Avocado (Persea americana Mill.) is a high-value subtropical fruit crop cultivated extensively in regions with Mediterranean and tropical climates (Alcaraz et al., 2013). The black-skinned Guatemalan–Mexican hybrid avocado cultivar Hass holds significant commercial value, currently leading global avocado production (Schaffer et al., 2013). For instance, in 2018, ‘Hass’ accounted for approximately 97% of the total avocado market in the United States (Ballen et al., 2021), due to its market-preferred characteristics, including optimal fruit size, thick peel, buttery flavor, and smooth, creamy texture (Silva and Ledesma, 2014). In recent years, optimizing avocado productivity has become increasingly important due to rising market demand, climate variability, and labor-intensive management practices. Among the major challenges facing avocado production are excessive vegetative vigor and inefficient fruit set, which results from intense competition between flowers and developing fruitlets and the newly emerging vegetative shoots above the panicles (Symons and Wolstenholme, 1990; Lovatt and Salazar-Garcia, 2006). Controlling tree size is a significant challenge in the avocado industry. Avocado trees grow vigorously, which complicates pest management, harvesting, and other cultural practices, increasing labor and production costs. In spring, substantial energy reserves are directed toward vegetative shoot growth, potentially reducing the resources available for fruit development (Köhne and Kremer-Köhne, 1989). Avocado productivity is further limited by alternate bearing, defined as variation in yearly orchard yields (Ziv et al., 2014; Cohen et al., 2023). In recent years, mitigating the adverse effects of climate change has become a major concern for growers, driven by the rising frequency of extreme weather events (Zhang et al., 2012; Shapira et al., 2021; Lahak et al., 2024). These challenges negatively impact both yield stability and fruit quality.

Plant-growth regulators are among the most effective tools for controlling tree growth and enhancing yield, offering practical solutions to production challenges in established orchards. These compounds, when applied at appropriate concentrations and timings, can modulate the balance between vegetative and reproductive growth, reduce excessive shoot growth, enhance flowering, and improve fruit size and yield uniformity (Lovatt and Salazar-Garcia, 2006). Among the most commonly used plant-growth regulators in avocado are gibberellin (GA)-biosynthesis inhibitors, particularly the triazoles uniconazole (UNI) and paclobutrazol (PBZ) (Menzel and Le Lagadec, 2014; Brogio et al., 2018). These plant-growth retardants function primarily by inhibiting GA biosynthesis, resulting in reduced shoot elongation and potentially shifting resource allocation toward reproductive organs (Rademacher, 2000; Mitra, 2018). Numerous studies in ‘Hass’ avocado orchards have shown that applying these growth retardants during the flowering period by foliar spray can suppress vegetative growth, enhance flowering, and significantly improve fruit yield, shape, and size, thereby increasing growers’ economic returns (Adato, 1990; Wolstenholme et al., 1990; Erasmus and Brooks, 1998; Brogio et al., 2018). Another possible method of application is through the irrigation system, i.e., soil drenching. However, the appropriate concentrations for this method are still unclear; highly variable doses of UNI (as the active ingredient) have been used in past studies, ranging from 0.001 to 1 g L^-1 (Bórquez-Lillo et al., 2015). This is a critical factor, particularly with irrigation-based applications, because the compound is delivered directly to the root zone and transported acropetally through the xylem (Davis et al., 1988). Moreover, due to the long-lasting persistence of these chemical agents in the soil, careful management of concentrations is essential to preventing prolonged suppression of tree growth (Menzel and Le Lagadec, 2014; Bórquez-Lillo et al., 2015).

The effect of the chemicals on fruit set and yield is less consistent than that on shoot growth; there have been many cases of the growth regulators decreasing shoot growth without any subsequent benefit in terms of productivity (Menzel and Le Lagadec, 2014). Moreover, the efficacy and long-term physiological consequences of such treatments in avocado remain inadequately understood, particularly under field conditions over multiple seasons. This study aimed to evaluate the effects of a UNI-based growth retardant applied by soil drenching via the irrigation system at increasing concentrations (8, 12, and 16 mL tree^-¹) on vegetative growth, reproductive performance, gas-exchange characteristics, and fruit yield in mature avocado trees. UNI was selected because it is generally less persistent than PBZ, thereby avoiding long-term tree stunting. Specific objectives included (i) monitoring changes in trunk diameter, flowering intensity, and inflorescence bud density; (ii) assessing seasonal dynamics of photosynthesis, stomatal conductance, and chlorophyll content; and (iii) quantifying yield components and fruit-size distribution over two consecutive seasons. The findings provide insights into the potential of growth-retardant applications to enhance yield quality while managing vegetative vigor in avocado orchards.

This study offers comprehensive insights into the effects of soil-applied UNI on vegetative growth, flowering, and physiological responses of commercial field-grown ‘Hass’ avocado trees under a Mediterranean climate. During the experimental period, the orchard experienced multiple extreme weather events, including heat waves with maximum air temperatures exceeding 40°C (peaking at 45.8°C in August 2023), and cold spells with minimum temperatures dropping below 4°C in winter, reaching -2.6°C in February 2025. These fluctuations in temperature, typical of the region’s Mediterranean climate, imposed potential environmental stresses that could negatively impact tree physiology, yield, and fruit quality (Shapira et al., 2021; Alon et al., 2022; Lahak et al., 2024), may have contributed to the interannual variability observed in yield and gas-exchange measurements, and may have modulated the effects of UNI treatments. UNI blocks cytochrome P450-dependent monooxygenases, thereby inhibiting the oxidation of ent-kaurene into ent-kauronic acid in the GA biosynthesis pathway, reducing endogenous GA levels (Rademacher, 2000). This suppression limits cell elongation in vegetative shoots, resulting in reduced trunk growth and potentially increasing the proportion of assimilates available for reproductive development (Katz et al., 2003). Compared to untreated controls, UNI-treated trees exhibited reductions in trunk diameter and an increase in floral bud density ( Figures 2 , 3 ), reflecting a clear suppression of tree vigor (Reddy et al., 2003; Chernoivanov et al., 2022; Lahak et al., 2024). These outcomes are consistent with prior research demonstrating the growth-inhibitory effects of UNI. For example, foliar application of UNI markedly curtailed shoot elongation in greenhouse-grown avocado seedlings, with the extent of inhibition increasing with application rate (Köhne and Kremer-Köhne, 1989). A comparable reduction in shoot growth following UNI foliar treatment was observed in non-irrigated ‘Hass’ avocado trees (Brogio et al., 2018). UNI soil drenching led to decreased shoot length and trunk cross-sectional area in ‘Hass’ avocado trees grafted on Mexicola rootstock 9 months after container establishment, relative to their untreated counterparts (Bórquez-Lillo et al., 2015). In avocado, as well as in other species, it is desirable to control vegetative growth and reduce canopy size in commercial plantations (Mitra, 2018). Effective vegetative control is particularly advantageous in high-density orchards and to reduce labor-intensive management practices (Kohne and Kremer-Kohne, 1990; Köhne and Kremer-Köhne, 1992; Menzel and Le Lagadec, 2014). It may also prove advantageous in the context of dual-use agricultural systems integrating over-canopy solar panels, as it can help restrain excessive vegetative growth of the trees under the panels (Trommsdorff et al., 2022). However, caution is required to avoid long-term oversuppression, which could negatively impact the source–sink balance and productivity.

The results of this study demonstrate a clear promotive effect of UNI treatment on flowering during the first season, with this effect partially persisting into the second season despite an overall reduction in reproductive activity ( Figure 3 ). Flowering intensity was negatively correlated with trunk diameter ( Figure 8 ), suggesting that trees allocating more resources to vegetative structural growth may do so at the expense of reproductive development. Such a trade-off between vegetative vigor and reproductive growth is well-documented in avocado (Wolstenholme et al., 1990), and may be influenced by hormonal regulation, particularly gibberellin-mediated growth promotion at the expense of floral initiation. Similar findings have been reported in other studies; for instance, UNI soil applications increased both the proportion of floral to total buds and overall flowering intensity in young ‘Hass’ avocado trees (Bórquez-Lillo et al., 2015). In litchi, enhanced flowering was observed with PBZ treatments only when vegetative growth was suppressed for 1 to 2 months before panicle emergence (Menzel and Simpson, 1990). Similar responses have also been documented in mango and citrus (Mitra, 2018). It has been suggested that inhibition of GA biosynthesis, particularly under conditions of reduced photosynthetic activity, limits vegetative growth and reallocates assimilates toward the shoot apex, thereby promoting floral induction (Katz et al., 2003). In addition, the export of GAs from developing seeds to nearby buds has been implicated in floral suppression in several fruit tree species, including avocado (Guardiola et al., 1982; Salazar-García and Lovatt, 2000). Indeed, canopy spray application of GA₃ at concentrations of 25 or 100 mg L^-¹ in November, prior to the “on” bloom, resulted in a reduced number of inflorescences, an increase in vegetative shoot production, and a 47% decrease in yield compared to untreated control trees (Salazar-García and Lovatt, 2000). Moreover, research in mango and citrus suggests that GA may inhibit flowering by downregulating the accumulation of the FLOWERING LOCUS T (FT)-like gene, hypothesized to act as a phloem-mobile florigen signal encoding mRNA accumulation in both leaves and buds (Nakagawa et al., 2012; Goldberg-Moeller et al., 2013; Ziv et al., 2014; De Sousa Lima et al., 2016). In the shoot meristem, FT interacts with the bZIP transcription factor FD to promote the floral transition by inducing the expression of MADS-box genes, including SUPPRESSOR OF OVEREXPRESSION OF CONSTANS1 (SOC1), FRUITFULL (FUL) and APETALA1 (AP1). These MADS-box transcription factors, in turn, activate floral meristem identity genes such as LEAFY (LFY), which encodes a unique plant transcription factor essential for flower development (Mouradov et al., 2002; Bäurle and Dean, 2006; Ziv et al., 2014). These floral molecular pathways were not examined in the present study; future research should incorporate molecular analyses to enhance the mechanistic understanding of the UNI treatments.

Physiological measures demonstrated that UNI-treated trees generally maintain higher chlorophyll content and enhanced rates of carbon assimilation and stomatal conductance compared to controls during the first year ( Figures 4 , 5 ). These effects were less pronounced in the second year, likely due to the reduced persistence of the UNI effects on the trees. Although carbon assimilation and stomatal conductance were slightly negatively correlated with chlorophyll content ( Figure 8 ), the increased photosynthetic efficiency can be related to the increase in leaf chlorophyll content (Lahak et al., 2024). In longan, application of UNI increased leaf chlorophyll contents and net photosynthetic rate (Nie et al., 2001). In mango, soil application of PBZ also increased leaf chlorophyll content (Mitra, 2018), and enhanced chlorophyll content and carbon-assimilation rates were observed in oil palms treated with PBZ (Chai et al., 2023). Notably, transcriptome analysis in the latter study showed that PBZ treatment upregulates the expression of geranylgeranyl diphosphate reductase, a key enzyme involved in chlorophyll biosynthesis. In contrast, GA₃ application downregulated genes associated with chlorophyll synthesis, as well as those encoding components and assembly factors of the photosystem I and II super complexes and Rubisco (Chai et al., 2023). In water-stressed soybean plants, foliar application of uniconazole resulted in higher chlorophyll content and photosynthetic rates compared to untreated controls. This effect was attributed to UNI-induced enhancement of antioxidant enzyme activities, increased accumulation of proline and soluble sugars, and reduced levels of malondialdehyde (MDA), collectively contributing to improved membrane stability and sustained photosynthetic performance under stress conditions (Zhang et al., 2007). The mechanistic pathways by which UNI enhances leaf chlorophyll content and photosynthetic rate in avocado warrant further investigation. Still, the increased carbon-assimilation rates observed in our study may offer added value for carbon credit systems, enabling avocado producers to generate additional income by selling carbon credits to industries or producers with higher greenhouse gas emissions (Lokuge and Anders, 2022; Trouwloon et al., 2023). The positive correlations observed between carbon assimilation, stomatal conductance, flowering intensity, and fruit weight ( Figure 8 ) suggest that photosynthetic performance is a central driver of both reproductive potential and fruit development in avocado trees under the conditions of this study.

Although flowering intensity in the first year of the 12 and 16 mL tree^-1 treatments was higher than the control, fruit yield was significantly lower ( Figures 3 , 6 ). In the second year, yields were notably low in both the control and 8 mL tree^-¹ treatments, likely reflecting an alternate bearing pattern (Lovatt, 2010; Pochamreddy et al., 2024). Although the 12 and 16 mL tree^-¹ treatments were expected to enter an “on” year with higher yields, their production remained low, indicating a potential suppressive effect of UNI on yield. This decoupling of flowering and yield corroborates reports of high floral intensity not necessarily translating to greater yield (Rubinovich et al., 2024). Still, data from both years show a positive correlation between flowering intensity and fruit yield ( Figure 8 ). The lower yields in the UNI-treated trees may result from reduced carbohydrate reserves due to suppressed vegetative vigor, increased fruitlet abscission, or heightened sensitivity to pathogens such as Botryosphaeria under the adverse environmental conditions of the experimental plot ( Figure 1 ), which may contribute to branch dieback and, consequently, increased fruitlet drop in these treatments (Valencia et al., 2019). These parameters warrant investigation in future studies. These findings contrast with those of Adato (Adato, 1990), who reported that early spring spraying of avocado cv. Fuerte increases yields. Similarly, mango trees treated with PBZ consistently outperformed untreated trees, with average 5-year yield of 123 kg in treated trees compared to 91 kg in controls (Mitra, 2018). Here, in the first year, the 12 and 16 mL tree^–1 treatment resulted in significantly higher average fruit weight compared to the control ( Figure 6 ). Moreover, these treatments exhibited a more uniform distribution across weight categories, with a noticeable shift toward heavier fruit grades, highlighting a clear effect of the treatment on enhancing fruit size ( Figure 7 ). This outcome aligns with the commonly observed negative correlation between fruit weight and yield (Mickelbart et al., 2012). In addition, Stern et al. (2009) demonstrated that applying UNI at 30% full bloom in cherry trees leads to a reduction in fruit-set percentage, and therefore a significant decrease in crop load, resulting in a higher proportion of larger fruit. Given the strong economic dependence of avocado prices on fruit size, the commercial advantage of such treatment-induced size enhancement could potentially offset yield reductions, contingent on market conditions. Thus, the impact of UNI treatments on yield components warrants careful consideration, and a financial analysis should be conducted to assess their overall commercial viability.

The results from this study showed significant effects of both treatment and season (year) on physiological and yield-related parameters ( Supplementary Table S2 ). This indicates that treatments had consistent effects across years, despite considerable seasonal variability. Significant interactions for g_s, flowering intensity, inflorescence bud density, fruit yield, and fruit weight indicate that the magnitude of UNI responses varied between years, possibly reflecting seasonal climate effects and alternate bearing. Although this study sheds light on the phenological and physiological effects of UNI on avocado trees, the molecular and physiological mechanisms of this growth retardant in avocado remain unclear. It should be noted that the lack of measurements on soil nutrient availability and UNI mobility and degradation rates limits our ability to fully explain the physiological responses observed. Given that soil texture and chemistry can influence the mobility and persistence of triazole growth retardants (Deng et al., 2012; Menzel and Le Lagadec, 2014), future studies should integrate molecular and physiological investigation and soil monitoring to improve mechanistic understanding.

This study demonstrates that soil application of UNI is an effective tool for managing vegetative vigor and modulating reproductive development in mature ‘Hass’ avocado trees under a Mediterranean climate. Taking together, the findings support the strategic use of UNI to optimize canopy structure and fruit quality in commercial avocado orchards. However, the observed trade-offs between vegetative suppression, yield quantity, and fruit size highlight the need for precise dose calibration and long-term evaluation. To aid decision-making, we present a comparative summary of the main advantages and disadvantages of each treatment ( Supplementary Table S3 ). The 8 mL tree^-¹ treatment provided a balanced outcome- moderate vegetative control, improved physiological performance, and partially enhanced fruit size- while maintaining relatively high yields. The 12 and 16 mL tree^-¹ treatments produced larger fruit and improved size distribution in the first year, but were associated with yield reductions, especially in the second year. The untreated control maintained the highest yield in the first year but produced the smallest fruit with poor size-class distribution and no vegetative growth control. Overall, the 8 mL tree^-¹ treatment can be considered the most balanced approach within the scope of this study, particularly when both yield and fruit size are economically relevant. Future studies should further investigate the long-term cumulative effects of repeated UNI applications on tree architecture, photosynthetic capacity, and yield sustainability, as well as their integration into high-density planting systems and climate resilience strategies.