Study site.

The study was conducted in 2013 and 2014, in a mature planting at Oregon State University’s North Willamette Research and Extension Center, Aurora, OR [lat. 45°16′47′′N, long. 122°45′23″W; USDA hardiness zone 8b (U.S. Department of Agricultural Research Service, 2014); the weather records for this site can be viewed at U.S. Department of Interior (2014)]. The soil is mapped as a Willamette silt loam, classified as a fine-silty, mixed, superactive, mesic, Pachic, Ultic, and Argixeroll.

Plants were established in Spring 2010 except for ‘Triple Crown’ (planted Oct. 2010) and ‘Columbia Star’ (planted Sept. 2011) at 1.5 m in the row with 3 m between rows (2222 plants/ha). A permanent grass cover crop grew between the rows. The in-row area was maintained (bare soil) with preemergent herbicides and hoeing, as needed. Plants were irrigated with a single line of drip tubing (UniRam; Netafim USA, Fresno, CA) containing pressure-compensating emitters (1.9 L·h⁻¹ in-line) spaced every 0.6 m.

Soil testing.

Soil samples were collected on 12 Nov. 2013 and 21 Oct. 2014 using a 2.4-cm diameter, 0.5-m long, slotted, open-side, chrome-plated steel soil probe (Soil Sampler Model Hoffer; JBK Manufacturing, Dayton, OH). Soil was sampled to a depth of 0.3 m at the center of the row, ≈0.3 m from the crown between plants and within the water emitter drip zone or fertilization area. A pooled sample was collected for the site (not replicated) and was sent for analysis of macronutrient and micronutrient concentration and pH to Brookside Laboratories Inc. (New Bremen, OH) (Table 1).

Table 1.

Soil nutrient pH, organic matter content, and nutrient levels when sampled in Nov. 2013 and Oct. 2014 at Oregon State University’s North Willamette Research and Extension Center, Aurora, OR (unreplicated).

View Table

Cultivars.

Four trailing blackberry cultivars (‘Black Diamond’, ‘Columbia Star’, ‘Marion’, and ‘Onyx’), one erect type (‘Ouachita’), and two semierect types (‘Chester Thornless’ and ‘Triple Crown’) were studied.

Production system.

The trailing cultivars were grown in an alternate year production system (Strik and Finn, 2012), but only plants in the fruiting or “on year” were sampled in 2013 and 2014. Plants were trained on a two-wire vertical trellis system in each row with the wires attached to steel posts at 1.0 and 1.6 m above the ground. New primocanes were trained along the ground in the row, under the floricane canopy during the growing season, per standard practice. In the erect and semierect blackberry cultivars, primocanes were tipped at a height of ≈1.0–1.5 m per standard practice to encourage branching. Floricanes were removed in autumn (after they had senesced). Primocanes were then trained to the same two-wire vertical trellis system as the trailing cultivars by wrapping and tying as necessary. Primocane branches were pruned to 1 m, in winter, as needed.

Fertilizer nutrients were applied per standard recommendations (Hart et al., 2006) and using results from soil analysis (Table 1). In 2013, 79 kg·ha⁻¹ N was applied in a split application with 34 kg·ha⁻¹ on 3 Apr. (8N–7.5P–14K) and 45 kg·ha⁻¹ on 28 May (40N–0P–0K–5Mg). In 2014, 90 kg·ha⁻¹ N was applied in two equal portions on 27 Mar. and 5 June (16N–7P–13K). An additional 28 kg·ha⁻¹ N was applied to erect and semierect cultivars on 7 July 2014 (16N–7P–13K). These granular fertilizers were broadcast within the in-row area and were washed into the soil through rainfall (data not shown). Copper fungicide (Nu-Cop; Albaugh Inc., Ankeny, IA) was applied to all plants (0.4 kg·ha⁻¹ of Cu) on 24 Mar. 2014 to control the cane diseases purple blotch [Septocyta ruborum (Lib.) Petr.] and cane rust [Kuehneola uredines (Link) Arthur].

Leaf sampling.

Tissue samples for nutrient testing were collected approximately every 2 weeks by choosing the most recent fully expanded leaves on primocanes (20–21 May to 6–7 Oct. 2013 and 2014) and floricane laterals (trailing: 6–7 May to 28–29 July; erect and semierect: 6–7 May to 6–7 Oct. 2013 and 2014) for a total of 11 samples in each year for primocanes (Table 2). ‘Black Diamond’ and ‘Ouachita’ primocanes grew earlier in the spring and were thus sampled 2 weeks before those of the other cultivars. Stages of plant development and fruiting season were recorded and are presented in Table 2. The rate of plant development and fruiting season did not appear to differ between years (data not shown). Yield data were not recorded, but the field was observed to have a good, typical commercial yield for these cultivars in Oregon.

Table 2.

Developmental stages for primocanes and floricanes through the season (2013 and 2014 averaged) for floricane-fruiting blackberry types and cultivars studied at Oregon State University’s North Willamette Research and Extension Center, Aurora, OR.

View Table

About 6 or 12 of the most recent, fully expanded primocane and fruiting lateral leaves, respectively, including petioles, were sampled per plot on each date and were left unwashed per standard recommendation (Hart et al., 2006). Sampled leaves were priority shipped to Brookside Laboratories for analysis. Leaf N was determined using a combustion analyzer with an induction furnace and a thermal conductivity detector (Gavlak et al., 1994). Other nutrients, including P, K, calcium (Ca), magnesium (Mg), aluminum (Al), B, copper (Cu), manganese (Mn), iron (Fe), and zinc (Zn) were determined using an inductively coupled plasma spectrophotometer after wet ashing the samples in nitric/perchloric acid (Gavlak et al., 1994).

Data analysis.

The treatments were arranged in a completely random design with three replicates of three-plant plots. Data were analyzed by tissue type (primocane or floricane) separately, as our goal was not to compare canes, but to determine changes in leaf nutrient concentration over the sampling period and assess optimal sampling times. In addition, there is evidence that floricanes and primocanes are independent in blackberry (Bryla and Strik, 2008). On each sample date, leaf nutrient data were first analyzed for the effect of year and cultivar using PROC MIXED (SAS version 9.3) with year as the main effect (n = 2) and cultivar as the subplot effect (n = 4) with a Satterthwaite approximation used, as needed, for main effect comparisons.

Three sample dates (from late July to late August) that bracketed the recommended tissue nutrient sampling time for blackberry in our region (Hart et al., 2006), were analyzed to determine the effect of year, cultivar, and sample date. Mean comparisons were performed using least-square means. Contrasts were used to determine differences between early- and late-fruiting types (trailing vs. erect and semierect) as well as differences between erect and semierect types during this period.

PROC UNIVARIATE and the Shapiro–Wilk procedure were used to assess normality of the data for all the aforementioned analyses. As the tissue concentration of many nutrients was not normally distributed, a log transformation was used to improve homogeneity of variance and to assess proportional effects. Data were back transformed for presentation.

Rate of development and fruiting season.

Key phenological stages and fruiting season are presented for the cultivars studied for each tissue sample date (Table 2). There were no observed differences among key phenological stages between years (data not shown). The trailing cultivars Black Diamond, Columbia Star, Marion, and Onyx had the earliest fruiting seasons, followed by the erect cultivar Ouachita and then the semierect cultivars Triple Crown and Chester Thornless (Table 2).

Year effect.

There was a significant main effect of year on primocane leaf nutrient concentration of the cultivars studied on many sample dates through the season for all nutrients (data not shown). In general, the concentration of N, S, B, Fe, and Al was higher and that of P, Mg, K, Ca, Mn, and Cu was lower on one or more sample dates in 2013 than in 2014.

There was an effect of year on floricane leaf nutrient concentration of the cultivars studied on many sample dates for all the nutrients (data not shown). The concentrations of K, B, Fe, Zn, and Al were higher, whereas N, P, Mg, Ca, S, and Mn were lower on at least one sample date in 2013 than in 2014.

The pattern of the change in primocane and floricane leaf nutrient concentration within each cultivar was relatively similar through the season between years (data not shown), but there was an effect of year and cultivar, and sometimes an interaction between year and cultivar found for several nutrients and sample dates. We have chosen not to show both years of data for seven cultivars and 11 primocane and 7–11 floricane sample dates. Instead, we are presenting data for the 2014 growing season to show the pattern of change over time. In Table 3, we present year by cultivar interactions for primocanes on select dates that bracket the range in current tissue nutrient sampling recommendations (Hart et al., 2006).

Table 3.

The effect of year and cultivar on primocane leaf nutrient concentration of floricane-fruiting trailing, erect, and semierect cultivars at the North Willamette Research and Extension Center, Aurora, OR and sampled on three dates in late July–early Aug. 2013 and 2014.

View Table

Primocane leaf nutrient concentration.

There was a significant effect of cultivar on primocane leaf N concentration in early spring and toward the later part of the growing season (Fig. 1). ‘Black Diamond’, ‘Columbia Star’, ‘Marion’, and ‘Onyx’ had lower leaf N in mid-June than the erect and semierect cultivars, whereas ‘Chester Thornless’ and ‘Triple Crown’ had a higher and lower leaf N, respectively, in the late summer and autumn. Leaf N declined from spring to summer in all the cultivars and was then variable (‘Chester Thornless’) or remained relatively stable (all others).

View Full Size

The effect of sample date over the growing season on the concentration of macronutrients in primocane leaves of floricane-fruiting, trailing (‘Black Diamond’, ‘Columbia Star’, ‘Marion’, and ‘Onyx’), erect (‘Ouachita’), and semierect (‘Chester Thornless’ and ‘Triple Crown’) blackberry at the North Willamette Research and Extension Center, Aurora, OR, 2014. A = nitrogen; B = phosphorus; C = potassium; D = magnesium; E = calcium; F = sulfur. Bars indicate standard error for cultivar.

Citation: HortScience 52, 6; 10.21273/HORTSCI11965-17

• Download Figure
• Download figure as PowerPoint slide

Leaf P declined through the season in ‘Ouachita’, ‘Triple Crown’, and ‘Chester Thornless’, whereas concentrations increased in late summer in most of the trailing cultivars to levels significantly higher than in the erect and semierect cultivars (Fig. 1). ‘Onyx’ was similar to the other trailing cultivars until mid-August when the P concentration in primocane leaves declined to levels similar to those measured in ‘Chester Thornless’.

Leaf K concentration declined from spring to autumn in ‘Ouachita’ (except for one increase in late July) and ‘Triple Crown’, as was reported for primocane-fruiting blackberry by Strik (2015), but in ‘Chester Thornless’ levels were the lowest in late summer. ‘Columbia Star’, ‘Marion’, and ‘Black Diamond’ had the highest leaf K concentration in late summer to autumn (Fig. 1), perhaps because the fruiting season was over in these cultivars (Table 2), although leaf K in ‘Onyx’ was highest in the spring.

Leaf Mg and Ca concentrations increased overall from spring to autumn in the erect and semierect cultivars, although levels did fluctuate during the summer, whereas in trailing cultivars levels were relatively flat through the season and were lower in late summer. Peaks in leaf Ca concentration during the season may have been related to variability in leaf sampling caused by summer pruning (tipping) as reported for primocane-fruiting blackberry (Strik, 2015). The trailing blackberry cultivars had a considerably lower leaf S concentration than the erect and semierect cultivars, especially from spring to midsummer (Fig. 1).

The pattern of change in leaf B was similar amongst the blackberry cultivars in the spring to midsummer (Fig. 2). However, leaf B increased rapidly from late July to the end of August in the erect and semierect cultivars to levels significantly higher than in the trailing cultivars, particularly in ‘Chester Thornless’ and ‘Triple Crown’. A similar pattern was observed for leaf Mn. Leaf Cu was relatively stable in the early season except for the semierect cultivars, but increased for all cultivars except ‘Ouachita’ in the late season. Leaf Zn concentration generally increased through the season and was highest (‘Marion’) and lowest (‘Columbia Star’) in the trailing cultivars on many sample dates.

View Full Size

The effect of sample date over the growing season on the concentration of micronutrients in primocane leaves of floricane-fruiting, trailing (‘Black Diamond’, ‘Columbia Star’, ‘Marion’, and ‘Onyx’), erect (‘Ouachita’), and semierect (‘Chester Thornless’ and ‘Triple Crown’) blackberry at the North Willamette Research and Extension Center, Aurora, OR, 2014. A = boron; B = manganese; C = iron; D = copper; E = zinc; F = aluminum. Bars indicate standard error for cultivar.

Citation: HortScience 52, 6; 10.21273/HORTSCI11965-17

• Download Figure
• Download figure as PowerPoint slide

Primocane leaf Fe and Al concentrations were much higher in the trailing cultivars than the other blackberry types likely because there was dust (soil) on these leaves and leaves are not washed before tissue analysis (Hart et al., 2006); trailing blackberry primocanes are not self-supporting and were laying on the soil during the sampling period as opposed to the erect growth habit of ‘Ouachita’, ‘Triple Crown’, and ‘Chester Thornless’.

Sampling time recommendations.

The current tissue sampling recommendations for blackberry vary by region (Bolda et al., 2012; Bushway et al., 2008; Fernandez and Ballington, 1999; Hart et al., 2006). Sampling is recommended for a time when tissue nutrient concentrations are relatively stable to minimize variation. Rather than a specific date, Strik (2105) recommended that primocane-fruiting, erect blackberry cultivars be sampled at the bloom to green-fruit stage of growth. In the erect and semierect cultivars we studied, the only nutrients that appeared to have relatively stable concentrations in the late July to early August sampling period were N, K, S, Mn, Zn, and Cu (Figs. 1 and 2). However, by late July to early August, some nutrients started to increase (Ca, Mg, B, and Al) or decrease (P) in concentration.

Tissue nutrient concentrations for the three primocane leaf sampling dates ranging from the end of July to late August of each year were compared for all cultivars (Table 3). We chose these dates to correspond to a range similar to what is currently recommended for sampling in our region (Hart et al., 2006). There was a significant year by sampling date interaction for all nutrients. When comparing sampling in late July to early August (dates 7 and 8 in Table 3), there was no sampling date effect for N, Mg, K, Ca, S, B, Mn, and Zn (only in 2014); P, Fe, and Al in both years; primocane leaf Cu was not stable during this period in either year. However, when the sampling period was mid to late August (dates 8 and 9, Table 3), primocane leaf N, Mg, Fe, Mn, and Al were stable in both years, leaf K, Ca, S, B, Cu, and Zn were stable in one of the years of study, whereas leaf P was not stable in either year.

Leaf N concentration declined from late July to mid-August, particularly in 2013 (Table 3), and leaf N, on average, was only within the current recommended sufficiency levels on the first sample date (both years). Leaf P was below current sufficiency levels on the first two sample dates in 2013, but was at the low end of the range on all dates. Average leaf K concentration was below current sufficiency levels on all sampling dates despite sufficient soil K (Table 1) and fertilization with K at recommended levels. Leaf K declined during the late July to late August sampling period in 2014 but only declined from late July to early August in 2013 before increasing again in late August (Fig. 1). Primocane leaf Fe and Mn were above recommended sufficiency levels on one or more later sampling dates (Table 3), perhaps because of the presence of dust on the leaves (Fe) and a soil pH that was at the low end of the recommended range for blackberry (Hart et al., 2006). All of the other nutrients were within current sufficiency levels during the late July to late August sample period.

Cultivar had a significant effect on the nutrient concentration of primocane leaves when sampled from late July to late August (Table 3). The trailing cultivars had a lower average leaf N than the erect and semierect cultivars (which did not differ). There was a cultivar × date interaction, because leaf N in ‘Onyx’, ‘Columbia Star’, ‘Triple Crown’, and ‘Ouachita’ declined rapidly from mid to late August, whereas leaf N changed relatively little during this period in the other cultivars (Fig. 1). All cultivars except ‘Chester Thornless’ and ‘Onyx’ had an average primocane leaf N concentration below the current recommended standards (when sampling from late July to late August; Table 3).

Average leaf P was below current sufficiency levels in all cultivars in 2013 and in ‘Ouachita’ in 2014 (Table 3). Trailing cultivars had higher leaf P than the erect and semierect cultivars, whereas the erect cultivar, Ouachita, had a lower leaf P than the semierect cultivars. Leaf P increased during the early sampling period for ‘Marion’, ‘Chester Thornless’, and ‘Onyx’, whereas levels remained relatively stable or declined in the other cultivars (Fig. 1).

Leaf Mg concentration was lower in the trailing cultivars than the other types and was higher in the semierect cultivars than in the erect cultivar, Ouachita (Table 3). Although blackberry type did not affect leaf K, there was a strong effect of cultivar. ‘Columbia Star’ had the highest leaf K and ‘Black Diamond’ and ‘Onyx’ the lowest, depending on year. All cultivars had a leaf K below the current sufficiency levels except for ‘Columbia Star’ and ‘Marion’ in 2014 (Table 3). Although there was no date × cultivar interaction during the late July to late August sampling period (Table 3), leaf K was higher in the late season for all trailing cultivars except ‘Onyx’ as compared with the erect and semierect cultivars that were still fruiting at this time (Fig. 1; Table 2). It is possible that uptake of K from the soil preferentially goes to the floricane during the fruiting period. On average, trailing cultivars had lower leaf Ca than erect and semierect cultivars and ‘Ouachita’ had a higher leaf Ca than the semierect cultivars (Table 3). There was a sampling date × cultivar interaction during the late July to late August period because leaf Ca increased from mid to late August in the erect and semierect cultivars considerably compared with the trailing cultivars (Fig. 1). Leaf Ca levels are affected by tipping date and resulting leaf age at sampling in erect blackberry (Strik, 2015).

The trailing blackberry cultivars had lower leaf B, Mn, and Zn and higher leaf Fe and Al than the erect and semierect cultivars from late July to late August (Table 3). The semierect cultivars had higher leaf B, Mn, and Cu, but lower Zn than the erect cultivar Ouachita. Leaf B and Mn increased in the erect and semierect cultivars on the late August sample date relative to the trailing cultivars (Fig. 2). Leaf Cu was significantly lower in ‘Triple Crown’ than the other cultivars on the late August sample date (Fig. 2). ‘Onyx’ and ‘Columbia Star’ had a lower leaf Al than the other cultivars on the late August sample date (Fig. 2).

Based on current sufficiency levels and the pattern of leaf nutrient change over the season, we do not recommend sampling the late-fruiting floricane cultivars ‘Ouachita’, ‘Triple Crown’, or ‘Chester Thornless’ at a particular developmental stage as was suggested by Fernandez and Ballington (1999; “following harvest”) or for primocane-fruiting blackberries (Strik, 2015). The primocane leaf nutrient concentrations measured in this study were below the published recommended standards for blackberries (Bolda et al., 2012; Bushway et al., 2008; Fernandez and Ballington, 1999; Hart et al., 2006) for P and K (all regions), N (only in Oregon; Hart et al., 2006), and Mg and S (only in the northeastern United States; Bushway et al., 2008). Leaf Fe (all regions) and Mn (California and northeastern United States; Bolda et al., 2012; Bushway et al., 2008) were higher than current standards.

Floricane leaf nutrient concentration.

The effect of cultivar on floricane (fruiting lateral) leaf tissue nutrient concentration over the 2014 growing season is shown in Figs. 3 and 4. Floricane leaf N concentration declined during the floricane season from early-fruiting lateral development (6 May) through fruiting and early floricane senescence in all of the blackberry types and cultivars (Fig. 3). Nitrogen may have declined as N was mobilized to the developing fruit. Leaf N levels were similar to those reported for floricane leaves of ‘Tupy’ and ‘Xavante’ erect blackberry in Brazil (Pereira et al., 2015) and ‘Black Diamond’ and ‘Marion’ in Oregon (Dixon et al., 2016). The pattern of change in leaf S in the floricanes was similar to leaf N (Fig. 3).

View Full Size

The effect of sample date over the growing season on the concentration of macronutrients in floricane leaves of trailing (‘Black Diamond’, ‘Columbia Star’, ‘Marion’, and ‘Onyx’), erect (‘Ouachita’), and semierect (‘Chester Thornless’ and ‘Triple Crown’) blackberry at the North Willamette Research and Extension Center, Aurora, OR, 2014. A = nitrogen; B = phosphorus; C = potassium; D = magnesium; E = calcium; F = sulfur. Bars indicate standard error for cultivar.

Citation: HortScience 52, 6; 10.21273/HORTSCI11965-17

• Download Figure
• Download figure as PowerPoint slide

View Full Size

The effect of sample date over the growing season on the concentration of micronutrients in floricane leaves of floricane-fruiting, trailing (‘Black Diamond’, ‘Columbia Star’, ‘Marion’, and ‘Onyx’), erect (‘Ouachita’), and semierect (‘Chester Thornless’ and ‘Triple Crown’) blackberry at the North Willamette Research and Extension Center, Aurora, OR, 2014. A = boron; B = manganese; C = iron; D = copper; E = zinc; F = aluminum. Bars indicate standard error for cultivar.

Citation: HortScience 52, 6; 10.21273/HORTSCI11965-17

• Download Figure
• Download figure as PowerPoint slide

Floricane leaf P declined rapidly in the early spring for all cultivars, but was then relatively stable from early June through floricane senescence, except in ‘Marion’ where leaf P dropped from early to late July. ‘Black Diamond’ had the greatest leaf P in June and early July, whereas ‘Chester Thornless’ had a significantly higher leaf P in the late summer than the other erect and semierect cultivars (Fig. 3).

Leaf K in the fruiting lateral declined from spring to summer in the early-fruiting trailing cultivars and the midseason erect cultivar Ouachita, whereas leaf K did not start declining much in the semierect cultivars until late July (Fig. 3) when fruit coloring began (Table 2). ‘Black Diamond’, ‘Columbia Star’, and ‘Onyx’ leaf K increased during the last 2 weeks of July just before the onset of senescence. Floricane leaf K concentrations were similar to those reported in other erect cultivars (Pereira et al., 2015) and in ‘Black Diamond’ (Dixon et al., 2016); however, leaf K for ‘Marion’ was higher in our study than what was previously reported by Dixon et al. (2016). Leaf K declined during fruiting likely because of the relatively high K content of blackberry fruit (Dixon et al., 2016; Harkins et al., 2014).

Floricane leaf Mg increased from spring to the beginning of the fruiting season in the erect and semierect cultivars, whereas leaf Mg stayed relatively stable in the early-fruiting trailing cultivars (Fig. 3). There was a very similar pattern in leaf Ca for all types and cultivars studied with increasing levels through the season. The reason for the decline in leaf Ca in the floricanes of erect and semierect cultivars in late August is not clear, but this decline was not observed in 2013 for these cultivars (data not shown). Floricane leaf Ca and Mg concentrations likely increased through the season because these leaves were aging, as has been reported in older leaves of raspberry (Hughes et al., 1979) and blackberry (Strik, 2015), and there are relatively low concentrations of these nutrients in trailing blackberry fruit (Dixon et al., 2016; Harkins et al., 2014).

In general, leaf B, Mn, Fe, and Al concentration increased during the floricane season, with the exception of lower values of leaf B and Mn for the erect and semierect cultivars either just after fruit harvest or during fruit harvest, respectively (Fig. 4). Leaf Cu concentrations decreased slightly in the early season, but increased and were more variable for the later-fruiting cultivars from August to October. Leaf Zn levels declined in May and then increased slightly or remained relatively stable during fruiting, depending on cultivar.

Nutrient sufficiency standards have not been developed for floricane leaves of any type of blackberry in North America. Because the nutrient concentration in floricane leaves of trailing cultivars just before fruit harvest (approximately mid-June to 1 July) was greater than published primocane leaf sufficiency levels (Hart et al., 2006) for most nutrients, floricane-specific standards would need to be developed. However, we do not recommend floricane leaves be used to manage fertilizer programs as the concentrations of most nutrients were rapidly changing during the lateral and fruit developmental period. By contrast, the floricane leaf concentrations of most nutrients in the erect and semierect cultivars were within the current primocane sufficiency levels (Table 3) when sampled in late July to early August (just before fruiting, Table 2), except for floricane leaf K (lower) and Fe and Mn (higher) (Fig. 4).