and soil organic matter dynamics. Soil Till. Res. 492 79, 7-31. 493 Smith P., 2016. Soil carbon sequestration and biochar as negative emission technologies. 494 Glob. Change Biol. 22, 1315-1324. 495 21 Totsche, K.U., Amelung, W., Gerzabek, M.H., Guggenberger, G., Klumpp, E., Knief, C., 496 Lehndorff, E., Mikutta, R., Peth, S., Prechtel, A., Ray, N., Kögel-Knabner, I., 2018. 497 Microaggregates in soils. J. Plant Nutr. Soil Sci. 181, 104-136. 498 van Groenigen, J.W., van Kessel, C., Hungate, B.A., Oenema, O., Powlson, D.S., van 499 Groenigen, K.J., 2017. Sequestering soil organic carbon: a nitrogen dilemma. Environ. 500 Sci. Technol. 51, 4738−4739. 501 Whalen J. K., Hu Q., Liu A., 2003. Compost applications increase water-stable aggregates 502 in conventional and no-tillage systems. Soil Sci. Soc. Am. J. 67, 1842–1847. 503 Zucconi F., de Bertoldi M.D., 1987. Organic waste stabilization throughout composting 504 and its compatibility with agricultural uses. In: Wise, D.L. (ed.), Global bioconversion. 505 CRC Press, Boca Raton, pp. 109–137. 506 507 22 FIGURE CAPTIONS 508 509 510 Figure 1. Organic C concentrations and isotopic signature in aggregate size classes. Bars 511 show confidence intervals at P<0.05, and different letters denote significant differences 512 between treatments at P<0.05 (when letters are not shown it means that no significant 513 differences were found). Black letters are for C contents and white letters for isotopic 514 signatures. 515 516 Figure 2. Nitrogen concentrations and isotopic signature in aggregate size classes. Bars 517 show confidence intervals at P<0.05, and different letters denote significant differences 518 between treatments at P<0.05 (when letters are not shown it means that no significant 519 differences were found). Black letters are for C contents and white letters for isotopic 520 signatures. 521 522 Figure 3. Contribution of each aggregate size fraction to soil OC stocks. The striped 523 section of the bars indicates the amount of soil OC that comes from the plant residues 524 added (please note that these values have been multiplied by a factor of ten, for clarity). 525 Bars show confidence intervals in the t-test at P<0.05. 526 527 Figure 4. Contribution of each aggregate size fraction to soil N stocks. The striped 528 section of the bars indicates the amount of soil N that comes from the plant residues 529 added (please note that these values have been multiplied by a factor of ten, for clarity). 530 Bars show confidence intervals in the t-test at P<0.05. 531 532 23 533 Figure 5. Fractions of OC and N remaining in the soil from plant residues (left, non-534 composted plant residues; right, composted plant residues) as a function of aggregate size. 535 536 24 Table 1. Composition of the soil used for the study. PR: plant residue, CPR: composted 537 plant residue. TOC: total organic carbon. 538 Time TOC (mg g-1) N (mg g-1) C/N δ13CVPDB (‰) δ15N (‰) Fraction of soil OC coming from plant residues (%) Fraction of soil N coming from plant residues (%) 0 Control 10.2 1.0 10.1 -25.0 4.8 - - 45 d Control 9.8 1.0 10.0 -25.5 6.3 - - PR 10.2 1.0 9.9 -12.5 119 4.5 7.7 CPR 9.6 1.1 9.0 -5.7 97 6.7 8.6 3 y Control 7.9 0.9 8.5 -25.0 5.1 - - PR 8.6 1.0 8.3 -17.2 125 2.7 8.2 CPR 8.9 1.1 8.2 -9.5 111 5.2 10.0 539 540 25 Table 2. Aggregate size distribution (% weight) and mean weight diameters (MWD). 541 Mean ± standard deviation (n=3). PR: plant residue; CPR: composted plant residue. 542 Different letters denote significant differences between treatments at p<0.05 (when letters 543 are not shown it means that no significant differences were found). 544 Aggregate size (mm) Time Treatment 2-3.15 1-2 0.5-1 0.2-0.5 0.1-0.2 0.05-0.1 < 0.05 MWD (mm) 0 Control 0.9 ± 0.4 9.5 ± 1.9ac 38.1 ± 3.4b 23.1 ± 3.0ab 4.2 ± 0.4a 3.2 ± 0.2ab 21.0 ± 7.5 0.55 ± 0.06ac 45 d Control 0.6 ± 1.0 16.3 ± 4.2ac 42.7 ± 5.2b 20.2 ± 3.2ab 3.9 ± 0.2a 3.1 ± 0.3ab 12.1 ± 8.0 0.58 ± 0.12ac PR 1.2 ± 0.4 21.1 ± 2.6bc 41.2 ± 2.4b 15.2 ± 2.6a 3.3 ± 0.4a 3.1 ± 0.4ab 14.8 ± 2.7 0.73 ± 0.17bc CPR 1.8 ± 0.9 22.6 ± 6.7c 43.1 ± 2.8b 16.6 ± 3.9a 3.6 ± 1.7a 2.6 ± 0.3a 9.8 ± 4.8 0.79 ± 0.10c 3 y Control 0.4 ± 0.2 3.8 ± 2.4a 22.7 ± 5.0a 32.5 ± 2.1c 7.7 ± 0.6bc 5.1 ± 0.3bc 27.8 ± 8.2 0.38 ± 0.07a PR 0.2 ± 0.2 8.8 ± 0.4ab 32.9 ± 3.2ab 26.6 ± 3.8bc 5.7 ± 1.1ab 5.0 ± 1.0bc 20.9 ± 2.2 0.49 ± 0.02ab CPR 0.2 ± 0.2 4.1 ± 3.6a 25.7 ± 4.7a 33.9 ± 3.0c 8.9 ± 1.3c 6.8 ± 1.7c 20.4 ± 3.2 0.40 ± 0.08a 545