Plant uptake of nitrogen adsorbed to biochars made from dairy manure

The conversion of dairy waste with high moisture contents to dry fertilizers may reduce environmental degradation while lowering crop production costs. We converted the solid portion of screw-pressed dairy manure into a sorbent for volatile ammonia (NH3) in the liquid fraction using pyrolysis and pre-treatment with carbon dioxide (CO2). The extractable N in manure biochar exposed to NH3 following CO2 pre-treatment reached 3.36 g N kg−1, 1260-fold greater extractable N than in untreated manure biochar. Ammonia exposure was 142-times more effective in increasing extractable N than immersing manure biochar in the liquid fraction containing dissolved ammonium. Radish and tomato grown in horticultural media with manure biochar treated with CO2 + NH3 promoted up to 35% greater plant growth (dry weight) and 36–83% greater N uptake compared to manure biochar alone. Uptake of N was similar between plants grown with wood biochar exposed to CO2 + NH3, compared to N-equivalent treatments. The available N in dairy waste in New York (NY) state, if pyrolyzed and treated with NH3 + CO2, is equivalent to 11,732–42,232 Mg N year−1, valued at 6–21.5 million USD year−1. Separated dairy manure treated with CO2 + NH3 can offset 23–82% of N fertilizer needs of NY State, while stabilizing both the solid and liquid fraction of manure for reduced environmental pollution.

Nitrogen (N) management is a major challenge in agricultural systems. Nitrogen fertilizer for crops is a significant cost, while the disposal of N-rich wastes such as dairy manure, can also be costly. The shift toward sustainable farming highlights the need for opportunistic waste management able to capture nutrients from liquid waste streams and transform them into high-value, dry fertilizers that can be safely transported and traded.
Dairy manure excretion in NY State alone averages 12,821,616 Mg per year 1,2 . Containing approximately 64,108 Mg N, 16,786 Mg P, 44,876 Mg K 3 , these excreted nutrients are sufficient to fertilize the state's extensive 17,321 hectares (ha) of corn production, requiring 51,360 Mg N calculated using the average application rate in 2018, 96 kg ha −14-6 . The benefits incurred by transforming dairy waste into agronomic inputs extend to dairy farmers and grain farmers alike in NY State; a farmer growing 81 ha of corn spends 28,000 USD year −1 for fertilizer 7 , while a dairy farmer with 550 cows spends 25,000 USD year −1 for manure storage 8 . To couple these processes, new technologies for recycling dairy waste products back into crop nutrient inputs are necessary.
While direct manure spreading has been the most common means of disposal and re-use of dairy waste, such practices can result in transport of N and P into waterways 9,10 , as nutrients from manure often exceed the agronomic demand 11 . Furthermore, manure storage for future manure application produces notable quantities of methane, a known greenhouse gas 5,12 . Moreover, long-distance transport of dairy manure for land application also has a number of drawbacks such as the spreading of pathogens 13 and costs associated with transporting a material with a water content of > 70% 14 . Alternatives to direct land application of dairy slurry are needed with less detrimental impact on the environment 15 , which utilize the high nutrient content of manure for agronomic purposes. Source-separation of manure followed by anaerobic digestion of the liquid portion is one alternative [16][17][18] that is being adopted across farms of all scales in NY State 19 and other regions.
Solid-liquid separation of manure is a first step in efficient re-use of waste nutrients 16 . The physical separation of manure into solid and liquid fractions (slurry) significantly lowers N leaching from the solid fraction 17,18 . Most of the inorganic N in dairy manure is found in the liquid portion, approximately 4400 mg NH 4 -N L −117 , and can be a significant source of N 2 O and NH 3 emissions from lagoons 5,20,21 . Thus, a technology is needed for removing N from stored slurry and converting it into a fertilizer. We see great potential in converting the solid portion of separated dairy manure into a biological charcoal, or biochar, with high sorption properties, able to remove N from the liquid portion of dairy manure.
. The incorporation of CO 2 molecules may further enhance NH 3 retention through the formation of NH 4 HCO 3 , creating NH 4 HCO 3 -intercalated biochar for use as a mineral-organic, slow-release fertilizer. The intimate association between the first layer of chemisorbed NH 3 on biochar surfaces and NH 4 HCO 3 precipitates projecting further out from the surface is expected to provide both long-term and immediately available N. Yet no study has evaluated the plant-availability of N incorporated into manure biochar through sequential NH 3 and CO 2 adsorption. Therefore, we quantify the plant uptake of N adsorbed to dairy manure biochar using either liquid NH 4 + or gaseous NH 3 with prior CO 2 conditioning. Crops grown in the greenhouse, tomato, marigold, and radish, were used in this small-scale study to demonstrate proof of concept. We benchmarked the performance of dairy manure biochar as an adsorber against wood biochar, a material reported to sorb up to 6 mg g −1 NH 3 -N 43,47 . We compare the plant-availability of N from biochars exposed to either liquid NH 4 + or gaseous NH 3 to the availability of N from urea fertilizer added in combination with each biochar. We expect greater plant-availability of N incorporated to dairy manure biochars compared to that incorporated into wood biochar. We expect greater N use efficiency from both biochars exposed to N compared to biochar added with inorganic fertilizer.

Results
The increase in total N of the amendments following NH 3 exposure was much larger for wood (1.13% point change in N, from 0.75 to 1.88% N) than manure biochars (a point change of 0.06% N, from 2.05 to 2.11% N). The increase in KCl-extractable N (sum of NO 3 -N and NH 4 + -N) was similar between wood (0.005-3.1 g N kg −1 ) and manure biochars (0.026-3.4 g N kg −1 ). The KCl-extractable, or plant-available N in manure biochar following NH 3 exposure increased 127 fold, in comparison to a 595 fold increase for wood biochar, with only 27% of the added N in wood biochar following NH 3 exposure being plant-available. This increase in plant-available N versus total N following NH 3 exposure was 21-fold greater for manure than wood biochar (Table 1). Furthermore, the increase in total N in manure biochar exposed to NH 3 (0.06%-points), was smaller than the increase in plant-available N (0.34%-points) ( Table 1). Exposure to NH 3 therefore increased the plant-available N in both biochars to the same extent, apparently, irrespective of the total N increase.
The change in total N in manure and wood biochars following immersion in the slurry was similar to that through NH 3 exposure, compared to unexposed biochars. However, in both biochars, a much smaller portion of this added N was immediately plant-available after immersion in slurry, 0.0-0.01 g N kg −1 , compared to after Table 1. Total carbon and nitrogen and KCl-extractable ammonium (NH 4 + -N) and nitrate (NO 3 -N) in amendments used for the greenhouse trial ± the standard deviation. ND not determined, as KCl extraction of slurry is not an appropriate measure plant-available N. Letters indicate significant differences between amendments using a one-way anova (p < 0.05; n = 3).  (Table 1). Thus, sorbed N on biochars following NH 3 exposure is more plant-available than sorbed N from the slurry. Manure biochar contained significantly greater total nutrients (acid-digestible) and plant-available (Mehlich III extractable) nutrients by mass than wood biochar, specifically Ca, Mg, and P (Tables 2, 3). This corroborates with previous reports of large amounts of ash minerals in biochars from manure feedstocks compared to woody feedstocks 23,36 . Calcium (Ca) comprised more than 15% of manure biochar mass and less than 1% of wood biochar mass. The high Ca concentration in manure biochar resulted from the regular liming of fresh manure solids after screw-pressing. Significantly greater total micronutrients (B, Cu, Fe, Mn, Zn) were observed in manure biochar compared to wood biochar, 3.14 vs. 0.80 g kg −1 , respectively. This trend was reversed for extractable elements: wood biochar contained 2.5-fold greater Mehlich III-extractable micronutrients (358.3 mg g −1 ) than manure biochar (146.1 mg g −1 ).
The total and Mehlich-III-extractable and therefore plant-available concentrations of heavy metals in both wood and manure biochars were below the EPA threshold values for biosolids intended for agriculture 40,48 . Manure biochar contained significantly greater total concentrations of heavy metals (Cd, Pb) than wood biochar, 15.3 vs. 3.0 mg kg −1 ( Table 2). However, Mehlich III-extractable heavy metals were 12-fold greater in wood biochar than manure biochar, reaching 9.8 mg g −1 vs. 0.84 mg g −1 , respectively ( Table 3).
Additions of wood biochar alone increased plant growth (dry weight of above and below-ground biomass) from 4.9 to 29% relative to no additions. Radish plants benefited from additions of manure biochar alone, with 18% greater plant growth than no additions. When urea alone (1×) was added to potting media, plant growth increased by 9-34% relative to unamended plants. Additions of manure or wood biochar together with urea (1×) increased plant growth by 14-63% relative to unamended plants ( Fig. 1, Table 4).
Different forms of N added with biochars had different effects on plant growth. Wood biochar applied together with urea (1×) promoted 0-18% greater plant growth than wood biochar treated with NH 3 despite the fact that the total N content in both types of amendments was identical, 1.88%. Furthermore, wood biochar and urea promoted 29-37% more plant growth compared to slurry-immersed biochar despite the similar total N values between the two types of amendments, 1.88% vs 1.84% (Tables 1, 4).
Differences in plant growth with the type of biochar and added N were also apparent. Overall, plant growth was 6-34% lower after adding manure biochar exposed to NH 3 compared to adding wood biochar exposed to NH 3 . These effects of adding manure biochar exposed to NH 3 on plant growth varied between plant type. In a one-way anova of the effect of amendment type on plant growth, between plant types, we observed that Table 2. Total nutrients measured in acid-digested (HClO 4 + HNO 3 ) amendments used for the greenhouse trial ± the standard deviation. Letters indicate significant differences between amendments using a one-way anova (p < 0.05; n = 3).  Table 3. Plant-available nutrients in amendments, extracted with Mehlich III ± the standard deviation. Letters indicate significant differences between amendments using a one-way anova (p < 0.05; n = 3). www.nature.com/scientificreports/ supplementing manure biochar with N by exposing it to NH 3 significantly improved tomato growth above tomatoes grown with only manure biochar, by 35%. Supplementing manure biochar with urea fertilizer significantly improved radish growth by 38% compared to additions of manure biochar alone ( Fig. 1, Table 4). Using a two-way anova of the effect of amendments and plant types on growth, plant growth in radishes amended with manure or wood biochar added with urea fertilizer and wood biochar treated with NH 3 were significantly higher compared to the growth of all plant types grown with manure biochar alone. Furthermore, no significant differences in plant growth were apparent between radish and tomato amended with manure or wood biochars treated with NH 3 or added with urea fertilizer, compared to adding urea alone, in any amount Table SI 5). Thus, supplementing wood or manure biochar with N from NH 3 exposure or urea fertilizer proved just as good for plants as adding urea fertilizer alone.

Analysis
Germination was affected by plant type and amendment, and was lower with greater nutrient additions for marigold and radish, and higher with the highest nutrient additions for tomato. Unamended and manure biocharamended marigold and radish had the highest germination rates, 80-90%, while tomato plants amended with the highest urea application rates, 1.5×, and wood biochar, reached 100% germination (Table SI 4).
After 40 days, the pH of pots amended with manure biochar (7.02-7.46) was significantly higher than the potting mix amended with urea (1×) or the unamended potting mix (6.11-6.48), based on both a one-way anova of the effect of amendment type on pH (Table 4) and a two-way anova of the effect of amendment type and plant type on pH (Table SI 5). The pH of pots with wood biochar immersed in the liquid manure (6.47-7.03) was highest among all wood biochar treatments (6.25-6.74). In contrast, wood biochar alone did not significantly increase the pH of the potting mix, relative to the unamended pots (Table 4, Table SI 5).
Plant N uptake increased with increasing N additions (Fig. 2). The range of urea additions encompassed N additions of all other amendments (Fig. 3) indicating positive growth responses across all rates of added N. Nitrogen from wood biochar exposed to NH 3 appears to be as plant available as urea 1×, since N uptake was slightly higher per unit N added (Fig. 3). Furthermore, the N uptake of plants grown with wood biochar treated with NH 3 was not significantly different than the N uptake of plants grown with urea 1×, based on either the The amount of plant-available N in each type of amendment in parentheses. Letters above the bars indicate significant differences between amendments within plant type (p < 0.05; n = 4; whiskers indicate standard errors). . Nitrogen from manure biochar exposed to NH 3 appears to be less plant available than urea, since N uptake was smaller per unit N added (Fig. 3), although these results varied between plant type. Radish plants grown with manure biochar treated with NH 3 had similar N uptake as when grown with the highest urea treatment. When added with urea, biochar did not increase N uptake, compared to N uptake with urea additions alone. However, adsorption of NH 3 onto biochars provided both immediately-available plant-available N and longer-term mineralizable N, allowing for greater plant uptake than equivalent amendments of urea. The initial KCl-extractable N in manure biochar exposed to NH 3 was comparable to urea 0.5× additions, indicating the mineralizability of N incorporated through NH 3 -exposure of manure biochar (Fig. 3). The difference in total N in wood biochar following NH 3 exposure was equal to the amount of N in urea 1× addition, yet the KClextractable N was comparable to the amount of N in urea 0.25×. Nevertheless, compared to urea 0.25×, plant N uptake in wood biochar sorbed with NH 3 was 12-26% greater. Moreover, in all experiments, NH 3 -exposed biochars outperformed biochars immersed in the liquid fraction, increasing N uptake by 29-55% for wood biochar and 25-75% for manure.

Discussion
The feedstock and pyrolysis conditions used to produce biochar for these experiments did not result in phototoxicity, as no difference in germination was detectable between plants with or without biochar. The moderately high temperatures used for pyrolysis, 500 °C may have reduced the phototoxicity reported from lower-temperature biochars 49 . The greater increase in KCl-extractable N than in total N in manure biochar exposed to NH 3 may be related to CO 2 exposure that increased extractability of the N already present in the biochar. Yet it is unclear whether CO 2 can solubilize trapped inorganic N species or cleave bonds of organic N. Either way, all N from NH 3 added to manure biochar was KCl-extractable. On the other hand, the increase in total N in wood biochar exposed to NH 3 was 19-fold greater than manure biochar exposed to NH 3 , even though both materials contained the same amount of KCl-extractable N. This means that the majority of N in wood biochar treated with NH 3 more strongly interacted with the biochar than the NH 3 -derived N in manure biochar, in alignment with previous studies 29,40 and was therefore not plant-available.  Table 4. The average ± standard deviation of the pH of potting mix and amendments after 40 days, total plant biomass (shoot and root combined) and relative increase in plant biomass relative to unamended plants (0×), total N plant uptake (shoot and root combined), and relative increase in biomass N uptake in comparison to unamended plants. Letters indicate significant differences from a one-way anova between amendments within plant type (p < 0.05; n = 4; results for a two-way anova include in supplementary online material www.nature.com/scientificreports/ Capturing volatile NH 3 from the liquid fraction of dairy waste onto sorbents such as biochars was more successful for recycling nutrients for use as potting media than immersing sorbents in the liquid fraction. Plant N uptake was 25-75% and 29-55% greater in manure and wood biochars, respectively, for the NH 3 treatment compared to slurry immersion. Lower plant growth and N uptake with biochar treated with the slurry could be attributed to a combination of both the lower amount of readily plant-available N (i.e., KCl-extractable) and lower rates of mineralization of organic N. The plant-availability of organic N in the liquid fraction is time-dependent and varies with manure age and processing. Pettygrove et al. 50 determined that 27-44% of total N in fresh lagoon slurry was mineralized after 6 weeks when incubated in a sandy loam soil. The low pH of our peat potting media may have slowed microbial mineralization rates of slurry N compared to rates reported for mineral soils 51 .
Delivering N from NH 3 -enriched manure biochar to plants was just as effective as adding urea with manure biochar. The increase in plant N uptake relative to the control was 43-55% vs. 47-52%, respectively. The high plant-availability of N in manure biochar doped with N may indicate that an NH 4 + salt such as NH 4 HCO 3 formed from repeated exposure to NH 3 and CO 2 , preserving all sorbed N in a plant-available form, as observed with CO 2 captured within NH 3 solvents 52-54 . Mineral forms of N were detected in human manure biochars and wood biochars exposed to repeated intervals of NH 3 and CO 2 using XPS, although two-fold more electrostaticallysorbed NH 4 + was detected in NH 3 -and CO 2 -exposed wood biochar compared to NH 3 -and CO 2 -exposed manure biochar 40 .
Using median international prices for fertilizer nutrients, the value of extractable N in biochars exposed to NH 3 is 1.7 USD Mg −1 for manure biochar and 1.5 USD Mg −1 for wood biochar 39 . The benefit of plant-extractable minerals (Mehlich-III extractable) in dairy manure biochar on plant growth was not evaluated experimentally, since equivalent extractable nutrients in manure biochar were added to all other treatments. Nevertheless, we can estimate the added monetary value of plant-extractable P and K in manure biochar, together with plantavailable N, as 15.4 USD Mg −1 for manure biochar exposed to NH 3 and 9.5 USD Mg −1 for wood biochar exposed to NH 3 using literature values 39 .
The increase in plant N uptake with additions of manure biochar exposed to NH 3 relative to unamended plants reached 20-87% or 7.1-25.7 mg N pot −1 , which, accounting for the amount of biochar added to each pot, www.nature.com/scientificreports/ is equivalent to 1-3.6 kg N Mg −1 manure biochar. If we account for more than 624,000 dairy cows in NY State 55 , each generating approximately 18.8 Mg (dry) manure year −156 , this amount of available N in NH 3 -exposed manure biochar scales to 11,232 Mg N year −1 or 6-21.5 million USD year −1 , based on the average price of N fertilizer (0.51 USD kg −139 ). With approximately 51,360 Mg N applied to grain corn in NY State in 2018 4,6 , separated dairy manure treated with NH 3 can offset 23-82% of N fertilizer needs while stabilizing both the solid and liquid fraction of manure for addressing both environmental pollution as well as recycling N to agriculture. A novel fertilizer has been developed from N-rich dairy waste which performs equally well compared to conventional urea fertilizer per unit applied N. Not only did NH 3 -sorbed wood biochar contain similar amounts of plant-available N as NH 3 -manure biochar, but promoted greater plant biomass growth and plant-N uptake than manure biochar or conventional urea fertilizer. Despite the greater NH 3 -derived N enrichment in wood than manure biochar, the precipitation of NH 3 -salts on manure biochar in comparison to chemisorption of NH 3 to wood, points at manure biochar being a more efficient approach to recycle manure N as a fertilizer.
This research demonstrates that it is possible to convert dairy manure solids into a biochar that can adsorb volatile NH 3 for use as a N fertilizer. Conversion of dairy manures with high water contents into a dry and N-rich soil amendment with N use efficiency by plants commensurate with urea fertilizer may provide life-cycle benefits to water quality and greenhouse gas emissions that should be studied in the future. Future studies should include scaling up CO 2 and NH 3 exposure of biochar. Multi-year field studies with crops such as corn should examine the long-term availability of N-doped biochars and potential differences in leaching and gaseous N losses. The feasibility of cost-effectively operating dairy manure pyrolysis as well as adsorption of N to biochars should be studied on farms and by small industry. Techno-economic studies should quantify the ability to optimize the production and distribution of such fertilizers at different scales and under different economic conditions.

Materials and methods
Enriching biochar amendments with nitrogen. We evaluated the effect of N-enriched dairy manure biochar and N-enriched wood biochar on plant growth. The first type of biochar was created from anaerobicallydigested dairy manure solids ('solid fraction'), screw-pressed at a dairy farm in upstate New York in April 2018. www.nature.com/scientificreports/ The solid fraction was charred in a modified muffle furnace with a rotating paddle at 500 °C for 30 min. The liquid fraction of the screw-pressed dairy manure was also collected, sieved with a 425 μm mesh sieve to remove solids, and stored at − 4 °C. The second type of biochar made from Douglas fir wood (Pseudotsuga menziesii) using high-temperature gasification, was provided by Green Tree Garden Supply (Ithaca, NY). Both biochars were sieved to below 2 mm particle size. Manure biochar and wood biochar were enriched with N through two methods: (1) repeated, sequential exposure to CO 2 and NH 3 , or (2) immersion in the sieved liquid fraction ('liquid fraction'), which contained a mixture of N species. The effect of these two N-enriched amendments on plant growth was compared to (3) separate additions of each biochar (manure or wood biochar) with urea fertilizer, (4) urea additions without biochar, (5) separate additions of each biochar without urea fertilizer, or (6) no additions of biochar or urea.
For the first N-enrichment method, 200 g of biochar were loaded in a 4-L Buechner funnel suspended upright inside a drying oven at 30 °C. The bottom of the funnel was connected to gas flow via silicone tubing, and the top covered with a lid and with parafilm. Manure biochar was first exposed to CO 2 gas (Instrument grade, Airgas, Ithaca, NY) for one hour. After one hour, CO 2 flow ceased, and NH 3 flow commenced for one hour. Ammonia gas was generated by pumping air at 4.72 × 10 -4 m 3 s −1 through a sealed Erlenmeyer flask containing 1 L of 2 M NH 4 OH (pH 12.43) kept on a hot plate at 30 °C. This process was repeated three times (manure biochar CO 2 + NH 3 ). To enrich wood biochar with N, we reversed the order of gas exposure, first NH 3 then CO 2 , also for three exposure intervals (wood biochar NH 3 + CO 2 ).
For the second N-enrichment method, biochars were immersed in the liquid fraction for 1.5 h at a ratio of 143 g:1 L. The biochar-slurry suspension was contained in a large glass beaker on a hot plate at 30 °C. After the immersion period, the suspension was sieved through a 425-μm mesh sieve to remove residual liquid. Biochars were dried at 80 °C for 2 days. We did not expose slurry-treated biochars to CO 2 .
Duplicate sets of each biochar-N mixture were homogenized and stored in sealed glass jars. The four biochar treatments evaluated were: (1) manure biochar treated with CO 2 + NH 3 ; (2) wood biochar treated with NH 3 + CO 2 ; (3) dairy manure biochar immersed in the liquid fraction; and (4) wood biochar immersed in the liquid fraction. To simplify notation, when describing biochars treated with CO 2 + NH 3 (manure) or NH 3 + CO 2 (wood), we will refer to NH 3 exposure without referring to CO 2 , as NH 3 uptake was our focus.
Chemical analysis of amendments. Subsamples of amendments were milled and analyzed for total C and N by dry combustion (Elementar; vario EL cube, Langenselbold, Germany). Non-milled amendment subsamples were extracted with 2 M KCl at a ratio of 0.1 g biochar mL −1 KCl and tested for NH 4 + and NO 3 − through a colorimetric method on an auto-flow analyzer (AA3 HR AutoAnalyzer, Seal Analytical, Mequon, WI).
Non-milled amendment samples were also analyzed for plant-available and total elements. Plant-available nutrients were extracted using a Mehlich III solution at a ratio of 0.1 g mL −1 . Total elemental analysis was conducted on 0.5 g of unmilled biochar spiked with 0.25 mg L −1 yttrium as an internal standard. Samples were dissolved in a mixture of 30% perchloric acid in nitric acid (70%) at 180 °C. Both Mehlich III extracts and digestate solutions were analyzed by inductively-coupled plasma optical emission spectroscopy (ICP-OES; Spectro Arcos, Ametek Materials Analysis, Kleve, Germany).
Greenhouse trial to evaluate amendments. A six-week growth trial was conducted to test amendment performance. Three types of plants, either marigold, radish, and tomato, were grown in a peat potting mix (TH6, Theriault and Hachey Peat Moss Ltd., Baie Sainte-Anne, New Brunswick Canada) in square pots 0.3 L by volume to which biochars were added. Seeds were obtained from commercial companies (radish and tomato from Burpee and marigold from Park Seed) and its use complies with relevant institutional, national, and international guidelines and legislation. Wood biochar amendments were added at 10% bulk volume of the square pots, while manure biochar amendments were added at an equivalent C amount as wood biochar. This resulted in unequal mass and nutrients additions (other than C) between wood biochar and manure biochar. To correct for the increased non-N nutrient addition from manure biochar a mixture of dry nutrients was added to all non-manure biochar treatments based on the Mehlich III extractable nutrient content of manure biochar (Tables SI 1, Table SI 2).
We tested the effect of urea additions to compare the effect of N source on plant growth (Table SI 2). Nitrogen sources included biochars enriched in N through exposure to NH 3 , biochars enriched in N through immersion into the liquid fraction, and urea fertilizer additions. Urea additions were adjusted to equal the increase in total N on wood biochar after sequential exposure to NH 3 and CO 2 (Table SI 3), 4.9 g N kg biochar −1 . We did not add slurry alone, as we focused on adding nutrients and fertilizers as dry materials.
For each plant type, five seeds were planted in each pot after filling with the respective media and nutrient additions. Seedlings were thinned to a single seedling after two weeks, and the germination rate recorded for each pot. Pots in which no seeds germinated received seedlings from replicate pots of the same treatment in which more than one seed germinated.
The N equivalency of the N added with the biochar was quantified in comparison to an N-response curve measured from plants grown with urea fertilizer but without biochar (Urea 99% reagent grade, Sigma Aldrich). Five urea application rates were tested based on the increase in total N on wood biochar + NH 3 (Table 1, Table  SI 3 www.nature.com/scientificreports/ sod history, legume mixture or manure additions), which is highly variable with soil type and with cropping history 57 . As mentioned above, we also tested the plant effect of each biochar added with urea (1×) (Table SI 3). All plants were irrigated daily with reverse osmosis water to 90% of field capacity, determined gravimetrically. The field capacity of TH6 media amended with wood biochar or manure biochar was calculated as the amount of water remaining in a 0.3 L PVC cylinder filled with soil after saturation and draining overnight. Irrigation was lowered from field capacity by 10% to prevent leaching during the experiment.
To overcome the hydrophobicity of the potting mixtures, pots were initially placed in trays of water to moisten them from the bottom-up. From the second day of the experiment until day 12, pots were misted from the top daily. After germination on day 12, pots were weighed to determine the amount of water needed to reach 90% of field capacity. A description of the thirteen potting mixtures and irrigation amounts is provided in the supplementary information (Table SI 2).
Plants were harvested after 40 days, and wet shoot and root biomass recorded. Shoots were cut at the soil surface, and roots were excavated from each pot. Roots were isolated via washing and sieving. Dry root and shoot biomass was determined after drying at 65 °C for 3 days. The potting mix was dried at 105 °C for 3 days. Total C and N contents in shoots and roots were determined by dry combustion. After the experiment, all potting mixtures were extracted with Mehlich III to determine changes in plant-available nutrients. The pH of potting mixtures was measured before and after the growth trial in 1 g potting mixture in 20 mL deionized water.
For each of three plant types (marigold = i 1 , tomato = i 2 , radish = i 3 ) and biomass type (roots = j 1 , shoots = j 2 ), the proportional increase in biomass and N uptake across four replicates (k = 1:4) of amended plants relative to the average of four replicates of unamended plants for the same plant and biomass type was calculated as follows: The N-fertilizer equivalency of amendments was calculated based on the difference in N uptake of plants grown with urea alone (1×) and plants grown with biochar amendments. Using market prices for mineral fertilizers 39 , we also calculated the replacement value of the plant-available N, P, and K in amendments.
Statistics. Data analysis was carried out in RStudio 58 and graphs created using ggplot2 59 . Least squares of treatment means (LS means) were calculated using emmeans 60 . Order-independent p values determined with the Student-t test were adjusted using Tukey's method for comparing a family of five estimates at the α = 0.05 threshold. Compact letter displays of pairwise comparisons for a significance level of p < 0.05 were created using mult-compView 61 . Type I analysis of variance (ANOVA) was calculated using an order-dependent F test within the emmeans package. We present a one-way anova of the effect of amendment types on plant growth (dry weight) and plant N uptake in the main manuscript, and include both one-way anova and two-way anova evaluating the effect of amendment type and plant type, and the interaction between them, on plant growth (dry weight) and N uptake in the supplementary material. All mention of 'significant differences' refers to the probability of observing an F ratio greater than 0.05 given the null hypothesis, Pr(> F), or a p value < 0.05.
Proportion of biomass N uptake increase i,j,k = biomass N uptake amended plants i.j.k − biomass N uptake unamended plants i,j,k biomass N uptake unamended plants i,j × 100 (%)