Method to reduce low temperature stress (LTS) in red raspberry (Rubus idaeus L.)

2.1The experimental field

The study was undertaken in an 7 m wide and 50 m long open polytunnel on farm land in mid-Norway (63°46’N, 11°16’E), with two raised beds 20 cm high and 90 cm wide spaced 3 m apart (Fig. 1). On each bed, two drip lines (Netafim Dripnet) were placed 15 cm apart. The drip line has a pressure-compensated drip nozzle placed at 50-cm intervals with discharge rate of 1.6 l h^–1 [Dryppvanning AS, Norway (Israel)]. The beds were mulched with woven black polyethylene plastic mulch (LOG ‘jorddekkeduk’, Norway). In July 2009, greenhouse-matured raspberry plants (30-cm tall) were established in circular openings (20-cm diameter) in the plastic mulch spaced 50-cm apart. The polytunnel was covered with roof sheeting from early May to early October each year of the experimental period which was from 2011 to 2014. Fertigation (irrigation and fertilization combined) were undertaken once a week in accordance with the plant development. Additional water was applied at a tension treshhold of 20 Cb. Weeding and control of diseases and pesticides were according to regional standard procedures for growing in polytunnel.

2.2The experimental design

The experiments are undertaken in a split plot design with two replications and two large plots in each block. One of the large plots was heated at threshold temperatures, while the other was not. The large plot contained six small plots each with seven plants. In the heated large plot two of the small plots were idle (reserve plots), that were treated similar to the other four small plots of the large plot all years, except the last year when four of the idle plots were added to the experimental design as treatment E (CBH, Table 1).

All recordings in a small plot were undertaken on the three central plants. Each plant was pruned leaving four canes in early spring. Treatment A = Genotype (‘Glen Ample’, ‘Stiora’, KV 91-39-7); Treatment B = B (Bent, bending and winter covering of canes); Treatment C = BH (Bent_H), similar to B except that heat was added under the winter cover; Treatment D = Erect canes (C or Control); Treatment E = CBH (Control_BH) in 2014; canes on idle plots, similar to treatment B except that the canes were trained directly vertical just before bud burst.

For Winter cover was used a double layer of Agrocover (Rovero, the Netherlands), which was removed each year in April/May when the chance of freeze injury on cane buds was absent. Heating was provided by one tube-formed oven (HYTH 454/56, 180 W; Hylite Manufacturing LTD, Heywood UK) per small plot. The ovens were connected to a thermostat inducing heating at ambient air critical minimum temperatures, which was decided to be –15°C, –10°C and –5°C (by experience) respectively in the periods 1 November to 28 February, in March and in April. A lamp connected to the thermostat visualized heating. The temperature was recorded under the winter cover. In April/May bent canes in treatments B and C were lifted to horizontal position allowing laterals to grow upwards. At anthesis (first flower) they were lifted into an upright position.

2.3The trellis

A special trellis was used to bend floricanes. The trellis used was a copy of the rotating cross arm trellis (RCA) system, available from Trellis Growing Systems (Fort Wayne, IN USA). The post was a square iron rod (4×4 cm²) driven 50 cm into the soil leaving 40 cm above soil (Fig. 2). At the top of the post, two iron metal plates were bolted with wing nuts. In between the two plates two cross-arms were inserted and their bases secured with a pin. Additional holes were drilled through the plates. Another pin was inserted through the plates and the cross-arm. The wires on long and short cross arms supported floricanes and primocanes, respectively, allowing them to move in wanted positions (bent to the ground, horizontal and upright). To support the canes a cane training system with two wires on the rotating cross arms was used to allow a hedge height of 160 cm (Fig. 2). The plants were topped at first bud above 160 cm height in early spring spaced ∼12.5 cm apart on the top wire.

2.4Data collection

Fruits were harvest twice a week from the end of July to the middle of September each year. Total fruit yield was recorded in grams and weighed fruit size was calculated based on fruit size and fruit yield at each harvest. Percentages salable, rotted and crumbly fruit were calculated, but not tabulated since levels of salable was almost equivalent to total yield because of few rotted and crumbly fruits. In spring 2012, 2013 and 2014 freezing injury was registered by evaluating the development of vegetative buds on the upper two thirds of the cane two weeks after bud burst, using a 0–9 scale where 0 is 100% dead buds/laterals and 9 is100% fresh buds/laterals. The percentage injury (dead buds) represented by scores between 9 and 0 was: score 8 = 0.1–13.1% , 7 = 13.2–26.2% , 6 = 26.3–36.3% , 5 = 36.4–47.4% , 4 = 47.5–60.5% , 3 = 60.6–73.6% , 2 = 73.7-86.7% , 1 = 86.8–99.9% . This is a simplification of a score table given earlier [12].

Because of experiences the first years, more detailed data collection was undertaken in 2014. Bud survival was recorded at the lower third, the mid third and the top third section of the cane. In addition number of buds destroyed by training the canes, total number of canes in the plot and number of canes lower than 160 cm, were counted in spring after the final lifting of the trellis.

Climatic data was from a nearby (300 m distance) meteorological station, available on the web (eklima, Meterological Institute, Norway).

2.5Statistics

Statistics, tables and graphs were undertaken using the SAS procedures GLM (Multivariate Anova with P-levels), tabulate and graph [13].

3.1Ambient temperatures in three winter seasons and the effect of cover on temperature

The effects of freezing injury of buds and subsequently on fruit yield parameters, are to a great extent influenced by the temperature patterns of the three test winters, in interaction with the treatments implemented to avoid freezing injury. The winter of 2011/2012 was relatively mild with temperatures that would not give freezing injury on normally developed cane buds. The temperature was not below the critical temperatures of –15°C and –10°C, respectively in the periods 1 November to 28 February and in March. However, at a few occasions the temperature went below the critical temperature of –5°C in April (Fig. 3). The snow cover was stable from 20 December to 20 March, but not thicker than 20 cm at any date. The temperature under BH (treatment C, Table 1) was not recorded that year, because of failure of the logger. The winter 2012/2013 was much colder and temperatures with bud freezing potential occurred in all months from December to 10 April. The snow cover was more unstable than the previous year, and in periods it blew off the top of the covering or rainfall melted the snow. However, the temperature under BH went down to –15°C in January 2013 while the ambient temperature went down to –16.2°C (Fig. 3; Fig. 4). The highest temperature in January was +5.0°C, while the temperature under cover and potential heat (BH) varied from+2.3 to –11.0°C. Since the heating was active at several occasions in 2013 this is one reason of higher minimum temperature. A thin snow cover could have strengthened the effect. In the period February and March 2013 the ambient temperatures varied between +6.6 to –22.8°C. Under BH in the same period the temperature varied from +3.2 to –11.0. The reduced maximum and increased minimum temperatures can be suggested a result of the covering itself, and of snow covering the BH treatment in periods combined with a heating effect in the cold spells of February and March. In April 2013 ambient temperature dropped to –8.8°C and to –8.2°C under BH. It is important to note that the cover prevented the high ambient temperatures going up to 14.5°C, which is important considering hardening and dehardening. The winter 2013/2014 was exceptional. It was mild, dry and windy and with no snow cover apart from a short period in early December. These conditions introduced a strong drought stress, which was the main stress factor. However, the temperature went below the critical temperature (–15°C) for a couple of days in mid-January 2014 and below the critical temperature (–10°C) for one day in mid-March. The drop in ambient temperature in mid-January 2014 to –17.5°C was answered by a drop to –14.2°C under BH and CBH, showing an improvement by the treatment (Fig. 3, Fig. 4). Similarly in mid-March BH and CBH improved the temperature with 6.1°C compared to ambient that dropped to –12.5°C; a combined effect of covering and heating.

It is shown that the large differences in climate between the years had potential to induce differences in injury of the cane buds by freezing or drought stress. It is also shown that covering and heating strongly influenced temperature. These factors would have a potential influence on fruit yield and yield parameters.

3.2Effects on cane and cane buds by treatment

Observations in spring any year did not show visible cracking of canes in the bent area. In 2014 more detailed examinations were introduced. These examinations showed that only two canes (of more than 1000) were torn off the mother plant during the bending and lifting procedures in the previous autumn and winter (not tabulated). Because of this it was enough canes evaluated as unharmed to achieve full plots for the growing season.

Freezing injury of cane buds was reduced by bending and covering of canes. However, the winter of 2011/2012 was mild with the lowest temperatures approximately at –13°C on three occasions in January and February, and it was close to critical temperature (–5°C) in April (Fig. 3). The heating was therefore activated only a few times in early April, and all treatments gave a freezing injury score higher than 7.5, but the differences between treatments were not significant (Fig. 5). The next winter was cold, and heating was activated at several occasions throughout the whole winter into early April (Fig. 3). Under these conditions erect canes (C) had more freezing injury of buds than bent and covered canes. Bending and covering reduced the freezing injury of buds to almost none, and no additional improvement caused by heating was observed (Fig. 5).

The winter 2013/2014 was not cold, and the temperature reached critical level on only one day in mid-January and also in mid-March (Fig. 3). However, there was no snow cower (and no rain for two months) and steady hard and dry winds were prevalent into March. Extensive drought damage was observed in the heather plants growing in surrounding areas. That year the bud injury on erect canes, especially in ‘Stiora’ (Fig. 5, Table 2), was attributed as a desiccation damage, rather than a freezing injury since injured buds were located along the entire two third top-length of the upright canes and not from the top and downwards as after freezing injury. It should be mentioned that ‘Stiora’ did not grow well in the clay loam soil of the experimental field, resulting in thin and short canes that often were less than 160-cm tall and therefore did not require pruning or only a short length of the tip was pruned (Table 4). Also, KV91-39-7 reacted similarly to the clay loam soil as ‘Stiora’ growing short canes. However, the canes were more sturdy and thicker than for ‘Stiora’. KV91-39-7 showed strong tolerance to the special climatic conditions with strong drought stress, and it adapted well to the bending treatments as did ‘Stiora’. The relatively high injury score for ‘Stiora’ compared with ‘Glen Ample’, could be explained by weaker and shorter canes (Fig. 5, Table 3). Selection KV91-39-7 had less bud injury than ‘Stiora’. On the lower third section the buds were resting for all cultivars. Number of buds that were torn off the cane by handling the trellis was reduced in this order ‘Glen Ample’ > ’Stiora’ >KV91-39-7 (Fig. 6, Table 4), and CBH and BH had less injury than B.

3.3Effect on fruit yield parameters by treatments

There was no interaction between year and freezing avoidance treatment for total yield and fruit size. Therefore annual results for these parameters are not tabulated. It was shown that standard training keeping the canes in an upright position the year round (Control) had higher fruit yield as a mean of genotypes and four years than for the two bending treatments (Table. 4). There were no differences in fruit weight between treatments, however because of the low yields of the bending treatments, compared to control, it would have been expected that the fruit weight should increase. The reduced fruit weight indicate a reduction in fruit number by bending. There was a tendency that the treatment with heating under the cover improved the winter survival as an average of all genotypes (Table 2). However, for ‘Glen Ample’ the added heat (BH) tended to reduce yield, probably because of the increased winter injury compared with no heat (B), in contradiction to ‘Stiora’ and KV91-39-7 where added heat improved yield. Anyway, standard trellising with erect canes (Control) had highest yield, though it was not significantly higher than the heat treatment for ‘Stiora’ and the selection KV91-39-7. The amount of salable fruit was from 91 to 98 percentage of total yield, and the rest was rotted and crumbly fruits. The effects by treatments on the percentage of these yield parameters were minor and generally not significant (not tabulated).

In 2014 idle plots (CBH) were introduced to examine if the result of bending till anthesis (B and BH) gave a general reduction in yield compared with CBH. This could be caused by the technique of forcing laterals in one direction at anthesis. For the two cultivars tested the standard erect canes (Control) still produced the highest yield, but the bent canes, given the opportunity to grow laterals naturally (CBH), tended to give higher yield than treatment B and C (respectively B and BH) for ‘Glen Ample’ (Table 5), the benefit was smaller and not significant for ‘Stiora’ (p = 0,099). There were no significant differences in fruit weight, but ‘Stiora’ had a strong tendency to develop larger fruit weight in control and CBH compared with B and BH. It could be suggested that forcing the laterals to one side partly could explain the reduced yield compared to standard trellis.