Orchard details

Redhaven trees on five clonal rootstocks (Rootpac^®-R, Rootpac^®-20, Rootpac^®-40, Rootpac^®-70) and Bailey seedling rootstock were planted at the University of Guelph Ontario Crops Research Centre – Simcoe, ON (42°51′40″ N, 80°16′8″ W) in the spring of 2016. They were trained to a central leader training system and spaced at 1.8 m within row and 5.0 m between rows (1111 trees·ha⁻¹). The experimental design consisted of a completely randomized block design with five replications of seven trees per block — a total of 35 trees per rootstock. Trees were planted in a Wilsonville sandy loam (Brunisolic Grey Brown Luvisol) (Presant and Acton 1984) with rapid drainage and soil textures consisting of mainly lacustrine gravelly sandy till (Hohner and Presant 1989). Trees were trickle-irrigated daily with the equivalent of ~2.5 cm of water weekly (adjusted for natural rainfall) on a schedule of four irrigation run times per day. Irrigation occurred every 6 h (20 min per event) using 2 L·h⁻¹ pressure-compensating emitters spaced 45 cm apart. Standard cultural and pest management practices for Ontario were used (OMAFRA 2015). Weeds were controlled within a 1-m strip on each side of the tree row using 1% (v/v) glyphosate applications in mid-May, June, and July and plastic tree guards were used to prevent contact with the bark. A permanent sod culture was established the year of planting between tree rows using a mixture of 40% perennial rye and 60% red fescue (Vineland Growers, Vineland, ON). All trees were propagated at the Simcoe Research Station in the spring of 2014. Rootpac^® rootstocks were derived from tissue culture and propagated by North American Plants (Lafayette, Oregon). Bailey rootstocks were grown from seed.

Horticultural measurements

Each autumn, trunk circumference was measured 30 cm above the graft union and trunk cross-sectional area (TCSA) was calculated. The date of full bloom was recorded annually beginning in 2017 when the trees bore fruit. In the autumn, root suckers were counted and removed and tree mortality and harvest date, yield (total fruit weight [FW]) and total number of fruits per tree were recorded. In addition, total marketable yield based on local commercial standards was measured by omitting any fruit smaller than 60 mm in diameter. Crop load per tree was calculated by dividing the total number of fruits by the TCSA, and average FW was calculated by dividing total FW by total number of fruits per tree. Cumulative yield was calculated as the sum of yield from 2017 to 2020. Cumulative yield efficiency (CYE) was calculated by dividing cumulative yield by TCSA in 2020. Following harvest and prior to pruning in 2020, the height and spread of the canopy were recorded.

Statistical analyses

Data were analyzed by the GLIMIX procedure of SAS (version 9.4, SAS Institute, Inc., Cary, NC) and mean separation performed using Tukey’s HSD test to separate means with treatments as fixed effects. To account for crop load effects on FW, crop-load adjusted average FW was analysed using crop load as a covariate. Shapiro–Wilk test was used to test the assumption that the residuals were normally distributed. Scatterplots of studentized residuals were observed visually to test the assumption that the errors were not heterogeneous. In cases where there were large deviations from assumptions, data were corrected by log- or square root-transformation prior to analysis.

Climate of orchard location

Climate normals (30 yr average 1981–2010) for Delhi, Ontario, located approximately 25 km from the orchard site, are presented in Table 2. This region experiences a warm humid continental climate with temperatures moderated by the Great Lakes (Köppen climate classification Dfb). On average, there are 145 frost free days per year with average winter and summer temperatures of −5.4 °C and 27.3 °C, respectively, 906 mm precipitation, and average monthly rainfall of 75 mm during 1 May to 31 Aug. On average, there are 38 d where winter temperatures fall below −10 °C, and 4 d where winter temperatures fall below −20 °C. Minimum winter temperatures recorded where the study was conducted were −10.9 °C, −13.0 °C, −22.7 °C, −24.2 °C and −16.6 °C for 2016–2020, respectively (data not shown).

Table 2.

Frost-free days, temperature, and rainfall parameters, based on 30 yr climate normals (1981–2010), Delhi, Ontario, Canada^a.

Tree survival

Tree survival was influenced by rootstock genotype in 2019 and 2020, but not in the prior formative years (Table 3). In 2019, tree survival was significantly lower on Rootpac-40 (88%; P = 0.0014) compared with the other rootstocks, which had no mortality. Tree survival of Rootpac-40 declined to 79% in 2020 (P < 0.0001), while no further mortality on any of the other rootstocks was observed. The cause of tree death of Redhaven on Rootpac-40 rootstocks was unclear. However, it was likely associated with a pathogen that infected the rootstock or caused by winter injury, leading to complete and rapid tree decline. In a rootstock study in Spain where Rootpac-70 and Rootpac-R were investigated among 16 rootstocks and no tree mortality was observed after 9 yr (Font i Forcada et al. 2020); Rootpac-40 was not included in that study. Mestre et al. (2015) observed no tree mortality of Rootpac-R after 13 yr in a study of 12 rootstocks using Big Top nectarine as the scion while Ben Yahmed et al. (2016) observed 20% tree mortality in Rootpac-70 after the first year of planting in an orchard in a Subirana (flat peach cultivar) orchard in northern Tunisia. In a Spanish study, Pinochet (2010) reported no tree mortality of a Japanese plum cultivar on Rootpac-R rootstock after year nine.

Table 3.

Tree survival (%) of Redhaven peach trees, as affected by rootstock over 5 yr. Trees planted in 2016 and trained to a central leader training system.

A unique aspect of Rootpac-40 and Rootpac-R compared with the other rootstocks examined in this study is that Rootpac-40 and Rootpac-R are peach × almond hybrid and plum × almond hybrid, respectively, with up to 50% almond genetics. In comparison, Rootpac-20 and Rootpac-70 have up to 25% almond parentage and Bailey no almond parentage (Font i Forcada et al. 2020). Almond (P. dulcis) is native to Persia and as such, it is better adapted to warmer growing regions than that experienced in southern Ontario. It is plausible that Rootpac-40 is more susceptible to winter injury for this reason, however that does not explain when Rootpac-R had no tree mortality. Another possibility is delayed incompatibility between the scion and Rootpac-40 rootstock, causing tree collapse. However, based on previous research by Çetinbaş et al (2018) who found no incompatibility with P. persica (peach and nectarine cultivars) within 1 yr of grafting, delayed incompatibility is a less likely cause of tree mortality.

Tree vigour

Tree vigour, as indicated by TCSA and canopy height and width, was influenced by rootstock selection from the time of planting in 2016 onward (Table 4). Based on TCSA, Rootpac-70 was consistently the largest rootstock (38% larger than Bailey) followed by Rootpac-R and Rootpac-40 (9% larger than Bailey). Rootpac-20 was the smallest of the Rootpac rootstocks (2% larger than Bailey). By year five (2020), all rootstocks produced trees with similar TCSA, except for Rootpac-70, which was significantly larger (P < 0.0001). Rootpac-70 was 38% larger than Bailey, while Rootpac-R and Rootpac-40 were both ~10% larger than Bailey, and Rootpac-20 was 2% larger than Bailey. It is difficult to compare tree vigour of the Rootpac rootstock series in this study with other studies who use a rootstock standard different than Bailey. Notwithstanding, comparisons can be made with the Rootpac rootstocks. In a Spanish study using Big Top as the scion, in the 11th year, Rootpac-20 and Rootpac-40 had each 31% less vigour than Rootpac-70 (Reig et al. 2020). These data are fairly consistent with the present study were Rootpac-20 and Rootpac-40 were 26% and 21% less vigorous than Rootpac-70, respectively. In a preliminary Italian study using Big Top and Rome Star as scions, trees on Rootpac-20 and Rootpac-40 had the least vigour among six rootstocks in year three. However overall tree vigour was influenced by planting location and training system (Scalisi et al. 2018). In a study on almonds, Lordan et al. (2019) found Rootpac-20 was the most dwarfing of ten rootstocks and had approximately 25% and 30% less vigour than Rootpac-R and Rootpac-40. No other studies in the current literature have compared any of the Rootpac rootstocks with Bailey to our knowledge.

Table 4.

Tree growth, as represented by tree trunk cross-sectional area, of Redhaven peach trees, as affected by rootstock over 5 yr. Trees planted in 2016 and trained to a central leader training system.

Tree height is an important component of training system. Target tree size in the central leader spindle trees used in this study is approximately 3.5 m. However, in a higher density system using a size controlling rootstock, developing a shorter, less vigorous bi-dimensional canopies with a ~3 m height would be desirable (Reig et al. 2020). Tree width and height in year five were both influenced by rootstock selection (Table 4). Trees on Rootpac-70 and Bailey produced the tallest trees compared with the other rootstocks, which were similar in height. Canopy width was also influenced by rootstock selection (P < 0.0001). Trees on Rootpac-70, Rootpac-20, and Bailey had the widest canopies, while trees on Rootpac-R were slightly narrower. Rootstock effect on tree width is confounded by the need to prune trees when they reach their allotted space of 1.8 m (i.e., to prevent encroachment on adjacent trees). Thus, both tree height and width data must be interpreted cautiously. On all rootstocks, Redhaven exceeded the 1.8 m spacing by the fifth leaf, indicating that Redhaven trees will require annual pruning thereon to contain them within this space. Overall, none of the Rootpac rootstocks provided a narrower canopy than Bailey. While tree height of Rootpac-R, Rootpac-20 and Rootpac-40 were slightly shorter than Bailey by year five, these rootstocks likely still impart excessive vigour to be used in a higher density system.

Rootstock suckers

The number of rootstock suckers per tree was significantly influenced by rootstock selection in all years in which there was sucker development. The quantity of cumulative root suckers for the years 2017 to 2020 was also influenced by rootstock selection (P < 0.0001) (Table 5). Rootpac-20 had the greatest propensity to sucker, followed by Rootpac-R. Low and insignificant suckering occurred on trees with Rootpac-40, Rootpac-70, and Bailey rootstocks. The higher tendency to sucker of Rootpac-20 and Rootpac-R could be related with its plum parentage (Table 1). Mestre et al. (2017) also reported that rootstock suckering is a drawback inherent with some plums. Reig et al. (2020) found similar rootstock suckering for Rootpac-20 and Rootpac-40 (on average 3 rootstock suckers per trees) and less suckering for Rootpac-70 using Big Top as the scion. Mestre et al. (2015) found very low suckering in Rootpac-R per tree during year 13 of production. Rootstock suckers are undesirable in the orchard because they can act as infection sites for disease and harbor pests. If suckers are profuse, they also can interfere with in-row weed management and can absorb systemic herbicides such as glyphosate, potentially injuring the tree (Johnson et al. 2011).

Table 5.

Number of suckers (no/tree) of Redhaven peach trees, as affected by rootstock over 4 yr. Trees planted in 2016 and trained to a central leader training system.

Yield, cumulative yield, yield efficiency

There was a strong rootstock effect on yield beginning in year three (2017), when the trees began bearing fruit (Table 6). There was also a strong rootstock effect on cumulative yield (P < 0.0001) and CYE (P < 0.001) at year five (Table 6). In 2017, Rootpac-70 yielded the most fruit, while Rootpac-20 yielded the least. Rootpac-70 continued to yield the most fruit in 2018 through 2020, whereas beginning in 2018 and thereafter, Rootpac-40 yielded the fewest fruit. In two of four years (2019 and 2020), Rootpac-R was among the highest yielding rootstocks. Cumulative yields were greatest on Rootpac-70, and 10% higher than Bailey. Bailey, the industry standard rootstock, was among the highest yielding rootstock in the last two years of cropping (2019 and 2020). However, cumulative yields on Bailey were intermediate amongst the rootstocks tested and 9% lower than Rootpac-70. The cumulative yields of Rootpac-R, Rootpac-20, Rootpac-40, and Rootpac-70 were 98%, 89%, 83% and 110% that of Bailey, respectively. There was also a significant rootstock effect on cumulative marketable yield; averaged over all rootstocks, approximately 10% of fruit were unmarketable because of undersized or immature fruit. In general, rootstock differences in cumulative marketable yield were similar to the trends in total cumulative yield. Trees on Rootpac-70 and Bailey had the greatest cumulative marketable yield, while trees on Rootpac-20 and Rootpac-40 had the least.

Table 6.

Total fruit yield, cumulative yield, and cumulative marketable yield (kg tree⁻¹), total and cumulative yield efficiency (kg·TCSA⁻¹) in year 2020 of Redhaven peach trees, as affected by rootstock over 4 yr. Trees planted in 2016 and trained to a central leader training system.

Whether these rootstock differences in yield will continue as the trees mature and continue to grow is unclear. However, it is likely that Rootpac-70 will become less productive in comparison to the other rootstocks as more pruning is required to restrict it to its planting space. The cumulative yield data are more indicative of the early yield potential of Redhaven on the examined rootstocks rather than the absolute yields that could be obtained at a particular location. This is because tree productivity is influenced by tree nutrient status and environmental and orchard management factors; when these factors are optimized, the full potential of the rootstock will be realized.

CYE was calculated using the sum of four years of yield (2017–2020) and the TCSA in year five (2020). This method was used to normalize yields amongst rootstocks that range in tree vigour. CYE was significantly influenced by rootstock selection (P < 0.0001), although the magnitude of the differences was small and likely of insignificant commercial importance (Tables 6). Trees on Bailey were more yield-efficient than Rootpac-20, Rootpac-40, and Rootpac-70. In turn, all of them did not differ significantly from Rootpac-R. Since the trees are still in their early years of production, the CYE data may not adequately predict cumulative yields of mature orchards. Once tree canopies fill their allotted space, rootstock effects on yield efficiency are influenced differentially by pruning severity. In other investigations reporting the yield performance of Rootpac rootstocks, Reig et al. (2020) found cumulative yields of Big Top on Rootpac-20, Rootpac-40 and Rootpac-70 rootstocks were similar after in their eleventh year, but the yield efficiency of Rootpac-70 was numerically lower (but not statistically different) - owing to its greater vigour.

Fruit weight

Higher tree crop loads can negatively influence FW, hence the requirement to thin peaches (Coneva and Cline 2006). To normalize any effect of vary crop loads among rootstocks, FW was adjusted using crop load as a co-variate using the method of Marini et al. (2012). Crop load-adjusted FW and overall average FW were influenced by rootstock selection every year from 2017 through 2020 (Table 7). Annually, Rootpac-40 consistently produced the largest fruit while Rootpac-20 consistently produced the smallest fruit. Rootpac-R, Rootpac-70 and Bailey produced fruit of intermediate weight. On average over the four years of cropping, Rootpac-40 produced fruit 16% larger than Bailey, while Rootpac-20, Rootpac-R and Rootpac-70 produced fruit 7%, 4% and 3% smaller than Bailey, respectively. Crop load averaged 1.0, 3.0, 2.7 and 1.8 fruit cm^–2 TCSA in 2017 through 2020 (after hand-thinning), respectively and did not exceed 3.2 fruit cm⁻² TCSA for any rootstock (data not shown). This finding suggests that crop load did not limit fruit weight for any of the rootstocks.

Table 7.

Adjusted mean weight of marketable fruit (g) of Redhaven peach trees, as affected by rootstock over 4 yr. Trees planted in 2016 and trained to a central leader training system^a.

In the study by Reig et al (2020), no difference in average FW of Big Top peach on Rootpac-20, Rootpac-40 and Rootpac-70 from year three to 11 were observed.

Notwithstanding the importance of the findings outlined above, this preliminary study had some limitations. While important, it was beyond the scope of this study to measure the abiotic and biotic factors influencing rootstock growth and performance. Further research examining the high mortality rate of Rootpac-40 is required should this rootstock be adopted commercially in this growing region. In addition, fundamental understanding and verification of resistance to nematodes and root rot (Rosellinia necatrix) are required should any of the Rootpac rootstocks become commercially important in Ontario.