The almond tree (Prunus amygdalus (L.) Batsch = P. dulcis (Mill.)) is a dominant global tree nut crop, accounting for 31% of total nut crop production and with 700 thousand hectares planted worldwide (Iglesias and Torrents, 2022). Its production has risen substantially worldwide, increasing by 60% in the last ten years (USDA, 2024). The almond tree is widespread throughout the Mediterranean basin and is part of the agricultural biodiversity and food production systems, especially in Italy and Spain (Fernández i Martí et al., 2009; Socias i Company et al., 2009; Alhajj Ali et al., 2023). Sustainability and efficiency are key factors in new agriculture, in which every cultivation technique must be innovated to improve the soil-plant-environment system and the efficient use of the input (e.g. irrigation water management and fertilizers) (Morugán-Coronado et al., 2020; Garofalo et al., 2024). A shift in cropping systems toward Sustainable Intensification (SI) pathways is imperative. SI is broadly defined as increasing agricultural production on existing land, reducing environmental impacts, and enhancing the provision of environmental services (Cano et al., 2023; FAO, 2024). The demand for innovation in fruit tree cultivation has led to the development of highly efficient and sustainable planting systems that move beyond the traditional open vase tree shape, already studied and applied to other species, such as olive, in which intensification from low to Super High-density (SHD) presented the greatest improvements, both for yield and Net Ecosystem Productivity (Mairech et al., 2020). These new agronomic models for almond production focus on growth-controlling rootstocks, enabling orchard intensification with narrow rows and canopies, thereby enhancing input efficiency, particularly labor and mechanization, while promoting orchard precocity (Ghelfi and Palmieri, 2015; Iglesias et al., 2021; Maldera et al., 2023). The SHD planting system, also known as edge or “sustainable and efficient systems” (SES), has gained popularity, primarily promoted by the Agromillora group nursery in Spain (Casanova-Gascón et al., 2019; Iglesias et al., 2021; Maldera et al., 2021a; Iglesias and Torrents, 2022). This new system has already been applied to other crops, such as olives, since 1994 (Tous et al., 2010; Connor et al., 2014), in which higher tree density per hectare and full mechanization of all cultivation operations are used. The system resulted in improved water use efficiency (Romero-Trigueros et al., 2019) and in decreased environmental footprints with respect to traditional cropping systems (Pellegrini et al., 2016; Camposeo et al., 2022). Maximizing branching density within the canopies of SHD trees is crucial to establish and maintain productive potential, especially for hedge-pruned trees (Rosati et al., 2013; Vivaldi et al., 2015; Lodolini et al., 2017; Montesinos et al., 2023).

To optimize light interception and achieve uniform crop yield, major temperate tree fruit industries have adopted higher densities in growing systems with smaller trees, which offer new possibilities for almonds (Thorp et al., 2021). Efficient light interception is critical for photosynthesis, and hence for all physiological processes of the plant; optimal tree architecture helps maximize the interception of sunlight, enhancing photosynthesis rates and overall tree growth (Maldera et al., 2021a, 2023). Tree architecture has been studied extensively for apple orchards. Rootstock effect was observed as a key factor in modifying the canopy architecture of these orchards (Tworkoski and Miller, 2007; Domi et al., 2012; Fallahi et al., 2018). Previous research has been also done on olive tree architecture and its relationship with micropropagation (Neri et al., 2020) or different planting systems. Rosati et al. (2013) observed that high branching and small diameters are important architectural characteristics to increase yield efficiency and affect the cultivar suitability for SHD olive orchards, while Lodolini et al. (2017) focused attention on the different varietal architectural characteristics of the canopy and primary branches in olive cultivars (Carella et al., 2022).

The management of almond trees significantly affects the distribution of shoots and leaf area as well as the mechanization of pruning. This technique, as observed in SES growing systems, can offer substantial benefits by ensuring a more consistent product through uniform pruning practices (Franzen and Hirst, 2016). Understanding how different scion/rootstock combinations respond to different planting systems is essential for optimizing orchard management. Moreover, selecting appropriate cultivars remains one of the most effective strategies for ensuring the agronomic and economic viability of high-density cropping systems (Vivaldi et al., 2015). A whole range of different cultivars could be used in SHD orchards, such as Guara Tuono (Dicenta et al., 2015; Felipe, 2017) and Lauranne^® Avijour (Grasselly et al., 1990; Legave et al., 1997), leading to different responses (Maldera et al., 2021b). A size-controlling rootstock allows the growth of any cultivar, including autochthonous ones like Filippo Cea, noted for its seed health benefits (Monastra et al., 1982; Summo et al., 2018). Considering the importance of scion-rootstock combinations in almond orchards, it becomes crucial to explore their influence on the architecture of almond trees within the context of SHD orchards (Montesinos et al., 2021a, 2021b, 2022). Although several studies have investigated pruning and canopy architecture in several fruit tree crops, including almonds (Negrón et al., 2013; Negrón et al., 2015), limited research has focused specifically on the effect of scion-rootstock combinations on tree architecture in SHD almond orchards (Montesinos et al., 2023). Cultivars, pruning practices, and row orientation influences the architectural development of almond trees, which in turn impacts light interception, yield potential, and overall orchard management (Maldera et al., 2021a, 2023; Negrón et al., 2013; Negrón et al., 2015).

This research aimed to fill this knowledge gap by examining how different almond cultivars grafted onto a new growth-controlling rootstock influenced tree architectural and growth parameters in a young SHD orchard. This open field research intended to provide valuable insights for development and application of new tools and methods for agricultural systems. It planned to help in optimizing the design and management of new almond orchards, contributing to increased productivity and sustainability in almond growing.

Figure 1 showed the typical mediterranean climate, characterized by an average annual rainfall of 380 mm/year; the lowest value of monthly rainfall was observed in May for 2021 (2.2 mm) and April for 2022 (8.0 mm), while the highest value was rainfall was in November for both years (82.6 and 93.9 mm respectively).

The results of this study show that the cultivar has a significant impact on the canopy growth ( Figure 2 ). The cultivar Guara Tuono consistently had the highest tree height (TH) out of the three cultivars. This difference in TH was maintained throughout the measurement period. Similarly, Tuono had the greatest canopy thickness (TT), followed by Filippo Cea and Lauranne. However, in the last measurement, Lauranne surpassed the other two cultivars in TT.

Canopy width (TW) showed similar patterns to TT, with Filippo Cea having the highest values in the first two measurements. However, from the third measurement, Tuono also reached the same width as Filippo Cea. Lauranne had the lowest TW initially but reached the same value as the other cultivars in the last measurement. This behavior was different from previous results, in which Lauranne Avijour showed higher canopy dimensions (Maldera et al., 2021b). The cultivar Filippo Cea had the highest canopy volume (VC), followed by Tuono, while Lauranne had the lowest volume. This trend in VC was consistent with the patterns observed in TH and TT. These results were surely influenced by mechanical pruning operations, equally performed throughout the orchard, that imposed certain dimensions to the three cultivars. It is already known that mechanical pruning is imperative in SHD orchards to fill and reduce the number of holes in the canopy and the costs of the operation (Fumey et al., 2011; Ferguson et al., 2012; Negrón et al., 2015; Vivaldi et al., 2015).

In terms of tree vigor, Filippo Cea consistently showed the highest values, indicating it was the most vigorous cultivar, while Lauranne exhibited the lowest vigor ( Figure 3 ). These data are in contrast with Maldera et al., (2021b), in which Lauranne showed more vigorous trees than Guara Tuono, which is in line with Montesinos et al. (2023) where Lauranne grafted on Rootpac^® 20 showed low vigor and lower canopy dimensions. Ben Yahmed et al. (2016) observed no differences in terms of vigor between Lauranne and Guara. Ben Yahmed et al. (2022) observed an opposite behavior in the same region, with Lauranne more vigorous than Guara and Tuono. In this case, as for other species, differences in pedoclimatic conditions and scion-rootstock combinations could lead to these responses (Warner, 1991; Weibel et al., 2003; Gullo et al., 2014; Lodolini et al., 2017; Fallahi et al., 2018; Ben Yahmed et al., 2022).

Canopy architecture plays a key role in the development of trees structure (Asai et al., 1996; Lauri et al., 2009, 2011; Jiménez-Brenes et al., 2017). The screening index (IS) revealed that Guara Tuono had the highest percentage of trees in class 3, followed by class 2 ( Figure 4 ). Filippo Cea and Lauranne had the highest percentage in class 4. These findings seem promising, considering a different behavior of Guara Tuono: this cultivar struggles to close the production wall, having a less open shape, with more gaps and less branching. The IS played an important role, showing a not well-adapted canopy structure of cv. Tuono. Montesinos et al. (2023) observed the same situation, with a better ramification of Lauranne, in all the different scion-rootstock combinations.

The branches parameters showed some differences among the cultivars. Guara Tuono had the highest percentage of trees in classes 2 and 3 for branch number (Br_N) ( Table 2 ). Filippo Cea had higher values in class 4 for branch diameter (Br_D), while Guara Tuono and Lauranne had similar distributions. Branch length (Br_L) showed higher values in class 3 for Filippo Cea and Lauranne, while Guara Tuono had a higher percentage in class 2. Branch angle (Br_A) differed among the cultivars, with Filippo Cea having more trees in classes 1 and 2, and Guara Tuono having more trees in classes 3 and 5. Montesinos et al. (2023) showed exactly the same values: Guara Tuono exhibited lower and shorter branches than Lauranne Avijour, while no comparisons can be made with Filippo Cea, because no bibliography references have been done yet on this cultivar.

For brachyblast parameters, all cultivars had fewer trees in classes 4 and 5 for brachyblast number (Bb_N) ( Table 3 ). Tuono had the highest percentage of trees in classes 2 and 3, while Lauranne and Filippo Cea had more distributed values, mainly in classes 1 and 2. Brachyblast length (Bb_L) and Brachyblast diameter (Bb_D) showed similar distributions among the cultivars, with higher values in class 1 for both parameters. These results are in contrast with previous ones, in which no relevant differences were observed (Montesinos et al., 2023).

Subterminal parameters were influenced by the cultivar. Tuono had the lowest number of subterminals (Su_N), mainly in classes 1 and 2. Filippo Cea had higher percentages in classes 2 and 3, while Lauranne had 20% of trees in class 4, indicating a higher ramification of the canopies ( Table 4 ). Subterminal diameter (Su_D) showed higher percentages in classes 3 and 4 for Tuono, while Lauranne had the most trees in class 2. For subterminal length (Su_L), Tuono had higher percentages in the higher classes, while Filippo Cea had more trees in the lower classes. Lauranne showed mid-values, with slightly higher values in class 2 than class 4. These findings are totally in line with IS data, observing a reiterated behavior of the Lauranne and Filippo Cea to be more branched, while Tuono is the one with the lowest and longest branches.

No statistical differences were observed for branch length, flower buds, and wood buds among the cultivars. However, the cultivar Tuono had higher values for both the total number of buds and bud density than Filippo Cea and Lauranne.

After the observation of all the examined parameters, an overview architectural model of the three cultivars can be done.

Tuono had lower growth than Filippo Cea but higher than Lauranne. It showed the lowest number of branches and, consequently, the highest number of pore spaces, leaving more holes in the hedgerow. These parameters are strictly related to its branching habit. Brachyblasts are more frequent with respect to the other cultivars but longer and bigger. Subterminals are the fewest, the biggest, and the longest, making more pruning necessary during the year. With the same length, the highest number of buds and, consequentially, bud density have been recorded, also confirmed by the correlations, in which the total number of buds was highly correlated to the number of wood buds and bud density.

Filippo Cea showed the highest growth and, more importantly, the highest vigor. It is characterized by the best screening index, due to an excellent distribution and occupation of space. This is related to its branching habit, which is characterized by more open branches, even if they are lower in number and with medium length and diameter. Average number and lowest diameter and length values characterized the brachyblasts. Subterminals are characterized by a medium number but the lowest diameter and length. Lower buds per shoot and bud density were observed, with the highest correlation between wood buds and total bud number.

Lauranne exhibited the lowest tree vigor. A medium screening index was observed, with a good distribution of the branches in the space. High number, average diameter, highest length, and a different angle characterized the branches. The fewest number of brachyblasts, with an average diameter and length, were observed. Subterminal shoots were the most numerous and showed the smallest diameter. A good number of buds per shoot and an average bud density were also observed. Interesting correlations were also observed between total buds and wood, flower buds, and shoot length.

In an SHD orchard, the best ideotype is a small tree with high branching density and small trunk diameters (Rosati et al., 2013; Vivaldi et al., 2015; Lodolini et al., 2017; Thorp et al., 2021; Montesinos et al., 2023). The cv. Lauranne seemed to fill the different spaces in the canopy better and faster, mostly with the lowest tree vigor of all the cultivars involved. This scion/rootstock combination could help to reach the goal of sustainable intensification, which has been the main pursuit of recent years (Cano et al., 2023).

The architectural approach can help growers to reach the goal of a full trained hedgerow better and faster, leading to the goal of a Sustainable and Efficient System (SES). The cultivar choice strongly influenced canopy dimensions, vigor, and architectural traits. Different cultivars exhibited variations in tree height, canopy thickness, width, volume, vigor, and various architectural traits such as branch parameters, brachyblasts parameters, and subterminal shoots parameters. Cv. Lauranne showed the best results in growth and in distribution of the branches, while Tuono seemed to be the worst distributed one. Cv. Filippo Cea showed the highest volume of the canopy, showing an intermediate behavior for the other architectural parameters. These findings provide valuable information for the development and application of tools and methods for agricultural systems design, assessment, and management. This work can help almond growers and breeders in selecting cultivars based on desired architectural traits and vigor, enhancing the productivity and sustainability of SHD orchards. Future research will explore the relationship between canopy architecture and yield parameters, considering different scion/rootstock combinations by using new size-controlling rootstocks like Pilowred^® or Rootpac-R^®.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

FM: Conceptualization, Investigation, Methodology, Project administration, Visualization, Writing – original draft, Writing – review & editing. SPG: Data curation, Formal analysis, Methodology, Software, Writing – original draft, Writing – review & editing. SC: Funding acquisition, Resources, Supervision, Validation, Writing – review & editing.