Jukuri, open repository of the Natural Resources Institute Finland (Luke) All material supplied via Jukuri is protected by copyright and other intellectual property rights. Duplication or sale, in electronic or print form, of any part of the repository collections is prohibited. Making electronic or print copies of the material is permitted only for your own personal use or for educational purposes. For other purposes, this article may be used in accordance with the publisher’s terms. There may be differences between this version and the publisher’s version. You are advised to cite the publisher’s version. This is an electronic reprint of the original article. This reprint may differ from the original in pagination and typographic detail. Author(s): Kimmo Rasa, Taina Pennanen, Krista Peltoniemi, Sannakajsa Velmala, Hannu Fritze, Janne Kaseva, Juuso Joona & Risto Uusitalo Title: Pulp and paper mill sludges decrease soil erodibility Year: 2021 Version: Published version Copyright: Rights: The Author(s) 2020 CC BY 4.0 Rights url: http://creativecommons.org/licenses/by/4.0/ Please cite the original version: Rasa, K, Pennanen, T, Peltoniemi, K, et al. Pulp and paper mill sludges decrease soil erodibility. J. Environ. Qual. 2021; 50: 172– 184. https://doi.org/10.1002/jeq2.20170 Received: 24 April 2020 Accepted: 7 October 2020 Published online: 8 December 2020 DOI: 10.1002/jeq2.20170 T E C H N I C A L R E P O R T S S u r f a c e Wa t e r Q u a l i t y Pulp and paper mill sludges decrease soil erodibility Kimmo Rasa1 Taina Pennanen2 Krista Peltoniemi2 Sannakajsa Velmala2 Hannu Fritze2 Janne Kaseva1 Juuso Joona3 Risto Uusitalo1 1 Natural Resources Institute Finland, Tietotie 4, Jokioinen FI-31600, Finland 2 Natural Resources Institute Finland, Latokartanonkaari 9, Helsinki FI-00790, Finland 3 Soilfood Oy, Viikinkaari 6, Helsinki FI-00790, Finland Correspondence Kimmo Rasa, Natural Resources Institute Finland, Tietotie 4, Jokioinen FI-31600, Finland. Email: kimmo.rasa@luke.fi Assigned to Associate Editor Laura Christianson. Funding information H2020 European Research Council, Grant/Award Number: 818290; Finnish Nutrient Recycling Pilot Programme, Ravinnekuitu-project; Tekes; NSSPPulp- project, companies involved Abstract Declining carbon (C) content in agricultural soils threatens soil fertility and makes soil prone to erosion, which could be rectified with organic soil amendments. In a 4-yr field trial, we made a single application of three different organic sludges from the pulp and paper industry and studied their effects on cereal yield, soil C content, and fungal and bacterial composition. In laboratory rainfall simulations, we also stud- ied the effects of the soil amendments on susceptibility to erosion and nutrient mobi- lization of a clay-textured soil by measuring the quality of percolation water passing through 40-cm intact soil monoliths during 2-d rainfall simulations over four con- secutive years after application. A nutrient-poor fiber sludge reduced wheat yield in the first growing season, but there were no other significant effects on cereal yield or grain quality. An input of ∼8 Mg ha−1 C with the soil amendments had only minor effects on soil C content after 4 yr, likely because of fast microbe-mediated turnover. The amendments clearly changed the fungal and bacterial community composition. All amendments significantly reduced suspended solids (SS) and total phosphorus (TP) concentrations in percolation water. The effect declined with time, but the reduc- tion in SS and TP was still >25% 4 yr after application. We attributed the lower ten- dency for particle detachment in rain simulations to direct interactions of soil min- erals with the added particulate organic matter and microbe-derived compounds that stabilize soil aggregates. In soils with low organic matter content, pulp and paper industry by-products can be a viable measure for erosion mitigation. 1 INTRODUCTION The pulp and paper industry produces large quantities of organic sludges as side streams (in Finland, 420,000 Mg dry matter per year) that are incinerated with low net energy recovery (Joona, Kuisma, Alakukku, & Kahiluoto, 2012). Abbreviations: CPMS, composted pulp mill sludge; DOC, dissolved organic carbon; DRP, dissolved reactive phosphorus; EC, electrical conductivity; FS, fiber sludge; LPMS, lime-stabilized pulp mill sludge; PP, particulate phosphorus; SS, suspended solids; TP, total phosphorus. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2020 The Authors. Journal of Environmental Quality © 2020 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America In a circular economy, these valuable organic side streams could be used elsewhere in new value chains, creating jobs, business opportunities, and environmental benefits. Agricultural use of organic sludges derived from the pulp and paper industry has been studied for different soil types and climates, with the results generally indicating positive impacts on crop production and soil physical properties (e.g., Chow, Rees, Fahmy, & Monteith, 2003; Gallardo, Cea, Tortella, & Diez, 2012; Sippola, Mäkelä-Kurtto, & Rantala, 2003). For example, Chow, Rees, Fahmy, and Monteith (2003) reported that pulp mill fibers improved soil 172 wileyonlinelibrary.com/journal/jeq2 J. Environ. Qual. 2021;50:172–184. https://orcid.org/0000-0003-2350-1839 https://orcid.org/0000-0002-4427-9053 mailto:kimmo.rasa@luke.fi http://creativecommons.org/licenses/by/4.0/ https://wileyonlinelibrary.com/journal/jeq2 RASA ET AL. 173 aggregation on a gravelly loam soil at application rates of 20–160 Mg ha−1. They also found that it took longer before surface runoff started and that soil loss was reduced up to 90% from the level of control when pulp fiber was used as soil amendment. However, less is known about whether these organic side streams can contribute to water protection (i.e., reduced nutrient transport) in clayey agricultural soils for a longer period by improving soil structure and protecting the soil surface against water erosion over several years. In general, organic matter in soils enhances aggregate sta- bility (Greenland, Lindstrom, & Quirk, 1962; Soinne, Hyvälu- oma, Ketoja, & Turtola, 2016; Tisdall & Oades, 1982). Also, organic soil amendments can directly form organo-mineral compounds that bind soil particles together. At the same time, mineralization of introduced organic matter is likely to cause changes in microbial activity and community composition, which affects soil structural properties (Abdi et al., 2017; Gallardo et al., 2012; Zibilske, Clapham, & Rourke, 2000). Microbiological decomposition changes the chemical com- position of the added organic matter and provides new types of compounds (e.g., extracellular polymeric substances, fun- gal hyphae, and microbial necromass), contributing to soil aggregation (Costa, Raaijmakers, & Kuramae, 2018; Peele & Beale, 1941). Decomposition of relatively slowly degradable organic compounds, such as cellulose and lignin present in pulp and paper industry sludges, provides a platform for a suc- cession of microbial substances potentially forming organo- mineral associations with clay-sized soil particles. If intro- duced organic matter increases the degree of soil aggregation and aggregate strength, the greater aggregate stability makes the soil less prone to erosion (Diacono & Montemurro, 2010; Le Bissonnais, 1996). The carbon (C) content of agricultural soils is currently declining in Finland (Heikkinen, Ketoja, Nuutinen, & Regina, 2013) and on a global scale (Lal, 2004). Organic soil amend- ments are an attractive option to increase and stabilize soil C storage in the long term (Poulton, Johnston, Macdonald, White, & Powlson, 2018; Six, Conant, Paul, & Paustian, 2002). Organic side streams from the pulp and paper indus- try have been shown to decompose rather quickly (Foley & Cooperband, 2002; Zibilske et al., 2000). Thus, the effect of a single addition of organic matter on soil structure and C con- tent is likely transient. However, recent literature highlights the importance of microbially derived organic material in sta- bilizing C via organo-mineral interactions and protecting it against decomposition (Kopittke et al., 2018, 2020). Carbon protected by organo-mineral associations leads to the forma- tion of small, water-stable micro-aggregates, contributing to soil structural stability over several years (Tisdall & Oades, 1982). In theory, enhanced microbial activity and potential changes in microbial communities due to added organic mat- ter can contribute simultaneously to long-term C storage and protection of soil structure against water-induced stresses. Core Ideas ∙ Erosion mitigation using pulp and paper mill sludges was tested in a 4-yr study. ∙ Intact soil monoliths were taken from field for lab- oratory rainfall simulations. ∙ Sludge addition reduced particle and P losses from soil to percolation water. ∙ Sludges decomposed quickly and had minor effects on soil C content after 4 yr. ∙ Sludge addition clearly altered soil bacterial and fungal community composition. The effect of pulp mill sludge on soil structure and the role of soil microbiology have been reported by Bipfubusa, Angers, N’Dayegamiye, and Antoun (2008). They measured increased aggregate stability after 2 yr of pulp mill sludge application (fresh and composted) on loam-textured soil con- taining 20% clay. They suggested that both stimulation of soil microflora and binding of humic substances with soil min- eral particles contributed to aggregate stabilization. Abdi et al. (2017) observed changes in microbial community structure for 3 yr after 9 yr of continuous application of pulp mill biosolids, whereas no changes were observed when soil was amended with liming materials. In a 4-yr field experiment, we examined the potential of three different pulp and paper industry organic sludges to reduce the susceptibility of a clay-textured soil to erosion and nutrient mobilization (rainfall simulation test), the poten- tial to increase soil organic matter content (soil sampling), and the potential to bring about changes in soil microbiol- ogy (DNA sequencing). Our starting hypotheses were that (a) organic matter input derived from pulp and paper indus- try side streams can improve soil structural stability, which reduces soil dispersion during rain events and the risk of asso- ciated off-site nutrient transfer through structured clay soils, and (b) the effect will vary depending on soil amendment properties and over time. We also tested the hypotheses that (c) a single, large input of organic soil amendments will pre- serve soil organic C over several years and (d) application of wood-derived amendments to an agricultural soil is reflected in soil microbial activity and community structure. Although research on the functionality of the soil microbial community and its responses in terms of soil functions is limited (Nan- nipieri et al., 2003; Yang et al., 2018), these interactions are increasingly highlighted when promoting sustainable man- agement of soils and attempts to maintain soil C stocks (Liang, Amelung, Lehmann, & Kästner, 2019). In the present study, we attempted to comprehend changes in erosion vulnerabil- ity due to pulp mill sludge amendments using a combination 174 RASA ET AL. T A B L E 1 Dry matter content (DM) of the three pulp and paper mill side streams used as soil amendments and amount of organic matter (OM), C, total N (TN), soluble N (Sol-N), and total P, K, S, Ca, and Mg supplied to soil with the amendments Sludgea DM OM C TN Sol-N P K S Ca Cd Cr % Mg ha−1 kg ha−1 g ha−1 CPMS 43.4 14.3 7.8 211 34 45 39 109 949 21 727 LPMS 49.7 17.3 9.0 253 32 53 30 131 2,181 16 528 FS 33.5 15.7 8.4 13 1 2 1 7 2,269 0.2 100 aCPMS, composted sludge; FS, fiber sludge; LPMS, lime-stabilized sludge. of research methods dealing with soil physical, chemical, and microbiological properties. 2 MATERIALS AND METHODS 2.1 Origin and quality of soil amendments The three soil amendments used in the experiment were pro- duced by Soilfood Oy (formerly Tyynelän maanparannus Oy) from organic side streams provided by Stora Enso’s Imatra Mill located in southeastern Finland. These were (a) com- posted pulp mill sludge (CPMS) and (b) lime-stabilized pulp mill sludge (LPMS), both derived from the process water treatment plant at the factory, and (c) fiber sludge (FS), which consists of cellulose fibers that are too short for the final prod- uct of the mill. The main difference between the materials is that those recovered from the mill’s wastewater treatment pro- cess contain phosphorus (P), nitrogen (N), and other nutrients added to the biological purification step to cut down oxygen demand of effluent waters, whereas FS is a nutrient-poor cel- lulose material from the pre-clarifier of cardboard machine process water, removed in a wire sieve as semi-dry mass. The origin, processing, analytical methods, and detailed proper- ties of the soil amendments derived from these side streams are given in the Supplemental Material (p. S2, Supplemental Table S1). 2.2 Field experiment The field experiment was established at Jokioinen in south- western Finland on a clay-textured soil classified as a Luvic Stagnosol (Eutric, Clayic, Protovertic) (IUSS Working Group, 2015). The experiment layout was a randomized complete block design with five replicates with a total of 20 plots, each mea- suring 6 by 15 m. Unamended plots served as the control treatment. The amendment rates used for the sludges at their original moisture content were 52, 51, and 72 Mg ha−1 for LPMS, CPMS, and FS, respectively (Table 1). These rates were based on an attempt to increase soil C content as much as possible without exceeding the soluble N (Sol-N) limit of 30 kg ha−1 allowed as autumn application. The sludges were spread on the soil in September 2015 as a single applica- tion in the beginning of the experiment, and the field was tilled immediately to a depth of approximately 10 cm. Details of establishment and maintenance of the experiment are given in the Supplemental Material (pp. S2–S4, Supplemental Table S2). All field plots were sampled at the 0-to-20-cm depth in autumn 2015, just before the amendments were applied, and again in autumn 2019, by taking from each plot three repli- cate samples that were combined for analyses. At the initial sampling in 2015, soil clay content, determined by a pipette method (Elonen, 1971), was 47%. Total C and total N (LECO CN-2000) content was 2.3 and 0.19%, respectively. Soil test P (1-h extraction at 1:10 vol/vol ammonium acetate at pH 4.65; Vuorinen & Mäkitie, 1955) was 10 mg L−l, which indi- cates that annual P fertilization would be unlikely to give yield responses (Valkama, Uusitalo, Ylivainio, Virkajärvi, & Tur- tola, 2009). Soil pH (H2O) was 6.4, and electrical conductiv- ity (EC) was 0.051 μS cm−1. Total C, total N, pH, and EC were determined for sam- ples taken from each field plot at the end of the experiment (autumn 2019). Additionally, soil total cadmium (Cd) content in samples from each plot was analyzed by using a graphite furnace atomic absorption spectrofotometer (AA280Z, Var- ian) in aqua regia digestate (SFS-ISO 11466:2007). In 2016 and 2019, the crop grown was wheat (Triticum aes- tivum L.); in 2017 and 2018, the crop was oats (Avena sativa L.). At the end of each growing season, the plots were har- vested with an experimental harvester, fresh weight of crop biomass was recorded, and grain samples were dried at 105 ˚C overnight to calculate moisture content. For analyses of nutri- ent and heavy metal content, grain samples were milled with a hammer mill and digested with 7 M HNO3, and the extracts were analyzed using inductively coupled plasma-optical emis- sion spectrometry. 2.3 Rainfall simulation test To investigate the ability of the soil amendments to stabilize soil aggregates, and thus reduce erosion and nutri- ent leaching through soil profile, large undisturbed soil RASA ET AL. 175 monoliths (30 cm in diameter, ∼40 cm deep) were extracted from all field plots in sections of polyvinyl chloride sewage pipe using a tractor-driven soil auger. Sampling was repeated in four consecutive springs (May 2016–May 2019). After coring and transport to the laboratory, the bottom of the monoliths was prepared to expose intact natural ped surfaces. Void spaces created during preparation were filled with washed 3-to-5-mm quartz sand. On the top side of each polyvinyl chloride cylinder, a 32-mm-diameter hole was drilled (center of the drill hole at the level of soil sur- face) to lead water out of the monolith surface if pond- ing occurs. The monoliths were then saturated from below for 1 d, maintained at saturation for an additional 2 d, and allowed to drain overnight. The monolith sampling and rain- fall simulation procedures are described in detail in Uusitalo et al. (2012). Simulated rain was applied for 5 h d−1 on two consec- utive days at an intensity of 5 mm h−1 (25 mm in both days). Four individual percolation water samples were collected from each monolith during the rainfall simula- tion. Samples were analyzed separately to detect possible within-simulation trends. The first water sample was taken from water draining after the saturation period, the second and fourth samples were taken during the consecutive simulated rainfall events, and the third sample was taken from water draining during the night between the events. The total volume of percolation water was close to equal in all treatments, and typically all rain applied percolated through the soil (Supplemental Figure S1). All samples were immediately analyzed for turbidity (2100 AN IS Turbidime- ter, Hach), and subsamples were passed through a 0.2-μm filter (Nuclepore, Whatman). All unfiltered samples and filtered subsamples were frozen and stored at −18 ˚C for later analysis. Dissolved reactive P (DRP) was analyzed in the filtered (0.2 μm) subsamples (Lachat QuikChem Method 10-115-01- 1-Q), and total P (TP) was analyzed after acid peroxodisulfate digestion of unfiltered samples in an autoclave (120 ˚C, 100 kPa, 30 min; Turtola, 1996) based on molybdate col- orimetry (Murphy & Riley, 1962). Particulate P (PP) was taken as the difference between TP and DRP. Water samples were further analyzed for suspended solids (SS) (material retained on a Whatman GF/A 1.6-μm filter), dissolved organic C (DOC) and total organic C (Shimadzu TOC-V CSH Total organic C analyzer), pH, EC (electrodes Mettler Toledo InLab Expert Pro-ISM and InLab 731-ISM, respectively), and total N (unfiltered), nitrate-N (NO3–N), ammonium-N (NH4–N; Lachat QuikChem Methods 10-107-04-2-C, 10-107-04-2-C and 10-107-06-2-B, respectively), calcium (Ca2+), potas- sium (K+), magnesium (Mg2+), sodium (Na+), sulfur (S), and Cd using inductively coupled plasma-optical emission spectrometry. 2.4 Microbiological analyses Samples for microbial analyses were taken from all 20 field plots (0-to-10-cm depth, composite samples comprising 10 subsamples) in spring and autumn 2018, 3 yr after the sin- gle application of the soil amendments. The methods used to analyze basal respiration, microbial biomass (amount of C and N in microbial biomass), phospholipid fatty acids, and soil-extractable C and N are presented in the Supple- mental Material (pp. S5–S6). The methods used for DNA extraction, amplicon sequencing, and bioinformatics are pre- sented briefly below. Glomalin-related soil proteins were ana- lyzed in triplicate 0.25-g samples of air-dry soil (autumn sam- ples only) as described in Moragues-Saitua, Merino-Martín, Stokes, and Staunton (2019). The DNA in both spring and autumn samples from all 20 plots was extracted using the NucleoSpin soil kit (Macherey) and sequenced at the Institute of Genomics, Tartu University, as paired-end 2 × 300 bp with the MiSeq platform (Illumina) using the MiSeq v3 kit, producing about 20–25 mil- lion reads per flow cell. For bacteria, the 16S SSU rRNA gene V4 region was amplified using primers 515F and 806R (Capo- raso et al., 2011, 2012). For fungi, the ribosomal internal tran- scribed spacer 2 region was amplified using primers gITS7 (Ihrmark et al., 2012) and ITS4 (White, Bruns, Lee, & Taylor, 1990) with 8 bp dual index for 24 cycles. Raw sequences of 40 samples have been stored in the NCBI genebank BioProject PRJNA607883 under accession numbers SAMN14150014- 53. Quality filtering, removal of artefacts, chimeric sequences, primer-dimers, and primers from raw 16S rRNA and inter- nal transcribed spacer 2 sequence reads and clustering and taxonomy annotation were conducted using the PipeCraft 1.0 pipeline (Anslan, Bahram, Hiiesalu, & Tedersoo, 2017) as described in Soinne et al. (2020). Details are provided in the Supplemental Material (pp. S5 and S6). 2.5 Statistical analyses Preliminary inspection showed that there were no clear within-simulation trends in SS or nutrient concentrations, so the results for the four individual water samples taken from each monolith were pooled for statistical analysis. The statis- tical analyses were performed using generalized linear mixed models. Treatment (control, FS, LPMS, CPMS) and year (2016–2019) and their interaction were used as fixed effects, and block and the interaction of block × year were used as random effects. Correlated observations between years were taken into account using the most suitable covariance struc- ture (homogeneous or heterogeneous compound symmetry or first-order autoregressive, or unstructured). The unstructured 176 RASA ET AL. covariance matrix is the most flexible because it imposes no pattern on the covariances, whereas CS assumes a constant covariance between all years. The lowest Akaike information criterion value was used as the most important criterion for selection of covariance structure, together with normality of the residuals (Gbur et al., 2012). Microbial variables (basal respiration, microbial C and N, extractable C and N, glomalin-related soil protein, and phos- pholipid fatty acids) were analyzed with the same model using two correlated seasons (spring and autumn) instead of years. Glomalin-related soil proteins were determined only for autumn samples. Unequal variances of treatments were allowed for total phospholipid fatty acids based on a lower Akaike information criterion value and a likelihood ratio test. Parameters measured in soil samples (C, N, Cd, EC, and pH) were analyzed from one depth (0–20 cm), using only treatment as a fixed effect and block as a random effect. Assumptions of gamma (with log link for PP, TP, DRP, and NH4–N) and Gaussian (with identity link for the other param- eters analyzed) distributions were used for all dependent vari- ables. All models were fitted by using the residual pseudo likelihood (for gamma) or the restricted maximum likelihood (for Gaussian) estimation method. The method of Westfall (1997) was used for pairwise comparisons of treatments with the control within the same year, depth, or season. Pairwise comparison of treatment effect over years was conducted with the Tukey–Kramer method. A significance level of α = .05 was used in all analyses. Degrees of freedom were calculated using the Kenward–Roger method (Kenward & Roger, 2009). The analyses were performed using the GLIMMIX procedure in the SAS Enterprise Guide 7.15 (SAS Institute). Operational taxonomic unit data from amplicon sequenc- ing were normalized using the GMPR method (Chen, Reeve, Zhang, Huang Wang, & Chen, 2018) in R 3.5.2. One out- lier sample, LPMS autumn, was removed because of its small library size. Permutational multivariate ANOVA was per- formed using distance matrices with function adonis from vegan 2.5–5 (Oksanen et al., 2019), with block as the stra- tum. Nonmetric multidimensional scaling was conducted with stable solution from random starts, axis scaling, and species scores with function metaMDS from vegan using the Bray– Curtis dissimilarity index and plotted with fitted environmen- tal variables envfit from vegan. 3 RESULTS 3.1 Rainfall simulation study Rainfall simulations carried out over four consecutive years after application of the different pulp and paper industry side streams suggested that all amendments tested significantly decreased soil susceptibility to particle mobilization and asso- ciated PP losses (Figure 1; Supplemental Table S3). In all individual years, the unamended control produced the highest SS, PP, and TP concentrations in percolation water (Figure 1), although differences between the control and treatments var- ied between years. In contrast, mobilization of DRP, which made up 11–33% of TP in percolation water, was not affected by application of the different organic amendments (Figure 1) despite the fact that they all slightly increased soil pH (Sup- plementary Material Table S4), which could have increased P solubility. Particle mobilization in amended soil was suppressed more strongly during the first 2 yr of the study, when particle con- centration in percolation water of the control treatment was twice as high (∼500 mg L−1), than in the latter 2 yr (mean SS, ∼250 mg L−1). Whether the effect of the amendments was gradually subsiding with time or if the less pronounced effect of the amendments was due to more stable soil structure in the drier conditions in the latter half of the study period remains unresolved. Of the three organic soil amendments studied, FS appeared to act most consistently throughout the years, with a mean 47–76% reduction in SS compared with the control. The low- est mean reduction in SS mobilization was associated with CPMS, but it still resulted in a 62% reduction in SS concen- tration compared with the control in the first year and a 30– 34% reduction during the last 3 yr. The highest single-year SS reduction (80–82%) was associated with LPMS in the first 2 yr; thereafter, the effect was 30–51% compared with the con- trol. Mobilization and transport of PP with percolation water closely followed the same trends as SS because of the natural close correlation between these parameters. Overall, the FS and LPMS treatments resulted in signif- icant increases in DOC mobilization (Figure 2). A flush of DOC was recorded in percolation water collected during the first year’s rainfall simulation (7 mo after amendment appli- cation) in all treatments, with 84–160% higher concentrations for the amended soils compared with the control. However, the water samples from the control treatment had highly vari- able DOC in the first year, and no statistically significant dif- ferences were established. In the second study year, DOC con- centrations in percolation water were significantly higher in the LPMS treatment (by 54%), whereas DOC concentrations in the other treatments were similar to that in the control. No differences between treatments and control were observed during the last 2 yr of the study (for inorganic and total C, see Supplemental Table S5). Total N and NO3–N concentrations tended to be lower in amended soils than in the control in the first study year but were typically similar or higher in the later years (Figure 2). The overall treatment effects regarding N species in perco- lation water were not significantly different from the control (Figure 2). The pH and EC of percolation water increased in FS and LPMS treatments, whereas no effect was observed RASA ET AL. 177 F I G U R E 1 Concentrations of suspended solids (SS), total phosphorus (TP), particulate P (PP), and dissolved reactive P (DRP) in percolation water from 40-cm deep monoliths of control and treated soil. Treatment effects over 4 yr are summarized in the right-hand panels, where the differences of means are represented in a link scale when the dependent variable was not normally distributed. Error bars denote 95% confidence intervals. Asterisks indicate significant difference between treatment and control (*p < .05, **p < .01, ***p < .001, Op < .1; ns [p ≥ .1]). Ctrl, control; CPMS, composted sludge; FS, fiber sludge; LPMS, lime-stabilized sludge in CPMS treatment (Supplemental Figure S2). For other ele- ments in percolation water, see Supplemental Table S6. 3.2 Effect on cereal yield No significant treatment effects on yield or yield quality were observed during the study (Supplemental Figure S3; Supple- mental Tables S7 and S8). However, in the first study year, the grain yield was 14% lower in the nutrient-poor FS treatment compared with the control. This was likely due to immobiliza- tion of N. In the second year, yield in the LPMS treatment was over 500 kg ha−1 higher than in the control treatment (p = .2). Due to very dry spring in the third year, the mean yields were exceptionally low (<1,500 kg ha−1). 3.3 Heavy metals One of the main concerns related to agricultural use of indus- trial side streams is whether they contain heavy metals or other harmful elements (Cd in particular) that may accumu- late in the soil and end up in plants. Of the soil amendments studied here, FS did not contain quantifiable amounts of 178 RASA ET AL. F I G U R E 2 Concentrations of dissolved organic C (DOC), total N, and NO3–N in percolation water from 40-cm-deep monoliths of control and treated soil monoliths. Treatment effects over 4 yr are summarized in the right-hand panels. Error bars denote 95% confidence intervals. Asterisks indicate significance difference between treatment and control (*p < .05, **p < .01, ***p < .005, Op < .1; ns [p ≥ .1]). For fiber sludge (FS) treatment year 2017, data on total N and NO3–N of one of the five replicate samples were removed as outliers. Ctrl, control; CPMS, composted sludge; LPMS, lime-stabilized sludge Cd, whereas the Cd content in CPMS and LPMS was 0.96 and 0.60 mg kg−1 DM, respectively (Supplemental Table S1). These concentrations where unexpectedly high, but they did not exceed the maximum permissible Cd concentration for soil amendments under Finnish legislation, which is 1.5 mg kg−1 DM. Due to high application rates (>20 Mg ha−1 DM), the total Cd loads were 20 and 14 g ha−1 for CPMS and LPMS, respectively. For these soil amendments, application exceeded the maximum permissible Cd load of 7.5 g ha−1 summed for a 5-yr period. However, soil Cd concentration did not change due to Cd application with the amendments (Supplemental Table S4). Moreover, Cd and lead (Pb) concentrations in grain did not exceed the maxi- mum permissible levels (0.1 and 0.2 mg kg−1 wet weight, respectively) set by European legislation (European Union, 2006; Supplemental Tables S7 and S8). Water samples from the rainfall experiment were also analyzed for Cd in the third and fourth years of the study, but all values were below the detection limit of 0.7 μg L−1 (data not shown). Concentrations of other harmful elements were low enough to allow even higher amendment rates than used in this study (Supplemental Table S1). 3.4 Effects of amendments on soil and microbes Four years after application of a single large dose of the organic soil amendments, only minor effects on soil C and N content were detectable (Supplemental Table S4), apart from the CPMS treatment having slightly higher (p = .053) soil C content than the control (by 0.18 percentage points). The soil amendments increased basal respiration in spring and microbial biomass (estimated by the fumigation- extraction technique) in autumn (Table 2). Clear changes in bacterial and fungal communities (Figure 3; Supplemen- tal Tables S9 and S10), revealed by DNA-based amplicon sequencing were observed in soils that received the amend- ments, although the coarser phospholipid fatty acid method indicated that microbial biomass remained rather unchanged RASA ET AL. 179 T A B L E 2 Basal soil respiration (BR), microbial biomass C (CMB) and N (NMB), and glomalin-related soil proteins (GRSP) in control and treatment plots (n = 5) Treatmenta BR CMB NMB GRSPb μg CO2 kg−1 h−1 mg kg−1 g kg−1 Spring samples FS 95.8 (82.7–111)*** 0.18 (0.16–0.21) 0.027 (0.022–0.032) LPMS 61.9 (53.4–71.7) 0.18 (0.16–0.20) 0.028 (0.024–0.033) CPMS 91.0 (78.6–106)** 0.17 (0.15–0.19) 0.027 (0.022–0.031) Control 56.6 (48.9–65.6) 0.16 (0.15–0.18) 0.028 (0.023–0.032) Autumn samples FS 42.0 (36.3–48.7) 0.25 (0.22–0.28)** 0.046 (0.041–0.051)** 1.71 (1.43–1.99)† LPMS 42.6 (36.8–49.4) 0.25 (0.23–0.28)** 0.046 (0.042–0.051)** 1.55 (1.27–1.83) CPMS 45.6 (39.3–52.8) 0.25 (0.22–0.28)** 0.042 (0.037–0.047)** 1.63 (1.35–1.91) Control 45.4 (39.2–52.7) 0.20 (0.18–0.22) 0.032 (0.027–0.037) 1.42 (1.14–1.70) Note. Values in parentheses are 95% confidence intervals. Asterisks indicate significant difference between treatment and control. aCPMS, composted sludge; FS, fiber sludge; LPMS, lime-stabilized sludge. bDetermined for autumn samples but not for spring samples. *Significant at the .05 probability level. **Significant at the .01 probability level. ***Significant at the .001 probability level. †Significant at the .1 pobability level. -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -1 .0 -0 .5 0. 0 0. 5 1. 0 NMDS1 N M D S 3 a) spring autumn CTRL FS LPMS CPMS -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -1 .0 -0 .5 0. 0 0. 5 1. 0 NMDS1 N M D S 3 b) pH Cmb Nmb pH Nmb F I G U R E 3 Plot of (a) bacterial and (b) fungal communities in control and treated soil, shown as 95% confidence interval ellipses of SD of the mean of five replicate samples (three-dimensional nonmetric multidimensional scaling [NMDS] with Bray–Curtis dissimilarity index, showing only axis 1 and axis 3). Grey arrows show increasing direction of significantly correlated (p < .01) environmental factors pH, microbial C (Cmb), and microbial N (Nmb). Axis 2 is not shown in the figure; it separated samples by sampling time (spring vs. autumn). Ctrl, control; CPMS, composted sludge; FS, fiber sludge; LPMS, lime-stabilized sludge (Supplemental Table S11). The treatments also affected the amount of K2SO4–extractable C in soil, but the changes inter- acted with the season (Supplemental Table S11). The levels of K2SO4–extractable N and glomalin-related soil proteins were only marginally affected by the treatments (Table 2). Sampling time and amendment type explained roughly 8 and 19% of the variation in microbial community compo- sition (p < .001) of both bacterial and fungal operational taxonomic units, respectively (Figure 3; Supplemental Table S12). All amendments increased soil pH (Supplemental 180 RASA ET AL. Table S4), which seemed to be an important environmental factor affecting the bacterial and fungal communities in the soil (Figure 3). The bacterial and fungal communities in the LPMS treatment, with the highest soil pH, differed most from the control and had the highest microbial N and microbial C (Figure 3; Table 2; Supplemental Table S4). Closer analysis of the microbial communities (Supplemental Table S9) revealed a distinct increase in plant root-associated fungal species, such as Sebacinales (around 300–700% increase in log-scale) and the arbuscular mycorrhizal fungus Funneliformis mossae, in plots receiving soil amendments. Operational taxonomic units representing various bacterial groups showed significant differences between the treatments and control plots (Supplemental Table S10). Bacterial representatives clustering into eight taxa (Anaerolineae, Chitinophagaceae, Demequinaceae, Microscillaceae, Myx- ococcales, Pedosphaeraceae, Rhodanobacteraceae, and Xanthomonadaceae) were more common in all amended plots, showing 200–700% increases (in log-scale, p Adj. ≤ .0001) compared with the unamended control. 4 DISCUSSION Rainfall simulation was conducted with an intensity of 5 mm h−1, which represents typical rains in Finland (Kuusisto, 1980). Although heavier storms naturally cause severe ero- sion events if occurring when soils are bare, the persistent but less spectacular losses of clay particles mobilized by lower intensity rains maintain turbidity of rivers flowing through the landscapes of southwestern Finland. This relatively flat area has predominantly clay-textured, structured soils and requires pipe drains to carry the weight of machines in spring. Thus, after frost disappears, most of the excess water discharges as subsurface drainage flow (Koskiaho et al., 2002; Turtola & Paajanen, 1995). Our rainfall simulation setup aimed to mimic the generation of subsurface drainage discharge during typical rainy periods. Amending soil with pulp mill side streams decreased parti- cle mobilization to percolation water and associated P losses over multiple years. As compared to other soil amendments used as erosion or PP control agents, our results suggest that the pulp mill wastes bring about as large reductions in SS and PP as gypsum or highly reactive lime [CaO/Ca(OH)2, so called “structural lime”] applications have done in earlier Finnish and Swedish studies (Ekholm et al., 2012; Svanbäck, Ulén, & Etana, 2014; Ulén & Etana, 2014). Ekholm et al. (2012) reported 64% reduction in PP exports from a 245-ha catchment having 100 ha of agricultural land that was almost all amended with gypsum at 4 Mg ha−1 and monitored after that for 3 yr. Ulén and Etana (2014) reported a 40–57% reduc- tion in TP losses from two field sites after application of reac- tive lime at a rate of 5 Mg ha−1 CaO equivalents. In the 6-yr study of Svanbäck et al. (2014), reactive lime applied to annu- ally plowed soil decreased PP losses by one-third. Because soil amendments used in our study had a large C/P ratio, mobilization of DRP was not observed but DRP con- centration in percolation water were in all treatments practi- cally the same as that of percolates from control soil. Ekholm et al. (2012) estimated that gypsum amendment would have decrease DRP loss by about 30% during their study, because marked elevation of EC of soil solution suppresses P desorp- tion from soil particle surfaces. For reactive lime, Ulén and Etana (2014) reported 10–40% reduction in DRP loss from the two field sites. Svanbäck et al. (2014) measured equal annual DRP losses for control plots and those treated with reactive lime. The mechanism behind the observed multi-year improve- ment in surface soil stability is likely more complicated with the addition of fiber sludge materials than in the case of inor- ganic, soluble soil-improving materials. In a previous study conducted using the same experimental set-up with intact soil monoliths cored for indoor rainfall simulations from gypsum- amended fields, Uusitalo et al. (2012) found that gypsum sig- nificantly elevated EC in the soil solution and in percolation water (>300 μS cm−1), which promoted aggregate stability and flocculation of clay particles. The effect on SS concen- tration was dependent on how long it took for gypsum to leach out of the soil profile. In the present study, addition of sludges from the pulp and paper industry also increased EC (Supplemental Figure S2) but not to the same degree as gypsum. We hypothesize that, in the case of organic soil amendments used in the present study, direct interactions of soil minerals with the added particulate organic matter and microbe-derived compounds stabilizing aggregates played a major role (see discussion below). The effect was also at least as long-lived as the effect brought about by gypsum applica- tion (Ekholm et al., 2012; Uusitalo et al., 2012). Due to their high C/nutrient ratios, these soil amendments can be used in much higher quantities than many organic materials that are more nutrient rich, such as manures and composted biosolids. However, attempts to increase soil C content may lead to increased leaching of C, especially on soils with higher C stocks. The present results indicate an ini- tial flush of DOC that subsides with time. Declining DOC in percolation water may indicate fast microbial decomposition of the organic matter added to soil and/or stabilization of par- ticulate organic matter within soil aggregates, thus immobi- lizing them in soil. Only a small proportion of the added C was recovered in the soil at the end of the 4-yr study (Supplemental Table S4), leaning to rapid microbial turnover of the added organic matter. The most easily degradable organic matter is bro- ken down during composting, whereas more recalcitrant com- pounds end up in the soil (Heikkinen et al., unpublished data, 2020; Hubbe, Nazhad, & Sánchez, 2010). In line with those RASA ET AL. 181 findings, a small increase in soil C concentration was only measured for the CPMS treatment. In boreal mineral soils, the mean annual C decline is estimated to be 0.4% relative to the existing C concentration (data from Finnish cropland soils for the period 1974–2009; Heikkinen et al., 2013). If the amendments tested in the present study had decomposed at this rate, almost all of the C applied would have been recov- ered after 4 yr. The almost 8 Mg of nonrecovered applied C indicated that losses of added C were two orders of magnitude greater than average decline of 0.4% in soil organic matter in Finnish soils. In our study, relative proportions of numerous microbial groups were changed compared with the control treatment 3 yr after addition of pulp and paper industry side streams. It is shown that changes even at the species level can alter the chemical composition of extracellular polymeric substances produced by microorganisms (Costa et al., 2018) and that dif- ferent microbial species have different effects on soil aggre- gation and erosion (Lehmann et al., 2020; Rigardo & Troeh 1979). In the present study, observed changes in microbial community were well in line with our hypothesis. However, to what extend microbiological interactions with soil minerals could explain improved soil stability against water-induced stresses remains open. Although the concept of aggregate stabilization due to microbial excretion was established almost a century ago (Peele & Beale, 1941), recent studies have revealed new details relating to stabilization of soil particles and C seques- tration. Kopittke et al. (2018, 2020) showed that microbe- derived, N-rich organic compounds in particular are capable of forming new organo-mineral associations on the surface of mineral particles. In the present study, microbial-bound N increased in all treated soils (Table 2). Lavallee, Soong, and Cotrufo (2020) addressed the impor- tance of conceptualization of soil organic matter pools in order to deepen understanding of organic matter functioning, per- sistence, and formation. In their approach, mineral-associated organic matter is distinguished from free particulate organic matter, and mineral-associated organic matter fraction is con- sidered to be physically protected against microbial turnover. Recently, Lehmann et al. (2020) reported that the fungal strains showing a dense growth of mycelia possessed the highest probability of aggregate formation, and most of these effective aggregator strains belonged to the phylum Ascomy- cota. Apart from Sebacinaceae, almost all of the most abun- dantly increased fungi in our study belonged to Ascomycota. For example, Tetracladium marchalianum, which showed a 230% increase (in log-scale) under all treatments (Supple- mental Table S9), was one of the most efficient aggregator fungi according to Lehmann et al. (2020). In addition, micro- bial necromass was recently estimated to make up more than half of soil organic matter in temperate soils (Liang et al., 2019). These new findings suggest that the effects of added organic matter on soil properties are largely mediated by soil microbiology. Thus, the reduction in particle mobilization observed in the present study might be associated with acti- vation of soil microbiota and a subsequent increase in the proportion of mineral-associated organic matter, which sta- bilizes soil aggregates. However, because the data only reveal positive associations rather than deeper mechanistic features, this remains speculation. More detailed studies are required to confirm whether the quality and stability of soil C is affected by organic soil amendments. The processing of the organic material (lignocellulosic material processed in elevated temperature) appeared to directly affect fungal species found in soil, as thermotol- erant species (e.g. Thermomyces lanuginosus; Singh, Mad- lala, & Prior, 2003) and species decomposing lignin and cellulose (e.g., Mycothermus thermophilus; Natvig, Tay- lor, Tsang, Hutchinson, & Powell, 2015) were abundant in LPMS and CPMS treatments. Among other abundant fungi, Sebacinales are suggested to indicate the less inten- sive land use typical of organic farming (Verbruggen et al., 2014), and F. mossae is found to improve nutrient sta- tus and biomass of ryegrass (Berthelo, Blaudez, Beguiris- tain, Chalot, & Leyval, 2018). Also, many bacteria groups that are beneficial for agricultural soils were detected, such as parasites on other bacteria (Bdellovibrionaceae; Starr & Baigent, 1966) and aerobic chemoheterotrophs mineralizing organic C from plant biomass (Chthoniobacteraceae; Kant et al., 2011). They were also abundant in the soils that received pulp and paper industry side streams (Supplemen- tal Table S10). It is tempting to speculate that their rise could be connected to qualitative changes in soil caused by the amendments. Overall, the studied organic side streams had only a minor effect on yields, and Cd present in amendments did not accumulate in soil or grains. In the study of Price and Voroney (2007), significant changes in soil heavy metal con- centration after multiple applications of papermill biosolids were not found either. However, caution should be exercised if repeated applications are planned. In practice, Cd con- centrations in side streams from the pulp and paper industry vary between individual mills, which allows selection of soil amendment materials with low Cd content for use in agricultural applications. 5 CONCLUSIONS This 4-yr field-scale experiment indicated that FS and com- posted and lime-stabilized sludge from the pulp and paper industry can be used to mitigate adverse effects of food pro- duction to the quality of discharge waters. The amendments showed a potential to reduce soil erosion through soil mono- liths over several years. They also increased the pools of 182 RASA ET AL. microbial-bound C and N in soil. A particularly interesting observation was an increased proportion of Sebacinales fungi, which are used as indicators of improved quality of agri- cultural soil in organic farming. However, the organic soil amendments tested had only minor effects on cereal yield and grain quality. The only concern with their use related to the Cd contents of the amendments. Although soil and grain Cd content was not affected in this study, care must be taken to select appropriate sources of materials with low Cd content. Further research on the contributions of industrial and forest- derived organic side streams to soil microbiological functions and their relations to food production and long-term C seques- tration are required. A C K N O W L E D G M E N T S This study was supported by the Finnish Funding Agency for Technology and Innovation (Tekes/Business Finland) and the companies involved in the NSPPulp project (UPM-Kymmene Oyj, Metsä Fibre Oy, Stora Enso Oyj, Biolan Oy, Ekokem Oy, Outotec Finland Oy, Tyynelän maanparannus Oy) and by the Ravinnekuitu-project (2018-2019), financed by the Nutrient Recycling Pilot Programme (Finnish government key project), and the European Research Council (ERC) under the European Union’s Horizon 2020 research and innova- tion programme, grant agreement 818290 (CIRCLES). We thank Helena Merkkiniemi for conducting rainfall simulations and water analyses, Tuija Hytönen and Juha-Matti Pitkänen for DNA analyses in the laboratory, and Helena Soinne and Riikka Keskinen help in soil sampling. AU T H O R C O N T R I B U T I O N S Kimmo Rasa: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administration; Writing-original draft; Writing- review & editing. Taina Pennanen: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing-original draft; Writing-review & editing. Krista Pel- toniemi: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing-original draft; Writing- review & editing. Sannakajsa Velmala: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Writing-original draft; Writing-review & editing. Hannu Fritze: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Visual- ization; Writing-original draft; Writing-review & editing. Janne Kaseva: Data curation; Formal analysis; Methodology; Software; Writing-original draft; Writing-review & editing. Juuso Joona: Funding acquisition; Resources; Writing- original draft; Writing-review & editing. Risto Uusitalo: Conceptualization; Data curation; Formal analysis; Funding acquisition; Investigation; Methodology; Project administra- tion; Visualization; Writing-original draft; Writing-review & editing. C O N F L I C T O F I N T E R E S T The authors declare no conflict of interest. O R C I D Kimmo Rasa https://orcid.org/0000-0003-2350-1839 Risto Uusitalo https://orcid.org/0000-0002-4427-9053 R E F E R E N C E S Abdi, D., Ziadi, N., Shi, Y., Gagnon, B., Lalande, R., & Hamel, C. (2017). Residual effects of paper mill biosolids and liming materials on soil microbial biomass and community. Canadian Journal of Soil Science, 97, 188–199. Anslan, S., Bahram, M., Hiiesalu, I., & Tedersoo, L. (2017). PipeCraft: Flexible open-source toolkit for bioinformatics analysis of cus- tom high-throughput amplicon sequencing data. Molecular Ecol- ogy Resources, 17, e234–e240. https://doi.org/10.1111/1755-0998.12 692 Berthelo, C., Blaudez, D., Beguiristain, T., Chalot, M., & Leyval, C. (2018). Co-inoculation of Lolium perenne with Funneliformis mosseae and the dark septate endophyte Cadophora sp. in a trace element-polluted soil. Mycorrhiza, 28, 301–314. https://doi.org/10. 1007/s00572-018-0826-z Bipfubusa, M., Angers, D. A., N’Dayegamiye, A., & Antoun, H. (2008). Soil aggregation and biochemical properties following the applica- tion of fresh and composted organic amendments. Soil Science Society of America Journal, 72, 160–166. https://doi.org/10.2136/sssaj2007. 0055 Caporaso, J. G., Lauber, C. L., Walters, W.A., Berg-Lyons, D., Lozupone, C. A., Turnbaugh, P. J., . . . Knight, R. (2011). Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proceedings of the National Academy of Sciences of the United States of America, 108, 4516–4522. https://doi.org/10.1073/ pnas.1000080107 Caporaso, J. G., Lauber, C. L., Walters, W. A., Berg-Lyons, D., Hunt- ley, J., Fierer, N., . . . Knight, R. (2012). Ultra-high-throughput micro- bial community analysis on the Illumina HiSeq and MiSeq platforms. The ISME Journal, 6, 1621–1624. https://doi.org/10.1038/ismej.20 12.8 Chen, L., Reeve, J., Zhang, L., Huang Wang, X., & Chen, J. (2018). GMPR: A robust normalization method for zero-inflated count data with application to microbiome sequencing data. PeerJ, 6, e4600. https://doi.org/10.7717/peerj.4600. Chow, T.l., Rees, H. W., Fahmy, S. H., & Monteith, J. O. (2003). Effects of pulp fibre on soil physical properties and soil erosion under simulated rainfall. Canadian Journal of Soil Science, 83, 109–119. https://doi.org/10.4141/S02-009 Costa, O. Y. A., Raaijmakers, J. M., & Kuramae, E. E. (2018). Microbial extracellular polymeric substances: Ecological function and impact on soil aggregation. Frontiers in Microbiology, 9, 1636. https://doi. org/10.3389/fmicb.2018.01636 Diacono, M., & Montemurro, F. (2010). long-term effects of organic amendments on soil fertility: A review. Agronomy for Sustainable Development, 30, 401–422. https://doi.org/10.1051/agro/2009040 Ekholm, P., Valkama, P., Jaakkola, E., Kiirikki, M., Lahti, K., & Pietola, L. (2012). Pietola Gypsum amendment of soils reduces phosphorus losses in an agricultural catchment. Agricultural and Food Science, 21, 279–291. https://doi.org/10.23986/afsci.6831 Elonen, P. (1971). Particle size analysis of soil. Acta Agralia Fennica, 122, 1–122. https://orcid.org/0000-0003-2350-1839 https://orcid.org/0000-0003-2350-1839 https://orcid.org/0000-0002-4427-9053 https://orcid.org/0000-0002-4427-9053 https://doi.org/10.1111/1755-0998.12692 https://doi.org/10.1111/1755-0998.12692 https://doi.org/10.1007/s00572-018-0826-z https://doi.org/10.1007/s00572-018-0826-z https://doi.org/10.2136/sssaj2007.0055 https://doi.org/10.2136/sssaj2007.0055 https://doi.org/10.1073/pnas.1000080107 https://doi.org/10.1073/pnas.1000080107 https://doi.org/10.1038/ismej.2012.8 https://doi.org/10.1038/ismej.2012.8 https://doi.org/10.7717/peerj.4600 https://doi.org/10.4141/S02-009 https://doi.org/10.3389/fmicb.2018.01636 https://doi.org/10.3389/fmicb.2018.01636 https://doi.org/10.1051/agro/2009040 https://doi.org/10.23986/afsci.6831 RASA ET AL. 183 European Union. (2006). Commission regulation (EC) no. 1881/2006. Retrieved from https://eur-lex.europa.eu/eli/reg/2006/1881/2016-04- 01 Foley, B. J., & Cooperband, L. R. (2002). Paper mill residuals and com- post effects on soil carbon and physical properties. Journal of Envi- ronmental Quality, 31, 2086–2095. https://doi.org/10.2134/jeq2002. 2086 Gallardo, F., Cea, M., Tortella, G. R., & Diez, M. C. (2012). Effect of pulp mill sludge on soil characteristics, microbial community and veg- etal production of Lolium Perenne. Journal of Environmental Man- agement, 95, S193–S198. https://doi.org/10.1016/j.jenvman.2010.11. 001 Greenland, D. J., Lindstrom, G. R., & Quirk, J. P. (1962). Organic materials which stabilize natural soil aggregates. Soil Science Soci- ety Proceedings, 26, 366–371. https://doi.org/10.2136/sssaj1962. 03615995002600040018x Gbur, E. E., Stroup, W. W., McCarter, K. S., Durham, S., Young, L. J., Christman, M., & Kramer, M. (2012). Analysis of generalized lin- ear mixed models in the agricultural and natural resources sciences. Madison, WI: ASA. Heikkinen, J., Ketoja, E., Nuutinen, V., & Regina, K. (2013). Declin- ing trend of carbon in Finnish cropland soils in 1975-2009. Global Change Biology, 19, 1456–1469. https://doi.org/10.1111/gcb.12137 Hubbe, M. A., Nazhad, M., & Sánchez, C. (2010). Composting as a way to convert cellulosic biomass and organic waste into high-value soil amendment: A review. Bioresources, 5, 2808–2854. https://doi.org/ 10.15376/biores.5.4.2808-2854 Ihrmark, K., Bödeker, I. T. M., Cruz-Martinez, K., Friberg, H., Kubar- tova, A., Schenck, J., . . . Lindahl, B. D. (2012). New primers to amplify the fungal ITS2 region: Evaluation by 454-sequencing of artificial and natural communities. FEMS Microbiology Ecology, 82, 666–677. https://doi.org/10.1111/j.1574-6941.2012.01437.x IUSS Working Group. (2015). International soil classification system for naming soils and creating legends for soil maps. Rome: FAO. Joona, J., Kuisma, M., Alakukku, L., & Kahiluoto, H. (2012). Carbon flows of waste biomasses applicable to agricultural soils in Finland. In A. Trémier, P. Dabert, C. Druilhe, M. N. Maudet, J. Barth, S. Siebert, & W. Bidlingmaier (Eds.), Global assessment for organic resources and waste management, international conference ORBIT 2012 (pp. 160–164). Weimar, Germany: Verlag Orbit. Kenward, M. G., & Roger, J. H. (2009). An improved approximation to the precision of fixed effects from restricted maximum likelihood. Computational Statistics & Data Analysis, 53, 2583–2595. Kopittke, P. M., Dalal, R. C., Hoeschen, C., Li, C., Menzies, N. W., & Mueller, C. W. (2020). Soil organic matter is stabilized by organo- mineral associations through two key processes: The role of the car- bon to nitrogen ratio. Geoderma, 357, 113974. https://doi.org/10. 1016/j.geoderma.2019.113974 Kopittke, P. M., Hernandez-Soriano, M. C., Dalal, R. C., Finn, D., Men- zies, N. W., Hoeschen, C., & Mueller, C. W. (2018). Nitrogen-rich microbial products provide new organo-mineral associations for the stabilization of soil organic matter. Global Change Biology, 24, 1762– 1770. https://doi.org/10.1111/gcb.14009 Koskiaho, J., Kivisaari, S., Vermeulen, S., Kauppila, R., Kallio, K., & Puustinen, M. (2002). Reduced tillage: Influence on erosion and nutri- ent losses in a clayey field in southern Finland. Agricultural and Food Science in Finland, 11, 37–50. https://doi.org/10.23986/afsci.5711 Kuusisto, E. (1980). On the intensity of rainfall in Finland. Aqua Fennica, 10, 3–11. Lal, R. (2004). Soil carbon sequestration to mitigate climate change. Geoderma, 123, 1–22. https://doi.org/10.1016/j.geoderma.2004.01. 032 Lavallee, J. M., Soong, J. L., & Cotrufo, M. F. (2020). Conceptualizing soil organic matter into particulate and mineral-associated forms to address global change in the 21st century. Global Change Biology, 26, 261–273. https://doi.org/10.1111/gcb.14859 Le Bissonnais, Y. (1996). Aggregate stability and assessment of soil crustability and erodibility: I. Theory and methodology. European Journal of Soil Science, 47, 425–437. https://doi.org/10.1111/j.1365- 2389.1996.tb01843.x Lehmann, A., Zheng, W., ryo, M., Soutschek, K., Roy, J., Rongstock, R., . . . Rillig, C. (2020). Fungal traits important for soil aggregations. Frontiers in Microbiology, 10, 2904. https://doi.org/10.3389/fmicb. 2019.02904 Liang, C., Amelung, W., Lehmann, J., & Kästner, M. (2019). Quantita- tive assessment of microbial necromass contributing to soil organic matter. Global Change Biology, 25, 3578–3590. https://doi.org/10. 1111/gcb.14781 Moragues-Saitua, L., Merino-Martín, L., Stokes, A., & Staunton, S. (2019). Towards meaningful quantification of glomalin-related soil protein (GRSP) taking account of interference in the Coomassie Blue (Bradford) assay. European Journal of Soil Science, 70, 727–735. https://doi.org/10.1111/ejss.12698 Murphy, J., & Riley, J. P. (1962). A modified single solution method for the determination of phosphate in natural waters. Analytica Chimica Acta, 27, 31–36. https://doi.org/10.1016/S0003-2670(00) 88444-5 Nannipieri, P., Ascher, J., Ceccherini, M. T., Landi, G., Pietramellara, G., & Renella, G. (2003). Microbial diversity and soil functions. Euro- pean Journal of Soil Science, 54, 655–670. https://doi.org/10.1046/j. 1351-0754.2003.0556.x Natvig, D. O., Taylor, J. W., Tsang, A., Hutchinson, M. I., & Powell, A. J. (2015). Mycothermus thermophilus gen. et comb. nov., a new home for the itinerant thermophile Scytalidium thermophilum (Torula thermophila). Mycologia, 107, 319–327. https://doi.org/10.3852/13- 399 Oksanen, J., Blanchet, J. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., . . . Wagner, H. (2019). Vegan: Community ecology package. R package version 2.5-5. Retrieved from https://cran.r- project.org/web/packages/vegan/vegan.pdf Peele, T. C., & Beale, O. W. (1941). Influence of microbiological activ- ity upon aggregation and erodibility of Lateritic soils. Soil Science Society Proceedings, 5, 33–35. https://doi.org/10.2136/sssaj1941. 036159950005000C0005x Poulton, P., Johnston, J., Macdonald, A., White, R., & Powlson, D. (2018). Major limitations to achieving “4 per 1000” increases in soil organic carbon stock in temperate regions: Evidence from long-term experiments at Rothamsted Research, United Kingdom. Global Change Biology, 24, 2563–2584. https://doi.org/10.1111/gcb. 14066 Price, Q. W., & Voroney, R. P. (2007). Papermill biosolids effect on soil physical and chemical properties. Journal of Environmental Quality, 36, 1704–1714. https://doi.org/10.2134/jeq2007.0043 Rigardo, R. R., & Troeh, F. R. (1979). Microbial effects in soil erodibility. Soil Science Society of America Journal, 43, 765– 768. Singh, S., Madlala, A. M., & Prior, A. (2003). Thermomyces lanuginosus: Properties of strains and their hemicellulases. FEMS https://eur-lex.europa.eu/eli/reg/2006/1881/2016-04-01 https://eur-lex.europa.eu/eli/reg/2006/1881/2016-04-01 https://doi.org/10.2134/jeq2002.2086 https://doi.org/10.2134/jeq2002.2086 https://doi.org/10.1016/j.jenvman.2010.11.001 https://doi.org/10.1016/j.jenvman.2010.11.001 https://doi.org/10.2136/sssaj1962.03615995002600040018x https://doi.org/10.2136/sssaj1962.03615995002600040018x https://doi.org/10.1111/gcb.12137 https://doi.org/10.15376/biores.5.4.2808-2854 https://doi.org/10.15376/biores.5.4.2808-2854 https://doi.org/10.1111/j.1574-6941.2012.01437.x https://doi.org/10.1016/j.geoderma.2019.113974 https://doi.org/10.1016/j.geoderma.2019.113974 https://doi.org/10.1111/gcb.14009 https://doi.org/10.23986/afsci.5711 https://doi.org/10.1016/j.geoderma.2004.01.032 https://doi.org/10.1016/j.geoderma.2004.01.032 https://doi.org/10.1111/gcb.14859 https://doi.org/10.1111/j.1365-2389.1996.tb01843.x https://doi.org/10.1111/j.1365-2389.1996.tb01843.x https://doi.org/10.3389/fmicb.2019.02904 https://doi.org/10.3389/fmicb.2019.02904 https://doi.org/10.1111/gcb.14781 https://doi.org/10.1111/gcb.14781 https://doi.org/10.1111/ejss.12698 https://doi.org/10.1016/S0003-2670(00)88444-5 https://doi.org/10.1016/S0003-2670(00)88444-5 https://doi.org/10.1046/j.1351-0754.2003.0556.x https://doi.org/10.1046/j.1351-0754.2003.0556.x https://doi.org/10.3852/13-399 https://doi.org/10.3852/13-399 https://cran.r-project.org/web/packages/vegan/vegan.pdf https://cran.r-project.org/web/packages/vegan/vegan.pdf https://doi.org/10.2136/sssaj1941.036159950005000C0005x https://doi.org/10.2136/sssaj1941.036159950005000C0005x https://doi.org/10.1111/gcb.14066 https://doi.org/10.1111/gcb.14066 https://doi.org/10.2134/jeq2007.0043 184 RASA ET AL. Microbiology Reviews, 27, 3–16. https://doi.org/10.1016/S0168- 6445(03)00018-4 Sippola, J., Mäkelä-Kurtto, R., & Rantala, P-R. (2003). Effects of com- posted pulp and paper industry wastewater treatment residuals on soil properties and cereal yield. Compost Science & Utilization, 11, 228– 237. Six, J., Conant, R. T., Paul, E. A., & Paustian, K. (2002). Stabiliza- tion mechanisms of soil organic matter: Implications for C-saturation of soils. Plant and Soil, 241, 155–176. https://doi.org/10.1023/A: 1016125726789 Soinne, H., Hyväluoma, J., Ketoja, E., & Turtola, E. (2016). Rela- tive importance of organic carbon, land use and moisture conditions for the aggregate stability of post-glacial clay soils. Soil & Tillage Research, 158, 1–9. Soinne, H., Keskinen, R., Heikkinen, J., Hyväluoma, J., Uusitalo, R., Peltoniemi, K., . . . Rasa, K. (2020). Are there environmental or agricultural benefits in using forest residue biochar in boreal agri- cultural clay soil? Science of the Total Environment, 731, 138955. https://doi.org/10.1016/j.scitotenv.2020.138955 Starr, M. P., & Baigent, N. L. (1966). Parasitic interaction of Bdellovibrio bacteriovorus with other bacteria. Journal of Bacteriology, 91, 2006– 2017. https://doi.org/10.1128/JB.91.5.2006-2017.1966 Svanbäck, A., Ulén, B., & Etana, A. (2014). Mitigation of phos- phorus leaching losses via subsurface drains from a cracking marine clay soil. Agriculture, Ecosystems & Environment, 184, 124– 134. Tisdall, J. M., & Oades, J. M. (1982). Organic matter and water-stable aggregates in soils. Journal of Soil Science, 33, 141–163. https://doi. org/10.1111/j.1365-2389.1982.tb01755.x Turtola, E. (1996). Peroxodisulphate digestion and fi ltration as sources of inaccuracy in determinations of total phosphorus and dissolved orthophosphate phosphorus in water samples containing suspended soil particles. Boreal Environment Research, 1, 12–26. Turtola, E., & Paajanen, A. (1995). Influence of improved subsurface drainage on phosphorus losses and nitrogen leaching from a heavy clay soil. Agricultural Water Management, 28, 295–310. https://doi. org/10.1016/0378-3774(95)01180-3 Ulén, B., & Etana, A. (2014). Phosphorus leaching from clay soils can be counteracted by structure liming. Acta Agriculturae Scandinavica, Section B — Soil & Plant Science, 64, 425–433. Uusitalo, R., Ylivainio, K., Hyväluoma, J., Rasa, K., Kaseva, J., Nylund, P., . . . Turtola, E. (2012). The effects of gypsum on the trans- fer of phosphorus and other nutrients through clay soil monoliths. Agricultural and Food Science, 21, 260–278. https://doi.org/10. 23986/afsci.4855 Valkama, E., Uusitalo, R., Ylivainio, K., Virkajärvi, P., & Turtola, E. (2009). Phosphorus fertilization: A meta-analysis of 80 years of research in Finland. Agriculture, Ecosystems and Environment, 130, 75–85. https://doi.org/10.1016/j.agee.2008.12.004 Verbruggen, E., Rillig, M. C., Wehner, J., Hegglin, D., Wittwer, R., & van der Heijden, M. G. A. (2014). Sebacinales, but not total root asso- ciated fungal communities, are affected by land-use intensity. New Phytologist, 203, 1036–1040. https://doi.org/10.1111/nph.12884 Vuorinen, J., & Mäkitie, O. (1955). The method of soil testing in use in Finland. Agrogeological Publications, 63, 1–44. Westfall, P. H. (1997). Multiple testing of general contrasts using logical constraints and correlations. Journal of American Statistical Associa- tion, 92, 299–306. https://doi.org/10.1080/01621459.1997.10473627 White, T. J., Bruns, T., Lee, S., & Taylor, J. W. (1990). Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, & T. J. White (Eds.), PCR protocols: A guide to methods and applications (pp. 315–322). New York: Academic Press. Yang, G., Wagg, C., Stavros, S., Veresoglou, D., Hempel, S., & Ril- lig, M. C. (2018). How soil biota drive ecosystem stability. Trends in Plant Science, 23, 1057–1067. https://doi.org/10.1016/j.tplants.2018. 09.007 Zibilske, L. M., Clapham, W. M., & Rourke, R. V. (2000). Multiple appli- cations of paper mill sludge in an agricultural system: Soil effects. Journal of Environmental Quality, 29, 1975–1981. https://doi.org/10. 2134/jeq2000.00472425002900060034x S U P P O R T I N G I N F O R M AT I O N Additional supporting information may be found online in the Supporting Information section at the end of the article. How to cite this article: Rasa K, Pennanen T, Peltoniemi K, et al. Pulp and paper mill sludges decrease soil erodibility. J. Environ. Qual. 2021;50:172–184. https://doi.org/10.1002/jeq2.20170 https://doi.org/10.1016/S0168-6445(03)00018-4 https://doi.org/10.1016/S0168-6445(03)00018-4 https://doi.org/10.1023/A:1016125726789 https://doi.org/10.1023/A:1016125726789 https://doi.org/10.1016/j.scitotenv.2020.138955 https://doi.org/10.1128/JB.91.5.2006-2017.1966 https://doi.org/10.1111/j.1365-2389.1982.tb01755.x https://doi.org/10.1111/j.1365-2389.1982.tb01755.x https://doi.org/10.1016/0378-3774(95)01180-3 https://doi.org/10.1016/0378-3774(95)01180-3 https://doi.org/10.23986/afsci.4855 https://doi.org/10.23986/afsci.4855 https://doi.org/10.1016/j.agee.2008.12.004 https://doi.org/10.1111/nph.12884 https://doi.org/10.1080/01621459.1997.10473627 https://doi.org/10.1016/j.tplants.2018.09.007 https://doi.org/10.1016/j.tplants.2018.09.007 https://doi.org/10.2134/jeq2000.00472425002900060034x https://doi.org/10.2134/jeq2000.00472425002900060034x https://doi.org/10.1002/jeq2.20170 Rasa et al 2021.pdf J of Env Quality - 2020 - Rasa Pulp and paper mill sludges decrease soil erodibility Abstract 1 | INTRODUCTION 2 | MATERIALS AND METHODS 2.1 | Origin and quality of soil amendments 2.2 | Field experiment 2.3 | Rainfall simulation test 2.4 | Microbiological analyses 2.5 | Statistical analyses 3 | RESULTS 3.1 | Rainfall simulation study 3.2 | Effect on cereal yield 3.3 | Heavy metals 3.4 | Effects of amendments on soil and microbes 4 | DISCUSSION 5 | CONCLUSIONS ACKNOWLEDGMENTS AUTHOR CONTRIBUTIONS CONFLICT OF INTEREST ORCID REFERENCES SUPPORTING INFORMATION