Index | Search | Home | Table of Contents

Rife, C.L. and J.P. Salgado. 1996. Selecting winter hardy oilseed rape for the great plains. p. 272-278. In: J. Janick (ed.), Progress in new crops. ASHS Press, Alexandria, VA.

Selecting Winter Hardy Oilseed Rape for the Great Plains

C.L. Rife and J.P. Salgado


  1. MATERIALS AND METHODS
    1. Field Studies
    2. Laboratory Freeze Tests
    3. The Effect of Light Duration During Hardening
  2. RESULTS AND DISCUSSION
    1. Field Studies
    2. Laboratory Freeze Tests
    3. The Effect of Light Duration During Hardening
  3. SUMMARY
  4. REFERENCES
  5. Table 1
  6. Table 2
  7. Table 3
  8. Table 4
  9. Fig. 1

Canola is an important oil crop worldwide and has significant potential for the central Great Plains. Both spring and winter types have been developed, but only those with the winter growth habit have suitable yield potential for the region. All commercial cultivars have inadequate winter hardiness and dormancy to survive consistently in the plains and production of the crop will not be economically feasible until these deficiencies are remedied. Sometimes the plants are killed by cold temperatures during the late fall-early winter (lack of winter hardiness), and other times the plants break dormancy during warm spells in the late winter-early spring and are killed by cold temperatures later (lack of spring survival). A crop of a hardy cultivar will survive some winters in Kansas, but attempts to grow canola consistently have failed. Seed yields of winter canola in Kansas test plots have been variable, ranging from near zero to over 3,000 kg/ha (Canola Production Handbook 1990). Low yields generally have been attributed to winter-kill. The wide diversity of genetic materials available should allow improved varieties with winter hardiness to be developed.

Breeding for better winter rapeseed began with population improvement and mass selection. A large number of plants were selected from a population based on plant phenotype and seeded together. This method is effective for highly heritable characters that can be determined before flowering, including winter hardiness in areas with more severe winters (Thompson and Hughes 1986). Field survival is the most commonly used method of evaluating winter plants for winter hardiness. This method is simple, requires minimum labor and is the ultimate test for winter hardiness (Fowler et al. 1981). Unfortunately, field trials are often inconclusive due to either complete death or complete survival of all or most plants within a trial (Fowler and Gusta 1979). Even when differential kill does occur, it is often irregular with large variations occurring within very short distances (Fowler 1979).

Generally, insufficient cold hardiness is also the main factor limiting the survival of cereals in regions with harsh winter environments (Fowler et al. 1981). Winter survival of crop plants is also influenced by many other factors (Levitt 1972). A University of Wisconsin study indicated winter survival in rapeseed was dependent on many morphological and physiological characteristics of the plant, soil conditions and weather fluctuations and that freezing tolerances differed between cultivars (Teutonico et al. 1993). In a winter canola study in Canada, sites with low winter survival suffered from late emergence, autumn drainage problems, freezing winter rains, or spring flooding conditions (Beaulieu and Hume 1987). As a result, screening methods that allow for accurate and precise assessment of cold hardiness potential are critical to winter crop research programs.

The inherent difficulties with field trials have stimulated interest in the development of cold hardiness prediction tests to complement field screening. Controlled freeze tests that employ a single medium temperature (Fowler et al. 1973), prolonged freezing (Thomas et al. 1988), or a range of temperatures to measure lethal dose temperature (Fowler et al. 1981) have been used to identify differences in plant cold hardiness. Controlled freeze tests have the advantage over field testing in that they are faster, give greater control of environmental conditions, and provide the opportunity for replication over time (Pomeroy and Fowler 1973). The most important advantage is that controlled freezing tests accelerate selection and provide positive results. They can be used year around to cause the differential cold injury that is needed for progress in breeding, where field trials can be used only once per year and frequently do not cause differential injury. A rapid screening test must be highly correlated with field survival, simple, repeatable, rapid, nondestructive, and require only a single plant for analysis (Fowler et al. 1981).

MATERIALS AND METHODS

Field Studies

Advanced experimental lines were derived from a number of field selections and advanced from segregating populations supplied from the University of Idaho in the late 1980s. These selections were originally made at Hays and Colby, Kansas, and Columbia, Missouri. An extremely diverse testing area was needed and established to increase the potential of having multiple locations were differential winter kill would occur (Fig. 1). Cooperators in nine states in the Great Plains region are growing this uniform yield test. The 1994-95 Advanced Canola Nursery (ACN) included 21 experimental lines and seven check cultivars. The 1995-96 ACN includes 12 experimental lines and 11 check cultivars. Tests are replicated three times and planted in a RCBD. Planting recommendations include seeding the tests six weeks before the first killing frost but individual cooperators use past experience and local conditions for making most management decisions. Data is collected for fall stand, winter survival, 50% bloom date, maturity date, plant height, lodging, shattering, seed yield, moisture content, and test weight. Total oil concentrations are also obtained.

Laboratory Freeze Tests

Seeds form nine lines were planted in the greenhouse to conduct one test. Seeds were placed in moist soil in each 2.5 by 5 cm cell of plastic six pack pots. After all plants were well established, each cell was thinned to one plant. Forty-eight plants of each line were established and maintained in eight six-pack pots in order to give eight experimental units (two replications of each line by four freeze combinations). After four weeks in the greenhouse, they were cold hardened in a growth chamber with an 11 h light period and the following temperature settings: Week 1, 7°C day and 4°C night; Week 2, 4°C and 1°C; Week 3, 1°C and -1°C; Week 4, 7°C and 4°C. At the end of the hardening period, the plants were placed in incubators for the freeze treatment. The temperature was set at -1°C for 10 h and then reduced at the rate of 1°C/h. When the temperature reached -9°C, that temperature was maintained. Plants were removed at 0, 12, 24, and 36 h. After one day recovery in mild temperatures, the plants were transferred to the greenhouse. After three week, survival counts were taken. In order for a plant to be considered alive, both active leaf and root growth was present. Percent survival for each experimental unit was calculated. LT50 (the time needed at -9°C to observe 50% death loss) were calculated using simple linear regression models. The same lines tested in this procedure were also grown in the field at three locations (Hays, Hutchinson, and Manhattan, Kansas) over the 1993-94 season and significant differences in winter survival were observed at all locations. Results obtained from this procedure (LT50 and mean percent survival) were compared to field data (Field Survival Indexes (Fowler and Gusta 1979) and percent winter survival). This procedure was repeated in order to test all lines grown in the field.

In addition to testing the cold hardiness of different lines, this procedure has been modified to select individual plants with increased cold tolerance. Segregating populations were exposed to -9°C for 30 h. Surviving plants were grown to seed and the procedure repeated. Seed obtained in this manner was planted along with seed obtained from plants of the same populations that were not exposed to lethal temperatures and the parents of the populations. This experiment was planted at two locations in Kansas in the fall of 1995 and survival data should be available in the summer of 1996.

The Effect of Light Duration During Hardening

Six rapeseed lines ('Cascade', 'Aspen', 'Cathy', PI458919, PI469737, and PI469892) were planted in six pack pots as described above. Three flats of 48 seedlings were established for each line. The lines were divided into three different treatments and planted at different times so plants would be at the same developmental stage for the freezing treatment. Four weeks after establishment, the cold hardened treatment (Chill) was placed in a vernalization chamber for four weeks. Three weeks after establishing the low light treatment (LLight), it was placed in the growth chamber. The light was reduced to 8 h with a day/night temperature regime of 18°/9°C. The unacclimated treatment (GrnHs) remained in the greenhouse (greenhouse lights were on 18 hours during the entire time, including the setup of the previous treatments). All treatments were thinned to 1 plants per pot two weeks after planting.

At the end of the acclimation periods, all treatments were placed in an incubator set at 0°C. After 5 h, the temperature was lowered to -1°C for 12 h. Chambers were then set to decrease 1°C/h until the temperature reached -5°C. They were then held at that temperature for 30 min and the -5°C treatment was removed. The temperature was reduced at the same rate and held at -8°C for 30 min and the -8°C treatment was removed. The temperature was decreased again and held at -11°C before the -11°C treatment was removed. The plants were initially placed in the growth chamber with 12 h light and 10°C day and 5°C night to recover. Plants were transferred to the greenhouse after two days. Plant counts were completed after three weeks.

RESULTS AND DISCUSSION

Field Studies

Although the winter of 1994-95 was milder than normal, seven of 12 locations still observed differential winter kill (Table 1). Over these seven locations, mean winter survival ranged from 90.1% for KS-M3579 to 62.2% for Cathy with an over all mean of 83.5%. Four additional locations had 100% survival and one location had near total deathloss. Very few lines had stands that were not harvestable at any location. Fourteen of the 15 lines in the first LSD group were experimental selections advanced in the Great Plains. Winter survival from two location in Kansas for 1993-94 of the lines grown in those tests is also included in Table 1. Conditions were much more severe at these tests and mean survival for all entries was 36.7%.

Although no experimental lines have proven themselves in field conditions, progress has been made in increasing winter survival by selecting in our environment. All released cultivars previously grown in the Great Plains have been developed in Europe, Canada, or the Pacific Northwest. These are areas with very different environments from that of the Great Plains. One experimental line KS3579, was released in the fall of 1995 as a germplasm. Canola quality breeder seed will be increased for four other lines with a decision on increasing foundation seed made in 1996.

Laboratory Freeze Tests

Significant differences in both freeze time and genotype were detected in both tests (Table 2). In all lines tested, as time at -9°C was increased, survival was decreased. Over all, survival averaged 76.9% at 0 h, 32.1% at 24 h, and 18.4% at 36 h. Survival ranged from highs of 100% at 0 h and 50% at 36 h to lows of 66.6% at 0 h and 0% at 36 h. Significant correlations were detected for both tests when freeze test data was compared to data from the field (Table 3).

The laboratory freeze test has the potential to be an important tool in screening germplasm for cold tolerance. The 1993-94 growing season was one where the primary death loss came over the winter and was due to cold temperatures. This may play a part in the strong correlations between the laboratory and the field data. This test was replanted in 1994-95 but due to mild conditions, significant death loss was not observed at any location. Many factors, both physiological and environmental, are involved in winter survival and cold tolerance is only one important factor. Advancing seed from surviving plants in segregating populations using this procedure should select for genes responsible for cold tolerance. Two generations per year are possible using this technique. In the field, at most, one generation is possible and factoring in the winters where survival is near 100% or near 0%, improvement will be made much less frequently.

The Effect of Light Duration During Hardening

Significant differences were detected for temperature, hardening method, genotype, and a temperature by hardening method interaction. All hardening methods survived well at -5°C (96%) but the Chill and LLight had a 48% and 37% increase in survival over no hardening at -8°C. At -11°C, only the Chill treatment had any survival (12%) (Table 4). Although a significant interaction involving lines was not present, PI469737 and PI469892 did tend to have increased survival in the LLight treatment as compared to the other lines tested. These two lines also showed increased survival in the field in Texas when a mild fall was followed by an abrupt cold front (Auld et al. 1992). This evidence supports the role that photoperiod may play in hardening oilseed rape lines and that there are genetic differences in a rapeseed line's ability to harden with changes in the photo period.

SUMMARY

Progress in increasing winter hardiness has been made for the Great Plains region. Areas of the world currently growing winter rapeseed are environmentally different and no cultivars adapted to the Great Plains have yet to be identified. Populations that have sufficient cold tolerance and dormancy that are adapted to the winter photoperiod south of the 40th parallel need to be developed and identified. This will only be accomplished by incorporating traits form different rapeseed lines that has been developed in different parts of the world. Using the tools outlined should help speed the development of populations that are stable and consistent in the environmental conditions of the Great Plains.

REFERENCES


Table 1. Mean results of the 1994-95 Advanced Canola Nursery planted at 12 locations and winter survival (WS) results from 1993-94 planted at 2 locations.

Genotype Yield (kg/ha) 10 Loc Yield (%Avg) 10 Loc WS (%) 7 Loc Bloom (April) 7 Loc Height (cm) 7 Loc Shatter (%) 5 Loc Lodge (%) 5 Loc Test wt. (kg/m3) 6 Loc Moist (%) 9 Loc Oil (%) 3 Loc WS (%) 1993-94
KS-M3579 1417 108 90* 15E 110 5.8 10* 533 9.6* 33.0 88*
KS-C3505 1388 106 89* 22L 129t 6.2 8* 565* 11.2 34.3 --
KS-C3504 1439 110 89* 21L 131t 7.5 15* 546 11.4 34.3 --
KS-M3311 1200 92 88* 19 117 4.5 9* 515 10.8 35.0 75*
KS-M3580 1593* 122* 88* 19 120 5.7 12* 556* 9.9* 34.5 80*
KS-C3104 1204 92 88* 20 119 6.7 12* 569* 10.0* 34.6 --
KS-C806 1369 105 87* 18 109s 5.0 14* 519 9.9* 33.2 --
MO-503-1 1509* 115* 87* 19 123 9.1 11* 556* 10.5 33.6 --
KS-C3208 1192 91 87* 20 120 4.1 10* 539 10.4 33.7 --
MO-503-24 1419 108 86* 20 124 5.7 9* 546 10.0* 35.1 --
KS-M3635 1477 113 86* 20 124 6.7 10* 550 10.8 34.2 68*
MO-513-9 1370 105 85* 19 122 7.7 12* 550 10.6 35.2 --
KS-C1701 1340 102 85* 20 125 4.6 5* 534 11.0 35.0 --
Ceres 1672* 128* 85* 21 123 4.5 8* 555* 11.7 34.2 30
MO-503-9 1343 103 85* 20 124 8.0 16* 529 10.5 34.6 --
MO-503-2 1327 101 84 19 123 6.4 15* 552 9.9* 36.0* --
KS-C2504 1308 100 83 16 110 5.9 23 528 10.4 33.6 --
KS-M3723 920 70 83 20 109s 8.8 25 529 9.8* 34.7 65*
Glacier 1312 100 83 21 128t 5.7 18 562* 10.7 34.8 58
KS-M3403 1151 88 83 20 117 6.8 15* 543 10.3* 33.1 70*
KS-C2403 1139 87 82 18 113 5.1 18 529 10.4 36.1* --
Winfield 1161 89 82 17 114 5.7 18 528 9.6* 37.0* --
KS-M3346 1282 98 81 19 118 3.9 10* 535 9.8* 35.0 78*
Bridger 1062 81 81 17 109s 8.3 37 538 10.1* 35.4* 58
KS-M3314 1250 96 79 19 116 6.1 15* 541 11.4 34.4 53
Liborius 1445 110 78 21L 125 6.0 9* 573* 11.0 36.2* --
KayStar 11 1575* 120* 73 21L 126t 5.7 10* 579* 11.4 36.2* --
Cathy 811 62 62 20 105s 5.8 37 537 9.4* 34.0 --
Mean 1309 100 84 19 119 6.2 15 544 10.4 34.7 37
LSD (.05) 193 15 6 1 5 NS 11 26 1.0 1.6 23
CV 29 29 12 8 7 87 102 7 17 5 31


Table 2. Survival from two freeze and three field tests.

Freeze test Field tests
Line LT50z Survival (%) FSIy Survival (%)
Test 1
Bridger 24 42 66 39
PI531284 23 33 67 40
PI305282 22 33 59 33
PI535849 19 30 63 36
PI469793 16 29 64 34
PI535868 14 27 53 26
PI458956 11 23 52 20
Ceres 9 24 59 31
Test 2
A112 -4 17 25 7
PI458959 42 65 59 31
Ceres 37 71 59 31
PI458925 28 68 50 23
PI535871 23 60 58 32
PI305282 23 57 59 33
PI470007 23 55 59 33
PI535849 20 55 63 36
zLT50=Time in hours at -9°C where 50% death is observed.
yFSI=Field Survival Index from Hays, Hutchinson, and Manhattan Kansas, 1993-94.


Table 3. Correlation between field and laboratory data for two freeze tests.

Test 1 Test 2
Freeze Test Field r p r p
% Survival % Survival 0.86 0.002 0.77 0.022
% Survival FSI 0.79 0.010 0.83 0.008
LT50 % Survival 0.92 0.000 0.82 0.010
LT50 FSI 0.90 0.001 0.88 0.003


Table 4. Survival from the artificial freeze test. Different pre-freeze hardening methods include chilling treatment; low light at growing temperatures, and greenhouse conditions.

Survival (%)
Freeze temperature
Hardening method Line -5°C -8°C -11°C Mean
Chill Casca 100 75 8 61
Chill Aspen 100 55 8 54
Chill Cathy 100 33 5 46
Chill Spr19 100 50 0 50
Chill Wnr37 100 75 25 67
Chill Wnr92 100 30 25 52
Chill Mean 100 53 12 55
LLight Casca 78 25 0 34
LLight Aspen 75 40 0 38
LLight Cathy 100 42 0 47
LLight Spr19 92 18 0 36
LLight Wnr37 92 58 0 50
LLight Wnr92 100 75 0 58
LLight Mean 89 42 0 44
GrnHs Casca 100 14 0 38
GrnHs Aspen 100 0 0 33
GrnHs Cathy 100 0 0 33
GrnHs Spr19 83 0 0 28
GrnHs Wnr37 100 17 0 39
GrnHs Wnr92 100 0 0 33
GrnHs Mean 97 5 0 34
LSD(.05) Hard & Temp 51 51 51 40
LSD(.05) Lines 28 28 28 16


Fig. 1. Test sites of the 1995 and 1996 ACN.


Last update August 19, 1997 aw