Differential Responses of Thai Fragrant Rice to Silicon Application Enhance Yield and Aroma Under Highland and Lowland Ecosystems

Abstract

Silicon (Si), a beneficial element accumulated by rice (Oryza sativa L.), enhances productivity and tolerance to biotic and abiotic stresses. Fragrance, primarily driven by 2-acetyl-1-pyrroline (2AP), is a key trait in premium rice markets. This study evaluated the effects of Si on grain yield, yield components, 2AP content, and Si accumulation in three Thai fragrant rice genotypes—BNM4, BNMCMU, and KDML105—under highland and lowland conditions. Plants received four Si application rates: 0 (control), 168, 336, and 504 kg Si ha⁻¹. Si significantly increased yield under lowland conditions, while responses in the highland were genotype-dependent, with only BNMCMU showing significant improvement at the highest Si rate. Silicon accumulation in shoot tissues was consistently higher in the highland than in the lowland across all genotypes. Nevertheless, Si application significantly increased shoot Si content under lowland conditions. A positive correlation between grain yield and shoot Si accumulation was observed under both environments, highlighting the role of Si in yield enhancement. The influence of Si on 2AP concentration was limited, with stronger effects from genotype and environment especially in the highland, where KDML105 consistently exhibited the highest 2AP levels. In the lowland, however, Si application significantly enhanced 2AP content in BNMCMU and KDML105. These findings underscore the significance of genotype × environment interaction and support precision Si application to enhance both yield and aroma in fragrant rice.

1. Introduction

Rice (Oryza sativa L.) is a staple food for more than half of the global population and plays a vital role in ensuring food and nutritional security. Among the diverse types of rice cultivated worldwide, fragrant rice varieties are particularly valued for their distinctive aroma and flavor. The compound 2-acetyl-1-pyrroline (2AP) is widely recognized as the principal contributor to the characteristic “popcorn-like” or “nutty” scent of fragrant rice [1,2,3,4]. Due to these unique sensory qualities, fragrant rice commands premium market prices and holds significant economic value in both domestic and international markets. Thailand, one of the world’s leading rice exporters, cultivates rice in both highland and lowland ecosystems, which differ substantially in altitude, soil characteristics, temperature, humidity, and other climatic factors. Lowland rice ecosystems, predominantly located in the Central Plain and parts of the Northeastern region of Thailand, are characterized by flat topography, fertile alluvial soils, relatively higher temperatures, and sufficient water availability through irrigation systems or seasonal rainfall. These environmental conditions are generally conducive to achieving high rice yields [5,6]. In contrast, highland rice ecosystems are mainly situated in the mountainous regions of Northern Thailand. These areas exhibit sloped terrains, well-drained but less fertile soils, lower nighttime temperatures, and limited water availability, often relying solely on rainfall during the growing season. Consequently, highland ecosystems face greater environmental constraints. Rice cultivation in highland areas typically involves traditional varieties, such as local aromatic or pigmented rice, which are highly valued for grain quality but generally produce lower yields [7]. These environmental differences can markedly influence rice growth, grain yield, and grain quality, including aroma [8,9,10,11]. These differences underscore the importance of promoting climate-resilient agriculture across diverse agroecosystems.

Silicon (Si) is widely recognized as a beneficial element for rice, a crop known for its high capacity to accumulate Si. Numerous studies have demonstrated that Si application can enhance plant tolerance to both biotic and abiotic stresses [12,13,14,15,16,17]. Elevated Si levels have been shown to mitigate damage from pathogens and insect pests, improve drought and salinity tolerance, enhance photosynthetic efficiency, and reduce lodging [16,18,19]. In rice, Si is absorbed at the roots and translocated via the xylem to aerial tissues, where it accumulates in the cell walls of leaves, husks, panicles, and other organs, providing both structural reinforcement and physiological benefits [20,21]. Rice absorbs silicon predominantly as monosilicic acid (H₄SiO₄) present in the soil solution. In this study, silicon dioxide (SiO₂) was used as the Si source. Under flooded paddy conditions, SiO₂ gradually dissolves to release H₄SiO₄, the form taken up by roots, with dissolution influenced by soil pH, temperature, and redox conditions. Although silicon is not classified as an essential nutrient, several research studies have shown that silicon enhances stress tolerance, improves nutrient use efficiency, and strengthens plant structures under various environmental stresses. For instance, silicon application can reduce the negative effects of drought, heat, and combined stress during the early grain-filling stage in both susceptible and tolerant wheat cultivars under controlled conditions [22]. Additionally, the application of silicon amendments has been found to improve the leaf chlorophyll index, agronomic efficiency, and nitrogen uptake in both maize and wheat, thereby promoting shoot and root development and ultimately increasing grain yield [23]. These effects align with the goals of eco-efficient nutrient management and low-input agronomic innovation to ensure sustainable intensification. Despite extensive research on the physiological and agronomic benefits of Si, its interaction with grain quality traits, particularly aroma, remains underexplored. Specifically, limited evidence exists on whether Si application can modulate the biosynthesis or accumulation of 2AP and how these effects vary across ecological zones such as highland and lowland environments. Moreover, most Si-related studies have primarily focused on yield enhancement and stress mitigation, with few addressing its potential influence on secondary metabolites, including volatile aroma compounds.

Three key research gaps remain. First, few systematic comparisons have been conducted on the effects of Si application in highland versus lowland environments, particularly concerning fragrant rice. Variations in elevation, temperature, and soil properties may significantly affect Si uptake, translocation, and tissue deposition, thereby influencing both grain yield and aroma expression. Second, genotype-specific responses to Si, especially regarding 2AP biosynthesis and Si accumulation, remain poorly understood. The interaction between genotype, environment, and management practices (G × E × M) is critical, yet comprehensive evaluations of Thai fragrant rice varieties under these conditions are scarce. Third, the partitioning of Si among different plant tissues (e.g., leaves, stem, flag leaf, brown rice and husk) and its relationship to aroma development has received limited attention. The primary objective of this study is to evaluate the effects of Si application on grain yield, yield components (e.g., panicle number, grain weight, spikelet fertility), 2AP concentration, and Si concentration in multiple Thai fragrant rice genotypes grown under highland and lowland conditions. By elucidating genotype-specific responses under contrasting environments. We hypothesize that the application of silicon can enhance both yield and grain aroma of Thai fragrant rice, with differential responses depending on the genotype and growing environment (highland vs. lowland). This hypothesis is grounded in the premise that silicon influences physiological processes such as nutrient uptake, stress tolerance, and secondary metabolite biosynthesis, which may interact with genetic and environmental factors to affect agronomic performance and grain quality traits. This research aims to support breeding programs targeting simultaneous improvements in yield and aroma, as well as to inform best practices for Si application. Furthermore, enhancing the quality and marketability of highland-produced fragrant rice could increase income for upland farming communities, contributing to smallholder farmer income enhancement and more inclusive and sustainable agricultural development.

2. Materials and Methods

2.1. Plant Materials and Silicon Treatment

Three fragrant rice cultivars were used in this study: Buer Ner Moo 4 (BNM4), a highland landrace traditionally cultivated by ethnic communities in northern Thailand; Buer Ner Moo CMU (BNMCMU), an improved variety developed by the Division of Agronomy, Faculty of Agriculture, Chiang Mai University; and Khao Dawk Mali 105 (KDML105), a nationally recognized elite aromatic variety. Field experiments were conducted during the wet season of 2022 (June–September) at two distinct locations: Highland site Tung Luang Village, Mae Win Subdistrict, Mae Wang District, Chiang Mai Province, Thailand (elevation 880 m). Lowland site Mae Hia Agricultural Research Station, Faculty of Agriculture, Chiang Mai University (elevation 336 m). Regional climate is monsoonal with distinct wet and dry seasons. The soil chemical properties at the experimental sites were as follows: Lowland site: pH 5.0 (1:1, soil/water), organic matter 1.14% (Walkley–Black method), total nitrogen 0.06% (Kjeldahl method), available phosphorus 49 mg kg⁻¹ (Bray II), and exchangeable potassium 45.52 mg kg⁻¹ (NH₄OAc, pH 7.0). Highland site: pH 5.76, organic matter 2.43%, total nitrogen 0.13%, available phosphorus 33.44 mg kg⁻¹, and exchangeable potassium 95.78 mg kg⁻¹. A factorial experiment was laid out in a randomized complete block design (RCBD) with three replications. Seeds were soaked in water in darkness for two days to initiate germination and then sown in nursery beds. After 30 days, seedlings were transplanted into field plots at a spacing of 25 × 25 cm (one seedling per hill). Silicon was applied as silicon dioxide (SiO₂) at four rates (0 (control), 168, 336, and 504 kg Si ha⁻¹; expressed as elemental Si equivalent derived from SiO₂). The material was finely ground and incorporated into the soil at the specified growth stages under flooded conditions to facilitate dissolution and release of monosilicic acid.

2.2. Sampling and Data Collection

At physiological maturity, rice plants were harvested and evaluated for the following agronomic traits: grain yield (adjusted to 14% moisture), plant height, number of tillers per plant, number of panicles per plant, grain filling percentage, unfilled grain percentage, 1000-grain weight, and straw yield. For silicon analysis, samples were collected from shoot tissues. For 2AP analysis and sensory evaluation, milled rice grains were used.

2.3. Silicon Concentration Analysis

Plant tissues were oven-dried at 70 °C for 7 days and ground into fine powder. The powdered samples were further dried at 60 °C for 2 days. A 0.1 g subsample was extracted using 3 mL of 50% NaOH and autoclaved at 121 °C for 20 min. The solution was transferred into a 50 mL volumetric flask and filtered through Whatman No. 42 filter paper (Cytiva Biotechnology (Hangzhou) Co., Ltd., Hangzhou, China). To 3 mL of the filtrate, 30 mL of 20% acetic acid and 10 mL of ammonium molybdate solution (pH 7.0) were added and left for 5 min. Next, 5 mL of 20% tartaric acid and 1 mL of reducing agent were added. After 30 min, absorbance was measured at 650 nm using a UV–Vis spectrophotometer (Biochrom Ltd., Cambridge, UK) (Modified from [24]).

2.4. 2-Acetyl-1-Pyrroline (2AP) Analysis

The concentration of 2AP in milled rice grains (pericarp, embryo, and aleurone layer removed) was determined using headspace gas chromatography with a nitrogenphosphorus detector (HS-GC/NPD). Each subsample was ground, sieved through a 35-mesh screen, and 1.0 g of the powder was placed into a headspace vial with 1.0 µL of 500 ppm 2,4,6-trimethylpyridine (TMP) as an internal standard. Vials were sealed with PTFE/silicone septa and aluminum caps and analyzed at Salana Organic Village (Social Enterprise) Co., Ltd., Nakhon Pathom, Thailand, following the method of [25].

2.5. Sensory Evaluation

To assess fragrance intensity, a simple sensory test was conducted. One gram of brown rice from each treatment was placed into a 1.5 mL microcentrifuge tube, followed by the addition of 1 mL of 1.7% potassium hydroxide (KOH) solution. After 10 min at room temperature, aroma intensity was evaluated by trained panelists and scored as follows: 0 = Non-aromatic, 1 = Moderately aromatic, and 2 = Strongly aromatic. Each sample was evaluated by 20 trained panelists in multiple sessions to confirm consistency (modified from [26]).

2.6. Statistical Analysis

All statistical analyses were performed using Statistix version 9 (Analytical Software, Tallahassee, FL, USA). Analysis of variance (ANOVA) was conducted to detect significant differences among treatments (p < 0.05). Least significant difference (LSD) was used for mean comparisons. Linear regression analysis was used to determine correlation coefficients and their significance.

3. Results

3.1. Effects of Location, Genotype, and Silicon Application on Agronomic Traits

Statistical analysis revealed significant variations in all measured traits across the two cultivation environments. Both genotype and silicon (Si) application had significant effects on most traits, except for grain-filled and unfilled percentages. Notably, significant interactions among location, genotype, and Si application were detected for several traits (p < 0.05), indicating that the response to Si application is dependent on both genotype and environmental conditions.

3.2. Grain Yield and Yield Components

Figure 1 illustrates the effects of Si application on grain yield and yield components across three fragrant rice genotypes BNM4, BNMCMU, and KDML105 grown under highland and lowland conditions. In the lowland environment, Si application consistently enhanced grain yield across all genotypes compared to untreated controls. At the highest application rate (504 kg Si ha⁻¹), BNM4 and BNMCMU showed marked yield increases of 65% and 94%, respectively. KDML105 responded most favorably at 336 kg Si ha⁻¹, with a 55% yield increase (Figure 1g). Under highland conditions, BNM4 exhibited the highest grain yields (6.52–5.86 ton ha⁻¹), while KDML105 produced the lowest (3.18–3.44 ton ha⁻¹). Conversely, in the lowland environment, KDML105 outperformed the other genotypes (3.35–5.33 ton ha⁻¹), whereas BNM4 recorded the lowest yields (2.16–3.80 ton ha⁻¹).

Figure 1. Effects of silicon (Si) application on plant height (a), number of tillers per plant (b), number of panicles per plant (c), grain filling percentage (d), grain unfilled percentage (e), 1000-grain weight (f), grain yield (g), and straw yield (h) in three Thai fragrant rice genotypes grown under highland and lowland conditions. Plants were subjected to four Si application rates: 0 (Control), 168, 336, and 504 kg Si ha⁻¹. Different letters above bars indicate significant differences according to the least significant difference (LSD) test at p < 0.05.

Silicon application also positively affected several yield components, including tiller number, panicle number, and straw yield, particularly in the lowland site. Most genotypes exhibited the strongest responses at 336 kg Si ha⁻¹. In the highland, only BNMCMU showed a significant increase in tiller number at 504 kg Si ha⁻¹ (Figure 1b,c,h). Plant height increased significantly with Si application at 504 kg Si ha⁻¹ in the lowland across all genotypes, but no such effect was observed in the highland. Other yield traits, including 1000-grain weight and grain-filled/unfilled percentages, were generally unresponsive to Si. However, in the lowland, BNM4 exhibited a 10.7% increase in 1000-grain weight at 336 kg Si ha⁻¹ compared to the control.

3.3. Silicon Accumulation in Plants

Silicon accumulation was analyzed in shoot tissues across both locations. Overall, Si accumulation was higher in the highland than in the lowland across all genotypes. In the highland, Si application did not significantly increase shoot Si content in BNM4 and KDML105; however, BNMCMU showed a significant increase at the highest application rate of 504 kg Si ha⁻¹. In contrast, under lowland conditions, all genotypes responded significantly to Si application. BNM4 showed increased shoot Si content at both 336 and 504 kg Si ha⁻¹, reaching 29.62 and 38.46 kg/m², respectively. BNMCMU and KDML105 responded positively at 336 kg Si ha⁻¹, with shoot Si contents of 27.71 and 31.13 kg/m², respectively (Figure 2). Correlation analysis revealed significant positive relationships between grain yield and shoot Si concentration in both environments. Grain yield was positively correlated with shoot Si levels across all three genotypes in both the highland and lowland (r = 0.71 and 0.79, respectively; p < 0.001; Figure 3), suggesting that Si accumulation in shoot tissues may play a key role in enhancing rice yield.

Figure 2. Silicon (Si) content in shoot tissues of three Thai fragrant rice genotypes grown under highland and lowland conditions and subjected to four levels of Si application: 0 (Control), 168, 336, and 504 kg Si ha⁻¹. Different letters above the bars indicate significant differences according to the least significant difference (LSD) test at p < 0.05.

Figure 3. Relationships between grain yield and silicon (Si) content in the shoots of three Thai fragrant rice genotypes grown under highland and lowland conditions with four levels of Si application: 0 (Control), 168, 336, and 504 kg Si ha⁻¹ (*** indicate significant difference at p < 0.001).

3.4. Grain Quality: 2-Acetyl-1-Pyrroline (2AP) Concentration and Content

Grain quality assessment focused on 2-acetyl-1-pyrroline (2AP), the principal aromatic compound in fragrant rice. Both 2AP concentration and total 2AP content were consistently higher in rice grown under highland conditions compared to those from the lowland across all Si levels and genotypes (Figure 4a,b). KDML105 exhibited the highest 2AP concentrations in both environments, ranging from 12.15 to 13.14 ppm (highland) and 5.29 to 6.58 ppm (lowland). Corresponding 2AP content values ranged from 3.92 to 4.26 mg/m² in the highland and 1.77 to 3.41 mg/m² in the lowland. BNM4 in the highland also recorded high 2AP content, ranging from 3.84 to 4.31 mg/m². In the highland, 2AP concentrations increased by 15.7% and 6.7% in BNM4 and KDML105, respectively, in response to 504 kg Si ha⁻¹. BNMCMU did not respond significantly under the same conditions. In the lowland, KDML105 exhibited an 18.1% increase in 2AP concentration with 168 kg Si ha⁻¹ (Figure 4a). Regarding 2AP content, no significant changes were detected in the highland. However, in the lowland, BNMCMU and KDML105 showed marked increases of 51.6% and 92.6%, respectively, at 168 and 336 kg Si ha⁻¹ (Figure 4b).

Figure 4. Concentration (a) and content (b) of 2-acetyl-1-pyrroline (2AP) in three Thai fragrant rice genotypes grown under highland and lowland conditions with four levels of silicon (Si) application: 0 (Control), 168, 336, and 504 kg Si ha⁻¹. Different letters above bars indicate significant differences according to the least significant difference (LSD) test at p < 0.05.

3.5. Sensory Evaluation and Correlation Analyses

Sensory evaluation revealed that Si application did not enhance aromatic intensity in any genotype under highland conditions. In contrast, in the lowland, Si application significantly improved the fragrance scores of BNM4 and BNMCMU (Table 1). Correlation analysis between grain yield and 2AP concentration revealed contrasting relationships under the two environments. As shown in Figure 5, a strong and significant negative correlation was observed in the highland site (r = –0.86, p < 0.001), suggesting that increased grain aroma was generally associated with a reduction in yield. This inverse relationship may reflect physiological trade-offs between carbon allocation to aroma biosynthesis versus grain filling under highland conditions. In contrast, no significant correlation was detected in the lowland site (r = 0.11, p = 0.63), implying that Si application enhanced aroma and yield independently, without a yield penalty. These findings suggest that the environment-specific physiological responses to Si application can decouple or reinforce the trade-off between quality and yield.

Table 1. Sensory evaluation scores for aroma intensity in three Thai fragrant rice genotypes grown under highland and lowland conditions with four levels of silicon (Si) application: 0 (Control), 168, 336, and 504 kg Si ha⁻¹. Aroma was assessed using a potassium hydroxide (KOH) test and scored as 0 (non-aromatic), 1 (moderately aromatic), or 2 (strongly aromatic), evaluated by 20 trained panelists in multiple sessions to confirm consistency.

Locations	Genotypes	Si0	Si168	Si336	Si504
Highland	KDML105	Moderately Strong aromatic	Strongly aromatic	Moderately Strong aromatic	Moderately Strong aromatic
	BNM4	Moderately Strong aromatic	Moderately aromatic	Non-Moderately aromatic	Moderately aromatic
	BNMCMU	Moderately aromatic	Moderately Strong aromatic	Moderately aromatic	Moderately aromatic
Lowland	KDML105	Strongly aromatic	Moderately Strong aromatic	Moderately Strong aromatic	Strongly aromatic
	BNM4	Non-aromatic	Moderately aromatic	Non-Moderately aromatic	Moderately aromatic
	BNMCMU	Non-Moderately aromatic	Moderately aromatic	Moderately aromatic	Moderately Strong aromatic

Figure 5. Relationships between grain yield and 2-acetyl-1-pyrroline (2AP) concentration in three Thai fragrant rice genotypes grown under highland and lowland conditions with four levels of silicon (Si) application: 0 (Control), 168, 336, and 504 kg Si ha⁻¹ (*** indicate significant difference at p < 0.001, and ns indicate no significant difference).

4. Discussion

This study investigated the interactive effects of cultivation location, rice genotype, and silicon (Si) application on grain yield, yield components, Si accumulation, and aromatic quality in fragrant rice varieties. The findings underscore the combined importance of genetic, environmental, and nutrient management factors in optimizing both productivity and grain quality, demonstrating that Si application may serve as a valuable agronomic strategy in fragrant rice cultivation. Cultivation location exerted a pronounced influence on grain yield, yield components, Si accumulation, and 2-acetyl-1-pyrroline (2AP) content across all genotypes. Highland and lowland environments represent distinctly different agroecosystems, characterized by contrasts in altitude, temperature, soil fertility, and microclimatic conditions, all of which can affect plant physiological responses, nutrient uptake, and secondary metabolite production [10,11]. Average temperatures during the growing season ranged from 24–31 °C in the lowland and 18–27 °C in the highland, consistent with regional norms. These differences were reflected in genotype performance across sites. These contrasting environments necessitate adaptive agronomic interventions to improve both yield and grain quality in regionally distinct ecosystems. Genotypic variation significantly influenced most measured traits, with the exception of grain-filled and unfilled percentages. This variability is likely due to differences in physiological and morphological characteristics such as root system architecture, nutrient uptake efficiency, stress resilience, and metabolic adaptability [10], which collectively contribute to genotype–environment interactions.

The contrasting responses of rice genotypes to Si application under highland and lowland conditions likely arise from complex physiological and nutritional mechanisms. Lower temperatures, higher baseline soil fertility, and potentially greater native Si availability in the highland may reduce the marginal benefit of external Si input, while simultaneously favoring the synthesis of secondary metabolites such as 2-acetyl-1-pyrroline (2AP) [10]. Conversely, in the lowland, Si application may play a stronger role in enhancing stress tolerance, nutrient uptake efficiency, and carbon allocation toward both yield and aroma biosynthesis [11,16]. These genotype-specific and environment-dependent processes highlight the importance of understanding plant physiology and soil–plant interactions when designing Si management strategies.

The varying responses of rice genotypes to silicon application across different environments reflect genotypes by environment interactions. These interactions are driven by genetic differences in physiological traits. Therefore, genotype selection and site-specific nutrient management are essential. In this study, variation in Si application response across genotypes and environments reflects such interactions, highlighting the need for location-specific genotype selection and nutrient management strategies [16,27,28].

Silicon application had a clear effect on multiple agronomic parameters, including grain yield, plant height, tiller number, panicle number, 1000-grain weight, and Si concentration in plant tissues. Under lowland conditions, all genotypes responded positively to Si application, with BNM4 and BNMCMU showing the most pronounced yield increases at 504 kg Si ha⁻¹, and KDML105 performing best at 336 kg Si ha⁻¹ [29,30,31]. In contrast, in the highland environment, yield improvements were observed only in the BNMCMU genotype at the highest Si application rate. These differences likely reflect genotype-specific physiological responses to both soil Si availability and inherent environmental advantages such as cooler temperatures and better baseline fertility in the highland. Although initial soil Si levels were not measured in this study, we acknowledge their importance in interpreting Si responsiveness and will incorporate this parameter in future work to improve our understanding of genotype–environment–nutrient interactions. In addition to enhancing yield, Si application also increased plant height, tiller number, and panicle number, especially in the lowland, contributing to greater biomass and improved reproductive capacity. Among yield components, thousand-grain weight increased only in BNM4 under lowland conditions following Si treatment. This suggests that while Si may enhance grain filling processes in specific genotypes, its effect may be conditional upon genotype environment compatibility and the physiological sink strength of the developing grain. Silicon accumulation patterns differed significantly between environments, with higher accumulation observed in the highland compared to the lowland across all genotypes. The application of Si enhanced Si accumulation in the shoots of all rice genotypes BNM4, BNMCMU, and KDML105 under highland conditions, while no significant response was observed under lowland conditions. The ability of rice plants to uptake and accumulate Si depends on their physiological and morphological characteristics, as well as genotype-specific traits. This pattern aligns with the structural role of Si in reinforcing cell wall rigidity, increasing disease resistance, and improving tolerance to abiotic stresses [13,32]. Previous studies have also shown that Si application promotes Si accumulation in rice; however, the extent of accumulation varies depending on plant part, genotype, and environmental conditions [31,33].

The aromatic quality of the rice, assessed through 2-acetyl-1-pyrroline (2AP) concentration and content, exhibited clear environmental dependence. Rice grown in the highland environment consistently exhibited higher 2AP concentrations than rice grown in the lowland, regardless of genotype or Si treatment. These findings are consistent with previous studies indicating that lower temperatures and highland-specific environmental conditions promote the biosynthesis of 2AP and other volatile aromatic compounds [25,34,35]. Nevertheless, Si application significantly increased 2AP content in lowland-grown rice, particularly in BNMCMU and KDML105, while no significant effect was observed under highland conditions. These results suggest that Si application may help mitigate suboptimal lowland conditions by enhancing 2AP biosynthesis, potentially through indirect mechanisms such as improved nitrogen and proline metabolism or alleviation of environmental stress [31,36,37]. The KDML105 variety consistently exhibited the highest 2AP concentration across both environments and responded positively to Si application, although the optimal Si level for 2AP enhancement differed depending on the location. In the highland, 504 kg Si ha⁻¹ resulted in a moderate increase in 2AP concentration in BNM4 and KDML105, while in the lowland, the most substantial improvement was observed in KDML105 at 168 kg Si ha⁻¹. These findings underscore the importance of aligning Si fertilizer regimes with environmental conditions and genotype-specific aroma potential to maximize quality outcomes.

Correlation analyses provided further insight into the complex relationships among grain yield, aroma, and silicon (Si) accumulation. In the highland environment, a strong negative correlation between grain yield and 2-acetyl-1-pyrroline (2AP) concentration suggests a potential trade-off between productivity and aromatic quality. This trade-off may reflect a physiological allocation dilemma, in which energy and metabolic resources are preferentially directed either toward reproductive development or aroma compound biosynthesis, but not both. In contrast, grain yield was positively correlated with Si accumulation under both highland and lowland conditions, reinforcing the role of Si in promoting overall plant health, structural integrity, and yield enhancement [13]. Although a formal correlation analysis was not conducted, observational trends suggest a positive relationship between 2AP content and sensory scores, particularly in the lowland environment. Genotypes such as BNMCMU and KDML105, which exhibited increased 2AP content in response to silicon application, also showed improved aroma scores in sensory evaluation. This supports the role of 2AP as a reliable biochemical marker for fragrance and highlights the potential of silicon application to enhance both measurable aroma compounds and consumer-perceived grain quality under lowland conditions.

These findings highlight the critical need for genotype and location-specific silicon (Si) application strategies. In lowland environments, where Si application positively influences both yield and aromatic quality, targeted Si management could significantly enhance crop performance and economic value. Furthermore, the observed genotype-dependent responses to Si, particularly in traits related to Si uptake, yield, and aroma, could be leveraged in breeding programs aimed at developing cultivars with improved Si use efficiency and enhanced physiological resilience.

5. Conclusions

This study demonstrates that cultivation environment, rice genotype, and silicon (Si) application interactively influence grain yield, yield components, Si accumulation, and aromatic quality in Thai fragrant rice. The results highlight distinct genotype-specific responses to Si application across highland and lowland ecosystems. In particular, lowland-grown rice showed significant improvement in both yield and 2-acetyl-1-pyrroline (2AP) content following Si application, while highland conditions inherently supported greater aroma production, albeit sometimes with a trade-off in yield. These findings underscore the potential of Si as a climate-adaptive, sustainable input that enhances both productivity and grain quality in location-specific ways.

Silicon application not only improved structural and physiological traits such as plant height, tiller number, and panicle formation but also increased Si accumulation in plant tissues. In the lowland environment, Si supplementation contributed to enhanced 2AP biosynthesis, improving grain aroma, particularly in BNMCMU and KDML105 genotypes. Correlation analysis revealed contrasting patterns: in the highland, grain yield was negatively associated with 2AP concentration, suggesting a trade-off between productivity and aroma, while in both environments, grain yield showed a positive association with shoot Si accumulation. These results suggest that silicon can improve both yield and aroma simultaneously in certain genotypes, depending on the environmental context.

This study highlights the potential of silicon application to improve rice aroma and yield under both highland and lowland conditions. The observed responses were strongly genotype- and site-dependent, emphasizing the significance of genotype × environment × silicon interactions. These findings support the use of targeted Si application, tailored to specific genotypic and agroecological contexts, as a promising strategy to enhance rice productivity and quality. Environmental differences between lowland and highland rice-growing areas influence growth dynamics, nutrient uptake, and responsiveness to Si. Although not essential, silicon enhances stress tolerance and grain quality. Understanding genotype-specific responses in each setting is crucial for optimizing yield and aroma. Furthermore, site-specific Si management offers the potential to enhance income for smallholder farmers and contribute to improved livelihoods, thereby supporting poverty alleviation efforts through agriculture in both highland and lowland regions.