Yield and quality traits stability of turmeric (Curcuma longa L.) genotypes under greenhouse conditions

1 Introduction

Turmeric (Curcuma longa L.) is one of the major spice crops cultivated across the tropics and subtropics of South Asia (Prasath et al., 2018). India is the leading producer of turmeric, with a contribution of 80% of global production, and is the leading exporter of turmeric, primarily to Bangladesh, UAE, Iran, Morocco, the USA, and Malaysia (PJTSAU, Agricultural Market Intelligence Centre, 2024). Turmeric is one of the valuable spices and medicinal plants with diverse uses in various industries like food, dye, pharmaceuticals, and textiles. Turmeric is known for its polyphenolic compound ‘curcuminoids’, which impart numerous health benefits and enhance its value. Increasing health consciousness and demand for natural supplements are propelling the curcumin market. Turmeric finds an important place in different industries along with food, coloring, textile, and pharmaceutical industries due to its numerous therapeutic uses, including antivenomous, antimalarial, antioxidant, antiaging, anticancer, antitumor, antimicrobial, anti-inflammatory, and antiprotozoal properties, along with the culinary uses (Amalraj et al., 2017). Turmeric is used as a vegetable in northern India and is also marketed in the form of turmeric oil, oleoresin, curcumin powder, and turmeric powder (Spices Board of India, 2024).

Numerous studies reported that under field conditions, the growth, yield, and quality of turmeric were greatly affected by geographical factors, production conditions, and climatic factors such as temperature, relative humidity, rainfall, and light (Aarthi et al., 2020; Akbar et al., 2016). The quality of the turmeric is mainly assessed by the shape, size, and color of the rhizomes (Prasath et al., 2018) and by the quantity of the polyphenolic pigments ‘curcuminoids’ (Aarthi et al., 2020). Such variability under open-field conditions necessitates alternative production systems that can ensure more consistent yield and phytochemical quality. There is a need for emphasizing the cultivation of turmeric under controlled greenhouse conditions, which are minimally influenced by limiting factors such as temperature, light, and humidity. Agricultural production in controlled environments is increasingly feasible and may play a significant role in providing nutrition and choice to growing urban centers. Controlled environment production systems are gaining attention to enhance yield and quality of different crops. They focus on high-value specialty crops because of their demand and economic viability. They offer solutions to enhance yield and metabolites, with minimal environmental influence on these factors (Folta, 2019). Soilless cultivation can improve quality, as well as total and marketable yield, compared with in-soil cultivation (Maboko et al., 2011).

Under field conditions, phenotypic stability of the genotype is paramount, as production must be consistent in an unpredictable and changing environment. However, the controlled environment affords focus on different breeding priorities, as environmental flux, pests, pathogens, and post‐harvest quality are less formidable barriers to production. On the contrary, breeding for controlled environments shifts the focus to a completely distinct set of plant traits, such as rapid growth, performance in low-light environments, and active manipulation of plant stature. Instead of breeding for phenotypic stability, plants need to be bred to maximize genetic plasticity, allowing specific traits to be presented as a function of the quality of the ambient light spectrum. Analysis of variance (ANOVA) and genotype plus genotype by environment (GGE) biplot techniques effectively delineate individual trait variations among genotypes across seasons; they fail to explain the complex multivariate associations and physiological trade-offs existing between yield and quality traits. Hence, modern multi-trait selection indices [multi-trait genotype–ideotype distance index (MGIDI) and factor analysis and ideotype-design based on best linear unbiased prediction (FAI-BLUP)] were employed in this study to simultaneously evaluate and integrate yield and quality traits, enabling the objective identification of superior turmeric ideotypes without compromising one trait for another (Gopi et al., 2026; Lavudya et al., 2026). The identification of stable, high-performing genotypes with balanced yield and elevated quality profiles will assist commercial growers in selecting ideal cultivars for protected cultivation systems, while aiding pharmaceutical industries in procuring raw materials tailored to specific trait requirements. Therefore, the primary objective of this study was to evaluate genotype × environment interactions, trait associations, and multi-trait selection efficiency to identify superior and stable turmeric genotypes under greenhouse conditions.

2 Materials and methods

Our present investigation was carried out for 2 years (2021–2023) with 21 turmeric genotypes representing different geographical locations of India, viz., IISR Pragati, IISR Prathiba, IISR Alleppey Supreme, Rajendra Sonali, Megha Turmeric 1, Waigon Turmeric, Roma, CIM Pitambar, Uttar Rangini, Chhattisgarh Haldi 2, NDH 8, CO 3, Acc. 1545, Erode Local, Mydukur Local, Acc. 849, Acc. 379, Acc. 14, Acc. 179, Acc. 214, and Acc. 69/5/22/I3. The trial was conducted under a greenhouse facility at Simply Fresh Pvt. Ltd., Arjunpatla, Hyderabad, Telangana, India (north latitude 17°52′12.70″; east longitude 78°55′16.319″; 525 m msl) (Table 1) under protected cultivation. The multi-span greenhouse was covered with a UV-stabilized diffuse polyethylene film, with insect-proof netting on all sides. The greenhouse microclimate was managed through natural ventilation using lateral roll-up openings, while retractable shade screens were used during peak hours to reduce solar radiation and heat load. Single sprouted rhizomes were planted in 24 L plastic pots (V G Plastech, Surat, Gujarat), measuring 14 in. in diameter and 12 in. in depth, with a capacity of 24 L, containing a mixture of virgin sterilized coir husk chips (Eco Coir Products, Tirunelveli, Tamil Nadu) under greenhouse conditions. The experiment was laid out in a completely randomized design (CRD) with two replicates. Each genotype was planted in 40 pots with one 30-day-old single sprouted plant, and 20 pots were considered as one replication. Standard agronomic practices were followed to raise the crop (Prasath et al., 2022), along with fertigation (Tamil Nadu Agricultural University, 2024). The conditions provided across the crop season are given in Table 1. The list of the genotype details is given in Table 2.

Table 1. Weather data during 2021–2022 and 2022–2023 in a greenhouse, Simply Fresh Pvt. Ltd., Arjunpatla, Telangana.

Table 2. List of turmeric genotypes used in the investigation.

2.1 Fresh rhizome yield (g clump⁻¹)

The fresh rhizome yield was assessed by weighing the harvest from five plants selected at random. The mean weight, reported as grams per clump (g clump^–1), was determined in accordance with the distinctness, uniformity, and stability (DUS) test guidelines (Protection of Plant Varieties & Farmers rights Authority (PPV & FRA), 2009).

2.2 Dry recovery (%)

To produce dried turmeric, fresh rhizomes were processed by boiling them in water until soft (approximately 45–60 min), as indicated by a mushy texture and the release of a characteristic aroma, as per standard methods (Protection of Plant Varieties & Farmers rights Authority (PPV & FRA), 2009). Subsequently, the boiled rhizomes were subjected to sun drying for 10 days, with exposure limited to 6–8 h daily, until a consistent moisture content of 10% was reached. The dry recovery percentage was subsequently determined by calculating the difference between the fresh and dry weights (Protection of Plant Varieties & Farmers rights Authority (PPV & FRA), 2009).

2.3 Essential oil (%)

Following the American Spice Trade Association (1968) guidelines, essential oil extraction was performed using a modified Clevenger apparatus. A 25-g powdered sample was transferred to a 1,000-mL round-bottom flask, which was subsequently filled with water to the halfway mark. The apparatus, including the condenser, was assembled, and the distillation process was run for 3 h. The essential oil, being lighter than water and immiscible, rises during boiling and accumulates as a distinct layer on the water’s surface. This oil was then collected in an Eppendorf tube, and the oil content was determined by the formula:

Essential oil (%) = volume of oil extracted (mL) / mass of sample (g)×100

2.4 Oleoresin (%)

The oleoresin content was assessed in the turmeric powder according to the American Spice Trade Association (1975) guidelines. A column was prepared by placing cotton at the bottom, into which 10 g of finely powdered sample was loaded, followed by the addition of roughly 50 mL of acetone. The setup was left undisturbed overnight to allow for thorough extraction. Subsequently, the column was drained into a pre-weighed beaker; draining continued until the yellow coloration was completely eluted from the cotton. The solvent was removed by evaporation in a water bath, leaving the oleoresin residue. The beaker was re-weighed, and the oleoresin percentage was determined using the formula:

Oleoresin (%)=weight of the residue (g) / weight of the sample (g)×100

2.5 Curcuminoids (%)

Curcuminoid content in the dried turmeric powder was determined according to the American Spice Trade Association (2004) guidelines. The dried rhizomes were first ground in a Cyclotec 1093 mill and sieved to a particle size of 325 μm. Using a Shimadzu UV-160 I spectrophotometer, the absorbance of the extracted curcuminoids was measured at 425 nm. The percentage concentration was calculated against pure crystalline curcumin (98%) as a reference. All measurements were conducted in triplicate, and the mean value was used for analysis. The following formula was applied:

Total curcuminoids (%)=(DS×AS×100)/(100×WS×1,650)

where.

• DS = dilution volume (100 mL).
• WS = weight of the sample in grams (1 g).
• AS = absorbance maxima.

2.6 Statistical analysis

To assess variability among genotypes across different environments, data were subjected to ANOVA. Prior to pooling the data across the two greenhouse seasons, homogeneity of experimental error variances was evaluated using Bartlett’s test on model residuals alongside Hartley’s maximum F-ratio test (Fmax=MSEmax/MSEmin; Hartley, 1950). Analysis of variance was conducted using a two-factor mixed-effects model. In this model, genotypes were treated as fixed effects, while seasons and replications nested within seasons were considered random effects. The significance of genotypic main effects was tested against the genotype × season interaction mean square (MSgs). Treatment means were subsequently compared using Duncan’s multiple range test (DMRT) at a 5% significance level, facilitated by the ‘doebioresearch’ package (Popat and Banakara, 2020). Variability parameters, genotypic and phenotypic coefficients of variation, broad-sense heritability, genetic advance, and genetic advance as a percentage of mean were estimated following Singh and Chaudhary (1985) from the formulas presented in Table 3. Multivariate stability was assessed for estimation of genotype × environment interaction (GEI) and presented graphically through GGE biplot analysis, and multi-trait ideotype selection was conducted using MGIDI and cross-validated with the FAI-BLUP index at a 20% selection intensity examined using the ‘metan’ package for multi-environment trial analysis (Olivoto and Lúcio, 2020). All statistical computations were performed utilizing RStudio, an integrated development environment for the R statistical software.

Table 3. Genetic variability parameter formulas from a two-factor pooled ANOVA across two seasons.

3 Results and discussion

3.1 Fresh rhizome yield (g clump⁻¹) and dry recovery (%)

Significant variation was observed among the genotypes for fresh rhizome yield under greenhouse conditions during both years of evaluation (Table 4). In 2021–2022, fresh rhizome yield ranged from 532.33 to 958.70 g clump⁻¹. Acc. 1545 (958.70) recorded the highest yield and was statistically comparable with IISR Pragati (923.17), Rajendra Sonali (921.50), CO 3 (929.67), and Mydukur Local (916.73). In contrast, lower yields were recorded in genotypes such as Uttar Rangini (532.33) and Acc. 69/5/22/I3 (554.37). During 2022–2023, wide genotypic variability was observed among the genotypes, with fresh rhizome yield ranging from 378.00 to 1,014.00 g clump⁻¹. NDH 8 exhibited superior performance (1,014.00), followed closely by IISR Pragati (995.00), indicating consistently high productivity across seasons. The pooled analysis across 2 years further confirmed significant genotypic variations, with a mean fresh rhizome yield varying from 499.33.00 to 959.08 g clump⁻¹. IISR Pragati recorded the highest fresh rhizome yield (959.08), followed by NDH 8 (938.83). The observed variation indicates substantial genetic diversity among the evaluated genotypes and suggests that greenhouse cultivation can effectively express yield potential while minimizing extreme environmental fluctuations. Dry recovery (%) also differed significantly among genotypes during both years of evaluation under greenhouse conditions (Table 4). In 2021–2022, dry recovery ranged from 16.92% to 25.79%. Chhattisgarh Haldi 2 exhibited the highest dry recovery (25.79%), followed by IISR Alleppey Supreme (23.93%) and Acc. 14 (23.65%), indicating superior dry matter accumulation potential. In 2022–2023, dry recovery (%) ranged from 15.48% to 27.93%, with Megha Turmeric (27.93%), Uttar Rangini (27.33%), and Erode Local (26.05%) observed to be superior. Chhattisgarh Haldi 2 maintained consistently high dry recovery (25.32%) in both years. Pooled analysis across two seasons revealed that dry recovery varied from 16.64% to 25.32%. Chhattisgarh Haldi 2 recorded the highest dry recovery (25.32%), followed by Uttar Rangini (24.71%) and Erode Local (24.08%). In contrast, IISR Pragati and Rajendra Sonali exhibited relatively lower pooled dry recovery despite their high fresh rhizome yield. This inverse trend between fresh yield and dry recovery in some genotypes indicates possible dilution effects due to higher moisture content in rapidly growing rhizomes. From a commercial perspective, genotypes with higher dry recovery are particularly desirable for processing industries, as they enhance dry rhizome output and improve economic returns per unit fresh produce. Yield was found superior for soilless culture under greenhouse than under field conditions due to the minimal changes in environmental factors such as temperature, light, and relative humidity, which was also reported in earlier studies (Ali et al., 2021; Balliu et al., 2021; Chiomento et al., 2021; Mouroutoglou et al., 2021; Pulela et al., 2022; Sinha et al., 2020) in different horticultural crops cultivated under greenhouse conditions.

Table 4. Mean performance of fresh rhizome yield (g clump⁻¹) and dry recovery (%) in turmeric genotypes under greenhouse conditions.

Means within the same column followed by the same lowercase superscript letter are not significantly different at p ≤ 0.05 according to Duncan’s Multiple Range Test (DMRT).

3.2 Quality traits [oleoresin (%), essential oil (%), and curcuminoid content (%)]

Significant genotypic variation was also observed for quality parameters including oleoresin (%), essential oil (%), and curcuminoid content (%) under greenhouse cultivation (Table 5). In 2021–2022, oleoresin content ranged from 3.27% to 12.24%, while in 2022–2023, it varied between 7.31% and 12.04%, indicating substantial genetic diversity in oleoresin content. Waigon Turmeric consistently maintained superior oleoresin content across both seasons and recorded the highest pooled mean (12.07%), followed by Roma and Acc. 69/5/22/I3. Essential oil content (%) exhibited a similar trend, with pooled values ranging from 3.20% to 8.50%. Waigon Turmeric (8.50%) and Acc. 69/5/22/I3 (7.95%) were identified as superior for essential oil content. However, slight year-to-year fluctuations were observed in certain genotypes, possibly due to differences in secondary metabolite accumulation influenced by micro-environmental conditions within the greenhouse. Curcuminoids are the major compounds that add pharmaceutical value and make turmeric more valuable (Prasath et al., 2018). Curcuminoids recorded significant variation among genotypes across both years. IISR Pragati (3.62) followed by Waigon Turmeric (2.81), IISR Prathiba (2.59), CIM Pitambar (2.47), Acc. 1545 (2.45), and Roma (2.38) were found to be superior for curcuminoids in the first year. Meanwhile, in the second year, IISR Prathiba (3.21) followed by Rajendra Sonali (2.96), IISR Pragati (2.86), CIM Pitambar (2.75) and Roma (2.72) were found to be superior. For the overall pooled analysis for 2 years, IISR Pragati (3.24) produced the maximum curcuminoid content followed by IISR Prathiba (2.90), CIM Pitambar and Rajendra Sonali (2.61), and Roma (2.55). Overall quality traits [oleoresin content (%), essential oil (%), and curcuminoid content (%)] were not found to be comparatively expected when compared with open-field conditions (Silaru et al., 2023). A decline in phytochemical composition of a few medicinal herbs, rhizome, and root crops under hydroponic and aeroponic production [perlite hydroponics, nutrient film technique (NFT), ebb and flow, and aeroponics] was reported (Hayden, 2006) and stressed the importance of work needed to improve the agronomic factors to enhance rhizome yield under hydroponic systems with a motive of enhanced phytochemical composition in high specialty crops like turmeric. Phytochemical compounds like curcuminoids are greatly affected by geographical factors, light, temperature, humidity, and agronomic factors (Akbar et al., 2016). A similar decline in capsaicinoids under controlled conditions was reported by Jeeatid et al. (2018). Pooled ANOVA across two seasons exhibits a highly significant variation (P<0.001) among genotypes for fresh rhizome yield, dry recovery, oleoresin, essential oil, and curcuminoid content (Table 6). However, the effect of season was non-significant for dry recovery (%), but significant for fresh rhizome yield and oleoresin content (%), essential oil (%), and curcuminoid content. Notably, genotypic variance accounted for most of the total variation, confirming the negligible role of environmental effects. Previous studies under open-field conditions have reported a stronger influence of GEI and environmental factors on yield and quality traits (Aarthi et al., 2020; Anandaraj et al., 2014; Lal et al., 2017). In contrast, the present greenhouse-based evaluation indicated a greater contribution of genotype, followed by GEI, while environmental influence was comparatively reduced. This suggests that controlled cultivation conditions enhance the expression of genetic potential by minimizing environmental fluctuations. Interestingly, high-yielding genotypes such as IISR Pragati also exhibited superior curcuminoid content, whereas others demonstrated a trade-off between yield and specific quality attributes. This differential response highlights the importance of multi-trait selection strategies in greenhouse turmeric cultivation.

Table 5. Mean performance of oleoresin content (%), essential oil (%), and curcuminoids (%) in turmeric genotypes under greenhouse.

Means within the same column followed by the same lowercase superscript letter are not significantly different at p ≤ 0.05 according to Duncan’s Multiple Range Test (DMRT).

Table 6. Pooled ANOVA for fresh rhizome yield (g clump⁻¹), dry recovery (%), oleoresin content (%), essential oil (%), and curcuminoids (%) in turmeric across two greenhouse seasons.

* Significant at p < 0.05%; Significant at p < 0.01%; * Significant at p < 0.001%; ns: non-significant.

3.3 Genetic parameters, correlation, and character association

Phenotypic coefficient of variation (PCV) was marginally higher than the corresponding genotypic coefficient of variation (GCV) across yield and quality traits, indicating minimal environmental masking under greenhouse conditions. In the pooled data, the highest GCV and PCV were recorded for curcuminoid content (39.88% and 42.36%, respectively), followed by essential oil content (19.67% and 21.83%), oleoresin content (15.71% and 17.08%), fresh rhizome yield (15.01% and 19.47%), and dry recovery (11.96% and 13.64%). High broad-sense heritability (hbs2>75%) coupled with high genetic advance as percent of mean (GAM > 20%) was observed for curcuminoids (hbs2=88.65%, GAM = 77.36%), oleoresin (hbs2=84.61%, GAM = 29.76%), essential oil (hbs2=81.24%, GAM = 36.53%), and dry recovery (hbs2=76.93%, GAM = 21.61%). Fresh rhizome yield exhibited moderate heritability (hbs2=59.45%) alongside high genetic advance (GAM = 23.85%) (Table 7). The combination of high heritability and substantial GAM across these traits confirms the predominance of additive gene action, indicating high selection efficiency. Correlation analysis revealed significant correlation among yield and quality traits across seasons. Fresh rhizome yield (g clump⁻¹) exhibited a significant positive association with curcuminoids (r = 0.23), but was negatively correlated with dry recovery (r = −0.41), oleoresin content (r = −0.47), and essential oil content (r = −0.46). Dry recovery was negatively correlated with curcuminoid content (r = −0.53) and positively correlated with essential oil content (r = 0.31). Oleoresin and essential oil exhibited a strong positive correlation (r = 0.589) (Figure 1). Genetic variability demonstrates the diversity of the germplasm, and a correlation will explain the relationship of the traits. In turmeric, yield and quality are the most important traits of economic use that eventually determine the value of the product, so selection of superior genotypes based on the genetic parameters will facilitate the breeding programs effectively. Similar results were reported by Aarthi et al. (2018a); Aarthi et al. (2018b); Gupta et al. (2016), and Neeraja et al. (2026).

Table 7. Genetic variability parameters for yield and quality traits in turmeric across two greenhouse seasons and pooled.

3.4 Environmental adaptability of superior and stable genotypes

Multivariate analysis was conducted to understand the interaction of genotypes by environment across 2 years to select the genotypes for stable yield and quality under greenhouse conditions. The ANOVA of the GGE biplot analysis is represented in Supplementary Table 2. It is clear that all the traits under study had a greater influence by genotype and contributed more variation than environment and GEI, with significance at P ≤ 0.005, explaining most of the variation. Notably, for fresh rhizome yield, PC1 and PC2 explained 75.69% and 24.31% of the total variation, respectively. It was evident that there was substantial contribution of the first two principal components in capturing the major sources of variation. The overall effect of environment in the studied traits was less compared with genotype and GEI. In controlled greenhouse production systems, where environmental conditions are regulated with minimal deviations, variations in yield and quality among genotypes are primarily due to genetic differences, followed by GEI. For fresh rhizome yield, IISR Pragati (G₁), NDH 8 (G₁₁), and Rajendra Sonali (G₄) were positioned towards the positive end of the AEC abscissa, indicating superior mean performance. Among these, Rajendra Sonali (G₄) showed relatively shorter projections onto the AEC ordinate, suggesting better stability across seasons (Figure 2A). In the case of dry recovery, Chhattisgarh Haldi 2 (G₁₀) combined desirable mean performance with comparatively stable responses (Figure 2B). Oleoresin content (%) was dominated by Waigon Turmeric (G₆) with superior performance and considerable stability, followed by Roma (G₇). IISR Prathiba and Acc. 69/5/22/I3 (G₂₁) were positioned ahead of the arrow with the shortest vectors, making them more stable than Waigon Turmeric (G₆) (Figure 2C). Meanwhile, Waigon Turmeric (G₆) outperformed all the other genotypes with higher essential oil content and moderate adaptability (Figure 2D). Curcuminoid content was the highest in IISR Pragati (G₁), followed by IISR Prathiba (G₂) that displayed moderate stability across seasons (Figure 2E).

Secondary metabolite accumulation in turmeric is known to be influenced by environmental factors such as moisture availability and temperature (Aarthi et al., 2020; Lal et al., 2017; Aydinşakir et al., 2025). Previous reports have highlighted strong environmental and stress-mediated influences on turmeric yield and quality under open-field conditions, particularly under water-limited environments (impact of growing conditions and water stress on turmeric). However, in the present controlled environment study, genotype contributed more prominently than environmental variation, indicating that protected cultivation can buffer stress effects and enhance stability (Aydinşakir et al., 2025). Nevertheless, optimization of microclimatic factors such as temperature, light intensity, humidity, and nutrient management remains crucial for maximizing secondary metabolite accumulation (Jeeatid et al., 2018). Similar observations have been reported in controlled environment studies where manipulation of agronomic factors significantly influenced metabolite production and yield (Retana-Cordero et al., 2022). Hence, there is a need for standardizing agrotechniques for maximizing yield with enhanced stable quality by altering agronomic and environmental factors.

3.5 Factor analysis and dimensionality reduction and multi-trait genotype–ideotype distance index

Factor analysis was performed to identify the major dimensions underlying the variation among the evaluated traits. Based on the eigenvalue criterion (>1), two principal factors were retained from the pooled mixed-model dataset, together explaining 82.82% of the total phenotypic variation (Table 8). Following varimax rotation, factor 1 (FA1; eigenvalue = 2.46) accounted for 49.14% of the total variation and was mainly associated with fresh rhizome yield (loading = 0.73), curcuminoid content (0.87), and dry recovery (−0.88). Factor 2 (FA2; eigenvalue = 1.68) explained a further 33.68% of the variation and was strongly associated with oleoresin content (−0.97) and essential oil content (−0.75). The mean communality of 0.83 indicated that the two retained factors explained a substantial proportion of the variation present across the measured traits. Notably, the factor structure separated oleoresin and essential oil (FA2) from fresh rhizome yield and curcuminoid content (FA1), rather than creating an absolute separation between all yield and all quality traits. The MGIDI was used to identify genotypes with a desirable combination of yield and quality traits. At a selection intensity of 20%, the four genotypes Roma, Waigon Turmeric, IISR Prathiba, and CO 3 were identified as the most promising genotypes across two seasons (Figures 3, 4; Table 9). The multi-trait selection resulted in considerable positive genetic gain (SG) for the major biochemical quality traits, particularly curcuminoid content (+33.56%; hbs2=0.86), oleoresin content (+11.54%; hbs2=0.65), and essential oil content (+9.37%; hbs2=0.83). In contrast, selection resulted in relatively small negative gains for fresh rhizome yield (−4.72%) and dry recovery (−3.30%). In protected cultivation economics, this modest sacrifice in rhizome yield can be compensated by the marked recovery in dry recovery, curcuminoid content per unit area along with consistent yield, and more predictability in quality traits. The factor contribution radar plot (Figure 5) demonstrates variations among the selected genotypes. Roma and IISR Prathiba showed relatively balanced contributions in both factors FA1 and FA2, indicating a favorable combination of yield and quality traits. In comparison, Waigon Turmeric showed a stronger contribution to FA2, reflecting its comparatively better performance for the oleoresin and essential oil contents.

Table 8. Principal component analysis, rotated factor loadings, communalities, and uniqueness values for five agro-quality traits in turmeric across evaluated seasons.

Table 9. Multi-trait genotype–ideotype distance index (MGIDI): selection differentials (SD), broad-sense heritability (h²), and predicted genetic gains (SG%) under 20% selection intensity.

3.6 Validation through FAI-BLUP and cross-seasonal concordance

The FAI-BLUP multi-trait index was subsequently used to validate the genotypes identified through MGIDI. Under ideotype 1, which targeted increases in yield and quality traits, Roma, IISR Prathiba, CO 3, and IISR Pragati were identified as the most desirable genotypes (Figures 3, 4; Table 10). From the two selection models, 75% similarity was observed between MGIDI and FAI-BLUP over the 2 years, with Roma, IISR Prathiba, and CO 3 observed consistently selected by both selection models. These results support that these genotypes were superior for their overall multi-trait performance. The stability of the selected genotypes was further examined using seasonal MGIDI analyses. Roma and IISR Prathiba remained among the superior genotypes for both seasons, indicating that their favorable balance of fresh rhizome yield and quality traits (oleoresin, essential oil, and curcuminoid) was affected negligibly by seasonal microclimatic variations under greenhouse production. Their consistent ranking across independent seasonal evaluations strengthens their potential as promising genotypes for controlled-environment cultivation. Simultaneously, improvement of rhizome yield and quality traits is complicated by competing physiological source–sink relationships. In the present study, the selected genotypes (Roma, IISR Prathiba, and CO 3) maintained substantial genetic gains for curcuminoids, oleoresin, and essential oil content, while sustaining viable fresh rhizome yields. This stability reflects high physiological adaptability to protected conditions, where regulated solar radiation and moderated thermal fluctuations enabled these cultivars to express their genetic potential for both primary biomass and secondary phenylpropanoid accumulation. Reports from the past for selection efficiency of MGIDI and FAI-BLUP in identifying superior multi-trait ideotypes have been reported in various crop improvement studies (Gopi et al., 2026; Neeraja et al., 2026; Olivoto et al., 2017).

Table 10. Concordance of superior turmeric genotypes identified via MGIDI and FAI-BLUP across individual seasons and pooled datasets.

4 Conclusion

The present study revealed considerable genotypic variation for fresh rhizome yield, dry recovery, and quality traits (oleoresin, essential oil, and curcuminoids) of turmeric under greenhouse conditions across two seasons. The relatively greater contribution of genotype to the observed variation, together with high heritability and genetic advance for quality traits, indicates good scope for selection in future breeding programs. IISR Pragati and NDH 8 were promising genotypes for fresh rhizome yield (g clump⁻¹), Chhattisgarh Haldi 2 and Uttar Rangini for dry recovery (%), and Waigon Turmeric for oleoresin and essential oil content (%). IISR Pragati and IISR Prathiba also showed high and stable performance for curcuminoids (%) across seasons. The negative correlation between fresh rhizome yield and quality traits highlights the difficulty of improving both traits simultaneously and supports the use of multi-trait selection. Based on MGIDI and FAI-BLUP, the genotypes Roma, IISR Prathiba, and CO 3 had better results without compromising rhizome yield and quality (oleoresin, essential oil, and curcuminoids) with appreciable predicted gains for curcuminoids (33.56%), oleoresin (11.54%), and essential oil (9.37%). Overall, the study identifies superior genotypes for yield and quality traits as well as those with a more balanced performance with more traits, providing useful guidance for genotype selection and future improvement of turmeric under greenhouse production systems.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. During the preparation of this manuscript, We used generative AI solely for the purpose of language refinement, grammatical correction, and readability improvement.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fhort.2026.1899925/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Koundinya A.V.V., Indian Institute of Horticultural Research (ICAR), India

Reviewed by: Sampath Lavudya, Indian Institute of Oilseeds Research (ICAR), India

Okti Purwaningsih, PGRI University of Yogyakarta, Indonesia

Anil Bhushan, Sher-e-Kashmir University of Agricultural Sciences and Technology of Jammu, India