Experimental site.

Field trials were conducted during Spring 2023 and 2024 at the University of Georgia Hort Hill Farm in Tifton, GA, USA (lat. 31°28′14.96″N, long. 83°31′53.11″W). The soil at the study site is classified as a sandy loam, with an average composition of 82% sand, 6% silt, and 12% clay at the depth of 0 to 15 cm. To establish baseline conditions, composite preplant soil samples (0–6 inches) were collected in 2023 from across the experimental field by combining 50 individual cores. Samples were analyzed using the Mehlich 1 extraction method at Waters Agricultural Laboratory (Camilla, GA, USA). The initial nutrient profile included P at 109.8 mg·kg⁻¹, K at 174.3 mg·kg⁻¹, magnesium (Mg) at 81.5 mg·kg⁻¹, and calcium (Ca) at 600.2 mg·kg⁻¹. The soil pH was 5.8, OM content was 0.64%, and CEC was 3.25 meq/100 g. Environmental conditions were relatively stable in terms of temperature, with May and June averages ranging from 27.4 to 32.4 °C across both years (Fig. 1A). In contrast, rainfall showed substantial variability: early-season precipitation (March–May) was considerably higher in 2024 (164.6–194.6 mm/month) compared with that in 2023 (73.2–88.1 mm/month), while June rainfall was lower in 2024 (45.2 mm) than in 2023 (183.6 mm) (Fig. 1B).

View Figure

• Download as Powerpoint
• Download Figure

Treatments.

The experiment included 10 treatments combining biochar at five application rates (0, 11.2, 22.4, 33.6, and 44.8 Mg·ha⁻¹) with two fertilizer types, conventional (granular and liquid) or poultry litter, in a 5 × 2 factorial design. Biochar was manually spread as a one-time application on 3 Mar 2023. Treatments were confined to the same parcel of soil across two consecutive years. In both years, fertilizer applications followed the University of Georgia Extension’s recommended N rate for tomato production, targeting 252 kg of available N·ha⁻¹ (Kissel and Sonon 2008). The trial had a randomized complete block design with four replications per treatment combination. Each plot had an area of 16.7 m² (1.83 m × 9.14 m). A 1.52-m alley was maintained between replicates to minimize soil movement and plot contamination. The biochar used in the study was sourced from Wakefield Biochar (Valdosta, GA, USA). Wakefield produces biochar from pine wood chips through pyrolysis at 600 °C. According to manufacturer data certified by the International Biochar Initiative Laboratory (Columbia, MO, USA), the biochar had a bulk density of 170 kg·m⁻³, pH of 8.84, organic carbon content of 21.1% (dry basis), hydrogen-to-carbon molar ratio of 0.18, total ash content of 57.0%, total N of 0.12%, electrical conductivity (EC) of 0.152 dS·m⁻¹, and a liming equivalent of 5.5% as CaCO₃.

The poultry litter used in this study was sourced from a broiler farm in Berrien County, GA, USA (lat. 31°18′45.51″N, long. 83°23′2.53″W) and consisted of uncomposted manure mixed with pine shavings as bedding material. Upon collection, litter samples were sent to Waters Agricultural Laboratory (Camilla, GA, USA) for nutrient analysis. Total N was measured using a combustion analyzer (LECO Corp., St. Joseph, MI, USA), and elemental concentrations were determined by inductively coupled plasma–optical emission spectrometry (iCAP series; Thermo Fisher Scientific, Madison, WI, USA) following open-vessel wet digestion. Application rates were calculated to supply 252 kg N·ha⁻¹. This N rate reflects recommendations for tomato production in Georgia and was applied equally across both fertilizer sources to ensure consistent N input among treatments. In Georgia, vegetable growers typically do not account for residual N from previous applications because of the high leaching potential and rapid mineralization. Therefore, residual N from 2023 was not factored into 2024 calculations. In 2023, poultry litter was applied at 2.21 Mg·ha⁻¹ and had an analysis of 37.3N–16.8P–51.0K–23.6Ca, with a pH of 8.27. In 2024, the application rate was 2.43 Mg·ha⁻¹, and the litter had an analysis of 33.9 N–24.4 P–59.0 K–41.2 Ca, with a pH of 7.52. Poultry litter was incorporated into the soil 21 d before transplanting in 2023 and 30 d before transplanting in 2024, following US Department of Agriculture guidelines requiring a 90-d preharvest interval for raw manure applied to aboveground crops (US Department of Agriculture 2025b). The inorganic fertilizer treatment included a preplant application of 56 kg N·ha⁻¹ using a granular fertilizer (10.0N–4.3P–8.3K; Rainbow Fertilizer LLC, Americus, GA, USA). Preplant applications for both poultry litter and inorganic fertilizer were performed on the same day (3 Mar 2023 and 4 Mar 2024, respectively). Raised beds were formed immediately after incorporating preplant poultry litter and granular fertilizer into the soil at a depth of approximately 15 cm. Drip irrigation tape (Typhoon™ Plus; Netafim, Fresno, CA, USA) and black totally impermeable film plastic mulch (Guardian TIF 1116; DNM Ag Supply, Inc., Calabasas, CA, USA) were installed using a tractor-mounted single-row mulch layer (RMC-172; Reddick Equipment, Williamston, NC, USA). This process was repeated in 2024, with mulch and irrigation systems removed on 28 Feb and reinstalled on 6 Mar following fertilizer application. For the inorganic fertilizer treatment, the remaining 196 kg N·ha⁻¹ was delivered via fertigation starting 1 week after transplanting. A liquid fertilizer (7.0N–0P–5.8K; Nutrien Ag Solutions, Tifton, GA, USA) was applied weekly and evenly split into 10 fertigation events at 19.6 kg N·ha⁻¹ per week. The fertilizer formulation contained a blend of N sources, including 3.17% nitrate-N, 1.25% ammonium-N, and 2.58% urea-N. While soil fumigation is a common practice in plasticulture systems, no fumigation was applied to the field in either 2023 or 2024. Following the final tomato harvest on 2023, the beds were left fallow and unirrigated until plastic mulch and drip tape were reinstalled the following spring.

Planting.

Tomato (Solanum lycopersicum) cultivar Summerhaven F1 (Seedway, Hall, NY, USA) was used for the trial. Seedlings were grown at Lewis Taylor Farms, a commercial transplant greenhouse nursery located in Tifton, GA, USA (lat. 31°26′49.32″N, long. 83°36′52.52″W), where seeds were sown on 17 Feb 2023 and 26 Feb 2024. After 6 weeks of greenhouse growth, seedlings were transplanted into the field on 24 Mar 2023 and 2 Apr 2024. Transplants were arranged in single rows on raised beds spaced 1.83 m apart, with plants spaced 45.7 cm within the row, resulting in a planting density of approximately 11,960 plants/ha. Each plot consisted of 10 plants. A conventional fresh-market stake-and-weave trellis system was used. Herbicide, fungicide, and insecticide applications were made according to standard University of Georgia recommendations (Horton et al. 2014).

Data collection.

Soil sampling followed the protocols recommended by Robertson et al. (1999). Baseline samples were collected before treatment application (day 0), followed by periodic sampling throughout the growing season. During each sampling event, eight soil cores were collected per plot using a manual soil probe at a depth of 0 to 15 cm to target the root zone at the center of each plot. Soil samples were placed into labeled paper bags corresponding to each plot. Samples were air-dried at room temperature for 1 week and subsequently sieved through a 2 mm mesh to remove roots, rocks, and other debris. The sieved soil was transferred into new labeled paper bags to prevent contamination and maintain sample integrity until analysis. To assess N retention, samples for nitrate (NO₃⁻) and ammonium (NH₄⁺) analyses were collected biweekly, air-dried, sieved, and stored in labeled paper bags using the same protocol. Additional samples for chemical and biological analyses were collected every 30 d.

Laboratory analysis.

Air-dried and sieved soil samples were submitted to Waters Agricultural Laboratory (Camilla, GA, USA) for soil health testing. Soil health encompasses a range of physical, chemical, and biological properties that influence soil’s ability to sustain plant productivity and ecological function (Lehmann et al. 2020; Nieder et al. 2018).

Chemical properties.

Soil pH was measured using a 1:1 soil-to-water ratio method (Eckert 1988; McLean 1982) with a LabFit pH analyzer equipped with Orion 815600 ROSS electrodes. Air-dried sieved soil (20 g) was mixed with 20 mL of deionized water, stirred for 5 s, and equilibrated for 10 min before measurement. The CEC was calculated as the sum of exchangeable H⁺, K⁺, Mg²⁺, and Ca²⁺ expressed in milliequivalents per 100 g of soil (meq/100 g). Soil N (NO₃⁻ and NH₄⁺) was determined using the KCl-cadmium reduction method. Samples were extracted with 2 M KCl, NO₃⁻ was reduced to NO₂⁻ via cadmium column, and concentrations were quantified via a flow injection analysis.

Biological properties.

Microbial activity was evaluated using the Solvita CO₂ Burst Method. Air-dried soil (30 g) was moistened with 9 mL deionized water, sealed in incubation jars with low-CO₂ Solvita probes, and incubated at 20 °C for 24 h. Evolved CO₂ was quantified using a Solvita Digital Color Reader. Active carbon, representing microbially available carbon, was analyzed via the permanganate oxidizable carbon method (Weil et al. 2003). Soil (2.5 g) was reacted with 0.2 M KMnO₄, shaken for 2 min, settled for 10 min, and absorbance at 550 nm was measured via ultraviolet-Vis spectrophotometry. Soil OM, a key indicator of biological substrate, was estimated by loss on ignition at 350 °C (Schulte and Hopkins 1996; Storer 1984) as follows: air-dried soil (2.5 g) was dried at 80 °C for 2 h, ignited in a muffle furnace for 1.25 h, and mass loss was recorded.

Data analysis.

We calculated TIN as the sum of NO₃⁻ and NH₄⁺ concentrations measured at each sampling event. To evaluate N retention over time, annual mean TIN values were derived by averaging TIN concentrations across all sampling periods within a year for each plot, replication, and treatment combination. This approach preserved the experimental replication structure and accounted for seasonal variability in N dynamics. The difference in annual mean TIN between 2023 and 2024 was calculated for each plot, replication, and treatment to assess year-to-year changes in N retention.

Statistical analyses were conducted using RStudio (Posit Team 2025). A linear mixed-effects model was used for the analysis of variance with the “lme4” package (Bates et al. 2015). The model included biochar application rate (0, 11.2, 22.4, 33.6, and 44.8 Mg·ha⁻¹), fertilizer type (inorganic or organic), and year (2023 and 2024) as fixed effects. Replication (four blocks) and plots were included as nested random effects to account for block-level variation and within-plot variability. When significant treatment effects (P < 0.05) were detected, Tukey’s honest significant difference test was used for mean separation at the 95% confidence level. Residual diagnostics were conducted to validate model assumptions. The Shapiro–Wilk test assessed normality, Levene’s test evaluated homogeneity of variance, and residual plots were visually inspected for independence. When significant interactions were present, only the interaction effects are presented and discussed; main effects were not interpreted in the presence of significant interactions. Estimated marginal means (EMMeans) were calculated using the “emmeans” package (Lenth 2016), and pairwise comparisons were adjusted using the Sidak method.

Effects of the interaction between biochar × year on pH.

The interaction between biochar application rate × year significantly affected soil pH (P = 0.0291; Fig. 2). In both years, soil pH increased with increasing biochar rates, but the magnitude of the increase varied. In 2023, pH increased from 5.54 at 0 Mg·ha⁻¹ to 6.61 at 44.8 Mg·ha⁻¹ (a difference of 1.07 pH units). In 2024, pH increased from 5.54 to 6.27 (an increase of 0.73 units). At the highest biochar rate, pH was significantly higher in 2023 than in 2024. No significant year-to-year differences were observed at lower biochar rates (0–33.6 Mg·ha⁻¹). Across both years, biochar rates ≥22.4 Mg·ha⁻¹ consistently resulted in the highest pH values.

View Figure

• Download as Powerpoint
• Download Figure

Effects of the interaction between fertilizer source × year on pH.

A significant interaction was observed between fertilizer source × year (P = 0.0001) (Table 1). Poultry litter consistently resulted in higher soil pH values than conventional fertilizer. In 2023, soil pH was 6.63 under poultry litter compared with 5.96 under conventional fertilizer (a difference of 0.67 pH units). This trend continued in 2024, with poultry litter producing a pH of 6.36 compared with 5.63 under conventional fertilizer (a difference of 0.73 units). Although soil pH under conventional fertilizer declined significantly from 2023 to 2024 (by 0.33 units), pH under poultry litter remained statistically stable between years.

Table 1. Interaction effect of fertilizer source × year on soil pH.

View Fullscreen

Effects on cation exchange capacity.

The CEC was significantly influenced by biochar rate, fertilizer source, and year; however, no significant interactions were observed among these factors. As shown in Fig. 3A, CEC increased with increasing biochar application rates. The highest value was recorded at 33.6 Mg·ha⁻¹ (5.15 meq/100 g), followed by 44.8 Mg·ha⁻¹ (4.91 meq/100 g). The 22.4 Mg·ha⁻¹ treatment resulted in a CEC of 4.57 meq/100 g, which was slightly higher than that of 11.2 Mg·ha⁻¹ (4.33 meq/100 g) and statistically similar to 44.8 Mg·ha⁻¹. The control (0 Mg·ha⁻¹) had the lowest CEC (3.82 meq/100 g). Compared with the control, the 33.6 Mg·ha⁻¹ rate increased CEC by 34.8%, while the 44.8 Mg·ha⁻¹ rate produced a 28.5% increase. The fertilizer source also had a significant effect on CEC (Fig. 3B). Poultry litter yielded a CEC of 5.22 meq/100 g, which was 34.1% higher than the CEC under conventional fertilizer (3.89 meq/100 g). Additionally, CEC was significantly greater in 2024 (4.69 meq/100 g) than in 2023 (4.43 meq/100 g), representing a 5.9% year-over-year increase (Fig. 3C).

View Figure

• Download as Powerpoint
• Download Figure

Effects of the interaction between biochar × fertilizer source on total inorganic nitrogen.

A significant interaction between biochar application rate × fertilizer source was observed for TIN levels (P = 0.0001) (Fig. 4). Under conventional fertilizer, TIN increased from 31.00 mg·kg⁻¹ at 0 Mg·ha⁻¹ to a peak of 46.82 mg·kg⁻¹ at 33.6 Mg·ha⁻¹ (a 51.1% increase) before declining to 32.18 mg·kg⁻¹ at the highest rate. In contrast, poultry litter treatments resulted in a more pronounced and consistent increase in TIN, increasing from 38.88 mg·kg⁻¹ at 0 Mg·ha⁻¹ to 95.12 mg·kg⁻¹ at 44.8 Mg·ha⁻¹, representing a 144.7% increase. Significantly higher TIN concentrations were observed in poultry litter treatments compared with conventional fertilizer at 22.4, 33.6, and 44.8 Mg·ha⁻¹, while no differences occurred at the lower application rates.

View Figure

• Download as Powerpoint
• Download Figure

Effects of the interactions between biochar rate × fertilizer source × year on CO₂ burst.

The three-way interaction between biochar application rate × fertilizer source × year significantly affected CO₂ burst (P = 0.0001) (Table 2). In 2023, microbial activity—as indicated by the CO₂ burst—was substantially higher than in 2024 across most treatments. Under poultry litter, biochar significantly increased CO₂ emissions at all application rates compared with the control (43.8 mg CO₂–C·kg⁻¹). The highest values were observed at 11.2 and 22.4 Mg·ha⁻¹, with 60.8 and 60.6 mg CO₂–C·kg⁻¹ representing increases of 38.8% and 38.3%, respectively. Under conventional fertilizer, biochar had a more moderate effect. The highest CO₂ burst occurred at 22.4 Mg·ha⁻¹ (46.6 mg CO₂–C·kg⁻¹), which was a 25.0% increase over the control (37.3 mg CO₂–C·kg⁻¹), although differences among rates were less pronounced. In 2024, overall microbial activity declined. Under poultry litter, the highest CO₂ burst was recorded at 33.6 Mg·ha⁻¹ (51.9 mg CO₂–C·kg⁻¹), which was a 46.5% increase over the control (35.4 mg CO₂–C·kg⁻¹). However, responses across other biochar rates were inconsistent. Under conventional fertilizer, CO₂ burst values ranged narrowly from 27.2 to 31.8 mg CO₂–C·kg⁻¹, with no significant differences among biochar rates, indicating limited microbial stimulation.

Table 2. Interaction effect of biochar rate × fertilizer source × year on CO₂.

View Fullscreen

Effects of the interaction between fertilizer source × year on active carbon.

The interaction between fertilizer source × year (P = 0.0001) had a significant effect on active carbon (Table 3). Under inorganic fertilizer, active carbon declined by 5.2%, from 370.0 mg·kg⁻¹ in 2023 to 350.8 mg·kg⁻¹ in 2024. In contrast, under organic fertilizer, active carbon increased by 5.6%, from 402.5 to 425.0 mg·kg⁻¹, over the same period. Organic fertilizer consistently produced higher active carbon levels than inorganic fertilizer, with an 8.8% increase in 2023 and a 21.1% increase in 2024.

Table 3. Interaction effect of fertilizer source × year on soil active carbon.

View Fullscreen

Effects of the interaction between biochar rate × year on organic matter.

A significant interaction between biochar application rate × year (P = 0.0262) was detected for OM (Table 4). In 2023, OM increased from 0.63% at 0 Mg·ha⁻¹ to a peak of 0.69% at 33.6 Mg·ha⁻¹, which was a 9.5% increase. In 2024, a similar trend was observed, with OM increasing from 0.61% at 0 Mg·ha⁻¹ to 0.70% at the same rate, corresponding to a 14.7% increase. At the highest biochar rate (44.8 Mg·ha⁻¹), OM was 7.9% higher in 2024 (0.68%) than in 2023 (0.63%). Despite these numerical increases, no statistically significant differences were observed among treatments in pairwise comparisons. This suggested that although biochar may promote gradual improvements in soil OM, the observed changes were not large enough within the study period to reach significance at the 95% confidence level.

Table 4. Interaction effect of biochar rate × year on organic matter.

View Fullscreen

Effects of the interaction between fertilizer source × year on organic matter.

A significant interaction between fertilizer source × year (P = 0.0003) was observed for OM (Table 5). Under organic fertilizer, OM increased from 0.67% in 2023 to 0.70% in 2024, representing a 4.5% increase over time. In contrast, OM under inorganic fertilizer remained relatively unchanged and declined only slightly, from 0.61% to 0.59%, during the same period. Organic fertilizer consistently resulted in higher OM than inorganic fertilizer in both years. The differences between fertilizer types were 9.8% in 2023 and 18.6% in 2024, indicating a widening gap in OM accumulation over time.

Table 5. Interaction effect of fertilizer source × year on soil organic matter.

View Fullscreen

Soil pH.

The consistent increase in soil pH with increasing biochar rates aligns with biochar’s inherent alkaline properties, which help neutralize acidic soils (Chintala et al. 2014). The biochar used in this study had a calcium carbonate (CaCO₃) liming equivalency of 5.50%, indicating a modest liming potential, further supporting its pH-buffering capacity (Berek and Hue 2016). Poultry litter amplified this effect, likely because of the mineralization of OM, which releases base cations and enhances pH stability (Cairo-Cairo et al. 2023; Feizi et al. 2017). Despite these buffering inputs, soil pH declined from 2023 to 2024 across all treatments. This reduction may be explained by cumulative acidification from N cycling—particularly nitrification, which converts ammonium (NH₄⁺) to nitrate (NO₃⁻) and releases hydrogen ions (H⁺)—contributing to the year-to-year decrease in pH (Strong et al. 1997; Zebarth et al. 2015). Elevated pH conditions can further stimulate nitrifying bacteria, accelerating nitrate production and associated acidity (Bergamasco et al. 2019; Kyveryga et al. 2004).

Cation exchange capacity.

Biochar significantly boosted the soil CEC. This improvement is attributed to the biochar’s high surface area and negative charge, which enhances its capacity to adsorb and retain cations (Hossain et al. 2020; Yuan et al. 2011). The addition of poultry litter further elevated the CEC, emphasizing the importance of organic inputs in providing additional exchange sites and improving soil structure (Lin et al. 2022). The year-over-year increase in CEC from 2023 to 2024 may be attributed to both the aging of biochar—through physical and chemical transformations that enhance its reactivity—and the cumulative effect of repeated poultry litter applications, which incrementally enriched the soil with OM and exchangeable cations (Dong et al. 2017a; Heitkötter and Marschner 2015; Masocha and Dikinya 2022). Together, these processes contributed to improved cation retention over time.

Total inorganic nitrogen.

The combination of biochar with poultry litter demonstrated a strong synergistic effect on TIN, especially at higher biochar rates. This effect is likely attributable to biochar’s capacity to adsorb and stabilize N released during poultry litter decomposition (Doydora et al. 2011; Oreoluwa et al. 2020). Poultry litter, as an N-rich organic amendment, undergoes sustained microbial mineralization, converting organic N into inorganic forms (Robinson and Sharpley 1995). Biochar may enhance this process by improving the microbial habitat and reducing N losses through leaching and volatilization (Sharpe et al. 2004; Shepherd and Bhogal 1998). At the highest biochar rate (44.8 Mg·ha⁻¹), the porous structure and high surface area of the biochar likely helped retain mineralized N in plant-available forms, contributing to the observed increase in TIN under organic fertilizer. In contrast, under inorganic fertilizer, the same high rate may have led to microbial immobilization or reduced mineralization, thus explaining the lower TIN levels (Nguyen et al. 2017; Yang et al. 2024).

CO₂ burst.

In this study, higher CO₂ burst values with biochar and poultry litter suggested enhanced microbial activity, likely because of the increased availability of carbon and nutrients from these amendments. Biochar porous structure provides a favorable habitat for microbial communities (Dai et al. 2021). However, poultry litter—a rich source of labile carbon, nutrients, and indigenous microbial populations—amplifies this effect by stimulating microbial activity and directly introducing microbes into the soil (Acosta‐Martínez and Harmel 2006). The decline in CO₂ burst observed in 2024 may be attributed, in part, to management-related disruptions in soil microbial activity, whereby the removal and reapplication of plastic mulch may have altered soil moisture dynamics and limited microbial access to water, potentially reducing microbial respiration (Schlüter et al. 2019; Siebielec et al. 2020). While Zhao et al. (2016) found that increased rainfall can stimulate microbial respiration, their findings were based on open-field systems. In contrast, this study was conducted under a plasticulture system, where impermeable plastic mulch restricts rainfall infiltration into the beds. As a result, the high precipitation in 2024 likely did not improve soil moisture within the root zone, thus helping to explain the observed reduction in microbial activity.

Active carbon.

The higher AC levels observed with poultry litter further highlight the importance of labile carbon inputs in driving microbial activity. Poultry litter contains complex organic compounds such as cellulose and lignin, which contribute to the labile carbon pool over time through microbial decomposition. Their gradual breakdown supports sustained microbial activity and enhances soil carbon cycling (Das et al. 2008; Mierzwa-Hersztek et al. 2018). In contrast, inorganic fertilizers lack organic carbon inputs, thus limiting their ability to sustain microbial-driven carbon cycling and resulting in lower AC levels (Dong et al. 2017b; Zhang et al. 2015). The interaction between fertilizer type × year underscores the dynamic nature of AC accumulation. Under organic fertilizer, AC increased over time, likely because of the continuous input of labile carbon and nutrients that support microbial activity. Conversely, the decline in AC under inorganic fertilizer suggested that the absence of organic carbon inputs limits microbial activity and carbon turnover over time.

Organic matter.

Poultry litter significantly increased soil OM levels compared with conventional fertilizer, highlighting the importance of organic amendments in improving soil carbon content. This effect can be attributed to the continuous input of organic carbon and nutrients from poultry litter, which fuel microbial activity and contribute to OM formation (Jarosz et al. 2022). As poultry litter decomposes, it supports the development of stable soil aggregates, which physically protect OM from microbial degradation and promote long-term carbon storage (Li et al. 2021). In contrast, conventional fertilizers lack organic carbon inputs and, thus, do not contribute to OM accumulation (Zhang et al. 2021).