Soil C, N and P bioavailability and cycling following amendment with shrub willow chips

Abstract Potato (Solanum tuberosum L.) crops are often cultivated in coarse-textured soils with low soil organic matter and high nitrate leaching risk. Incorporating shrub willow chips into soil could enhance soil properties, while temporally immobilizing N and thus reducing N leaching. We performed a laboratory incubation study and a field experiment to evaluate the effects of shrub willow chips applied at increasing rates in the fall after the potato harvest on C, N and P cycling, soil pH and moisture, and on barley (Hordeum vulgare L.) yield in the following year. In comparison with the control, willow chip incorporation at the rates of 40 and 60 Mg ha−1 increased total C content, but it did not affect the activity of C cycling enzymes. Willow chip addition at these rates also induced nitrate immobilization and reduced barley grain yield and total N uptake, but increased the activity of N cycling enzymes (β-1,4-N-acetylglucosaminidase and leucine aminopeptidase). Mehlich-3 extractable P content and phosphomonoesterase activity were not affected by willow chip addition. Our results suggest that shrub willow chips increased total organic C and immobilized N following their incorporation and can thus mitigate nitrate leaching after the potato harvest. The N immobilization was short lived and was not observed over second winter. We recommend to seed a forage legume in the spring following shrub willow chip incorporation. Willow chip incorporation is an effective means of increasing soil organic carbon.


Introduction
Potato production is characterized by a high degree of soil disturbance caused by intensive and frequent tillage, return low residues resulting in soil organic matter (SOM) depletion over time (Gasser et al. 1995;Grandy et al. 2002;Nyiraneza et al. 2017).The addition of exogenous C inputs may be an effective way to mitigate soil degradation.A study by Gagnon et al. (2001) suggested that materials with a high lignin content, such as wood chips, may enhance soil organic carbon.
Intensive potato production has been linked to relatively low N use efficiency, resulting in high nitrate accumulation after harvest (Liang et al. 2019).This excess N is susceptible to leaching and may lead to contamination of surface and ground waters and eutrophication of receiving streams, rivers, and estuaries (Jiang et al. 2011;Zebarth et al. 2015).High nitrate levels in drinking water cause a serious threat to the health of Prince Edward Island (PEI) residents since the groundwater is the only source of drinking water (Liang et al. 2019).Moreover, the loading of nitrate to receiving waters contributes to anoxic events and degrades the ecological functioning of the associated water bodies (Liang et al. 2019).The addition of shrub willow (Salix spp.) chips to soils may mitigate nitrate leaching by immobilizing mineral N during fall and winter and enhance soil properties such as soil aggregate stability (Marmier et al. 2022;Nyiraneza et al. 2022).
Willow shrub species are commonly grown along riverbanks on erodible and marginal farmland to control the movement of soils, nutrients and herbicides from agricultural fields to water bodies, and to control soil erosion by wind, as well as to offset greenhouse gas emissions through C sequestration (Wilkinson 1999;Amichev et al. 2014;Prosser et al. 2020).Shrub willows have a lifespan of around 30 years and are harvested on a 3-year cycle (Keoleian and Volk 2005).Although harvested shrub willow chips are commonly used as fuel wood, biofuel, biochar, animal bedding and mulch (Labrecque et al. 1997;Karp et al. 2011;Hangs et al. 2016;Ssegane et al. 2016;Rasa et al. 2018;Esperschuetz and Bloomberg 2022), they could be alternatively used as a soil amendment.
Positive, negative or neutral impacts of wood chips on soil pH have been reported (Tahboub et al. 2008;Mirzaei et al. 2021;Nyiraneza et al. 2022).Wood chip amendments have been reported to retain soil water, thereby increasing soil moisture content (Kraus 1998;Davis et al. 2000;Chalker-Scott 2007;Gruda 2008;Li et al. 2018).Wood chip amendments have also been reported to improve total C and N contents (N' Dayegamiye and Angers 1993;Lalande et al. 1998;Tahboub et al. 2008;Li et al. 2018;Nyiraneza et al. 2022), as well as other soil micro and macronutrients (Holtz et al. 2004;Sanborn et al. 2004;Nyiraneza et al. 2022).However, a major concern about using wood chip amendments is that their high C to N ratio can cause temporal inorganic N immobilization, potentially resulting in N deficiency for subsequent crops.The impact of wood chips on soil N immobilization is inconsistent in the literature.Some studies observed N immobilization following wood chip applications (Mackensen et al. 2003;Holtz et al. 2004;Gebhardt et al. 2017;Espinosa et al. 2020;Bourdon et al. 2021;Dessureault-Rompré et al. 2022;Marmier et al. 2022).Others found no N immobilization (Burgess et al. 2002;Tahboub et al. 2007).Immobilization of inorganic N and possibly other nutrients (e.g., P) by soil microorganisms increases competition with plants for inorganic N and may result in N deficiency, stunted growth and reduced crop productivity.A number of studies have associated decreased crop growth and yield with N deficiency after wood chip incorporation into soils (N' Dayegamiye and Dubé 1986;Eldridge et al. 2012;Li et al. 2018;Dessureault-Rompré et al. 2022).The magnitude and duration of wood chip effects on soil properties are related to the amount applied and decomposition rate.
Extracellular enzymes are mainly secreted by soil microbes and plants and play a key role in soil nutrient cycling by performing several biochemical reactions that are necessary for the life of soil microorganisms and for soil function maintenance (Neemisha and Sharma 2022).For instance, the biochemical reactions of C cycle are accomplished by different enzymes such as α-glucosidase (AG), β-glucosidase (BG) and cellobiohydrolase (CB) through their abilities to degrade cellulose, and by phenol oxidase (PO) and phenol peroxidase (PP) through lignin degradation (Ouyang et al. 2014;Sharma et al. 2020).The N and P cycling are performed by various enzymes including β-1,4-N-acetylglucosaminidase (NAG), leucine aminopeptidase (LAP) and phosphomonoesterase (PME) through chitin, proteins and phosphomonoester degradation, respectively (Sinsabaugh et al. 2005(Sinsabaugh et al. , 2008;;Stark et al. 2014).These soil enzymes serve as sensitive indicators of soil quality and health (Gelsomino et al. 2006) as their activities respond rapidly to soil management induced changes (Nyiraneza et al. 2018;Uwituze et al. 2022) and because the measurement of their potential activities is relatively simple and quick (Neemisha and Sharma 2022).Soil extracellular enzyme activity can also be used to indicate nutrient availability.A number of studies showed that activities of enzymes involved in N-and P-mineralization are negatively related to soil available N and P, respectively (Sinsabaugh et al. 1993;Kang and Freeman 1999;Uwituze et al. 2022).
Stimulation of extracellular enzyme activity by wood chip incorporation into soils has been reported by Martens et al. (1992).Gebhardt et al. (2017), found increased potential activity of C cycling (β-d-cellubiosidase, β-xylosidase, AG and BG), N cycling (NAG and LAP) and P cycling (phosphatase) enzymes following the incorporation of wood chip amendments into soils.In addition, Gagnon et al. (2001) reported increased activity of acid phosphatase, arylsulfatase and urease in soils with raw and composted pulp residues applied at rates of 45 and 90 Mg ha −1 .Increased phosphatase activity has been reported after wood chip addition by Lalande et al. (1998).Shrub willow chips can potentially be used as soil amendments to release nutrients and maintain SOM, but more studies are needed to better understand C, N and P cycling following their incorporation.
The objective of this study was to assess C, N and P cycling under willow-chip amended soils.Specifically, this study aimed to: (i) determine the potential decomposition rate of willow chips and soil C mineralization rate under laboratory conditions before applications at field scale; (ii) assess the effects of increasing willow chip application rates in the field on the soil pH, moisture content, total C and N content, nutrient availability and extracellular enzyme activity at different sampling times; and (iii) determine the impact of willow chip application rates on the following barley yield and N and C uptake.

Shrub willow chips production and general characteristics
Willow chips were obtained from locally grown willow shrubs (Viminalis 5027 and SV1 Salix Viminalis).In December 2018, a 3-year shrub willow plantation (first coppiced cycle) was harvested with the stems cut into small pieces (1-2 cm) using an electric wood chipper (Fig. 1A).After willow harvesting and again 1 week prior to the field application, a sample of shrub willow chips was collected and dried in the oven at 60 • C for 48 h.Approximately 15 subsamples were collected, ground, sieved and analyzed for total C and N contents using an Elementar analyzer (Vario Max, Elementar Analyzer, Hanau, Germany).Willow chip moisture content was also measured.On average, the willow chips had a moisture content of 52%, and contained 470 g total C and 9.7 g total N dry matter, thus a C to N ratio of 48.The willow chips were stored in a warehouse until application following the potato harvest in a field that occurred in October 2019 (Fig. 1B).

Laboratory incubation
Prior to the application of willow chips in the field, a controlled microcosm experiment was initiated to assess the decomposition rate of willow chips.Soils for the incubation experiment were collected from the field experiment plots in August 2018 by collecting three to four samples per plot (n = 36; 0-15 cm soil depth) using a Dutch auger and combined into one composite soil sample.The composite sample was passed through a 2 mm sieve and stored at 4 • C for the incubation study.After 72 h, 200 g (dry weight equivalent) of sieved soils were placed in 1-L mason jars and willow chips were added at an equivalent rate of 0, 10, 30 or 50 Mg ha −1 on oven-dry weight basis.Lids were fitted with septa to allow for gas sampling.
The jars were incubated in a completely randomized design blocked by sampling date with four treatments (willow chips at 0, 10, 30 and 50 Mg ha −1 ) × four reps × four sampling times (0, 71, 161 and 395 days) for a total of 64 samples.Soil-wood mixture was maintained at ∼50% water-holding capacity and To determine the soil organic C mineralization rate, the jars were flushed with zero air on days 0, 71, 161 and 395, and the CO 2 concentration was measured after 72 h on days 3, 74, 164 and 398.From each experimental jar, 30 mL of air was collected with a 60 mL syringe and 23 GI needle.Samples were then injected into an infrared gas analyzer and the CO 2 concentration was determined by a gas chromatograph using a Li-COR-820 gas analyzer (COR Biosciences, Lincoln, Nebraska, USA).Nitrogen (N 2 ) was used as a carrier gas at 0.7 mL min −1 .Duplicate samples collected from each jar were combined for an average value.The peak area was recorded and used to determine CO 2 concentration by reference to standards of known CO 2 concentrations.The total CO 2 was calculated and expressed as μg CO 2 -C g −1 soil according to the Li-COR gas analyzer manual.
Mass loss was also used as an alternative method to estimate willow chip decomposition rate.Remaining mass was measured on days 71, 161 and 395.For each treatment, the soil amended with the willow chips was passed through a 2-mm sieve and retained chips were washed, oven-dried at 60 • C for 72 h, and weighed for the remaining mass weight determination.The values were subtracted from the initial willow chip mass weight added to the jars to determine the weight loss by decomposition.
Soils were also analyzed for pH, gravimetric moisture content, total C, total N, and mineral N (NH 4 -N and NO 3 -N) at the end of the incubation.We measured soil pH using a 1:1 soil-to-water ratio solution (Hendershot et al. 2008) using a glass electrode (Thermo Scientific Orion 5 star pH meter).Soil samples were oven-dried for 48 h at 105 • C to determine soil moisture content.Total C and total N were quantified by a dry combustion method (Skjemstad and Baldock 2008) using an Elementar analyzer (Vario Max, Elementar Analyzer, Hanau, Germany).Ammonium (NH 4 -N) and nitrate (NO 3 -N) were extracted with 2 mol L −1 KCl (Maynard et al. 2008), and then soil filtrates were analyzed calorimetrically using flow injection analysis on a Lachat QuickChem 8500 (Lachat Instruments, Loveland, CO).

Field experiment description
The site for field experiment was located at the Agriculture and Agri-Food Canada Harrington Research Farm (46 • 20 37N, 63 • 10 11W) in PEI, Canada.The soil is a Charlottetown fine sandy loam, classified as Ferro-Humic Podzol in the Canadian soil classification system.The 0-15 cm layer has 74 g kg −1 clay, 417 g kg −1 silt and 509 g kg −1 sand.The soil pH, total C, total N and Mehlich-3 extractible P and K were 5.73 g kg −1 , 17 g kg −1 , 1.9 g kg −1 , 142 mg kg −1 and 217 mg kg −1 , respectively.The experimental layout was a completely randomized block design with four willow chip application rates of 0, 20, 40 and 60 Mg ha −1 (wet basis), replicated nine times for a total of 36 plots.The experimental unit dimensions were 8 m × 6 m.During 2019, the trial plots were planted with potatoes (Solanum tuberosum L., CV Russet Burbank), and one week after the potato harvest, the willow chips were surface applied, raked and disked into the soil at a depth of approximately 10 to 15 cm.In spring 2020 (28 May), all plots were planted with barley (Hordeum vulgare L.) underseeded with red clover (Trifolium pratense L.) and, in 2021, the red clover was allowed to regrow.Barley was fertilized with 51 kg N, kg P 2 O 5 and K 2 O kg ha −1 as recommended by PEI Department of Agriculture and Land based on soil analysis (PEI Department of Agriculture and Fisheries 2017).

Field experiment sampling and analysis
Soil samples were collected in the field one month after willow chip incorporation in the fall (November 2019) and monthly throughout the following growing season (May, June, July, August and September 2020) at 0-15 cm soil depth.Samples from each plot represented a composite of four subsamples taken randomly using a Dutch auger.As a result of pandemic restrictions on lab access at the time of sampling, all soils were immediately placed in −20 • C freezers until processing.A subsample of field-moist soil was sieved through 4.75-mm sieve.Moisture content and NH 4 /NO 3 extractions were completed on fresh soils.The remainder of the sample was air-dried for soil pH and Mehlich-3 extractable P measurements.A portion of the air-dried sample was ground and sieved through a 500-μm sieve for total C and N analyses.Soil samples for enzyme assays were frozen at −20 • C until analysis.Soil pH, moisture content, total C and N, and NH 4 -N and NO 3 -N were quantified and analyzed as described above.Soil extractable P was determined with Mehlich-3 extraction solution (Mehlich 1984;Ziadi and Tran 2008) followed by inductively coupled plasma optical emission spectroscopy (ICP, VARIAN 820-MS, Varian Inc. Scientific Instruments, Mulgrave, Australia).

Extracellular soil enzyme activity analyses
Potential extracellular enzyme activity assays were previously described in detail by Uwituze et al. (2022).Briefly, we measured the activity of two extracellular oxidase enzymes (PO and PP) in soil suspensions prepared by thoroughly mixing 1.0 g of soil in 50 mL of 50 mmol L −1 acetate buffer solution (pH 5.0) following the protocol described by Center for Dead Plant Studies (2000) with minor modifications.After 20 h of incubation at 25 • C in the absence of light, the absorbance was read using a microplate spectrophotometer reader at 460 nm (Synergy HTX, BioTek Instruments, Inc., Winooski, Vermont, USA).Potential oxidative activities were then calculated and expressed as μmol h −1 g −1 dry soil.
We also assayed the activity of six extracellular hydrolase enzymes (AG, BG, CB, NAG, LAP and PME) using soil slurries prepared by homogenizing 2.75 g of field moist soil and 91 mL of 50 mM sodium acetate buffer solution (pH 5.7), according to the fluorometric microplate protocol by Bell et al. (2013).The incubation period was 1.5 h in the absence of light at 25 • C, except for LAP, which was incubated at 4 • C for 24 h.The fluorescence intensity was read with a microplate reader that had a 460 nm filter (center wavelength/bandpass: 360/40 nm excitation, 460/40 nm emission).From these fluorescence values, we calculated potential hydrolytic activity as the rate of substrate converted in nmol h −1 g −1 dry soil.Detailed information on studied extracellular enzymes and their general functions are found in Table 1.
In-situ soil nitrate and phosphate dynamic using anion exchange membranes Soil nitrate and phosphate released during fall and winter periods following willow chip incorporation (in 2020 and 2021) were also measured using anion exchange membranes (AEMs).The AEMs strips of 6 cm × 2.5 cm (type AR 204, Ionics, Watertown, MA) were prepared as described by Ziadi et al. (1999).They were inserted into the soil (four in winter 2020 and two in winter 2021 per plot) at a 15-cm depth through an opening sliced with a shovel, and the soil was firmly pressed to maintain good contact between the AEMs and the soil.To facilitate their retrieval, each AEM was attached to a fishing line attached to a flag.After a contact period of 208 days (30 October 2019-25 May 2020) and 186 days (6 November 2020-5 May 2021), each AEM was washed with distilled water to remove adhering soil particles, and then placed into individual tubes containing 25 mL of 1 KCl (Ziadi et al. 1999).Nitrate and phosphate adsorbed on AEMs were desorbed in the laboratory by 2-h shaking in a solution containing 1 mol L −1 KCl, filtered through Whatman No. 5 filters, and analyzed for nitrate and phosphate concentrations.Nitrate was measured using a Lachat QuickChem 8500 (Lachat Instruments, Loveland, CO, USA), and phosphate by colorimetry, following the ascorbic acid-molybdate method (Murphy and Riley 1962).The nitrate and phosphate fluxes from AEMs (expressed as μg NO 3 -N cm -2 day -1 , μg PO 4 -P cm -2 day -1 ) were calculated by multiplying the nitrate or phosphate concentrations by the volume of extractant (25 mL), and then dividing by the total surface area of the membrane (two sides, 30 cm 2 ) and the number of days in contact with the soil (Ziadi et al. 1999).Values were then multiplied by 100 to express them per 100 cm −2 .

Crop harvest
Barley was harvested in late summer 2020 by separating grain and straw, and analyzing C and N concentrations in each sample.For each plot, straw biomass harvested from 12 m 2 (1.5 m × 8 m) was weighed, and a subsample of at least 500 g was taken, the rest of the straw biomass samples being returned to their respective plots.Grain and straw biomass samples were oven-dried (60 • C for 48 h) and weighed for yield.Oven-dried tissue samples were finely milled to pass through a 0.15-mm sieve for total C and N analysis using an element analyzer (Vario Max, Elementar Analyzer, Hanau, Germany).Total C and N accumulation was obtained by adding together the C and N accumulation of grain and straw calculated by multiplying C and N concentration by their respective dry matter yield.

Kinetic model description and statistical analysis
The percentage of willow chip dry mass remaining (MR) on each sampling date was calculated as: where X 0 is the initial mass of willow chips at the beginning of the incubation experiment (t = 0) and X t is the remaining mass at time t.The percentage of mass loss was calculated as the difference with the dry MR: Decomposition rates for the different willow chip application rates were calculated by fitting dry MR from each treatment to the three-parametric single exponential decay model with an asymptote (Dessureault-Rompré et al. 2020) using the nonlinear regression function in SigmaPlot software version 14.5.The model was described as: where Y is the fraction of the initial mass of willow chips remaining at time t (days); k d is the decomposition rate constant (day −1 ); a is the fraction of the initial mass of willow chips that is subjected to loss; Y 0 is the asymptotic value when time is infinite and t is time.As the values of Y 0 and a are interrelated, a constraint was imposed on that equation so that the sum of Y 0 + a corresponded to the initial amount of willow chips (Dessureault-Rompré et al. 2020).
To estimate the potentially mineralizable organic C pools (C O ) and the first-order rate constant (k C ) of soil organic C min-eralization, the nonlinear regression approach for N mineralization of Smith et al. (1980) was used in SigmaPlot software.The model was described as: where C min is the cumulative evolved CO 2 at specific time (t); C O is the potentially mineralizable C; and k C is the rate constant (day −1 ).
Statistical analyses were performed using the MIXED procedure of SAS (SAS Institute Inc. 2014) for a completely randomized block design by treating willow chip application rates and sampling dates as fixed factors, and replicate as a random factor.Repeated measurement was done for parameters that were analyzed at different sampling dates.All data were tested for normality using the Shapiro-Wilk test, and logarithmically transformed to meet the assumptions of ANOVA for variables that were not normally distributed.The DIFF option of SAS was used for means of comparison when the main treatment and sampling date effects were significant at the probability level of 0.05.

Laboratory incubation experiment
The remaining mass of willow chips significantly declined over the incubation time in all treatments with dry MR increasing proportional with the willow chip rate (Fig. 2A).Willow chip decomposition in all rates was represented by an exponential decay curve (Fig. 2A, Table 2).The decomposition rate constants (k d ) were 0.0232, 0.0083 and 0.0027, respectively, at the rates of 10, 30 and 50 Mg ha −1 , indicating a decreasing decomposition rate with increasing rates.
Willow chip rates (Mg ha The cumulative CO 2 evolution significantly increased over the incubation time in all treatments with willow chip amended soils releasing significantly more CO 2 than the unamended soils (Fig. 2B).Soil organic C mineralization in all treatments was represented by an exponential growth curve (Fig. 2B, Table 3).Potentially mineralizable C (C O ) rose with increasing willow chip rates.The C mineralization rate constants (k C ) were 0.0154, 0.0195, 0.0173 and 0.0164, at the rates of 0, 10, 30 and 50 Mg ha −1 , respectively, indicating a decreasing rate of C mineralization when willow chips were added at high rates.
Soil pH, moisture content, total C and NH 4 -N contents were not significantly influenced by willow chip application rates after 395 days of incubation.However, total N and specif-ically NO 3 -N contents were significantly reduced by willow chip rate (Table 4).

Field experiment Soil properties over time
There was no significant effect of willow chip rate and interaction treatment × date for the soil pH (Table 5).Overall, the soil pH measured at two-time points in 2020 indicated greater values on 25 May than on 29 September.
Soil moisture content was significantly affected by willow chip rate, with higher values observed at rates of 40 and 60 Mg ha −1 compared to 20 Mg ha −1 and the control Table 3. Regression models of cumulative CO 2 released over the incubation.
Willow chip rates (Mg ha   (Table 5).Soil moisture content varied significantly according to the sampling date, with the highest value recorded on 14 November 2019, while the lowest values were observed on 23 July and 20 August 2020.Soil moisture content was significantly different between willow chip rates for four of six sampling dates (significant treatment × date interac-tion), with comparable values in November and September (Fig. A1).
Total C was significantly affected by willow chip rate, with greater content observed at the rates of 40 and 60 Mg ha −1 compared to 20 Mg ha −1 and the control (Table 5).Total C did not vary according to the sampling date, but there was a significant treatment by date interaction.Mean total C contents were significantly different between willow chip rates for four of six sampling dates with comparable values early after material application, in November and May (Fig. A2).Willow chip at higher rates could mitigate moisture stress in the drier months.
There was no significant effect of willow chip rate and interaction treatment × date for the total N (Table 5).Total N changed significantly according to the sampling date, with the highest content occurring on 23 July and 20 August 2020, and the lowest content occurring on 14 November 2019, 25 May 2020 and 29 September 2020.
Ammonium (NH 4 -N) was unaffected by willow chip rate, but it varied significantly according to the sampling date, with the highest NH 4 -N content occurring on June 23 and July 23 in 2020, and the lowest level occurring on 25 May and 29 September 2020 (Table 5).There was a significant treatment by date interaction on NH 4 -N content.Mean NH 4 -N contents were significantly different between willow chip rates for three of six sampling dates with comparable values observed in November, July and August (Fig. A3).
Nitrate (NO 3 -N) was significantly affected by the willow chip rate, with the lowest NO 3 -N content found at rates of 40 and 60 Mg ha −1 compared to 20 Mg ha −1 which was significant lower than the control (Table 5).The NO 3 -N also varied significantly according to the sampling date, with the NO 3 -N level peaking on 23 June 2020, following fertilizer application, and decreased thereafter.A significant treatment × date interaction indicated that the soil NO 3 -N levels were more strongly reduced in May and June 2020 (Fig. A4).
There was no significant willow chip rate effect and interaction treatment × date on Mehlich-3 P, but it varied significantly according to the sampling date, the highest value recorded in July 2020 whereas the other dates were comparable (Table 5).
There were no significant willow chip rate effects on hydrolase enzyme activity, with the exception of higher NAG and LAP activities observed at the rates of 40 and 60 Mg ha −1 compared to 20 Mg ha −1 and the control (Table 6).Activity of all hydrolase enzymes varied significantly with the sampling date, with the highest activity occurring on 20 August and 29 September 2020 (for all hydrolases), 14 November 2019 (BG) and 23 June 2020 (BG and CB), and the lowest activities occurring on 23 July 2020 (AG, BG and CB), and on 25 May 2020 (NAG and LAP) or November 2019 (AG).There was a significant treatment by date interaction for AG and NAG activities.Mean activity of AG and NAG was significantly different between willow chip rates for respectively two (Fig. A5) and four (Fig. A6) of six sampling dates, especially during summer months.
Oxidases were also unaffected by willow chip rate but varied significantly according to the sampling date (Table 6), with both PO and PP potential activities peaking on 20 August 2020, and at the lowest point on 14 November 2019 (PO) and on 29 September 2020 (PP).There was, however, significant treatment × date interaction for PP potential activity, with three of six sampling dates showing significant differences between willow chip rates, all before July 2020 (Fig. A7).
Results of nitrates desorbed from AEMs are shown in Fig. 3A.Over winter 2019-2020, AEM-nitrate flux was significantly affected by willow chip rate, and ranged between 3.3 to 61.1 μg NO 3 -N 100 cm −2 day −1 with the lowest values being observed at rates of 40 and 60 Mg ha −1 compared to the 20 Mg ha −1 which was significant lower than the control.Over winter 2020-2021, AEM-nitrate content was also significantly impacted by the willow chip rate while both highest rates (40 and 60 Mg ha −1 ) gave twice higher fluxes than 20 Mg ha −1 rate.
In addition, the AEM-phosphate fluxes were significantly affected by willow chip rates only over the winter 2019-2020 sampling period (Fig. 3B).During that period, AEMphosphate ranged between 1.0 and 6.2 μg PO 4 -P 100 cm −2 day −1 with highest fluxes being only observed at the highest rate (60 Mg ha −1 ).

Barley yield and nutrient uptake
One year after willow chip incorporation, barley grain yield was significantly affected by willow chip rate with a decrease from 3.60 to 2.81 Mg ha −1 at the 40 and 60 Mg ha −1 rates compared to the control (Table 7).However, the barley straw yield was not affected by willow chip application.
Total N uptake was lower at the rates of 40 and 60 Mg ha −1 than in other treatments (Table 7).In this case, similar trends were observed for both grain and straw N uptake.Total barley C uptake was significantly affected by willow chip rate, with rates of 40 and 60 Mg ha −1 associated with lower values than 20 Mg ha −1 and the control.

Shrub willow chip decomposition in lab incubation study
In the present study, mass loss ranged from 21% to 60% across willow chip application rates and incubation time.Results showed that both mass loss and constant decay rate (k d ) decreased as the rate of willow chip addition increased.This decrease could be explained by higher N immobilization as the amount of willow chips increases.Moreover, high amount of plant residues applied to the soil can produce anaerobic conditions that limit residue decomposition (Repullo et al. 2012).
Willow chip rates 0 Mg ha  slower decomposition rate than in our study for corn (Zea mays, 0.0005 day −1 ) and spring wheat (0.0008 day −1 ), even though those residues had initial C/N ratios (40 for corn and 48 for spring wheat) comparable to that of willow chips in this study.This indicates that the initial C/N ratio of residues is not the sole factor in explaining the decomposition rate.Consequently, willow chip decomposition rate from different studies may be affected by many other factors, including Table 7. Least square means (LSMEANS) and variance analysis (ANOVA) for the effect of willow chip rate on barley yields and C and N uptake.
Barley yield (t ha −1 ) Nitrogen uptake (kg N ha −1 ) Carbon uptake (kg C ha chips quality (i.e., lignin and carbohydrate contents), initial wood chip density, diameter, moisture content, plant species and plant parts included in chipping as well as age of chips as reported previously (Garrett et al. 2007).In addition, characteristics of decomposer community, environmental factors (temperature, moisture), soil properties (texture, SOM and depth) as well as handling of willow chips before application (e.g., storage and stockpiling) may influence the decomposition rate (Garrett et al. 2007).
Results of this study showed a rapid evolution of CO 2 at the beginning of the incubation period, particularly when willow chips were added at high rates.This indicated initial acceleration of native SOC decomposition that can be attributed to the priming effect induced by fresh willow chip addition.Also, this initial high potential of C mineralization can reveal the presence of labile C compounds in willow chips and their fast degradation in the soils.An increase in CO 2 evolution was observed as willow chip application rates increases which could be attributed to the increase in soil microbial respiration (Sissoko and Kpomblekou-A 2010; Nyiraneza et al. 2022).
We observed low constant rates (k C ) of C mineralization with increasing rate of willow chip application as explained above.Comparable values of k C were observed between unamended control and amended soils likely due to greater respiration in soils with high sand content (Six et al. 2001;Xu et al. 2016).The k C observed in our study ranged between 0.0154 day −1 and 0.0195 day −1 and was 1.7-fold to 1.3-fold lower than that of sandy loam soils (0.026 day −1 ) amended with different cover crop residues in Ontario (Katanda 2022).

Effect of willow chip application on soil ph and moisture
Soil pH was unaffected by willow chip rate in both our field study and our incubation laboratory.This was also the case in a study by Tahboub et al. (2008) following one application of pecan wood chips.In contrast, some previous studies observed soil pH decreases following wood chips addition likely due to acidifying processes during the mineralization of wood residues, the nitrification of co-applied N fertilizers or root exudation (Tahboub et al. 2008;Mirzaei et al. 2021).
Another study reported increased soil pH after the addition of wood residues (Mirzaei et al. 2021), which can be explained by the decarboxylation of organic anions and release of OH − or the high concentrations of basic cations including Ca, Mg, and K when residues are decomposing.In our study, soil pH varied according to the sampling date, and the mean pH value decreased by 0.16 units from May to September 2020 in all plots due probably to the N fertilizer added in spring.
For the incubation study, soil moisture content was comparable among willow chip rates.However, in the field, soil moisture increased by 11% and 18% with willow chips at rates of 40 and 60 Mg ha −1 , respectively.This increase is likely associated with the direct effect of incorporated wood residues on reducing evaporation losses, disruption of soil water flow paths, as well as absorption of water into the wood matrix (Li et al. 2018), thereby increasing the soil water holding capacity.Our results are aligned with those of Gagnon et al. (1997) who reported a linear relationship between the increase in soil water content and the amount of C added from wood composts.Various other studies observed enhanced soil water retention after wood chip application to soils (Kraus 1998;Davis et al. 2000;Chalker-Scott 2007;Gruda 2008).

Effect of willow chip application on carbon cycling
At rates of 40 and 60 Mg ha −1 , total C content increased by 4% and 5%, respectively, compared to the control, while in the incubation this was not observed.This increase may be attributed to the direct effect of willow chips on SOM as the content of C macromolecules such as lignin and cellulose, and other polysaccharides increase with the amount of wood chips applied.A positive impact of wood chip application on SOM was reported in previous studies (N 'Dayegamiye and Angers 1993;Lalande et al. 1998;Tahboub et al. 2007;Li et al. 2018).Increased CO 2 is aligned with increased C content and increased moisture content related to increasing willow chip rate.Enhanced microbial respiration rate after wood chips incorporation in this study corroborates previous studies (Zibilske 1987;Gebhardt et al. 2017;Espinosa et al. 2020;Nyiraneza et al. 2022).
The increases in total C content and soil respiration under soil amended with willow chips at high rates were not associated with the increases of enzyme activity involved in C cycling (AG, BG, CB, PO and PP) which might be attributed to inorganic N limitation as previously reported by Bowles et al. (2014).A study by Gebhardt et al. (2017) also found the lowest potential hydrolytic activity of C degrading extracellular enzymes in wood chip amended soils.In fact, soil microorganisms may spend their energy in enzyme production to acquire a given nutrient from complex organic compounds only when that nutrient is limiting (Harder and Dijkhuizen 1983;Uwituze et al. 2022).Nitrogen limitation was evidenced by decreased NO 3 -N concentration in soil amended with willow chips at high rates for both field and lab experiments, and AEM-nitrate measured over winter 2020.A temporal change in potential hydrolytic and oxidative activity related to C cycling was observed, which was mostly associated with seasonal fluctuations in substrate availability, soil temperature and moisture content (Wallenstein and Weintraub 2008).

Effect of willow chip application on nitrogen cycling
The immobilization of soil mineral N following incorporation of willow chip corroborates previous findings (Holtz et al. 2004;Gebhardt et al. 2017;Espinosa et al. 2020;Bourdon et al. 2021;Dessureault-Rompré et al. 2022;Marmier et al. 2022).This N immobilization implies that willow chip addition can potentially mitigate nitrate availability for leaching, especially after potato harvest when high residual nitrate coexists with excessive moisture (Homyak et al. 2008;Moorman et al. 2010;Li et al. 2013;Marmier et al. 2022).However, this mineral N immobilization may be short-lived as demonstrated by AEMs where AEM nitrate fluxes were strongly lower in plots amended with willow chip than in the unamended control during the first winter but comparable fluxes among treatments were observed in the second winter.
Increasing soil C supply with willow chip under N limited conditions resulted in higher activity of two N cycling enzymes (NAG and LAP).Bowles et al. (2014) reported that N degrading enzyme activity increased with C availability.At rates of 40 and 60 Mg ha −1 , willow chip incorporation boosted potential NAG activity by 20% and 40%, respectively compared with the control.This increase is likely attributed to the increase in fungal biomass when large amounts of willow chips are incorporated into the soil.In fact, the first step of decomposition of wood chips that contain cellulose and lignin is preferentially performed by fungi (Lalande et al. 1998), suggesting that, increased NAG activity in this study was driven by the presence of suitable substrate (chitin) that is found in fungal cell wall.Our results corroborate the results of Gebhardt et al. (2017) who found an increased NAG activity after the addition of chipped wood from Juniperus monoserma trees into soils.
LAP enzyme is involved in peptides and proteins degradation to release leucine residues (Stark et al. 2014).Willow chip addition at rates of 40 and 60 Mg ha −1 increased potential LAP activity by 12%−13%.When N is limiting, soil microbes can be stimulated to produce N mineralizing enzyme such as LAP to decompose organic N from SOM to obtain inorganic N that compensates for the high C/N ratios because the degradation of complex wood chip substrates requires more N. Increased LAP activity under wood chip-amended soils was also reported in other studies (Tian et al. 2016;Gebhardt et al. 2017).Temporal variations in N cycling potential activity were observed, and their highest levels occurred on August and September 2020, at a time that ammonium and nitrate concentrations were low.This indicated that, in the short term, a high willow chip C/N ratio limited N availability to soil microbes triggered the production of N-acquiring enzymes.

Effect of willow chip application on phosphorus cycling
Soil Mehlich-3 P was unaffected by the willow chip rate, implying that willow chips are not a significant source of P. A previous study by Tahboub et al. (2007) also found no significant effect of wood chip application on available P. In addition, Dessureault-Rompré et al. ( 2022) found a very lower P supply rate in miscantus straw and willow chip amended plots than in the control.Similarly, willow chip application did not impact PME activity.PMEs hydrolyze phosphate esters to release inorganic P which increases PME production under soil P deficiency (Deng and Tabatabai 1996;Uwituze et al. 2022).

Effect of willow chip application on barley production
Willow chip incorporation at 40 and 60 Mg ha −1 in the previous fall reduced barley grain yield compared with the control due to soil N immobilization.Our results are in agreement with findings by Dessureault-Rompré et al. (2022) who found a decreased lettuce (Lactuca sativa) yield by 35% and 14%, respectively, in soils amended with miscantus straw and willow chips.Li et al. (2018) reported decrease in wheat growth and vigor with high concentrations of poplar chips.Total N and C uptake in barley tissues followed the grain yield trend.Our results contradict, however, those of N'Dayegamiye and Dubé (1986) who reported higher cereal and forage yields in wood chip amended plots than in the control because they combined wood chips with pig manure as a source of nitrogen in a long-term study.Therefore, the application of willow chip in fall may immobilize N over the winter and in the following spring and a higher N fertilizer rate may need to be applied if a cash crop is seeded in the following spring.However, N immobilization from willow chips was short lived and was not observed during the second winter.

Conclusion
This study assessed the willow chip decomposition rate when applied following a potato harvest and its effects on C, N and P cycling, and on barley productivity.Willow chip decomposition rates decreased with increasing application rates.Willow chip addition at high rates did not affect soil pH, but it increased soil moisture content.The addition of willow chips at high rates increased soil total C and soil respiration, but this increase was not associated with higher activity of C-cycling related enzymes due to N limitation to soil microbes.Furthermore, willow chip addition at high rates significantly decreased soil nitrates which resulted in enhanced NAG and LAP activity to accelerate the decomposition of soil organic N compounds for more N acquisition.The addition of willow chips did not impact soil extractable P contents and PME potential activity.Mineral N immobilization by willow chips addition at high rates reduced barley grain yield and total N uptake.Overall, results from our study demonstrated that willow chips applied in fall may immobilize N over the following winter and thus can potentially mitigate nitrate availability for leaching following high N demanding crops such as potatoes.Higher supplemental N fertilizer rates may be needed on a cash crop planted in the following spring to compensate for N immobilization.We would recommend planting a legume crop following willow chip application.Immobilization of N in willow-chip amended soils was shortlived as it was not noticed in the second winter.Willow chips are therefore an effective means to increase soil organic carbon and soil moisture holding capacity.

Fig. 2 .
Fig. 2. (A) Mass loss of willow chips and (B) cumulative CO 2 evolution over a 395-day incubation period.Decay and growth curves were fitted for each application rate with exponential decay and growth functions.Related kinetic model estimated parameters are found in Tables2 and 3. Y 0 is the asymptotic value when time is an infinite; a is the biomass of willow chips that is subjected to loss; k d is the decomposition rate constant (day −1 ); C O is the potentially mineralizable C; k C is the rate constant for C mineralization (day −1 ).Different letters indicate statistically different values at the 0.05 probability level.

Fig. 3 .
Fig. 3. (A) Anion exchange membrane-nitrate (AEM-N) and (B) anion exchange membrane-phosphate (AEM-P) fluxes during winter 2019-2020 and 2020-2021 following willow chip incorporation.Different letters within a group indicate statistically different values at a 0.05 probability level.Vertical bars represent standard error of the mean.
Fig. A1.Decomposing the interaction of the effects of rate and sampling date on soil moisture content.Different letters within a group indicate statistically different values at a 0.05 probability level.Vertical bars represent standard error of the mean.

Table 1 .
Extracellular enzymes assay, their EC number, substrates, general function and indicator of microbial activity.
Y is the dry willow chip remaining mass at time t (days); Y 0 is the asymptotic value when time is infinite; a is the biomass of willow chips that is subjected to loss; k d is the decomposition rate constant (day −1 ) and R 2 is the nonlinear coefficient of determination. Note: C O is the potentially mineralizable C; k C is the rate constant (day −1 ) for C mineralization and R 2 is the nonlinear coefficient of determination. Note:

Table 4 .
Least square means (LSMEANS) and analysis of variance (ANOVA) for the effect of willow chip application rate on soil pH, moisture content, total C and N, ammonium (NH 4 -N), and nitrate (NO 3 -N) at 395 days of incubation.
Note: ns--not significant at a 0.05 probability level.* , * * * , significant at 0.05 and 0.001 probability levels, respectively.Values within a column followed by different letters are statistically different at the 0.05 probability level.

Table 5 .
Least square means (LSMEANS) and ANOVA for the effect of willow chip application rates and sampling dates on soil pH and moisture content, total C and N, ammonium (NH 4 -N), nitrate (NO 3 -N) and Mehlich-3 extracted phosphorus.
Values within a column followed by different letters are statistically different at 0.05 probability level.ns--no significant difference; * , * * * , significant at 0.05 and 0.001 probability levels, respectively.(Sampling dates are November 2019, May, June, July, August and September 2020). Note:

chip rates (Mg ha −1 )
Note: ns, * * , * * * , not significant at 0.05 probability level, significant at 0.01 and 0.001 probability level, respectively.Values within a column followed by different letters are statistically different at a 0.05 probability level.