Abstract
Background and goals Grapevine rootstocks significantly influence scion vigor, yield, and fruit quality, making their appropriate selection vital for vineyard management. This study introduces two indices, the Rootstock Efficiency Index (REI) and Rootstock Stability Index (RSI), to assess rootstock performance based on scion vigor, yield, fruit quality, and vine balance.
Methods and key findings Sixteen years of data from Moscato bianco vines grafted onto 10 different rootstocks in Piedmont (Italy) were examined; metrics included yield, pruning weight, vine balance, fruit quality, and disease incidence. Significant rootstock effects were observed for yield, pruning weight, total soluble solids, and titratable acidity, while disease susceptibility was similar. Rootstock 420A had lower yield, organic acids, and pruning weight, while 157-11 had a higher Ravaz index, indicating greater efficiency but more environmental sensitivity. More stable rootstocks such as SO4 and 41B had lower cumulative yield than variable rootstocks like Du Lot and 1103P, suggesting a tradeoff between stability and productivity.
Conclusions and significance These results highlight the value of REI and RSI for rootstock selection. Because target levels for sugar accumulation, organic acids, and vine balance vary by cultivar and region, the indices use flexible weighting and normalization to account for local conditions. REI and RSI offer a standardized yet adaptable framework to benchmark performance against local expectations while supporting broader comparisons. As multiseason, multisite data accumulate, benchmarks can be refined to provide actionable guidance to improve vineyard efficiency, sustainability, and resilience across diverse environments.
Introduction
Grapevine rootstocks play an essential role in shaping the performance and sustainability of viticulture by influencing key traits such as disease resistance, vine vigor, yield, and fruit quality attributes, including sugar content and organic acid composition (Chen et al. 2024). As a result, selecting an appropriate rootstock has become a cornerstone of modern viticultural practices. Rootstocks affect grapevine physiology by modulating nutrient uptake, water-use efficiency, and photosynthetic performance, all of which contribute to enhancements in scion vigor and yield. For instance, improvements in nitrogen uptake capacity (Keller et al. 2001a, 2001b, Cochetel et al. 2017), water-use efficiency (Pou et al. 2022, Bernardo et al. 2025), and photosynthetic capacity (Soar et al. 2006) have been associated with rootstock selection, demonstrating their influence on grapevine performance. A growing body of research has evaluated the effect of rootstock on scion vigor, yield, and fruit quality traits (Walker et al. 2007, Kviklys et al. 2012, Kodur et al. 2013, Tecchio et al. 2022). However, the interaction between rootstock and scion often exhibits phenotypic variability such as higher yield accompanied by reduced sugar content, which complicates generalized evaluations of rootstock performance. In response to this challenge, efforts have been made to more systematically quantify rootstock contributions; for example, an index based on nitrogen uptake capacity has been proposed (Kulmann et al. 2020). While such indices provide a valuable starting point, they focus on isolated performance traits, emphasizing the need for more comprehensive frameworks that capture the multifactorial nature of rootstock performance. Environmental factors, including annual climatic variability and scion genotype, further compound the complexity of evaluating rootstock efficacy. It has been demonstrated that environmental conditions can exert effects on vine vigor, yield, and fruit quality that are 1.9 to 6.3 times greater than the influence of rootstock (Migicovsky et al. 2021), highlighting the critical importance of conducting long-term studies to ensure reliable and transferable insights. This variability underscores the necessity of adopting performance indices that can integrate multiyear data and account for environmental heterogeneity. To address these gaps, this study introduces two novel indices, the Rootstock Efficiency Index (REI) and the Rootstock Stability Index (RSI), designed to provide a holistic and accessible approach to evaluating grapevine rootstock performance. These indices synthesize data on scion vigor, yield, and fruit quality, offering a comprehensive yet practical means to assess rootstock performance under diverse environmental conditions. To validate their applicability, this study used a 16-yr data set collected in the Piedmont region of Italy, focusing on Muscat à petits grains blancs, a commercially significant scion variety that was grafted onto 10 rootstocks. As the trial was conducted in Italy, where the variety is locally known as Moscato bianco, this designation is used hereafter for consistency. Moscato bianco is used widely in sparkling wine production and is valued for its naturally high acidity and low pH, qualities that contribute to freshness in sparkling wines (Kerslake et al. 2014). Measured traits included yield components, pruning weight, vine balance (yield-to-pruning weight ratio), fruit quality metrics, and disease incidence, thereby capturing a wide array of factors that influence vine performance. By leveraging this long-term data set, the proposed indices can bridge the gap between trait-specific assessments and comprehensive performance evaluations. This approach not only provides a robust framework for rootstock selection, it also facilitates development of rootstocks tailored to meet the challenges of climate variability and evolving viticultural demands.
Materials and Methods
Vineyard site, plant material, and management
The experiment was conducted over 16 yr (1984 to 1999) in a vineyard located in Castiglione Tinella, Piedmont, Italy (44°43′N; 8°11′W; 405 m asl). The project was driven and funded by the industry association Vignaioli Piemontesi, which identified and selected the rootstocks in accordance with outcomes from their proprietary research initiatives. The vineyard, planted in 1981, featured rows oriented north-south on a south-facing slope of less than 2%. Vines were spaced at 1 m within rows and 2.2 m between rows. The soil was well-drained, uniformly deep (>1 m), and of mixed texture, with a pH of 7.8 and 1.2% organic matter content. The study evaluated Moscato bianco grapevines (clone R2) grafted onto 10 rootstocks: 420A, 5BB, SO4, Cosmo 2, Cosmo 10, 157-11, 41B, Du Lot, 1103P, and 140Ru (Table 1). Every experimental year, vines were trained to a vertical shoot-positioned system and cane-pruned using the Guyot method, leaving 14 buds per vine. Shoots were positioned loosely between two pairs of foliage wires at 35 and 70 cm above the cordon, which was at 80 cm above the soil. Drip irrigation was employed using pressure-compensated emitters with a flow rate of 2 L/hr, spaced 90 cm apart. Irrigation was applied as needed between budbreak and bloom to prevent water stress, withheld through mid-July to regulate shoot growth, and resumed weekly from mid-July to August. Irrigation frequency decreased during the cooler ripening period. Annual fertilization was applied at rates of 40 kg/ha nitrogen (N), 20 kg/ha phosphorus (P), and 60 kg/ha potassium (K). Harvest occurred on a single date each year, scheduled when fruit on the regionally predominant rootstock, SO4, reached 16 Brix. No shoot thinning, crop load adjustment, or leaf removal was implemented during the growing season. Canopies were mechanically hedged once or twice near veraison, depending on vegetative growth. These practices were consistent with standard regional management of Moscato bianco vines for sparkling wine production. As hedging reduces canopy biomass, this practice may have influenced pruning weight and consequently, the calculated Ravaz index (RI). While the effect of hedging was applied uniformly across all treatments, its potential to reduce differences in vigor expression among rootstocks cannot be excluded. Weed management was uniform across the trial and consisted of pre-emergence herbicide applications within the vine row, combined with mechanical cultivation and between-row mowing during the season.
The origin of the rootstocks examined in this study and their degree of tolerance to Phylloxera, drought, and salinity. Symbols represent relative scion vigor or resistance to Phylloxera and abiotic stress: 〇, high; △, moderate; ×, low (from Zhang et al. 2016). The study was conducted over 16 yr (1984 to 1999) in a vineyard located in Castiglione Tinella, Piedmont, Italy.
Experimental design and analysis
The trial employed a randomized complete block design with five replications. Each block consisted of 27 vines, including guard or border vines. The experimental treatments included 10 rootstocks and data were collected from one vine per rootstock per block, centrally positioned where possible and planted randomly within each replicate. This resulted in a total of 50 vines used for analysis, based on 10 rootstocks and five replications. Statistical analysis was conducted using a split-plot analysis of variance (ANOVA), evaluating interactions among year and rootstock effects. ANOVA, followed by the Tukey-Kramer multiple range test, Pearson’s correlation analysis, and heatmap were performed using R (ver. 4.3.1) and Statcel 4 (an add-in application in Microsoft Excel 2019 for Windows).
Measurements
Weather data were obtained from a nearby AgWeather-Net station operated by the Piedmont Region (https://www.arpa.piemonte.it) located 250 m from the experimental vineyard (44°04′N; 8°01′W; 410 m asl). For each of the 16 experimental years, daily maximum temperature (Tmax), minimum temperature (Tmin), and mean temperature (Tmean) were recorded and calculated for the growing season, defined as the period from April to October. Additionally, growing degree days (GDD) were calculated for the same period using a base temperature of 10°C, according to the formula GDD = ∑ (Tmean−10), where daily values less than 10°C were set to zero to exclude non-contributory data. These metrics were used to characterize the climatic conditions during the study period and to assess their potential impact on vine development and fruit composition. Vine growth, disease incidence, and fruit composition (including total soluble solids [TSS], titratable acidity [TA], and pH) were assessed at harvest. Botrytis bunch rot and sour (acid) rot were assessed independently on a per-vine basis. For each disease, both incidence and severity were recorded. Incidence was defined as the number of clusters per vine exhibiting visible symptoms of infection, while severity was defined as the estimated percentage of affected cluster area showing disease symptoms. All clusters on each vine were evaluated; no subsampling was conducted, ensuring comprehensive assessment across all replicates. Although the two diseases were scored separately, both Botrytis cinerea and sour rot symptoms could occur simultaneously on the same cluster, particularly under environmental conditions favorable to co-infection. Disease identification was based on visual and olfactory cues, with gray mold indicative of B. cinerea and sour rot characterized by berry browning, leakage, and acetic odor. To evaluate the multidimensional performance of the rootstock, we measured 14 traits related to yield, berry quality, and vine balance, from which we calculated three indices (Equations 2, 3, and 4). Vine vigor was assessed annually by measuring cane pruning weight during winter pruning. Yield data were collected from 10 vines within each replicate, with five replicates per rootstock. Yield per vine and cumulative yield per vine were calculated for each rootstock and experimental year. Thirty-cluster samples were collected randomly from each replicate just prior to harvest for quality analyses. Individual cluster weight was recorded. Five hundred grams of berries from each sample were juiced using a fresh squeeze juicer for chemical analysis. TSS were measured using an Atago refractometer and pH was determined using a Thermo Orion 370 pH meter. TA was quantified using an automatic titrator with an autosampler and control unit (Titroline 96). Harvest cluster rot was assessed as incidence (percentage of infected clusters per vine) and severity (percentage of infected berries per cluster). The percentage of berries with signs of sour rot and Botrytis disease was also evaluated. The net yield was calculated using the following formula:
where Yield is yield per vine and D is the proportion of diseased berries to total berries. In addition, REI was calculated using the following formula:
where REI is equal to 0 if RI < 2 × RIopt and TSS < 2 × TSSopt and TA < 2 × TAopt, and where RIopt, TSSopt, and TAopt represent the desired benchmarks of variety-specific RI as suggested by Howell (2001), and TSS and TA, respectively, as suggested by Di Stefano (1981). RI, TSS, and TA represent the actual value. In this study, the values of RIopt, TSSopt, and TAopt of Moscato bianco scion were set to 5, 17 Brix, and 8 g/L, respectively. The REI equation integrates multiple traits (RI, TSS, TA) to evaluate rootstock performance. The inclusion of “if” conditions ensures that deviations from the optimal values (RIopt, TSSopt, TAopt) are handled consistently regardless of whether they are above or below the benchmark, reflecting an equal penalty for any departure from the ideal benchmark. Since REI incorporates multiple traits, the “if” logic in Equation 2 ensures that each trait’s contribution to the index is unbiased and comparable. Next, to visualize the fluctuation of REI, ΔREI was calculated using the following formula:
where REIn denotes REI in year n (n = 1987 ÷ 1998).
Finally, the RSI was defined as the standard deviation of ΔREI as an indicator of the stability of rootstock performance over years, calculated using the following formula:
where
Results
Climatic conditions
During the 16-yr study period, annual maximum temperatures ranged from 29.8 to 32.8°C (Table 2), while annual minimum temperatures varied between −4.3 and 0°C. Despite these variations, no significant frost damage was recorded. GDD during the grapevine growing season (April to October) averaged 3068°C across the study period, with 13 of the 16 years falling within ±5% of this mean. The highest GDD was observed in 1984 (3294°C) and the lowest (2573°C), in 1994. Precipitation during the growing season averaged 554 mm, but annual variation was substantial, with a 3.3-fold difference between 1996, the wettest year (1013 mm), and 1991, the driest year (307 mm). These climatic variations highlight the environmental heterogeneity that the rootstocks experienced, providing a robust context for evaluating their performance under diverse conditions.
Weather traits during the experimental period conducted over 16 yr in a vineyard located in Castiglione Tinella, Piedmont, Italy. Tmax, daily maximum temperature; Tmin, daily minimum temperature; Tmean, daily mean temperature; GDD, growing degree days.
Yield and its components, berry quality, vigor, and vine balance
Data on yield and its components (cluster number and cluster weight), berry quality (TSS, TA, pH), vigor, and vine balance (pruning weight and RI) over 16 yr are summarized (Table 3). No significant interactions were detected between year and rootstock for any trait, however, significant main effects were observed individually for both year and rootstock. The magnitude of the mean squares indicated that all traits were more strongly influenced by year than by rootstock. Interannual variations were pronounced, with yield ranging from 7.5 to 16.6 t/ha (a 2.2-fold difference), cluster number per vine ranging from 12.3 to 20.7 (1.7-fold), and cluster weight ranging from 105 to 273 g (2.6-fold). Berry quality traits also exhibited variability between years, with TSS ranging from 15.2 to 19.9 Brix (1.3-fold) and TA ranging from 7.8 to 14.0 g/L (1.8-fold). Interestingly, must pH did not always correspond to TA although must sugar content and acidity were correlated (data not shown). Pruning weight ranged from 1.82 to 3.84 t/ha (a 2.1-fold difference), while the RI varied from 2.43 to 8.57. There was a weak positive correlation between pruning weight and TSS (r = 0.34, p < 0.001, n = 620). Comparisons among rootstocks revealed significant differences. Among the 10 rootstocks, 420A consistently showed the lowest values for yield, cluster number, TA, and pruning weight. In contrast, 157-11 exhibited the highest yield, cluster number, and RI. Du Lot displayed the highest TA, must pH, and pruning weight, but had the lowest RI. TA showed a moderate positive correlation with GDD and precipitation during the growing season (Supplemental Tables 1 and 2). Cluster number and pruning weight showed a moderate negative correlation with precipitation during the growing season (Supplemental Table 2).
Effect of rootstock on yield, fruit (must) quality, and vigor of Moscato bianco grafted onto 10 rootstocks in a vineyard located in Castiglione Tinella, Piedmont, Italy. Each value represents mean ± standard error (n = 5). TSS, total soluble solids; TA, titratable acidity; ANOVA, analysis of variance; df, degrees of freedom; MS, mean square.
Disease incidence, severity, and net yield
Strong annual main effects were observed for both Botrytis and sour rot incidence and severity, indicating that climatic conditions played a significant role in disease dynamics during the study period (Table 4). For sour rot-infected clusters, a significant main effect of rootstock was detected. Among the rootstocks, 420A showed a trend toward more severe sour rot than others, however, this trend did not reach statistical significance in the multiple comparison tests. Additionally, a marginal (10% level) interaction between year and rootstock was observed for sour rot severity. In some years, significant differences in severity were evident among rootstocks, but no consistent patterns emerged across the entire 16-yr data set, suggesting that the interaction between rootstock and environmental conditions strongly influenced sour rot severity in specific years. Net yield, calculated as total yield minus the weight of diseased berries, demonstrated significant variation among rootstocks. Rootstocks 157-11 and Cosmo 10 consistently had the greatest net yields, reflecting their strong overall performance and lower susceptibility to disease-related losses. In contrast, 420A exhibited lower net yields, consistent with its lower total yield and greater sour rot severity (Tables 1 and 2). These findings highlight how rootstock selection affects not only total yield but also the proportion of vinifiable, disease-free berries. Correlations between disease severity, yield, and climatic variables revealed additional insights. Sour rot severity was positively correlated with GDD, suggesting that warmer growing seasons tended to exacerbate the severity of this disease (Supplemental Table 1). Conversely, net yield was negatively correlated with GDD, indicating that warmer conditions may reduce the proportion of vinifiable berries, possibly due to both increased disease pressure and physiological vine impairment. These results underscore the importance of considering both disease resistance and environmental adaptability during rootstock selection. Rootstocks that balance high yields with reduced disease susceptibility (e.g., 157-11 and Cosmo 10) may offer significant advantages in regions prone to warmer growing seasons or variable climatic conditions.
Effect of rootstock on disease rates, disease severity, and net yield of Moscato bianco grafted onto 10 rootstocks in a vineyard located in Castiglione Tinella, Piedmont, Italy. Each value represents mean ± standard error (n = 5). ANOVA, analysis of variance; df, degrees of freedom; MS, mean square.
REI and RSI, indices of rootstock efficiency and stability
REI showed no significant interactions between year and rootstock, but main effects for both factors were detected (data not shown). Consistent with other traits analyzed, the effect of year on REI was far greater than that of rootstock. This result underscores the substantial influence of annual climatic variability on rootstock performance. A clear difference in REI was observed between 1988 to 1992 and 1997 to 1999 (Figure 1), suggesting that certain years experienced environmental conditions that either enhanced or hindered rootstock efficiency. To better illustrate year-to-year variability, the annual change in REI was calculated and expressed as ΔREI (Figure 2). The ΔREI values for the 10 rootstocks showed similar patterns in most years, with year-to-year changes ranging from −50 to +100%. However, certain periods such as 1993/1994 and 1996/1997 revealed divergent trends among rootstocks. During these years, rootstocks were divided into three distinct groups: those with increased REI, those with decreased REI, and those with minimal change. These variations highlight the differential response of rootstocks to specific environmental conditions, suggesting that some rootstocks are more resilient to climatic shifts, while others are more susceptible. RSI further emphasized the variability in REI among the rootstocks. Du Lot, 1103P, and 5BB exhibited high annual variation in RSI, indicating lower stability (Table 5). Conversely, 41B and SO4 showed much greater stability, maintaining relatively consistent RSI values throughout the study period. These findings point to the suitability of 41B and SO4 for vineyards where climatic conditions are unpredictable, as their performance is less influenced by environmental variability. In contrast, the high variability observed for Du Lot, 1103P, and 5BB suggests that these rootstocks may be more suited to regions with consistent climatic conditions, or where management practices can mitigate their sensitivity. Cumulative total and net yield for Moscato bianco scion grafted onto each rootstock over the 16-yr period are shown (Figure 3). Both total and net yields followed similar trends across rootstocks, allowing them to be classified into two distinct groups: high-yielding and low-yielding rootstocks. High-yielding rootstocks included 157-11 and Cosmo 10, which consistently demonstrated superior performance in terms of total production and net yield after accounting for disease-related losses. In contrast, Moscato bianco scion on rootstocks such as 420A consistently exhibited lower total and net yields, reflecting both lower overall productivity and higher susceptibility to disease-induced losses. The patterns of ΔREI and RSI combined with the yield data reveal important trade-offs between productivity and stability (Table 6). While high-yielding rootstocks such as 157-11 and Cosmo 10 delivered strong cumulative yields, their year-to-year REI variability was notable, suggesting that these rootstocks may be less predictable under fluctuating climatic conditions. Conversely, more stable rootstocks like 41B and SO4 maintained consistent REI values but tended to fall into the low-yielding category. This trade-off between stability and productivity is a key consideration for growers, depending on the specific goals and environmental conditions of their vineyard. Overall, these results demonstrate that rootstock performance is influenced not only by inherent traits but also by interactions with climatic variability. Rootstocks with greater stability like 41B and SO4 may be a reliable option in regions with unpredictable climate, while high yielding but less stable rootstocks like 157-11 and Cosmo 10 could be valuable in regions with more consistent growing conditions.
Annual change in the rootstock efficiency index (REI) of the 10 rootstocks examined in this study, conducted in a vineyard located in Castiglione Tinella, Piedmont, Italy.
Year-to-year change in the rootstock efficiency index (ΔREI) of the 10 rootstocks examined in this study, conducted in a vineyard located in Castiglione Tinella, Piedmont, Italy.
Annual variation in rootstock efficiency index (REI) of Moscato bianco grafted onto 10 rootstocks in a vineyard located in Castiglione Tinella, Piedmont, Italy. The REI for each rootstock in each year was calculated according to Equation 2.
Cumulative total and net yield of Moscato bianco grafted onto 10 rootstocks in a vineyard located in Castiglione Tinella, Piedmont, Italy.
Year-to-year change in the rootstock efficiency index (ΔREI; %) of Moscato bianco grafted onto 10 rootstocks in a vineyard located in Castiglione Tinella, Piedmont, Italy. ΔREI in each rootstock was calculated according to Equation 3.
Discussion
Influence of rootstock on yield, berry quality, vigor, and disease severity of Moscato bianco
In this study, the yield, berry quality, scion vigor, and disease severity of Moscato bianco grafted onto 10 different rootstocks with diverse genetic backgrounds were assessed over a 16-yr period, under varying climatic conditions (Tables 1 and 2). The results demonstrate that rootstock significantly influenced the yield of Moscato bianco, with 420A being the least productive rootstock and Cosmo 10 and 157-11, the most productive rootstocks (Table 3). This variation in yield was driven primarily by significant differences in the number of clusters per vine among the rootstocks. Some Moscato bianco scions on different rootstocks consistently produced more clusters, which directly contributed to greater overall yield, while others had fewer clusters, resulting in lower productivity. This suggests that the capacity of a rootstock to influence inflorescence initiation and fruit set plays a critical role in determining final yield, highlighting the importance of cluster number as a key yield component in rootstock performance evaluation. Previous studies have similarly highlighted the influence of rootstock on scion yield, attributing these effects to differences in water and nitrogen uptake efficiency, as well as photosynthetic capacity (Keller et al. 2001a, 2001b, Soar et al. 2006, Pou et al. 2022).
Rootstocks also had a notable impact on Moscato bianco fruit quality measures, particularly TA. Among the tested rootstocks, scions on 420A produced berries with the lowest TA, while 5BB, Du Lot, and 1103P were associated with higher TA (Table 3). The influence of rootstock on fruit acidity of the scion was consistent with findings reported in prior studies (Keller et al. 2001a, Kodur et al. 2013). Variations in fruit acidity were attributed primarily to differential potassium (K⁺) uptake by the root system, which subsequently affects the pH and TA of the berries (Kodur 2011). Interestingly, a study using Red Alexandra as a scion reported that TA in vines grafted onto 420A was comparable to other rootstocks (Cheng et al. 2017). This discrepancy underscores the complex interaction between rootstock and scion, where the same rootstock can have different effects on berry quality, depending on the rootstock-scion combination (Keller et al. 2012, Kidman et al. 2013, Clingeleffer et al. 2019, Migicovsky et al. 2021). This finding highlights the importance of scion-specific rootstock evaluations when interpreting results or making recommendations for rootstock selection.
In addition to yield and fruit quality, rootstock significantly influenced scion vigor. Rootstocks such as 5BB and Du Lot promoted greater scion vigor, while 420A was associated with reduced scion vigor, consistent with previous findings (Zhang et al. 2016). The interplay between scion vigor and other traits such as yield and berry quality further highlights the importance of selecting rootstocks based on specific vineyard goals. The incidence of sour rot also varied among rootstocks (Table 4). For instance, Cosmo 10 showed a lower incidence of sour rot compared to 420A in some years, suggesting a potential influence of rootstock on disease resistance or tolerance. Variations in scion disease incidence across different rootstocks may be attributed to differences in fruit compactness and vine vigor. Previous studies have shown that rootstock selection can influence fruit compactness (Keller et al. 2001a). Rootstocks influence fruit compactness by altering vine vigor, nutrient and water uptake, and hormonal balance, all of which affect berry size and how tightly berries are arranged within a cluster. More vigorous rootstocks often produce larger berries and more compact clusters, while less vigorous rootstocks tend to result in looser, more open clusters that are less prone to disease. Scion vigor and cluster compactness are factors known to affect susceptibility to B. cinerea. More vigorous rootstocks tend to produce denser canopies, creating microclimatic conditions favorable for fungal infection. However, in the present study, rootstock 420A exhibited a higher incidence of disease despite its low scion vigor (Tables 3 and 4), which contrasts with earlier findings. This inconsistency suggests that fruit compactness, although not assessed in this study, may have played a critical role in modulating disease susceptibility. However, the differences in sour rot severity among rootstocks were relatively small, and their impact on net yield mirrored the patterns observed for total yield (Tables 3 and 4). These findings indicate that while rootstock may influence disease incidence, other factors such as environmental conditions likely play a more significant role. Notably, while rootstock significantly affected all measured traits, the effect of year was consistently greater than that of rootstock. This aligns with findings from prior studies, emphasizing the dominant influence of climatic variability and the necessity for long-term evaluations to accurately assess rootstock performance (Kidman et al. 2013, Kulmann et al. 2020, Migicovsky et al. 2021).
These results reinforce the importance of multiyear trials in providing robust conclusions about rootstock efficiency and stability under diverse environmental conditions. The relationship between traits such as sour rot severity, net yield, and climatic variables further underscores the importance of environmental factors in rootstock performance. Sour rot severity showed a positive correlation with GDD, while net yield exhibited a negative correlation with GDD (Supplemental Table 1). Additionally, TA demonstrated a positive correlation with precipitation (Supplemental Table 2). These findings align with prior research suggesting that reduced sunlight associated with more precipitation can suppress acid decomposition, leading to higher TA in the berry. Overall, this study provides valuable insights into the long-term performance of rootstocks and emphasizes the need to balance productivity, berry quality, and environmental adaptability. The findings stress the importance of understanding rootstock-specific responses to climatic variability, which can inform the selection of rootstocks tailored to the specific goals of vineyard management in diverse growing regions.
Role of the REI in rootstock research
Rootstocks exert multifaceted effects on scion yield, fruit quality, and scion vigor, making evaluation of their overall performance a complex and challenging task. To address this, we developed the REI as a comprehensive metric for rootstock performance. Conceptually defined as “the efficiency with which rootstocks promote high-quality berry production while maintaining vine balance,” REI is calculated using Equation 2, which incorporates key traits such as net yield, TSS, TA, and RI by penalizing deviations from desired benchmark values. Although RI does not influence immediate market value directly, it was incorporated into the REI due to its importance in assessing vine balance, a key determinant of long-term yield sustainability and fruit quality (Howell 2001). The multiplicative structure of the REI was specifically chosen to ensure that suboptimal performance in any single component, whether yield, fruit composition, or vine balance, results in a proportional reduction in the overall index. This approach reflects the inherent interdependence among these viticultural traits, thereby providing a more integrated measure of vine performance over multiple seasons, and ensures that the REI reflects a balance between productivity, berry quality, and vine vigor.
Among the 10 rootstocks studied, 157-11 exhibited the highest REI (9.1 ± 0.57), primarily due to its high net yield, which compensated for its slightly higher-than-optimal scion vigor (Figure 1). This result suggests that 157-11 effectively balances productivity with acceptable levels of fruit quality and scion vigor, making it a high-performing rootstock in this study. In contrast, 420A showed the lowest REI (7.3 ± 0.51), reflecting its low yield despite having average TA and scion vigor values that were closer to the desired benchmarks (Table 3; Equation 2). These results highlight the trade-offs inherent in rootstock performance, where superior outcomes in one trait may not compensate for deficiencies in others. The REI offers a significant advantage by consolidating the performance of rootstocks across multiple traits into a single, interpretable value. This simplification allows for the rapid comparison of complex data sets and facilitates decision-making for rootstock selection. For instance, growers and researchers can quickly identify high-performing rootstocks like 157-11 based on their balanced contributions to yield, berry quality, and vine vigor, while recognizing the limitations of lower-performing rootstocks like 420A.
Moreover, the REI framework has practical applications beyond comparative evaluations. By quantifying rootstock performance in a standardized manner, the REI can guide the development and improvement of new rootstock varieties. Breeders and researchers can use this index to evaluate candidate rootstocks in trials, refine their breeding objectives, and streamline the selection process for varieties that meet specific viticultural goals. Additionally, the REI provides a robust tool for long-term monitoring of rootstock performance, enabling growers to assess how rootstocks respond to changing climatic or management conditions over time. Importantly, the interpretation of REI values must consider the underlying components that contribute to the index. For example, while 157-11 achieved the highest REI due to its superior net yield, its slightly higher scion vigor suggests that careful management practices may be needed to optimize its long-term performance. Conversely, while 420A had the lowest REI, its closer-to-optimal acid and scion vigor values suggest potential suitability for low-yield scenarios, where fruit quality and vine balance are prioritized.
Overall, the REI not only simplifies rootstock evaluation, it also provides actionable insights for growers, researchers, and breeders. Its ability to integrate complex, multifactorial data into a single value ensures that critical tradeoffs in rootstock performance can be easily identified and addressed, making it an invaluable tool for modern viticulture. In its current formulation, the REI reflects the relative variability of its component traits, such that traits with greater year-to-year fluctuation (e.g., TA) may disproportionately influence the index compared to more stable traits (e.g., TSS), regardless of their actual commercial relevance. This introduces a limitation inherent in the unweighted framework. However, the REI is designed to be adaptable: weighting factors can be explicitly incorporated to adjust the contribution of each trait in accordance with its agronomic or economic significance. Such customization would allow the REI to be tailored to specific cultivars, viticultural regions, or market objectives. The present study serves as a conceptual validation of the index’s structure.
Evaluation of the rootstocks by combining REI, RSI, and cumulative yield
The REI, like other metrics, exhibited substantial year-to-year variation during the study period (Figure 1). To better understand the stability of REI, year-to-year changes were evaluated by calculating and visualizing the ratio of REI to its value in the previous year (Equation 3; Figure 2). Furthermore, the range of variation was quantified by normalizing these changes using Equation 4 (Table 5). The analysis revealed significant annual fluctuations in the REI of some rootstocks (Du Lot, 1103P, and 5BB), while SO4 and 41B demonstrated remarkable stability (Table 5).
When examining the genetic background of these rootstocks in relation to their RSI, a measure of annual variation in performance, several important patterns emerged. Du Lot, which exhibited the highest RSI and therefore the greatest instability, is pure Vitis rupestris. Similarly, hybrids involving Vitis berlandieri × V. rupestris, such as 1103P and 140Ru, also showed substantial REI fluctuations (Tables 1 and 5). This suggests that V. rupestris-derived rootstocks may be more sensitive to environmental conditions due to their vigorous growth characteristics. This aligns with prior research, as V. rupestris is well-documented for its high scion vigor and susceptibility to environmental variability (Dias et al. 2017). Conversely, the most stable rootstock identified in this study was 41B, a hybrid of Vitis vinifera and V. berlandieri. The moderate scion vigor of 41B likely contributes to its stability, making it less prone to environmental stress-induced fluctuations. In this context, scion vigor plays a pivotal role in rootstock performance. Excessive scion vigor, as seen in Du Lot, can surpass the optimal threshold for vine productivity. While moderate scion vigor supports high yields, excessive vegetative growth can divert resources away from fruit development, leading to a decline in yield (Dias et al. 2017). The findings of this study support this concept, as Du Lot, despite its robustness, likely exceeded the optimal scion vigor threshold, resulting in yield instability (Table 3).
Interestingly, the total cumulative yield did not consistently align with the REI or RSI values. For instance, Du Lot, 41B, and 157-11 had the greatest total cumulative yields (Figure 3A), yet these rootstocks varied in their efficiency and stability: Du Lot had high vigor and low stability (high RSI), 41B showed high stability (low RSI) but lower REI, and 157-11 achieved both high REI and moderate RSI. Conversely, 420A and SO4 had the lowest cumulative yields despite SO4 exhibiting strong performance stability. This suggests that high yield alone does not necessarily indicate high efficiency or stability, as yield itself does not capture the full picture of stability or efficiency. These results highlight the complexity of rootstock performance evaluation; using integrated indices like REI and RSI offers a more nuanced approach to evaluating rootstock performance.
A closer analysis of the study’s findings underscores the exceptional performance of 157-11. This rootstock achieved a high total cumulative yield (Figure 3A) and a high REI while maintaining a moderate and stable RSI. These characteristics make 157-11 the optimal rootstock for Moscato bianco under the conditions of this study. In the Piedmont region, the most used rootstocks for Moscato bianco are SO4 and Kober 5BB, due to their proven adaptability to local soil conditions and consistent agronomic performance. In contrast, 157-11C is seldom used in commercial vineyards despite its potential, and it remains largely underrepresented in regional viticultural practice. The high REI of 157-11 reflects its ability to effectively balance yield, fruit quality, and vine balance. The moderate RSI of 157-11 further indicates that this rootstock maintains consistent performance across variable environmental conditions, minimizing year-to-year fluctuations (Table 7). The performance of 157-11 demonstrates its potential to achieve a favorable balance between maximizing yield and minimizing annual variability (Table 8). These results demonstrate the importance of using metrics like REI and RSI in tandem with cumulative yield data to identify rootstocks that meet the dual goals of productivity and stability in viticulture.
Rootstock stability index (RSI) of each rootstock examined in this study, conducted in a vineyard located in Castiglione Tinella, Piedmont, Italy.
Cumulative total and net yielda (t/ha) of Moscato bianco grafted onto 10 rootstocks in a vineyard located in Castiglione Tinella, Piedmont, Italy.
Conclusions
This 16-yr study offers a novel evaluation of 10 grapevine rootstocks grafted with Moscato bianco under varying climatic conditions, emphasizing the value of long-term data in assessing rootstock-scion-environment interactions. Two novel indices—REI and RSI—were developed to assess rootstock performance. REI integrates yield, fruit quality, and vine balance into a single efficiency metric, while RSI measures performance stability across years. These tools help decouple rootstock effects from climatic variability. Among the rootstocks, 157-11 proved most efficient, with high cumulative yield, top REI, and moderate RSI—highlighting its strong balance of productivity and stability. In contrast, 41B and SO4 were more stable but less productive, making them better suited to variable climates. Vigorous types like Du Lot and 1103P showed high variability (high RSI), likely due to exceeding optimal vigor thresholds. The correlations between climatic traits and rootstock performance further emphasize the influence of environmental variability. For instance, sour rot severity was positively correlated with GDD, while net yield decreased with higher GDD, suggesting that warmer seasons may exacerbate disease pressure and physiological stress on vines. These findings highlight the critical importance of integrating both environmental adaptability and inherent physiological traits in rootstock selection. The study demonstrates that long-term field-based evaluations are essential to identify rootstocks capable of maintaining high productivity, vine balance, and resilience under environmental variability. The REI and RSI developed here offer a robust and scalable framework for rootstock assessment, translating complex physiological and agronomic data into practical decision-making tools. These indices enable growers, researchers, and breeders to make informed selections based on multidimensional performance criteria, particularly in the context of climate change and shifting production goals. By advancing a data-driven approach to rootstock evaluation, this research supports development of sustainable viticultural strategies, improves vineyard adaptability and efficiency, and enhances the industry’s capacity to respond proactively to climatic and economic challenges.
CRediT Authorship Contributions
YK, RB, SS, and PS: Data Curation; YK, SS, and PS: Formal Analysis and Writing – Original Draft; RB: Investigation and Methodology; VN, SS, and PS: Writing – Review & Editing; PS: Conceptualization
Supplemental Data
The following supplemental materials are available for this article in the Supplemental tab above:
Supplemental Table 1 Correlation analysis between each trait/rootstock combination and growing degree days (GDD). The table is color-coded according to the magnitude and direction of the correlation coefficient, with warm colors (e.g., red) indicating strong positive correlations, cool colors (e.g., blue) indicating strong negative correlations, and intermediate colors representing weak or near-zero correlations. The study was conducted over 16 yr (1984 to 1999) in a vineyard located in Castiglione Tinella, Piedmont, Italy. TSS, total soluble solids; TA, titratable acidity; REI, rootstock efficiency index.
Supplemental Table 2 Correlation analysis between each trait/rootstock combination and precipitation. The table is color-coded according to the magnitude and direction of the correlation coefficient, with warm colors (e.g., red) indicating strong positive correlations, cool colors (e.g., blue) indicating strong negative correlations, and intermediate colors representing weak or near-zero correlations. The study was conducted over 16 yr (1984 to 1999) in a vineyard located in Castiglione Tinella, Piedmont, Italy. TSS, total soluble solids; TA, titratable acidity; REI, rootstock efficiency index.
Data Availability
The data underlying this study are available on request from the corresponding author.
Footnotes
This work was partially supported by JST SPRING (Grant number: JPMJSP2119). This publication is part of the project NODES which has received funding from the MUR – M4C2 1.5 of PNRR funded by the European Union - NextGenerationEU (Grant agreement no. ECS00000036). The authors thank Dr. Albino Morando (Vit.En. Di Morando Albino & C. S.A.S.) for generously providing the historical data series used in this study. We also express our gratitude to Dr. Daniele Eberle (Consulting Agronomist) for his critical reading of the manuscript and valuable suggestions.
Kita Y, Baldovino R, Novello V, Savoi S and Sabbatini P. 2026. Evaluating rootstock performance: Efficiency and stability insights from a 16-year Moscato bianco study. Am J Enol Vitic 77:0770008. DOI: 10.5344/ajev.2026.25034
By downloading and/or receiving this article, you agree to the Disclaimer of Warranties and Liability. If you do not agree to the Disclaimers, do not download and/or accept this article.
- Received July 2025.
- Accepted January 2026.
- Published online April 2026
This is an open access article distributed under the CC BY 4.0 license.
References
Vol 77 Issue 1
Jump to section
Related Articles
Cited By...
More from this TOC section
Similar Articles
