Art02_Franzaring.indd Journal of Applied Botany and Food Quality 85, 134 - 143 (2012) Institute for Landscape and Plant Ecology (320), University of Hohenheim, Germany Effects of increased autumn temperatures and sub-acute levels of ammonia on post-winter development of four cultivars of winter oilseed rape (Brassica napus L.) J. Franzaring, I. Holz, A. Fangmeier (Received February 14, 2012) Summary Early planting dates, warm autumns and increased N mineralisation rates may advance the growth of winter crops and also a high supply of nitrogen poses risks of excess biomass accumulation before the winter leading to reduced frost hardiness and excessive leaf loss. Although the number of frost days is currently declining in tempe- rate Europe, events of frost may still create damage in the future in winter crops that have accumulated too much biomass in the prece- ding autumn. Here we report on a study in which the responses of four cultivars of winter oilseed rape (OSR) to elevated autumn temperatures and sub- acute levels of ammonia (NH3) were compared. It has been suggested that high concentrations of the gas, e.g. in livestock intensive regions may both act as an airborne fertiliser and reduce the frost hardiness of plants. Advancing treatments were imposed for 88 days in the autumn and consisted of pot grown plants kept in greenhouses with- out (ELVT) and with elevated concentrations of ammonia (ELVT+A; 195 µg m-3 NH3). Reference plants (AMB) were raised outdoors, where temperatures were 1.4°C lower than in the greenhouses. After the treatments plants were all overwintered outside to study whether the pre-winter growth advancement was still discernible in the following spring. Shed leaves were collected weekly to fol- low how much shoot biomass was lost during and after the winter. Cultivars responded differently to warming and exposure to ammo- nia. Up to the winter shoot biomass was strongly increased by the advancing treatments. However, fi nal shoot mass in the following summer did not differ between cultivars and was unaffected by the higher temperatures in the preceding autumn. Nevertheless, sig- nifi cantly more biomass was observed in ammonia fumigated plants. Higher autumn temperatures increased leaf shedding and advanced fl owering and senescence in the next spring so that plants showed a signifi cantly reduced seed mass, harvest index and oil yield at the fi nal harvest. Obviously, the growth advancement in the preceding autumn by elevated temperatures negatively affected the availability of resources in the following spring. In contrast, plants that were grown at both elevated ammonia and temperature in the autumn showed a delayed fl owering, higher shoot and seed mass, increased harvest index and oil yield. We conclude that growth advancement by elevated autumn temperatures without the re-supply of nutrients increases leaf shedding during winter. Nevertheless, the loss of re- sources in winter for re-growth in spring will certainly be of minor importance for yield formation as compared to the frost damage resulting from late spring frosts. Introduction Oilseed rape (OSR) is currently the most important bioenergy crop in the temperate regions and much effort has been made to improve yields and the oil quality of this species in recent decades. Climatic changes may shift the current OSR crop belt in the near future and in its current cultivation range increasing temperatures may affect the phenology, speed up the crop development and eventually reduce the seed output due to the shorter crop growth cycle. At the same time, seed maturation, yield and oil quality may directly be affected by planting dates and high temperatures (MORRISON and STEWART 2002; AKSOUH-HARRADJ et al., 2006; PELLET et al., 2008; BAUX et al., 2008; FARAJI et al., 2009). It may also be hypothesised that with the predicted climatic changes, new pests and diseases will occur in the future, while the rising winter temperatures may not be suited to the vernalization requirements of some winter crop cultivars any more. Fig. 1 shows time series for OSR yield increments since 1961 and changes in the average duration of the culture period between the years 1951 and 2008 for Germany. Because sowing and harvest dates have been moved to earlier and later dates, respectively, average winter OSR culture periods increased from the 1950s to the 1980s but then decreased since the 1990s. At the same time OSR fl owering occurred 15 days earlier in the decade 1999 to 2008 as compared to the climate normal period (Fig. 2). It is unclear, however, whether and how strong the introduction of new cultivars has had an effect on the observed phenological changes and the development of yields. There is also no information available whether yield increments over time and changes in crop duration are interrelated. Fig. 1: Changes in the average duration of the winter OSR cultivation period (○, y-axis to the left) and yields (■, y-axis to the right) in Germany. Phenological records (1951-2008) stem from the database of Deutscher Wetterdienst (DWD) and data on oilseed rape yields in Germany (1961-2008) from FAOSTAT. Effects arising from changes in the fl owering window due to de- viations from the normal climatic program can be manifold. Earlier onset of fl owering may be especially critical if pollinators are still absent and if late frosts occur. The fl owering window of OSR is about one month in total and adverse climatic conditions may well occur during this rather long time period affecting at least part of the 250 275 300 325 350 1950 1970 1990 2010 D ay s fr o m so w in g to h a rv es t 0 20 40 60 80 d t h a -1 Effects of autumn growth advancement in winter oilseed rape cultivars 135 fl owers. Because the climate in central Europe may become more variable in the future, the uniform phenological development may be impaired if the heat accumulation over time and the photoperiodic response of crops and cultivars do not match any more. Besides changes in the beginning of fl owering, the alteration of the plant phe- nological program by climatic changes may also affect the timing of root and leaf development, which drive the resource acquisition and nutrient use effi ciency (NUE) of plants (NORD and LYNCH, 2009). While the timing of fertilizer application may be adapted to these new conditions, the availability of water in the future will depend on the timing and intensity of rainfall. However, variations in precipita- tion patterns under climatic change are currently much more diffi cult to forecast than changes in temperatures. Responses of oilseed rape to higher spring temperatures PELTONEN-SAINIO et al. (2009 a, b) addressed the current and future potential of OSR farming in Northern Europe suggesting that crop- ping of winter instead of spring varieties may be more favourable in a warmer climate. Information is lacking, however, on problems with OSR cropping in the southern European regions that might arise from climatic changes. The suitability of irrigated oilseed rape cultivation as a winter rotational crop in semiarid regions of the South-Western US was addressed by ADAMSEN and COFFELT (2005). Planting the same spring varieties as used in the Great Plains in November at Maricopa Agricultural Center (Arizona) lead to acceptable yields but plants produced too much vegetative biomass (straw) at the expense of seed mass. A reason for the lower harvest index may be the high temperatures in the study region that affected fl owering and seed fi lling. MORRISON and STEWART (2002) found that fl oral fertility and seed fi lling were signifi cantly reduced at high air temperatures above 29.5°C. It may thus be suggested that the increasing frequency of hot spells even in the cool and temperate regions may reduce the seed set of OSR. This could be the case in both winter and spring cultivars and, indeed, PELTONEN-SAINIO et al. (2007) have attributed the yield losses observed in Finnish OSR over the past 15 years to the increased occurrence of summer heat. In contrast to this study, the halt in OSR farm yields since 1985 in the UK has been attributed to crop management factors (shorter rotations, direct drilling, lower ap- plication rates of nutrients) rather than to changing climatic factors and failures in germplasm improvement (BERRY and SPINK, 2006). Still, the authors suggest that an increase in temperatures by 3°C will reduce the seed fi lling period by 13 days. Using historical crop yield and climate data from Canada, KUTCHER et al. (2010) found that the beginning of July, i.e. the beginning of the fl owering period, was the critical period in the determination of OSR yields. Low yields were strongly associated with high July temperatures whereas high yields were associated with greater-than-average precipitation, and to a lesser extent, cooler-than-average nocturnal temperatures. In contrast to the negative effects of high temperatures during fl ower- ing and the seed determination period 19-25 days after mid-fl owering the earlier fl owering caused by warmer springs may also be regarded as a positive response. This is due to the fact that light refl ection by the fl owers will be reduced earlier so that leaf area and the assimi- lation of carbohydrates may overall profi t. One would then expect higher yields provided the absence of thermal stress later on. BERRY and SPINK (2006) state that bringing forward fl owering by one week could improve yields but it is unclear whether the earlier fl owering and increased yields in Germany (Fig. 1) are interrelated. Still, the strong yield depressions by 20% in the hot summer 2003 and the warm spring in 2007 indicate adverse climatic infl uences on OSR and give some idea on the effects of thermal stress in the future. Fig. 2: Beginning of fl owering of winter OSR in Germany during the WMO climate normal period 1961-1990 (left) as compared to the warmer decade 1999- 2008 (right). Phenological data for over 3500 stations (black dots) from Deutscher Wetterdienst (DWD). Maps were produced on the basis of a digital elevation model (1 km x 1 km) with local trend surface prediction (BIVAND et al., 2008) using the statistical programming environment R. Numbers in scale represent days of the year (DOY, with DOY 1 = 1 January). In most regions fl owering occurs 10 to 15 days earlier nowadays than during the climate normal period. 136 J. Franzaring, I. Holz, A. Fangmeier Responses of oilseed rape to elevated autumn and winter tem- peratures Despite heat stress and heat spells probably occurring more fre- quently in the future, frost will still play a signifi cant or even in- creasing role in a warmer world (GU et al., 2008). This is because more severe damage or post-freeze setback may turn up in plants that experienced a strong pre-freeze phenological advancement due to warmer conditions. In winter crops, suffi cient biomass and cryoprotective compounds must be accumulated between the sowing and the onset of the win- ter to increase their frost hardiness. On the other hand, too much shoot biomass and premature shooting will increase the danger of frost damage due to the exposure of aerial plant parts and the occur- rence of frost drought. In order to prevent premature shooting under climate change, farmers may apply plant growth regulators to keep plants in the rosette stage but this investment may narrow the bene- fi ts. To respond to the increased likelihood of shooting before the onset of winter, sowing of winter OSR cultivars could eventually be moved to later dates in the near future. In spring cultivars, however, sowing dates should be as early as possible because late planting will result in lower yields, a reduced harvest index and lower oil contents (GROSS, 1964; KONDRA, 1977; JOHNSON et al., 1995). While the concept of accumulated thermal time (growing degree days, GDD) is successfully applied to predict the development of spring and winter OSR (MORRISON et al., 1989) in spring, not much information is available on the relationship between heat accumulation and pre winter growth of winter OSR. According to DIEPENBROCK (2006a) winter OSR should be sown in Germany between 15 August and 15 September at a density of 60 to 90 seeds per square metre. Until winter plantlets should be in the 6- to 8-leaf stage and have an optimum shoot dry weight of approximately 1 g per plant. This will give plants a high chance to survive the winter. Growth completely ceases below 2°C and low temperatures will induce the accumulation of phospholipids in plant leaves, which renders them more resistant to frost dehydration (RAPACZ, 1998, 1999, 2002; CHRISTEN and FRIEDT, 2007). Another prerequisite to withstand frost is the existence of a thick root collar containing large amounts of soluble carbohydrates, which will also benefi t a rapid re- growth in spring (LÄÄNISTE et al., 2007). During the winter a mean of 50% of the dry shoot mass will normally be lost, but higher losses may occur due to frost damage (DIEPENBROCK, 2006b). LÄÄNISTE et al. (2007) have addressed the importance of sowing dates on winter hardiness of OSR in Estonia, which is currently the northern border for the winter cultivars of this crop. Based on three years trials they found that winter survival was strongest in plants that had received 416 growing degree days (GDD above a base temperature of 5°C). While plants from late sowing treatments did not survive the winter also the plants from the early sowing variants were severely affected by frost damage. The authors found that at 500 GDD the percentage of overwintering plants was below 50%. Effects of N-availability on the frost hardiness of plants Another important non-climatic factor, which may strongly advance the development of winter crops, is of course the availability of fer- tilisers. While P and K fertilisers may be applied to the seedbed to increase yields, nitrogen availability in autumn does not have pro- nounced effects on the fi nal yields as long as the fertilisation is ade- quate in the following spring. According to COLNENNE et al. (2002) even severe autumn N defi ciencies will allow enough shoot mass accumulation to ensure suffi cient re-growth in spring. In fact, excess fertilisation in autumn may increase the danger of frost damage in crops due to the stronger built up of shoot biomass and the exposure of aerial plant parts to frost desiccation. Apart from soil-N, elevated concentrations of ammonia near live- stock farms may signifi cantly contribute to the N-supply of plants (SOMMER, 1988, SOMMER et al., 1991, 2009). The uptake rates of gaseous ammonia have been shown to be elevated at higher soil fertilisation rates, which may be explained by the higher leaf area index at higher availability of soil N that will lead to higher N depo- sition to the crop. Also BÖHME et al. (2003) found that crop biomass and canopy area determine how much nitrogen will be scavenged from the atmosphere and made available for additional plant growth. While N utilisation from airborne ammonia will be largest in grow- ing stands, plants will switch from sinks of NH3 to active sources during the senescence due to shifts in compensation points. Bi-direc- tional fl uxes of ammonia between the atmosphere and OSR leaves have been studied by HUSTED et al. (1995, 1996) and MASSAD et al. (2009) under controlled conditions and by NEMITZ et al. (2000) and SCHJOERRING et al. (2000) in the field. Whereas it has been suggested that even low environmental con- centrations of ammonia may reduce the frost hardiness of plants from acidic heathlands (e.g. Calluna, SHEPPARD and LEITH, 2002; SHEPPARD et al., 2008), exposure to extremely high concentrations (1000 nl L-1) of ammonia did not affect low-temperature hardening of conifers and wheat (CLEMENT et al., 1995, 1999). Physiological mechanisms potentially involved in the direct effects of ammonia on the frost hardiness of plants relate to the reduced accumulation of osmotic compounds (polyamines and carbohydrates) and phos- pholipids allowing plant tissues to adapt to and withstand low temperatures. While these relationships have not been well estab- lished, indirect effects of ammonia on the frost sensitivity of plants may occur due to the increase in shoot biomass. Scope of the study We assumed that higher autumn temperatures and sub-acute con- centrations of ammonia as an additional N-source would fi rst stimu- late pre-winter plant growth and later increase the risk of leaf loss in winter oilseed rape. We further postulated different responses in different cultivars and that carry-over effects from the autumn treatments would be recognisable in the formation of yields and the quality of seeds in the following spring. To test these hypotheses, we grew different cultivars of OSR in two consecutive experiments with an early and a late sowing date under ambient and elevated tempera- tures to quantify the additional biomass production of selected culti- vars under warmer autumn conditions. We used a third treatment, in which plants were exposed to both elevated temperatures and sub- chronic concentrations of ammonia to study the additional stimula- tion of shoot growth. Plants from the late sowing were over-wintered to observe how the treatments affected the shedding of leaves during winter and how much biomass and seeds were produced in the fol- lowing spring. Performing such pot-based studies under controlled climatic conditions and collecting the shed leaves at fi xed intervals may serve to address the resource use effi ciency as well as the progress of senescence in higher temporal resolution as compared to fi eld trials, in which shed biomass cannot be accounted for. Materials and methods Plant cultivation Two individual experiments with different sowing dates were per- formed to investigate the effects of elevated autumn temperatures and sub-acute levels of ammonia on the growth of winter oilseed rape (OSR). While plants from the fi rst experiment (early sowing date) were harvested before the winter to address the plant responses in late autumn, only the plants of the second experiment (late sow- ing date) were overwintered to study the effects of elevated autumn Effects of autumn growth advancement in winter oilseed rape cultivars 137 temperatures and ammonia exposures on yield formation and seed quality in the following spring. In the fi rst experiment, OSR was sown on 15 August 2008 (DOY 228), which is three days earlier than the sowing date in Hohenheim in recent years. In the second experiment the crop was sown fi ve weeks later on 22 September 2008 (DOY 266). After sowing, each of the 48 pots (4 per cultivar * 4 cultivars * 3 treatments) was fi xed in packs of 6 pots on polystyrene racks placed above individual plastic water containers to be able to follow water consumption over time. Water consumption was registered for each of the pots and water was re- fi lled on demand to the plastic containers at known amounts. The four cultivars used in the experiments ‘Asgard’ (A), ‘Express’ (E), ‘Lorenz’ (L) and ‘Viking’ (V) were provided by Norddeutsche Pfl anzenzucht Hans-Georg Lembke KG (NPZ Holtsee). Seed lots of cultivars A, E, L and V had thousand seed weights (TSW) of 4.9, 4.5, 4.7 and 4.3, respectively. Five seeds per pot were sown at a depth of 1.5 cm in plastic containers (truncated cone, L=13 cm, B=13 cm, H=13 cm, A=169 cm2, V=1.7 L) and reduced to one plant per pot representing a fi nal density of 60 plants m-2. Pots were fi lled with 0.65 kg standard soil (Fruhstorfer LD 80, Archut-Hawita Gruppe, Vechta) and equipped with glass fi bre wicks (Type 6000, Hecker Werke GmbH, Weil im Schönbuch) for automatic watering. At water saturation the substrate has a maximum water content of 73.6 wt. % and 27.7 vol. %. The soil substrate is approved after EN 12580 and contains bark humus, volcanic clay and peat (35% organic matter in dry mass and pH 6.1). It is enriched with Radigen® slow release micronutrients and Aquafl ow® slow release NPK fertilisers and has a macronutrient content of 150 mg L-1 N and P2O5 each as well as 250 mg L-1 K2O. Accounting for the pot volume and its area the initial nitrogen availability represented a value of 150.9 kg ha-1. Until winter no further nutrients were supplied, because plants were har- vested in the rosette stadium. In the late sowing experiment, which lasted until seed maturation, plants received 100 mL of a nutrient solution (VDI, 2008) on 23rd of March 2009, representing a second gift of 0.5 g N per pot. The VDI standard solution is used for curly kale (Brassica oleracea L.) cultures and for its suffi cient sulphur concentrations was believed to be well suited for OSR pot cultures. The total N availability (slow release fertiliser plus the nutrient so- lution) amounted to 0.755 g pot-1 and this quantity was taken into account to derive the N economy parameters nitrogen use effi ciency (NUE) and nitrogen recovery in the seeds. Treatments and monitoring of ammonia and climate Three treatments were applied in the experiments. One set of plants was kept under ambient conditions (AMB) in a fenced premise which was covered by a transparent roof to exclude rainfall. The two other treatments were grown in greenhouses (Floratherm® Alltop KA 200 Krieger GmbH Herdecke, L 200, B: 185 cm, H: 210 cm, V: 7 m3) to raise the air temperatures by ca. 2°C above the ambient con- ditions (ELVT). In addition to the elevated temperatures one of these warm treatments was subjected to elevated ammonia concentrations (ELVT+A). Both greenhouses, i.e. treatments ELVT and ELVT+A, were equipped with a charcoal fi lter unit, a ventilator and fl exible tubes (Ø15 cm), through which air was lead into the greenhouse. Air was exchanged 54 times per hour in both of the greenhouses. To enrich the ELVT+A treatment with ammonia, pure NH3 (quality 3.8, 40 L fl ask, SWF Friedrichshafen) was lead via a Tefl on tube into the air stream. Outlet pressure of the fl ask was reduced from 8 to 1 bar resulting in an average fl ow of 1.5 mL min-1 of the gas. This supply was calculated to yield a set ammonia concentration of about 200 µg m-3 (285 ppb) in the ELVT+A treatment. In order to avoid placement effects the treatments were switched between the two greenhouses bi-weekly. At these dates also the pots were randomly re-arranged. Ammonia concentrations were monitored daily in the ELVT+A and weekly in the other treatments using the Radiello® passive sampler method (Sigma Aldrich, FSM, 2006). The adsorbing cart- ridge is coated with phosphoric acid, which adsorbs ammonium at a sampling rate of 235 mL min-1. The quantity of ammonium was determined against standards at 635 nm by visible spectrometry (Beckman Coulter DU 640B, Fullerton) in the form of indophenol which develops from the reaction between phenol and sodium hypo- chlorite in the presence of cyanoferrate. In order to calculate ammo- nia concentrations, the NH4+ mass per sampler was divided by the product of the duration (minutes) of the exposure and the sampling rate. In order to obtain information on the temperatures and relative air humidity in the three treatments calibrated Tinytag temperature loggers (Gemini Data Loggers, UK) were operated on a 10 min rou- tine inside and outside the greenhouses. Measurements on oilseed rape In the early sowing experiment, germination and numbers of germi- nated seedlings were recorded. On day 7 after germination (7 DAG) one seedling per pot was harvested at random and the number of seedlings per pot was reduced to 3 individuals to reduce plant com- petition. Two more harvests took place 14 and 21 days after germina- tion (14 DAG and 21 DAG) leaving just one plant per pot. Leaf area and fresh weight of plant fractions leaf and stem were determined. Harvested plants were put into labelled paper bags and were dried at 80°C for 24h in a drying cabinet. After drying, dry matter (DM) of the seedlings was determined and absolute (AGR) and relative growth rates (RGR) were calculated accordingly. Calculations of growth rates were based on average plant weights (n=4 per treatment and cultivar). Leaf greenness was determined weekly with a SPAD- meter (Minolta SPAD 502) to follow differences in chlorophyll contents in the different treatments. Five measurements were taken on the top of the youngest completely unfolded leaf and averages of these measurements were recorded. In the late sowing experiment the same growth related parameters were studied. Like in the early sowing experiment root biomass was not analysed. However, much effort was put on the harvest of dead leaf material, which was present after the winter and which accumu- lated during the senescence of the plants. Shed leaves were collected every week and shoot length and fl owering phenology were recorded thoroughly. At the onset of fl owering, SPAD values were determined on a weekly basis, with measurements being performed on marked leaves at a height of 80 cm. We chose a leaf from a high position instead from the rapidly senescing rosette to be able to get extended information on the whole process of senescence. At the fi nal harvest shoot biomass was separated into stem, leaves, pod walls and seeds. While the fractions stem, leaves and pod walls were dried at 80°C, seeds were air dried to constant weight. Finally, the dry weight of all fractions was determined. Seeds of each plant were counted with a Contador seed counting machine (Pfeuffer GmbH Kitzinigen, Germany) to derive thousand seed mass. Seed oil contents were analysed using the near infrared spectroscopy (NIRS) standard procedure outlined in VDLUFA (2002). N concentrations of leaves that had developed before the winter and seeds harvested the following spring were determined by elemental analyses and seed PK- concentrations were determined by ICP-OES following acid digestion. Data evaluation and statistics From the original biomass data determined at several harvests, specifi c leaf area (SLA), relative growth rates (RGR) and absolute growth rates (AGR) were calculated after HENDRY and GRIME 138 J. Franzaring, I. Holz, A. Fangmeier (1997). All the biological data were subjected to descriptive statisti- cal analyses and one or two-factorial ANOVA to identify signifi cant treatment effects as well as treatment* cultivar interactions. Posthoc Tukey Honest Signifi cant Differences (HSD) tests were used for multiple comparisons to study whether there were signifi cant diffe- rences in the response of the chosen cultivars. All statistical analyses and most of the graphical representations were performed using the open source programme R and different libraries (R DEVELOPMENT CORE TEAM, 2009). Results and discussion Climate and ammonia concentrations The daily mean temperatures recorded inside the greenhouses (treat- ments ELVT and ELVT+A) and outside (AMB) are presented in Fig. 3. The temperature curve from 15 August 2008 (early sowing of OSR) to 24 June 2009 (harvest of OSR from the late sowing experi- ment) is also compared to the long term daily temperature means at Stuttgart, Germany. As can be seen in Tab. 1, ambient temperatures in the autumn 2008 were slightly higher than the long term average. However, in December 2008 and January 2009 mean temperatures remained below the long term average for those months. The ex- tended frost spell early in the year therefore proved to be well suited to study the effect of advanced autumn growth on the winter hardi- ness of OSR. During the warm autumn treatment (ELVT) in the early sowing ex- periment (15.08.-13.10.2008) temperatures were 2.4°C higher than in the AMB treatment. In the late sowing experiment, in which the treatments lasted from 22.09. to 19.12.2008 (88 d) the realised tem- perature increase was only 1.4°C. Mean temperatures did not differ between the two greenhouses, which served to increase the tempera- tures (ELVT and ELVT+A). Mean daily ammonia concentrations during the early sowing ex- periment were 2.5 (± 0.7) µg m-3 outside (AMB), 7.3 (± 6.4) µg m-3 inside the untreated greenhouse (ELVT) and 275 (± 99) µg m-3 inside the ammonia treated greenhouse (ELVT+A). In the late sowing ex- periment the mean daily NH3-concentrations ranged from 6.8 (± 6.4) µg m-3 outside (AMB) to 6.4 (± 3.2) µg m-3 inside the untreated greenhouse (ELVT) and 195 (± 93) µg m-3 inside the ammonia Fig. 3: Mean daily temperatures during the growth experiments as compared to the long term daily averages and changes in day length. Warm autumn treat- ments lasted from 15 August to 19 December 2008. Grey bars indicate the duration of the experiments I (early sowing) and II (late sowing). Long term temperatures are based on the 1991-2005 time series from Stuttgart-Airport (Data available from Deutscher Wetterdienst, DWD). treated greenhouse (ELVT+A). The ammonia treatments resulted in sub-acute ammonia concentrations comparable to those measured after slurry applications. Responses during the autumn treatments Seeds germinated at a rate of almost 100% with no signifi cant differences between the cultivars and the three treatments in both the early and late sowing experiments. However, there were slight cultivar differences in the speed of germination in the follow order A>V>L>E, but the different speed of germination was unrelated to growth differences subsequently evolving in the three treatments. Nevertheless, the availability of seed resources had slight effects in the seedling establishment phase. In the AMB treatments of the early sowing experiment, thousand seed masses (TSM) of the cul- tivars were positively associated with their absolute growth rates (AGR) between 7 and 21 days after germination, while the opposite relationship was observed in the plants that were exposed to higher temperatures. At the same time, the OSR cultivar ‘Viking’ with low availability of seed resources (low TSM) responded more strongly to the atmospheric nitrogen fertilisation in the ELVT+A treatment than the other lines (data on growth performance in intermediate harvests not shown). OSR plants from all cultivars and from both the early and late sow- ing experiments were able to absorb ammonia via their shoots and achieved a signifi cantly higher shoot mass than in the other treat- ments. Fig. 4 indicates that shoot mass was highest in the ELVT+A treat- ments and reached an average of 20 g DM in the early and 3.5 g DM in the late sowing experiment. The results confi rm that advancing the plant development by higher temperatures and elevated ammonia concentrations may strongly increase the shoot biomass produced before the onset of the winter. In both of the experiments shoot bio- mass of the ELVT and ELVT+A treatments was well above a shoot dry weight of 1 g, which has been considered the minimum plant biomass for the winter survival by DIEPENBROCK (2006a). Even the plants from the late sowing experiment in the AMB treatment had approached this value indicating that OSR sowing could be moved to the beginning of September. Effects of autumn growth advancement in winter oilseed rape cultivars 139 Tab. 1: Mean temperatures during the experiments, temperature sums accumulated until the individual harvests and mean shoot biomasses recorded across the four different OSR cultivars. Temperature sums (Growing Degree Days, GDD) refer to daily temperature means above a threshold of 5°C. DAG represents days after germination. Long-term data (1991-2005) are based on daily records from Stuttgart-Airport (data from DWD). Day Mean temperatures (°C) Heat units (GDD>5°C) Shoot biomass (g DM) length Long- outside inside Diffe- outside inside Diffe- outside inside Increment (hrs) term AMB ELVT rence AMB ELVT rence AMB ELVT (%) Experiment 1 Sowing date 15.08.08 DAG 0 Harvest 1 01.09.08 DAG 7 13.9 18.1 19.8 22.9 3.1 266 330 64 0.055 0.10 96.5 Harvest 2 08.09.08 DAG 14 13.7 17.4 19.5 22.6 3.1 363 447 84 0.35 0.54 54.5 Harvest 3 15.09.08 DAG 21 13.5 16.8 18.9 21.7 2.8 442 538 96 1.46 2.44 67.4 Harvest 4 13.10.08 DAG 49 12.7 14.7 15.9 18.3 2.4 652 795 143 15.1 16.5 9.3 Experiment 2 Sowing date 22.09.08 DAG 0 Harvest 1 06.10.08 DAG 7 11.8 12.5 11.2 13.1 1.9 83 109 26 0.01 0.01 0.0 Harvest 2 13.10.08 DAG 14 11.6 11.9 12.6 14.6 2.0 156 193 37 0.03 0.05 111 Harvest 3 20.10.08 DAG 21 11.4 11.1 12.9 15.0 2.1 207 265 58 0.08 0.26 201 Harvest 4 13.11.08 DAG 45 10.7 9.3 11.0 12.9 1.9 261 339 78 0.89 2.49 180 Harvest 5 24.06.09 DAG 268 11.5 7.1 8.8 9.2 0.4 1263 1341 78 45.7 43.2 -5.3 Mean temperature (22.09. to 19.12.) 6.5 7.7 9.1 1.4 Fig. 4: Box plots showing the mean shoot biomass and harvest index of four OSR cultivars grown in early and late sowing experiments. Autumn treatments refer to ambient temperatures (AMB: 1), elevated temperatures (ELVT: 2) and elevated temperatures plus ammonia (ELVT+A: 3). Winter shoot biomass from the early sowing experiment (a) was harvested on 13 October 2008 (49 DAG) and that of the late sowing experiment (b) one month later (45 DAG). Final shoot biomass of the late sowing experiment (c) was harvested on 24 June 2009 (268 DAG) and harvest index (d) relates to the contribution of fi nal seed to shoot mass. 140 J. Franzaring, I. Holz, A. Fangmeier The strong increase in shoot mass due to warmer autumns and a luxurious supply of fertilisers (here airborne ammonia) may induce premature shooting and increase the frost sensitivity of plants. The maximum leaf number prior to the onset of the winter period should not exceed 8 leaves according to DIEPENBROCK (2006a), but in the early sowing experiments plants had 11.4 in the AMB and 13.3 leaves on average in both the ELVT and ELVT+A treatments. It may thus be expected that plants would have not survived the follow- ing cold winter. However, plants were harvested prior to the winter. Large differences in leaf numbers were found between the cultivars. While V had only 8.2 leaves in the AMB treatment, the other culti- vars produced 11.7 (E), 12.2 (A) and 13.2 (L) leaves on average in the early sowing experiment. While A, E and V did not signifi - cantly increase leaf numbers due to warming in the ELVT treatment, cultivar L signifi cantly increased leaf numbers to 18. It may thus be followed that L should not be sown too early, whereas V is much less temperature sensitive. The plants from the late sowing experiment produced less than 8 leaves before the winter but all plants from the late sowing experiment survived the cold winter of 2008/2009. Leaf numbers were 5.1, 7.4 and 7.7 in the AMB, ELVT and ELVT+A treatments, respectively and like in the early sowing experiment cul- tivar L responded strongly to elevated temperatures with 8.25 leaves on average (data not shown). The accumulated heat sum (GDD above a threshold of 5°C) until the fi nal harvest of the early sowing experiments was 652 and 795 in the AMB and the ELVT treatments, respectively. According to LÄÄNISTE et al. (2007) such thermal sums would be much too high for the survival of OSR under cool winter conditions. However, these fi ndings were made in Estonia and might not be representative for milder oceanic climates. Winter hardiness of OSR in Estonia was strongest in plants that had received 416 GDD between sowing and the onset of the winter. The plants from present late sowing experi- ments received only 261 and 339 GDD in the AMB and ELVT treat- ments between sowing (21.09.2008) and 31 December but all of the plants survived the following cold winter. Responses during the following spring The development of plants subjected to the three autumn treatments was followed thoroughly in the following spring including the regu- lar collection of shed leaves, the determination of shoot length and growth stages and weekly measurements of leaf greenness (SPAD). At the fi nal harvest (24 June 2009), the biomass allocated to leaves, stems and the reproductive organs (pod walls and seeds) was de- termined and oil contents were analysed in the dry seeds. Total dry shoot mass averaged to 45 g plant-1 DM. This is comparable to the average shoot weight of 17 - 34 g DM per plant, which results from the shoot mass production of 100 to 200 dt ha-1 (=1 to 2 kg m-2) at a density of 60 plants m-2. Average seed mass averaged to 12 g plant-1, which is also comparable to the seed yield in the fi eld. At a fi nal plant density of 40 plants per square metre, seed production per plant ranges between 8 and 14 g DM per plant yielding 320 to 560 g m-2 (32 to 56 dt ha-1). Fig. 5 shows a comparison of the leaf shedding and the leaf greenness in the four cultivars over time. The cultivars A and V had somewhat higher initial SPAD values than the other two cultivars, indicating that there were slight differences in the chlorophyll concentrations between different cultivars. Results of two-way ANOVAs indi- cating the existence of signifi cant differences between the cultivars and cultivar * treatment interactions are shown in Tab. 2. Chlorophyll degradation and probably the onset of senescence in the spring occurred earlier in A than in the other cultivars despite its higher initial SPAD values. At the same time, cultivars V and E had some- what higher initial SPAD values in the plants that were treated with ammonia in the previous autumn. However, signifi cant cultivar * treatment interactions were only observed on 28 April. A highly sig- nifi cant effect of the warming treatments was found for the leaf mass that was shed during winter and the fi nal cumulative leaf mass (see fi rst and last observation 61 and 143 DOY in Fig. 5). In the warm treatment cultivars A and L shed more biomass during the winter than in the other treatments, but only in cultivar A the strong leaf loss during the winter could not be compensated in the following spring (see Fig. 5). Furthermore, highly signifi cant cultivar differences and a strong cultivar * treatment interaction were observed in fi nal shoot length. While cultivar E reduced shoot length upon advanced growth by elevated autumn temperatures and ammonia concentrations, cul- tivar L produced taller plants in both of the advancing treatments and cultivar A in the ELVT+A treatment. Signifi cant cultivar differences and effects of the elevated tempera- tures and ammonia concentrations were also observed in the onset and duration of fl owering, while a signifi cant cultivar * treatment interaction across the treatments was only found in the onset of fl owering. Plants of cultivar V grown at elevated autumn tempera- tures fl owered three days earlier than those grown at ambient tem- peratures. Nevertheless, differences in the onset of fl owering and changed fl owering windows did not explain differences in fi nal seed output. While signifi cant cultivar differences were found for seed mass, with cultivar L being the most productive, signifi cant treatment effects were found for elevated temperatures and elevated ammonia. Elevated autumn temperatures reduced seed mass, but elevated am- monia compensated this negative effect. Also the growth traits stem mass, harvest index and oil yield showed opposite responses, with higher values when plants were subjected to a combination of ele- vated temperatures and elevated concentrations of ammonia. Interestingly, thousand seed mass was unaffected by the autumn treatments, while ANOVA across all treatments revealed signifi cant cultivar differences with cultivar A having the smallest seeds. Pod numbers and weights of pod walls (siliques) were unaffected by the treatments and did not signifi cantly differ between cultivars, sug- gesting that these traits had not been changed much (or selected for) during the breeding history. Nevertheless, the harvest index (Fig. 4) showed highly signifi cant cultivar differences with the cultivars E and L having the highest and cultivar V the lowest values. Adverse effects of the autumn treatments on the harvest index are primarily due to the growth enhancement of the vegetative plant parts at the expense of the seeds. Finally, the seed oil contents appeared to be unaffected by the elevated temperatures and ammonia concentra- tions supplied in the previous autumn. However, signifi cant cultivar differences were observed, with cultivar L having the highest oil contents and the tallest cultivar V the lowest. When addressing rela- tionships between the oil contents and other traits across treatments and cultivars a slight negative relationship (R2 = 0.49) was identifi ed between the mean oil contents and the seed N concentrations. The existence of a negative relationship between these variables and a trade-off between the synthesis of oils and proteins has also been found in other studies (refer to the review of RATHKE et al., 2006). Also the SPAD values recorded on 16 April 2009 showed a slightly negative relationship (R2 = 0.15) to the oil contents that were deter- mined at the end of the season. It therefore seems that plants with a high spring chlorophyll and presumably N status produce seeds with lower oil content. Conclusion The pot-grown plants from present experiments produced compa- rable fi nal shoot and seed mass as plants grown in the fi eld. Also Effects of autumn growth advancement in winter oilseed rape cultivars 141 SCHULTE AUF’M ERLEY et al. (2007) found that pots-in-greenhouse experiments are a simple but suited tool to study the performance of different OSR cultivars under a variable N supply. When sown late, plants of the tested four OSR cultivars strongly increased pre-winter shoot mass under elevated temperature (+180%) and even more so under elevated ammonia concentrations (+270%). Relative shoot mass stimulations were lower when plants were sown earlier. The results indicate that OSR was able to utilise gaseous ammonia to pro- duce signifi cantly more biomass than the untreated plants. Still we are not sure how much of the ammonia-N that was absorbed in the autumn was lost later on due to leaf shedding or due to outgassing. Although the different treatments were ceased before the winter, spring growth of the plants remained different in the three treat- ments. Final shoot mass was still 13% (pments. Final shoot mass was still 13% (pments. Final shoot mass was still 13% ( <0.05) and seed mass 16% (p(p( <0.05) higher in the plants that were fumigated with ammonia in the previous autumn. In contrast, the plants that were subjected to higher autumn temperatures had a 5% (ns) lower shoot mass and a 13% (p13% (p13% ( <0.05) lower seed mass than those grown under ambient tem- peratures. The fi ndings show that pre-winter growth enhancement has discernible effects on the performance of plants in the following spring. Reasons are the loss of resources due to the shedding of leaf mass in the winter. Nevertheless, the effects of autumn growth ad- vancement by elevated temperatures and the leaf loss due to winter frost will be of minor importance as compared to late frost events eventually occurring in spring. Furthermore, we suggest that growth advancement due to atmospheric ammonia will probably not play a signifi cant role at ambient concentrations, which are normally lower than 50 µg m-3. Acknowledgements Student assistants Zorica Kauf, Felix Frey and Sebastian Weller are thanked for taking care of the plants and performing part of the analyses. Time series of the phenological data were obtained from Deutscher Wetterdienst (DWD), Offenbach. References ADAMSEN, F.J., COFFELT, T.A., 2005: Planting date effects on flowering, seed yield, and oil content of rape and crambe cultivars. Ind. Crops Prod. 21, 293-307. AKSOUH-HARRADJ, N.M., CAMPBELL, L.C., MAILER, R.J., 2006: Canola response to high and moderately high temperature stresses during seed maturation. Can. J. Plant Sci. 86, 967-980. BAUX, A., HEBEISEN, T., PELLET, D., 2008: Effects of minimal temperatures on low-linoleic rapeseed oil fatty-acid composition. Eur. J. Agron. 29, 102-107. BERRY, P.M., SPINK, J.H., 2006: A physiological analysis of oilseed rape yields: past and future. J. Agr. Sci. 144, 381-392. BIVAND, R.S., PEBESMA, E.J., GOMEZ-RUBIO, V., 2008: Applied spatial data Fig. 5: Development of leaf loss during winter and spring (above), cumulative leaf loss (centre) and leaf greenness (below) in four OSR cultivars subjected to Development of leaf loss during winter and spring (above), cumulative leaf loss (centre) and leaf greenness (below) in four OSR cultivars subjected to warming treatments in the previous autumn. X-axis represents time in days of the year (DOY). Treatments refer to AMB (○), ELVT (●) and ELVT+A (●). For the existence of signifi cant treatment effects refer to Tab. 2. Note that the fi nal cumulative leaf loss represents the dry mass of all leaves that were produced by a plant from sowing to harvest. 142 J. Franzaring, I. Holz, A. Fangmeier T ab . 2 : E ff ec ts o f g ro w th e n h an ce m en t b y i n cr ea se d a u tu m n t em p er at u re s an d e le va te d c o n ce n tr at io n s o f th e ga s am m o n ia i n f o u r cu lt iv ar s o f w in te r o il se ed r ap e. D at a re fe r to t h e p la n t p er fo rm an ce ( p h en o lo g y, S PA D a n d b io m as s) i n t h e fo ll ow in g s p ri n g w it h m ea n s ac ro ss f o u r re p li ca te s p er c u lt iv ar i n e ac h o f th e th re e tr ea tm en ts . C u lt iv ar s: ‘ A ’ - A sg ar d , ‘E ’ - E x p re ss , ‘L ’ - L o re n z, ‘ V ’ - V ik in g . R es u lt s o f th e A N O V A i n d ic at e si g n ifi c an t d if fe re n ce s b et w ee n c u lt iv ar s (C ), t re at m en ts ( T ) an d i n te ra ct io n s b et w ee n T ) an d i n te ra ct io n s b et w ee n T C * T . S ig n ifi c an t te m p er at u re e ff ec ts w er e te st ed u si n g t h e co m p ar is o n b et w ee n A M B a n d E L V T , C * T . S ig n ifi c an t te m p er at u re e ff ec ts w er e te st ed u si n g t h e co m p ar is o n b et w ee n A M B a n d E L V T , C * T w h il e te st s fo r am m o n ia e ff ec ts r es te d o n t h e co m p ar is o n b et w ee n t h e E L V T a n d E L V T + A t re at m en ts . S ig n ifi c an ce l ev el s: * p < 0 .0 5 , * * p < 0 .0 1 a n d * * * p < 0 .0 0 1 . a re fe rs t o l ea f m as s th at h ad b ee n s h ed d u ri n g t h e w in te r; l ea ve s w er e co ll ec te d o n 1 M ar ch 2 0 0 9 . b r el at es t o a ir d ri ed s ee d s. T re at m en ts A M B E L V T E L V T + A T em p . N H 3 C om b in ed A m b ie n t te m p er at u re E le va te d t em p er at u re E le va te d t em p er at u re E ff ec t E ff ec t E ff ec t p lu s am m on ia C u lt iv ar s A E L V A E L V A E L V L ea f lo ss d ur in g w in te r (g D M )a 1. 21 1. 58 1. 27 1. 27 4. 21 3. 30 3. 67 3. 56 3. 42 3. 18 2. 50 4. 53 ** * ** * O ns et o f F lo w er in g (D O Y ) 10 2 10 2 10 2 10 5 10 0 10 1 10 2 10 2 10 2 10 2 10 3 10 2 ** ** * ** * * F lo w er in g du ra ti on ( da ys ) 26 26 25 27 26 25 25 25 25 24 25 27 ** ** * F in al s ho ot l en gt h (c m ) 14 3 14 8 13 9 15 5 14 0 14 0 15 0 15 4 15 3 13 0 15 4 15 5 * ** * * ** * ** * S PA D ( 16 A pr il 2 00 9) 52 .8 46 .9 46 .7 49 .1 50 .6 46 .0 49 .2 50 .0 52 .3 48 .9 46 .5 53 .5 * * ** S PA D ( 21 A pr il 2 00 9) 49 .1 46 .3 45 .3 47 .5 50 .8 46 .2 47 .7 50 .5 48 .2 48 .6 46 .4 51 .1 * * ** S PA D ( 28 A pr il 2 00 9) 39 .1 44 .3 43 .8 45 .7 48 .5 42 .0 45 .2 48 .1 44 .4 44 .7 44 .9 53 .0 * * ** * * ** * S PA D ( 5 M ay 2 00 9) 40 .2 41 .8 42 .0 43 .9 42 .0 41 .1 42 .3 41 .3 38 .6 42 .2 43 .3 50 .0 * * * S PA D ( 13 M ay 2 00 9) 21 .6 35 .5 34 .2 40 .3 1. 0 35 .7 28 .2 30 .6 11 .2 35 .4 36 .1 36 .6 ** * * ** * ** * L ea f m as s to ta l (g D M ) 8. 1 7. 2 7. 8 8. 0 9. 2 10 .0 10 .4 9. 8 10 .8 10 .6 10 .2 11 .6 ** * ** * S te m m as s (g D M ) 15 .8 14 .1 17 .0 16 .8 12 .3 12 .9 15 .8 14 .4 15 .9 14 .1 17 .8 14 .7 * ** * * ** P od n um be r 18 8 18 4 18 8 15 5 15 1 16 9 16 5 15 3 18 5 19 5 16 3 16 3 P od w al ls ( g D M ) 10 .0 9. 7 9. 8 8. 8 8. 4 8. 9 9. 5 8. 3 10 .2 10 .8 9. 7 8. 5 S ee d m as s (g D M )b 12 .7 12 .1 13 .2 11 .7 10 .3 11 .2 12 .0 9. 7 12 .4 13 .2 13 .6 10 .8 * * * * S ho ot m as s (g D M ) 46 .6 43 .1 47 .8 45 .3 40 .2 43 .0 47 .7 42 .2 49 .3 48 .7 51 .3 45 .6 * H ar ve st i nd ex 0. 27 0. 28 0. 28 0. 26 0. 26 0. 26 0. 25 0. 23 0. 25 0. 27 0. 27 0. 24 ** * ** * ** * * ** * ** T ho us an d se ed m as s (g ) 2. 9 2. 8 3. 2 3. 0 3. 1 3. 3 3. 1 2. 9 2. 8 3. 2 3. 5 3. 0 * S ee d oi l co nt en ts ( % ) 50 .2 53 .3 53 .5 50 .5 52 .6 52 .7 53 .6 51 .6 50 .2 53 .0 54 .3 50 .4 ** ** * ** * O il y ie ld s (g p la nt -1 ) 6. 4 6. 5 7. 1 5. 9 5. 4 5. 9 6. 4 5. 0 6. 2 7. 0 7. 1 5. 4 * * * ** S ee d N c on ce nt ra ti on s (% ) 2. 3 2. 2 2. 0 2. 4 2. 1 2. 2 2. 0 2. 3 2. 2 2. 2 1. 9 2. 3 ** * ** * ** * N U E S ( g se ed g -1 a va il ab le N ) 16 .8 16 .0 17 .5 15 .4 13 .6 14 .8 15 .9 12 .9 16 .4 17 .5 18 .1 14 .3 * * * * N U E O ( g oi l g- 1 av ai la bl e N ) 8. 4 8. 6 9. 4 7. 8 7. 2 7. 8 8. 5 6. 6 8. 3 9. 3 9. 4 7. 18 * * * ** Cultivar Treat. C*T Cultivar Treat. C*T Cultivar Treat. C*T Effects of autumn growth advancement in winter oilseed rape cultivars 143 analysis with R. Springer Verlag, New York. BÖHME, F., MERBACH, I., WEIGEL, A., RUSSOW, R., 2003: Effect of crop type and crop growth on atmospheric nitrogen deposition. J. Plant Nutr. Soil Sci. 166, 601-605. CHRISTEN, O., FRIEDT, W., 2007: Winterraps. Das Handbuch für Profis. DLG Verlag, Frankfurt. CLEMENT, J.M., VENEMA, J.H., VAN HASSELT, P.R., 1995: Short-term expo- sure to atmospheric ammonia does not affect low-temperature hardening of winter wheat. New Phytol. 131, 345-351. CLEMENT, J.M.A.M., VAN HASSELT, P.R., VAN DER EERDEN, L.J.M., DUECK, T.A., 1999: Short-term exposure to atmospheric ammonia does not affect frost hardening of needles from three- and fi ve-year-old Scots pine trees. J. Plant Physiol. 154, 775-780. COLNENNE, C., MEYNARD, J.M., ROCHE, R., REAU, R., 2002: Effects of nitrogen defi ciencies on autumnal growth of oilseed rape. Eur. J. Agron. 17, 11-28. DIEPENBROCK, W., 2006a: Raps. Saat und Bestandsbildung. In: Heyland, K.V., Hanus, H., Keller, E.R. (eds.), Handbuch des Pfl anzenbaus: Ölfrüchte, Band 4, 87-90. Ulmer Verlag, Stuttgart. DIEPENBROCK, W., 2006b: Raps. Physiologie und Ertragsbildung. In: Heyland, K.V., Hanus, H., Keller, E.R. (eds.), Handbuch des Pfl anzenbaus: Ölfrüchte, Band 4, 54-68. Ulmer Verlag, Stuttgart. FARAJI, A., LATIFI, N., SOLTANI, A., RAD, A.H.S., 2009: Seed yield and water use effi ciency of canola (Brassica napus L.) as affected by high temperature stress and supplemental irrigation. Agr. Water Manage. 96, 132-140. FSM, 2006: Radiello Manual. Centro di Richerche Ambientale Padova. Fondazione Salvatore Maugeri IRCSS. Supelco – Sigma-Aldrich Edition. GROSS, A.T.H., 1964: Effect of date of planting on yield, plant height, flower- ing, and maturity of rape and turnip rape. Agron. J. 56, 76-78. GU, L., HANSON, P.J., POST, W.M., KAISER, D.P., YANG, B., NEMANI, R., PALLARDY, S.G., MEYERS, T., 2008: The 2007 eastern US spring freeze: Increased cold damage in a warming world? BioScience 58, 253-262. HENDRY, G.A.F., GRIME, J.P., 1993: Methods in comparative plant ecology. A laboratory manual. Chapman and Hall, London. HUSTED, S., SCHJOERRING, J.K., 1995: Apoplastic pH and ammonium con- centration in leaves of Brassica napus L. Plant Physiol. 109, 1453- 1460. HUSTED, S., SCHJOERRING, J.K., 1996: Ammonia flux between oilseed rape plants and the atmosphere in response to changes in leaf temperature, light intensity, and air humidity. Interactions with leaf conductance and apoplastic NH4 and H concentrations. Plant Physiol. 112, 67-74. JOHNSON, B.L., MCKAY, K.R., SCHNEITER, A.A., HANSON, B.K., SCHATZ, B.G., 1995: Infl uence of planting date on canola and crambe production. J. Prod. Agric. 8, 594-599. KONDRA, Z.P., 1977: Effect of date of planting on rapeseed. Can. J. Plant Sci. 57, 607-609. KUTCHER, H.R., WARLAND, J.S., BRANDT, S.A., 2010: Temperature and pre- cipitation effects on canola yields in Saskatchewan, Canada. Agr. Forest Meteorol. 150, 161-165. LÄÄNISTE, P., JÕUDU, J., EREMEEV, V., MÄEORG, E., 2007: Sowing date influ- ence on winter oilseed rape overwintering in Estonia. Acta Agr. Scand. B, Soil Plant Sci. 57, 342-348. MASSAD, R.-S., LOUBET, B., TUZET, A., AUTRET, H., CELLIER, P., 2009: Ammonia stomatal compensation point of young oilseed rape leaves during dark/light cycles under various nitrogen nutritions. Agr. Ecosyst. Environ. 133, 170-182. MORRISON, M.J., MCVETTY, P.B.E., SHAYKEWICH, C.F., 1989: The deter- mination and verifi cation of a baseline temperature for the growth of Westar summer rape. Can. J. Plant Sci. 69, 455-464. MORRISON, M.J., STEWART, D.W., 2002: Heat stress during flowering in sum- mer Brassica. Crop Sci. 42, 797-803. NEMITZ, E., SUTTON, M.A., SCHJOERRING, J.K., HUSTED, S., WYERS, P.G., 2000: Resistance modelling of ammonia exchange over oilseed rape. Agr. Forest Meteorol. 105, 405-425. NORD, E.A, LYNCH, J.P., 2009: Plant phenology: a critical controller of soil resource acquisition. J. Exp. Bot. 60, 1927-1937. PELLET, D., BAUX, A., GROSJEAN, J., HEBEISEN, T., HILTBRUNNER, J., HUNZIKER, H., 2008: Temperature and rapeseed oil nutritional quality. Agrarforschung 15, 480-485. PELTONEN-SAINIO, P., JAUHIAINEN, L., HANNUKKALA, A., 2007: Declining rapeseed yields in Finland: how, why and what next? J. Agr. Sci. 145, 587-598. PELTONEN-SAINIO, P., JAUHIAINEN, L., VENÄLÄINEN, A., 2009a: Comparing regional risks in producing turnip rape and oilseed rape – Today in light of long-term datasets. Acta Agr. Scand. B, Soil Plant Sci. 59, 118-128. PELTONEN-SAINIO, P., HAKALA, K., JAUHIAINEN, L., RUOSTEENOJA, K., 2009b: Comparing regional risks in producing turnip rape and oilseed rape – Impacts of climate change and breeding. Acta Agr. Scand. B, Soil Plant Sci. 59, 129-138. RAPACZ, M., 1998: Physiological effects of winter rape (Brassica napus var. oleifera) Prehardening to frost. II. Growth, energy partitioning and water status during cold acclimation. J. Agron. Crop Sci. 181, 81-87. RAPACZ, M., 1999: Frost resistance and cold acclimation abilities of spring- type oilseed rape. Plant Sci. 147, 55-64. RAPACZ, M., 2002: Cold-deacclimation of oilseed rape (Brassica napus var. oleifera) in response to fl uctuating temperatures and photoperiod. Ann. Bot. 89, 543-549. RATHKE, G.-W., BEHRENS, T., DIEPENBROCK, W., 2006: Integrated nitrogen management strategies to improve seed yield, oil content and nitrogen effi ciency of winter oilseed rape (Brassica napus L.): A review. Agr. Ecosyst. Environ. 117, 80-108. R DEVELOPMENT CORE TEAM, 2009: R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-project.org. SCHJOERRING, J.K., HUSTED, S., MÄCK, G., NIELSEN, K.H., FINNEMANN, J., MATTSSON, M., 2000: Physiological regulation of plant-atmosphere am- monia exchange. Plant Soil 221, 95-102. SCHULTE AUF’M ERLEY, G., WIJAYA, K.-A., ULAS, A., BECKER, H., WIESLER, F., HORST, W.J., 2007: Leaf senescence and N uptake parameters as selection traits for nitrogen effi ciency of oilseed rape cultivars. Physiol. Plant. 130, 519-531. SHEPPARD, L.J., LEITH, I.D., 2002: Effects of NH3 fumigation on the frost hardiness of Calluna – Does N deposition increase winter damage by frost? Phyton – Ann. Rei Bot. 42, 183-190. SHEPPARD, L.J., LEITH, I.D., CROSSLEY, A., VAN DIJK, N., FOWLER, D., SUTTON, M.A., WOODS, C., 2008: Stress responses of Calluna vulgaris to reduced and oxidised N applied under ‘real world conditions’. Environ. Pollut. 154, 404-413. SOMMER, S.G., 1988: A simple biomonitor for measuring ammonia deposition in rural areas. Biol. Fert. Soils 6, 61-64. SOMMER, S.G., JENSEN, E.S., 1991: Foliar absorption of atmospheric ammonia by ryegrass in the fi eld. J. Environ. Qual. 20, 153-156. SOMMER, S.G., ØSTERGÅRD, H.S., LØFSTRØM, P., ANDERSEN, H.V., JENSEN, L.S., 2009: Validation of model calculation of ammonia deposition in the neighbourhood of a poultry farm using measured NH3 concentrations and N deposition. Atmos. Environ. 43, 915-920. VDI, 2008: VDI guideline 3957 Part 3. Biological measuring procedures to determine and assess effects of air pollutants on plants (bioindication) – Procedure for standardised exposure of curly kale. VDLUFA, 2002: Arbeitsanleitung für NIRS-Untersuchungen an Körner- raps im Rahmen des Qualitätssicherungssystem NIRS/NIT. Verband Deutscher Landwirtschaftlicher Untersuchungs- und Forschungsanstalten, http://www.vdlufa.de/nirs. Address of the authors Dr. Jürgen Franzaring, Dr. Ingo Holz, Prof. Andreas Fangmeier, Institute for Landscape and Plant Ecology (320), University of Hohenheim, Öko- logiezentrum 2, August-von-Hartmann-Str. 3, D-70599 Stuttgart, Germany.