Stepwise Precision Fertilization To Quantify Nutrient Accumulation And Partitioning In Double-season Sweet Potato Systems

Summary

This protocol details a stepwise precision fertilization method synchronized with physiological growth stages to optimize nutrient supply in double-season sweet potato systems. It includes procedures for quantifying organ-specific nutrient accumulation, fitting nonlinear dynamic models to absorption trajectories, and calculating apparent nutrient balances to mitigate environmental risks.

Abstract

Double-season fresh sweet potato production often experiences a mismatch between soil nutrient supply and crop demand, leading to suboptimal yield and inefficient resource use. Traditional fertilization often relies on basal applications that fail to meet the crop's dynamic requirements under distinct spring and autumn meteorological conditions. This article presents a comprehensive protocol for a stepwise precision fertilization strategy that synchronizes nitrogen, phosphorus, and potassium inputs with critical phenological nodes. The method partitions total nutrient inputs into five distinct application events corresponding to specific growth stages: seedling establishment, vine expansion, tuber initiation, rapid bulking, and maturation. The protocol further describes a rigorous destructive sampling regime to dissect whole-plant nutrient partitioning. By employing logistic nonlinear regression models, researchers can quantify the maximum accumulation capacity, duration of the rapid uptake phase, and inflection points of nutrient uptake. This approach enables precise identification of critical absorption windows and evaluation of apparent nutrient balances. The application of this protocol facilitates the optimization of fertilization regimes, enhancing both storage root yield and commercial quality while reducing the risks associated with nutrient deficits or surpluses in intensive double-cropping systems.

Introduction

Sweet potato (Ipomoea batatas. L.) has evolved from a subsistence crop into a high-value functional food source and industrial raw material^1,2. In intensive agricultural regions, double-cropping systems—where two crops are harvested annually from the same land—are increasingly adopted to maximize land equivalent ratios and economic returns³. However, the sustainability of these high-intensity systems is frequently compromised by inefficient nutrient management strategies that overlook the temporal variability in crop nutrient demand and seasonal climatic variations^4,5.

The fundamental challenge in double-season cultivation lies in the distinct meteorological contexts of the spring and autumn growing seasons⁶. Spring crops develop under increasing temperatures and day length, whereas autumn crops face gradually declining thermal energy and solar radiation⁷. These environmental disparities significantly alter the phenological progression, biomass accumulation rates, and source-sink translocation efficiencies of the crop^5,8. Conventional fertilization practices, which typically involve a heavy basal application of compound fertilizers followed by irregular top-dressing, often result in a supply-demand mismatch². This mismatch manifests as either luxury consumption during the early vegetative phase, leading to excessive vine growth at the expense of tuber formation, or hidden hunger during the late bulking phase, where premature senescence limits final yield potential^2,9.

Nitrogen (N) management presents a specific challenge in sweet potato production due to its dual physiological effects². While essential for canopy establishment and photosynthetic capacity, excessive nitrogen availability prior to tuber initiation significantly suppresses storage root development, shifting the partitioning of assimilates toward foliage rather than underground storage organs^10,11,12. Conversely, nitrogen deficiency during late growth stages accelerates leaf senescence and shortens the effective photosynthetic period. Recent meta-analyses indicate that optimizing nitrogen application rates and timing is the single most influential factor in stabilizing yields across diverse soil types^6,13.

Phosphorus (P) and Potassium (K) dynamics present equally complex management requirements. Phosphorus is critical for root primordia initiation and energy transfer processes, yet it is often over-applied in baseline fertilization, leading to soil accumulation and potential environmental runoff^1,14. Potassium, often referred to as the quality element for root crops, is required in substantial quantities during the rapid storage root enlargement phase to facilitate the phloem loading and transport of sucrose from leaves to sink organs^15,16. A deficiency in potassium during this critical window compromises not only the total yield but also the commercial marketability of the tubers, leading to poor shape and reduced starch content^17,18.

To address these agronomic complexities, there is a pressing need for a standardized, quantifiable methodology that moves beyond empirical rule-of-thumb fertilization. Precision agriculture requires protocols that link specific management actions to physiological responses. Recent advancements in modeling nutrient uptake trajectories using nonlinear logistic functions have provided new insights into the temporal patterns of crop nutritional requirements¹⁹. By fitting these models to field data, researchers can mathematically determine the onset and cessation of rapid uptake phases, thereby identifying the precise windows where fertilization intervention yields the highest marginal return²⁰.

Specifically, while traditional split applications rely on calendar days, SPF determines fertilization events based on quantifiable physiological inflection points. By extracting parameters such as the maximum accumulation rate (R_max) and the inflection point time (t₀) from the logistic models, practitioners can pinpoint the exact biological window for intervention. This minimizes nutrient surplus and directly links temporal application strategies to enhanced commercial quality.

This article introduces a Stepwise Precision Fertilization (SPF) protocol designed to synchronize nutrient delivery with the physiological rhythm of double-season sweet potatoes. Unlike static fertilization, SPF employs a dynamic allocation strategy that distributes nutrients across five key growth stages. This method integrates rigorous sampling and analytical frameworks to quantify organ-specific nutrient partitioning and whole-plant accumulation trajectories. The objective is to provide a reproducible, scientifically robust workflow that enables researchers and advanced practitioners to optimize nutrient use efficiency, enhance commercial yield, and maintain soil fertility in intensive cropping systems^21,22.

Traditional fertilization models often estimate nutrient content using static linear assumptions or fixed calendar days, which differ fundamentally from our approach. Such conventional methods fail to account for the dynamic, non-linear biological rhythms of crop growth and nutrient utilization efficiency constraints²³ . Consequently, a significant research gap remains in precisely synchronizing nutrient supply with actual, time-varying physiological demands, particularly in double-season systems where spring and autumn present distinctly different meteorological and developmental challenges²⁴. Our Stepwise Precision Fertilization (SPF) protocol addresses this gap by integrating the logistic growth model to dynamically estimate nutrient accumulation. Unlike static methods, SPF uses quantifiable inflection points to track temporal shifts in nutrient levels. Based on this method, we provide a critical recommendation: practitioners should transition from arbitrary split applications to physiologically-driven interventions, dynamically adjusting nutrient weights at specific phenological nodes to optimize seasonal partitioning and minimize environmental surplus²⁵.

Specifically, while traditional split applications rely on calendar days, SPF determines fertilization events based on quantifiable physiological inflection points. By extracting parameters such as the maximum accumulation rate (R_max) and the inflection point time (t₀) from the logistic models, practitioners can pinpoint the exact biological window for intervention. This minimizes nutrient surplus and directly links temporal application strategies to enhanced commercial quality. By following this protocol, users can generate high-resolution datasets that elucidate the mechanistic links between fertilizer timing, plant physiological status, and final agronomic outcomes.

Protocol

1. Experimental site preparation and design

1. Select an experimental site with flat terrain, efficient drainage, and homogeneous soil fertility to minimize spatial heterogeneity.
2. Collect composite soil samples from the top 0–20 cm layer using an Edelman auger. Take samples from five random points within each plot and combine them into a single plastic bag prior to land preparation for each growing season (spring and autumn).
3. Air-dry the soil samples at room temperature, grind them mechanically, and pass them through a 2 mm stainless steel sieve to analyze baseline physicochemical properties, including pH, organic matter, total nitrogen, available phosphorus (Olsen-P), and available potassium (NH4OAc-extractable K).
4. Establish a randomized block design for the experiment with a minimum of four replicates per treatment. Establish four treatments: Conventional fertilization (CON), Stepwise Precision Fertilization (SPF), Nitrogen-Focused (NF), and Potassium-Boosted (KB). Maintain the exact same plot allocations for both spring and autumn seasons to account for potential legacy effects.
5. Demarcate individual plots with an area of approximately 25 m² (planting density of 40,000 plants per hectare), ensuring the inclusion of buffer rows (minimum 1 m width) between plots to prevent fertilizer cross-contamination and lateral root migration.
6. Exclude the outer two rows of each plot as border rows to minimize spatial bias. Designate the next 10 consecutive plants strictly as the Destructive Sampling Zone. Leave the remaining central core of the plot (minimum 20 plants) completely untouched as the Harvest Zone, ensuring that in-season sampling does not compromise the final plant population density or yield integrity.

2. Calculation and implementation of stepwise fertilization

1. Determine the total seasonal target dosage by calculating the difference between the nutrient requirement for a predefined regional target yield and the indigenous soil nutrient supply. For example, aiming for a target storage root yield of 35 t/ha, the absolute target values applied in this study were 150 kg·ha^⁻1 for N, 75 kg·ha^⁻1 for P₂O⁵, and 225 kg·ha^⁻1 for K₂O.
2. Split the total nutrient input into five distinct application events (m = 5) corresponding to the following phenological nodes: basal application (at transplanting); vine expansion phase (diagnosed when rapid internode elongation begins and seedling survival is stabilized); tuber initiation phase (diagnosed upon the first visible swelling of adventitious roots into white primordia); rapid bulking phase (diagnosed when canopy closure reaches ≥80% and primary roots exhibit rapid thickening); late bulking/maturation phase (diagnosed when basal leaves begin early physiological yellowing).
3. Calculate the specific dosage for each application event using the ratio formula
   F_x,t = F_x × R_t (1)
   Where F_x is the total seasonal amount of nutrient x, and R_t is the allocation proportion for the t^th event.
   NOTE: To provide a physiological rationale, early luxury N must be avoided to prevent tuberization suppression, while K must be sustained during filling to support starch translocation. The exact R_t proportions recommended for SPF across the five stages are: Basal (N: 20%, P: 50%, K: 10%), Vine Expansion (N: 20%, P: 20%, K: 10%), Tuber Initiation (N: 20%, P: 20%, K: 20%), Rapid Bulking (N: 30%, P: 10%, K: 40%), and Maturation (N: 10%, P: 0%, K: 20%).
4. Weigh the specific fertilizer amount for each plot individually using a precise electronic balance.
5. Apply the fertilizer.
   1. For furrow application: Open a shallow trench (10 cm depth) along the ridge side, distribute the fertilizer evenly, and cover immediately with soil.
   2. For drip fertigation systems: Dissolve the calculated fertilizer completely to a maximum concentration of 2 g/L to prevent emitter clogging. Inject the solution at a standardized volume of 150 L per plot. Run the system with pure water for 15 min before and after injection to flush the lines, avoiding salinization and ensuring uniform spatial distribution.
6. Record the exact date, meteorological conditions, and soil moisture status at the time of each application.

3. Destructive sampling and organ partitioning

1. Schedule eight sampling time points per season to capture the full growth trajectory: seedling survival, vine elongation, canopy closure, tuber initiation, early bulking, mid-bulking, late bulking, and pre-harvest.
2. Using a stratified random sampling method, select three plants strictly from within the 10-plant Destructive Sampling Zone, continuously bypassing the border rows.
3. Excavate the entire plant carefully, using a spade to loosen the soil around a 30 cm radius to preserve the fine root system and small developing tubers.
4. Wash the harvested plants with tap water to remove all adhering soil and debris.
5. Separate the plant immediately into four distinct organ fractions: leaves (including blades), petioles, stems (vines), and storage roots (tubers).
   NOTE: Separate fine fibrous roots from storage roots based on diameter (>5 mm for storage roots).
6. Weigh the fresh mass of each organ fraction immediately to avoid moisture loss.
7. Cut the stem and tuber samples into small (<1 cm) cubes to facilitate uniform drying; keep leaves and petioles intact but separate.
8. Place each organ sample into a labeled paper bag.

4. Dry matter determination and chemical analysis

1. Place the sample bags in a forced-air oven.
2. Deactivate enzymes by heating at 105 °C for 30 min.
3. Reduce the temperature to 75 °C and dry the samples until a constant weight is achieved (typically 48–72 h).
4. Weigh the dry samples to determine the dry matter accumulation (DMA) for each organ.
5. Grind the dried samples using a stainless steel mill and pass the powder through a 0.25 mm sieve.
6. Digest 0.2 g of the powdered samples using a 5:1 (v/v) mixture of concentrated sulfuric acid (H₂SO₄) and 30% hydrogen peroxide (H₂O₂) at 350 °C for nutrient extraction.
7. Analyze the digest for total nitrogen using the Kjeldahl method or an automated elemental analyzer; total phosphorus using the vanadium molybdate yellow colorimetric method; total potassium using flame photometry.
8. Calculate nutrient accumulation ($NA$, kg·ha^-1) using the formula:
   NA = (2)

5. Data analysis and model fitting

1. Calculate the Harvest Index (HI) and Root Partitioning Coefficient to quantify biomass allocation to economic organs.
2. Fit the whole-plant and storage root nutrient accumulation data to the Logistic growth equation using the non-linear least squares ('nls') function in R software:
   Y =  (3)
   Where Y is accumulation, A is the maximum theoretical accumulation, k is the growth rate coefficient, and t₀ is the inflection point time.
3. Derive secondary parameters programmatically: duration of the rapid phase (T_rapid) and the maximum rate (R_max) using the derived formulas: t₁₀ = t₀ – ; t₁₀ = t₀ + ; T_rapid = t₉₀ – t¹⁰; R_max =  
4. Calculate the Apparent Nutrient Balance (B) for each plot, where 'Total Crop Removal' is defined strictly as the total nutrient mass permanently exported from the field via the harvested commercial and non-commercial storage roots (excluding vines left in the field):
   B = Total input – Total crop removal    (4)
   NOTE: "Total Crop Removal" is strictly defined as the total nutrient mass permanently exported from the field via both commercial and non-commercial storage roots. Aboveground shoots left in the field as crop residue are excluded from the removal calculation.
5. Perform statistical analysis using a Linear Mixed-Effects Model (LMM) in R (e.g., 'lme4' package) to account for the two factors (treatment × season) and repeated measures over time (8 collections), defining 'Treatment' and 'Season' as fixed effects, and 'Plot' as a random effect.

Results

Baseline soil characteristics and seasonal variations

Table 1 presents the physicochemical properties of the topsoil (0–20 cm) determined prior to transplanting in both the spring and autumn growing seasons. The analysis reveals that fundamental soil characteristics, including pH (maintained within a weakly acidic range of 6.38–6.42), organic matter content, and bulk density, remained statistically stable between the two seasons. This stability confirms that t...

Discussion

The implementation of the Stepwise Precision Fertilization (SPF) protocol demonstrates that high-yield formation in double-season sweet potato systems relies heavily on the synchronization of nutrient delivery with crop physiological demand^26,27. The results confirm that the supply rhythm is as critical as the total dosage. By utilizing the stepwise protocol, we effectively mitigated the common trade-off between vegetative growth and reproductive storage, a pheno...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Analytical sieve (0.25 mm)	Retsch	AS200	Particle size standardization
Calcium superphosphate	Sinofert Holding Co., Ltd. (China)	≥12% P2O5	Phosphorus source
Compound fertilizer (N–P2O5–K2O)	Sinofert Holding Co., Ltd. (China)	Custom formulation	Used for stepwise precision fertilization
Digestion block system	Gerhardt	Kjeldatherm	Acid digestion of plant material
Drip irrigation and fertigation system	Netafim	Standard agricultural system	Optional fertigation treatment
Drying oven (forced-air)	Memmert	UF110	Sample drying at 75–105 °C
Electronic balance	Mettler Toledo	MS204S	Used for precise fertilizer weighing
Flame photometer	Sherwood Scientific	Model 420	Potassium determination
Kjeldahl nitrogen analyzer	Foss	Kjeltec 8400	Determination of total nitrogen
Paper sample bags	Whatman	Grade 1	Drying and storage of plant samples
Potassium sulfate	Sinofert Holding Co., Ltd. (China)	≥50% K2O	Potassium source for bulking stage
Soil auger	Eijkelkamp	Edelman auger	Soil sampling (0–20 cm depth)
Spade and hand tools	Local agricultural supplier	—	Destructive sampling of whole plants
Stainless steel plant grinder	Retsch	ZM200	Grinding dried plant samples
Statistical analysis software	R Foundation	R version 4.3.0	Logistic model fitting and ANOVA
Sweet potato planting material (Ipomoea batatas L.)	Local certified seed supplier	—	Uniform cultivar used for double-season field experiment
Urea (N fertilizer)	Sinofert Holding Co., Ltd. (China)	≥46% N	Nitrogen source for split application
UV–Vis spectrophotometer	Shimadzu	UV-1800	Phosphorus colorimetric analysis