The experimental site was located in Kamphaeng Saen district of Nakhon Pathom province, central Thailand (Latitude: 14.036028, Longitude: 99.958806). It covered approximately 0.8 ha and was a part of the experimental plots of Kasetsart University on the Kamphaeng Saen campus. The plot soil belonged to the Kamphaeng Saen (Ks) soil series, being classified as Typic Haplustalfs with a loam-textured soil. The sugarcane plot used for the experiment was a first ratoon crop of the Khon Kaen 3 variety of a Saccharum spp. hybrid, continuing the study reported by Radasai et al. (11). Figure 1 presents the environmental data including the rainfall and air temperature during the growing season of planted sugarcane (Feb 2022–Feb 2023) and of first ratoon sugarcane (Feb 2023–Jan 2024). The annual rainfall levels for the 2022–2023 and 2023–2024 growing seasons were 1,316 mm and 613 mm, respectively, and the mean temperatures were 28.3°C and 29.3°C, respectively.

Following the harvesting period of plant sugarcane (pre-experiment), composite topsoil samples were randomly collected from a depth of 0–15 cm to analyze various physical and chemical properties: particle size distribution, pH, electrical conductivity, cation exchange capacity, organic matter, total nitrogen (N), available phosphorus (P), extractable potassium (K), calcium (Ca), magnesium (Mg), sulfur (S), iron (Fe), manganese (Mn), zinc (Zn), copper (Cu), and silicon (Si) (12–15). Table 1 presents the properties of the soil prior to conducting the experiment. Additionally, soil samples were collected from each subplot after harvest to analyze the amount of extractable K in the soil post-experiment, using the same method as for the pre-experiment soil samples.

The experiment was designed as a 2 × 7 factorial in a randomized complete block design with three replications, totaling 14 treatment combinations across 42 subplots. The first factor was the application of soil K fertilizer (basal fertilizer), with K0 representing no application and K1 representing application. The second factor was the type of foliar-supplemented K fertilizer, consisting of seven solutions: F0-water (control); F1-KCl; F2-K₂SO₄; F3-K₂SiO₃; F4-KNO₃; F5-molasses; and F6-vinasse. The foliar chemical fertilizers used were of commercial fertilizer grade and applied at a concentration of 2.5% weight by volume (w/v), while the molasses and vinasse were obtained from the sugarcane-sugar-ethanol industry and applied with 5× dilution. Each subplot was 7.5 × 10 m² and consisted of five rows sugarcane, along with other planting details according to Radasai et al. (11). Basal chemical fertilizer was applied approximately 2 months after harvesting the plant crop at rates of 93.75 kg N ha^-1, 18.75 kg P₂O₅ ha^-1, and 75 kg K₂O ha^-1 for K1; however, no K fertilizer was applied for K0. These fertilizer rates were based on recommendations by Thai Department of Agriculture for ratoon sugarcane, using soil analysis data (16). N fertilizer was applied in two splits (at 2 and 4 months) and foliar fertilizer was sprayed when the ratoon sugarcane was aged about 5 months, during the grand or vegetative growth phase (120–240 days), according to Tayade et al. (17). The foliar fertilizer was sprayed uniformly over the subplots at a rate of 2,000 L ha^-1 (15 L per plot). An anionic, straight-chain leaf binder was used at a rate of 1 mL per 2 L of fertilizer solution in all treatments to enhance absorption and reduce any loss from rain wash-off. Table 2 lists the properties of the foliar-applied fertilizer solution and the amounts of nutrients provided to each treatment through foliar supplementation. Irrigation was provided using a furrow system solely to prevent water deficiency, while pest control measures were applied uniformly across all subplots.

The first ratoon sugarcane from 9 m² plots was harvested 11 months after the initial harvest, focusing on the three central rows, each 2 m in length. The evaluation included determining the fresh weight and the number of millable canes, with 10 representative millable canes being randomly selected from each plot to measure length and diameter. Sugarcane juice quality parameters—fiber (%), Brix (%), polarity (%) or sucrose concentration, commercial cane sugar (CCS, %), purity (%), sugar recovery (%), and sugar yield (Mg ha^-1), were analyzed and calculated as described by Radasai et al. (11).

Two representative millable canes were randomly collected from each plot and oven-dried at 70°C until a consistent dry weight was achieved. Then the dried samples were homogenized using grinding. The total N content was determined using the Kjeldahl method with wet oxidation (H₂SO₄-NaSO₄-Se), according to Bremner (18). The resulting solution was used to measure the total concentrations of P, K, and Si. Total S concentrations were measured using an acid digestion mixture (3:1 HNO₃:HClO₄), according to Kongtawee et al. (19). The concentrations of P and S were determined via UV-visible spectrophotometry, while the K and Si concentrations were analyzed using atomic absorption spectroscopy. Then, these concentration data were used to calculate the nutrient uptake of sugarcane stalks per unit area.

All collected data (cane length, cane diameter, fresh yields, juice quality, and nutrient uptake) were checked for normality using the Kurtosis and Skewness tests. All data were normally distributed, except for juice purity, which required data transformation using reflection and logarithmic methods prior to analysis. Analysis of variance with F-tests was conducted to statistically assess mean differences. Post hoc multiple comparisons were performed using Duncan’s test at a significance level of 0.05 (p < 0.05) to identify significant differences among treatments, with the IBM SPSS Statistics 27.

K use efficiency (KUE), defined as the mass of harvested products relative to the mass of K applied to the sugarcane, and agronomic efficiency (AE) of applied K were adapted from the method for N use efficiency described by Thorburn et al. (20). KUE was calculated based on the yields for each treatment (K rate applied through basal and foliar applications, expressed as K₂O ha^-1), and AE was calculated as the increase in yield per kg of K₂O applied through basal and foliar applications, as follows:

where Sugarcane yield_t and Sugarcane yield₀ refer to the sugarcane yields for each treatment and the control, respectively.

The growth of the first ratoon sugarcane variety Khon Kaen 3 in loamy soil, whether with or without soil-applied K combined with foliar K supplementation, did not significantly affect the growth parameters of the ratoon sugarcane (cane length, cane diameter, and the number of millable canes). Additionally, soil K application and foliar K supplementation did not significantly affect these growth parameters ( Table 3 ). However, the sugarcane yield was affected by the interaction between soil K levels and foliar K supplementation at p < 0.05. The combination of no soil-applied K with foliar K₂SiO₃ (K0×F3) resulted in the highest cane yield, which was not significantly different from those achieved with no soil-applied K combined with foliar KNO₃ (K0×F4) and with soil-applied K combined with foliar water (K1×F0). Furthermore, the absence of soil-applied K combined with foliar K supplementation using KCl (K0×F1), vinasse (K0×F6), and K₂SO₄ (K0×F2) resulted in ratoon sugarcane yields, which were not significantly different from those with soil-applied K and foliar water (K1×F0). The significantly lowest ratoon sugarcane yield was in the treatment without soil-applied K and with foliar water application (K0×F0), likely due to insufficient K supply. In the case of soil with sufficient K application, foliar K supplementation with K₂SO₄ (K1×F2), KNO₃ (K1×F4), and KCl (K1×F1) did not significantly affect ratoon sugarcane yield compared to foliar water (K1×F0). Conversely, foliar K supplementation with K₂SiO₃ (K1×F3), molasses (K1×F5), and vinasse (K1×F6) significantly reduced the ratoon sugarcane yield compared to foliar water (K1×F0), as shown in Table 3 .

The study results indicated that the soil-applied K fertilizer and foliar K supplementation in various forms had no effect on juice quality parameters (fiber, Brix, polarity, and CCS), nor on sugar quantity, as measured by sugar recovery. Additionally, there were no interaction effects between these two factors for any of these juice quality parameters and sugar recovery. However, soil-applied K fertilizer did significantly increase the purity of the sugarcane juice compared to the treatment without soil-applied K ( Table 4 ).

The interaction between soil K fertilization and foliar K supplementation impacted the sugar yield. Non-soil-applied K combined with foliar KNO₃ (K0×F4) produced the highest sugar yield. However, this was not significantly different from the combination treatments of non-soil-applied K with foliar K₂SiO₃ (K0×F3), soil-applied K with foliar water (K1×F0), and foliar KNO₃ (K1×F4), ( Table 4 ). With the non-soil-applied K, foliar K supplementation helped to maintain sugar yield levels, preventing any decrease. Based on the study results, non-soil-applied K combined with foliar K supplementation resulted in sugar yields that were not significantly different from treatments with soil-applied K combined with foliar water (K1×F0), except when foliar molasses was applied (K0×F5). On the other hand, the ratoon sugarcane that was treated using soil-applied K, foliar K supplementation with K₂SiO₃ (K1×F3), molasses (K1×F5), and vinasse (K1×F6) resulted in a significant reduction in the sugar yield, similar to the reduction in overall sugarcane yield ( Table 3 ).

The interaction between soil K application and foliar K supplementation influenced the uptake of N, P, K, and Si, but had no effect on the uptake of S in the cane stalk of ratoon sugarcane. In addition, based on the results, foliar K supplementation induced ratoon sugarcane without soil-applied K to take up N at levels comparable to those with soil-applied K. Notably, the treatment without soil-applied K combined with foliar water (K0×F0) had the lowest N uptake, which was significantly different from the other treatments ( Table 5 ). The combination of non-soil-applied K with foliar supplementation of KNO₃ (K0xF4) produced the significantly highest uptake of P in ratoon sugarcane. However, this value was not significantly different from those of the combinations of soil-applied K with foliar KNO₃ (K1×F4), non-soil-applied K with foliar supplementation of K₂SiO₃ (K0×F3), and K₂SO₄ (K0×F2) ( Table 5 ).

The combination of non-soil-applied K with foliar supplementation of KNO₃ (K0×F4) produced the significantly highest K uptake in the ratoon sugarcane stalk. However, this was not significantly different from the treatments of soil-applied K combined with foliar K₂SiO₃ (K1×F3), KCl (K1×F1), and KNO₃ (K1×F4), and non-soil-applied K combined with K₂SiO₃ (K0×F3) ( Table 5 ). The combination of soil-applied K with foliar supplementation of K₂SO₄ (K1×F2) produced in the significantly highest Si uptake in the cane stalk of ratoon sugarcane. This was higher than the treatments where molasses (F5) and water (F0) were sprayed, regardless of whether soil-applied K was used, as well as being higher than the treatment with soil-applied K combined with foliar KCl (K1×F1). There were no significant differences in S uptake across treatment combinations, although there was a trend toward higher S levels with the combination of soil-applied K and foliar supplementation of K₂SO₄ (K1×F2), likely due to the additional S supplied by the K₂SO₄ foliar application ( Table 5 ).

Based on the results, KNO₃ and K₂SiO₃ foliar supplementations were more effective among the various types of K foliar fertilizers. Foliar supplementation with 2.5% w/v of KNO₃ and K₂SiO₃ in ratoon sugarcane grown in soil with an extractable K level of 72.4 mg kg^-1 without additional soil-applied K, maintained ratoon sugarcane productivity—cane yield and sugar yield—at levels comparable to treatments with soil-applied K. When soil K levels are limited, foliar K supplementation can enhance K absorption and utilization within the plant, particularly during the grand growth period, which typically has higher nutrient demands. This effect was particularly notable in the uptake of nutrients such as N, P, and Si. Furthermore, foliar K application (whether or not soil-applied K was used) promoted K uptake in the cane stalk at levels equal to or higher than from soil-applied K alone. Foliar fertilization optimizes nutrient absorption, translocation, and assimilation within the plant, thereby increasing fertilization efficiency (21).

In soils where K fertilizer was applied, foliar K supplementation did not lead to further increases in cane yield and sugar yield compared to soil-applied K without foliar K supplementation (foliar water). In contrast, combining foliar K₂SiO₃, molasses, and vinasse with soil-applied K had a negative impact, resulting in lower cane and sugar yields than for soil-applied K without foliar supplementation. It was likely that these foliar K applications created some unfavorable conditions for the ratoon sugarcane, leading to a decrease in cane and sugar yields. The exact cause is still unknown; even considering the properties of the fertilizer solutions applied, as shown in Table 2 , there was no evidence that the pH and EC of these fertilizers affected sugarcane growth. The pH of these three fertilizer solutions was low (pH 4.37–4.81) but still higher than that of the K₂SO₄ solution (pH 2.87), while their EC range (13.25–17.34 dS m^-1) was lower than that of the other fertilizer solutions (26.56–38.59 dS m^-1). Additionally, they had a lower concentration of K than the other solutions but contained Si, which has been shown to have positive effects on sugarcane growth and yield (22–24).

The response of ratoon sugarcane that had received sufficient K through soil application differed from that of plant sugarcane, despite foliar application reducing the translocation distance of K from roots to leaves. These findings contrasted with Radasai et al. (11), who reported no negative effects of foliar K supplementation on sugarcane yield and sugar yield. In their study, foliar supplementation with K₂SiO₃ and KNO₃ significantly increased plant sugarcane yields. Additionally, in plant sugarcane, foliar K supplementation, especially with molasses, enhanced nutrient uptake. It was possible that foliar K supplementation had a more pronounced effect on plant sugarcane compared to ratoon sugarcane that had already received adequate soil-applied K. Alternatively, environmental factors may have influenced the results. For example, during the 2023–2024 growing season in the study, the rainfall was one-half of that recorded in the 2022–2023 season, while the average temperature was slightly higher ( Figure 1 ). This was consistent with a general decrease in the average sugarcane yield in Nakhon Pathom province, Thailand, which fell from 53.8 Mg ha^-1 in the 2022–2023 season to 50.6 Mg ha^-1 in the 2023–2024 season (25, 26). Robertson et al. (27) observed that water deficits during the tillering phase significantly impacted leaf area, tillering, and biomass accumulation, but had minimal effect on final yield. In contrast, they reported that water deficits during the canopy establishment phase had a much greater negative effect on final yield, including total biomass, stalk biomass, and stalk sucrose.

The KUE of first ratoon sugarcane revealed that withholding soil-applied K while applying foliar K supplementation resulted in a KUE range of 2,493–19,149 kg cane kg⁻¹ K₂O, significantly higher than soil-applied K alone (K1F0), which measured at 1,798 kg cane kg⁻¹ K₂O. Similarly, soil-applied K combined with foliar K supplementation produced a lower KUE (987–1,349 kg cane kg⁻¹ K₂O) than foliar K alone ( Figure 2 ). Among foliar K sources, K₂SiO₃ exhibited the highest KUE despite its low K content ( Table 2 ), as it generated substantial ratoon cane yields, followed by foliar vinasse and KNO₃. This suggests that K₂SiO₃ may enhance K efficiency by delivering Si, which could synergistically promote plant growth under limited K conditions. Likewise, AE values confirmed that foliar K application without soil-applied K produced consistently higher results than soil-applied K alone, with AE values ranging from 351 to 6,129 kg cane kg⁻¹ K₂O, compared to 446 kg cane kg⁻¹ K₂O for soil-applied K alone (K1F0). When soil-applied K was combined with foliar K, AE values dropped further to between 36 and 249 kg cane kg⁻¹ K₂O ( Figure 2 ). These findings suggest that, when soil K levels are limited, foliar K supplementation, especially with K₂SiO₃, can effectively enhance KUE and AE in first ratoon sugarcane. Traditional soil-applied K may be less effective in ratoon cane, possibly due to soil fixation or leaching losses that limit K availability. Foliar applications may bypass these issues by delivering K directly where it is needed for growth.

While foliar K application incurs additional costs due to increased labor requirements and the need for specialized spraying equipment, the same equipment used for pesticide application can be utilized if thoroughly cleaned beforehand. Foliar K application offers higher KUE by requiring less K fertilizer, which can lead to cost savings by reducing K inputs while maintaining yields. However, scaling up foliar K application presents practical limitations, including the need for appropriate equipment, favorable weather conditions to ensure effective application, and sufficient farmer training. Educating farmers is essential, as foliar application requires precise timing and technique to maximize its benefits.

Maintaining adequate plant nutrient levels in the soil is crucial for sustainable crop production. Although foliar supplementation effectively maintains productivity—both cane and sugar yields—without soil-applied K, relying solely on foliar K supplementation over the long term could potentially affect soil nutrient status. Analysis of soil available K after the experiments ( Table 6 ) shows that, although there were no significant differences between treatment combinations, treatments without soil K application generally exhibited lower available K levels compared to those with soil K application. This indicated a need for further research, particularly into continuous cropping systems relying solely on foliar K without soil K application. Continuous cropping can deplete soil nutrients, alter microbial communities, and change soil properties (28). Despite these concerns, using foliar KNO₃ and K₂SiO₃ in soils with insufficient available K can be a valuable strategy for farmers, especially when fertilizer prices are high. Foliar fertilizer application can reduce the total amounts of fertilizer applied while achieving high fertilizer efficiency (29). This approach offers a way to reduce K fertilizer usage or to manage soils with high K accumulation from long-term use, where foliar application provides an efficient means of nutrient delivery without further increasing soil K saturation. Foliar application of 2.5% w/v KNO₃ and K₂SiO₃ can serve as a practical and effective alternative to soil K fertilizers, providing a viable option for maintaining productivity and managing nutrient levels.

Comparing the productivity of plant sugarcane (from 11) with first ratoon sugarcane (from the current study) of the same variety and at the same site, it was observed that the fresh yield of first ratoon sugarcane decreased by 7.7–23.5%, with an average reduction of 14.0%. This reduction was consistent with the general trend of decreasing yields in ratoon sugarcane due to factors such as variety, nutrient management, and environmental conditions (5–7). Historical data from 35 experimental plots of the Khon Kaen 3 variety in Northeastern Thailand (1995–2007) showed average yields of 113.13 Mg ha^-1 for plant sugarcane and 103.13 Mg ha^-1 for ratoon sugarcane, an 8.8% reduction (30). Global comparisons show similar reductions: Indonesia with 3.5–24.8% (31); India with 27.4% for first ratoon and 31.4% for second ratoon (32) and 7.0–9.6% in calcareous soils (33); South Africa with 12.3–24.4% for the first-to-fifth ratoons (5); and Brazil with 0.9–11.1% (34).

The sugar yield of first ratoon sugarcane, with various foliar K supplementation treatments, decreased by 8.7% to 25.5%, averaging a 14.1% reduction compared to plant sugarcane. Typically, this decrease was due to the reduced fresh yield as the number of ratoon crops increased. Environmental factors may also play a role. For example, rainfall is crucial during the growth stage for higher-quality yields, and sucrose accumulation occurs post-primary growth (17). Historical data for the Khon Kaen 3 variety shows that sugar yield was 16.94 Mg CCS ha^-1 for plant sugarcane and 15.38 Mg CCS ha^-1 for ratoon sugarcane, indicating a 9.23% decrease (30). In the current study, based on the CCS calculations, the sugar yield decreased by 5.2% to 24.9%, averaging a 12.2% reduction. These reductions aligned with findings from other regions: India (6.7–9.2%, 33) and Indonesia (1.3–38.7%, 31). The variation in sugar yield reduction may also have been due to differences in sugar yield calculation, as some studies base their calculations on sugar recovery, CCS, or sucrose content.

A comparison of nutrient uptake between planted sugarcane (11) and the ratoon (current study) variety Khon Kaen 3 revealed significant reductions in N, P, K, S, and Si in the ratoon crop, averaging 57.0%, 53.6%, 60.7%, 37.1%, and 72.8%, respectively. These reductions were primarily due to the lower cane yield and reduced nutrient concentrations in the ratoon stalks, leading to a notable decrease in nutrient uptake per unit area. Singh et al. (35) suggested that ratooning induces rhizospheric changes, including pH alterations, soil enzymatic activities, and phenolic contents during the ratoon growth stage, which impact ratoon root properties. These changes result in significant impediment to nutrient acquisition and uptake, ultimately limiting growth and biomass production. Furthermore, field management practices such as irrigation, nutrient management, burning, and trashing, along with diseases, pests, increased competition between tillers, and subsequent tiller mortality, are associated with reduced ratoon cane yields (6). Additionally, nutrient uptake in the first ratoon cane in the current study was similar to that of planted sugarcane of the Khon Kaen 3 variety, as reported by Amonpon et al. (36) and Whangrattanacharoen et al. (37), grown in sandy soil and loamy sand soil, respectively. However, the nutrient uptake was lower than reported by Juntahum et al. (38), likely due to differences in environmental conditions, soil properties, and management practices. Compared to other regions, N and K uptake in the current study was typically lower, while P uptake was similar (33, 39, 40). These variations were likely influenced by differences in sugarcane varieties and environmental factors.