Introduction

Sugar beet (Beta vulgaris L.) is an important field crop primarily grown for its high sucrose content, which is used to produce edible sugar. Sugar beet accounts for about 22% (37.4 MT) of the world's total edible sugar production (170.5 MT; sugar cane + sugar beet). Türkiye produces around 2.6 MT of crystal sugar annually, responsible for roughly 1.5% of the world’s total sugar production (FAO 2023).

In semi-arid and cold climatic conditions (from 38°N to 60°N), sugar beets are typically grown during March–October (about 180–200 days growing period). Water scarcity is the most critical factor limiting sugar beet production during the conventional summer growing season (Yazici and Taner 2023) due to the high irrigation requirement of 900–1200 mm (Fabeiro et al. 2003), and it is irrigated 5 or 6 times during the growing period (Albayrak et al. 2010). Temperature and precipitation patterns in all climate zones are projected to change remarkably as a result of global climate change (IPCC -Intergovernmental Panel on Climate Change- 2021). In particular, regions with cold or hot arid and semi-arid climates and the Mediterranean climate are expected to be strongly affected by climate change (Giorgi 2006; Vicente-Serrano et al. 2014).

In semi-arid and cold climates of the Mediterranean region (between 35°N and 45°N latitudes), sugar beet could be sown in autumn, and harvesting starts in early summer, like the end of June (Rinaldi and Vonelda 2006; Hoffmann and Kluge-Severin 2010; Tayyab et al. 2023). It was emphasized that temperature, solar radiation and precipitation are the most important climatic parameters determining sugar beet adaptation in a region (Kenter et al. 2006). The yield of sugar beet grown in winter is higher than in conventional sugar beet cultivation (Jaggard and Werker 1999; Rinaldi and Vonella 2006; Hoffman and Kluge-Severin 2010; Webster et al. 2016; Stephan et al. 2020). Higher yields of autumn-sown sugar beet have been reported due to higher light interception in May and June and a longer vegetation period (Jaggar and Werker 1999; Hoffmann and Kluge-Severin 2010). On the contrary, Ertürk and Ağır (2022) revealed that the yield of sugar beet sown in the autumn was 23% lower, and the sugar content was 9% higher than that sown in the spring, whereas the production cost of sugar beet grown in the autumn–spring period was lower than in the summer growing season. However, despite the economic and ecological benefits of autumn-sown, bolting resistance (Hoffmann and Kluge-Severin 2010; Deihimfard et al. 2019), frost tolerance (winter hardiness) (Loel and Hoffmann 2014) and sufficient oxygen supply in the soil (Hoffmann and Kluge-Severin 2011) are prerequisites for successful sugar beet cultivation in autumn. When sugar beet was sown in autumn exposed to cold temperatures between 8 and 10 °C for a prolonged period (Milford and Limb 2008), the beets vernalize and start to bolt in spring under long-day conditions (12–14 h light exposure; Mutasa-Göttgens et al. 2010).

Adaptive crop management in agricultural systems to mitigate the negative impacts of climate change and adapt to its inevitable consequences is among the topics that have been widely studied (Lehmann et al. 2013; Bönecke et al. 2020). In winter crop cultivation in the Aegean region and the Mediterranean coastal zone, waterlogging due to excessive rainfall between December and March negatively affects productivity (Yavas et al. 2012; Unay and Simsek 2020; Simsek and Unay 2022). Ridge sowing has been recommended to avoid waterlogging (Gan et al. 2013; Mo et al. 2017; Fischer et al. 2019), and some researchers have obtained successful results under waterlogging conditions (Ren et al. 2016; Du et al. 2021). The second important factor limiting sugar beet sowing in the autumn is bolting. Mohammadian et al. (2023) emphasized that the GDD should exceed the sum of 300–400 when temperatures reach 6–8 °C to prevent bolting. Among the most essential advantages of ridge sowing, the increase in temperatures in the root zone was emphasized (Stathakos et al. 2006; Krause et al. 2009). No sugar beet variety yield trials have been conducted in this region before. For this reason, as a result of the test studies carried out by the Variety Registration and Seed Certification Center, which carries out multi-location studies in Türkiye, the three most compatible varieties with high adaptability to all environments in yield and sugar content were evaluated. The genotype × environment interaction plays a critical role in determining plant growth, development and productivity. Root yield and sugar content, the most important commercial traits of sugar beet, are strongly influenced by the environment (Hassani et al. 2018; Hemayati et al. 2024).

The most crucial cause of leaf senescence caused by stress is chlorophyll degradation (Barickman et al. 2019; Zhao et al. 2024), and decreases in SPAD and Fv/Fm values with the acceleration of leaf senescence negatively affect photosynthetic capacity (Ren et al. 2016; Shah et al. 2017). In addition, with poor conditions in sugar beet, especially in moist soils, abnormalities that will affect the yield and sugar content in the root shape appear. It has been suggested that the easiest way to observe this negativity is to measure the forking rate, root length and root diameter (Çınar and Ünay 2021; Bojović et al. 2022).

In the coastal zone of the Aegean region, wheat, maize and cotton agriculture dominate, and sugar beet is not included in the cropping system. Autumn-sown crops are adversely affected by waterlogging, especially due to excessive rainfall in December, January and February. Moreover, there is a risk of bolting in autumn sowing due to low temperatures that may occur in the following periods. In order to avoid these disadvantages, the ridge sowing method should be tried.

Cotton and maize as the main crop or cotton and silage maize as the second crop after wheat are intensely cultivated in the coastal zone of the Aegean region, such as Izmir, Aydın and Manisa. The main objective of the study was to test the feasibility of growing sugar beet without irrigation during the winter growing period in the Mediterranean climate zone, where continuous wheat, cotton and maize rotation is practised. Secondly, the study aimed to determine the performance of either ridge or flat sowing methods, particularly in terms of bolting and waterlogging prevention. This study will help to give an idea about the possibilities of growing irrigated crops, which are becoming increasingly problematic during the summer growing period with climate change, without irrigation during the winter growing period in suitable climates.

Materials and Methods

Field Experiments

A 2-year field experiment was carried out at the Research and Training Farm (37° 45′ 30″ N, 27° 45′ 19″ E) of the Aydın Adnan Menderes University Faculty of Agriculture during the 2020/2021 and 2021/2022 winter growing seasons. The experiment was arranged according to a split-plot design with four replications in a Randomized Complete Block Design (RCBD). The sowing method was the main factor as flat and ridge, whereas cultivar was the subplot factor. Based on the test studies carried out by the Variety Registration and Seed Certification Center of Türkiye, the cultivars Terranova and Aranka from KWS and Dionetta from Betaseed, which are tolerant to bolting (NZ type) and highly adapted to all environments regarding yield and sugar content, were used.

The preceding crop grown in the experiment area was cotton in both years. Ridges were formed a four-row ridge former with diabolo rolls. A rotor loosened and lifted the soil, while the following metal plates and rollers formed and compacted the ridges. The height of the ridges was about 30 cm, and the width of the upper surface was about 25 cm. After seedbed preparation in both sowing methods, pelleted seeds were sown with a pneumatic precision planter in 4-row plots of 6 m length, 13 cm apart in the row and 70 cm row distance (~ 110 thousand seeds ha−1) on 1 December 2020 and 30 November 2021. Before sowing, sugar beets were supplied with 75 kg ha−1 nitrogen, phosphorous and potassium. Then, another 92 kg ha−1 nitrogen was applied as urea when the plants were in the 4–6 leaf stage.

In both years, no diseases or pests that would require chemical plant protection were detected. In addition to that, weed control of the experimental area was carried out manually during the growing season. The experiment area was not irrigated in either year. Remarkably, the expected bolting due to vernalization was not significantly observed in either year. (No bolted plants were observed in the first year, while only one bolted plant was observed in the second year.) The experiment was harvested by hand on 07.06.2021 in the first year and 01.06.2022 in the second year. Since the region's climate warms rapidly from June onwards and enters a period without rainfall, harvesting was done before the soil was completely dry to avoid damaging the root. The difference in harvest dates resulted from this situation. The total vegetation period from sowing to harvest was 189 days in the first year and 184 days in the second year.

The soil of the experiment area has a sandy-loamy structure and a slightly alkaline pH (7.75). Besides, the soil has low salt (0.005%), lime (2.18%), organic matter (0.98%) and potassium content (163 ppm), while phosphorus content (14 ppm) is moderate. Weather data were obtained from a meteorological station less than 0.5 km from the experiment area. Growing degree days (GDD) from the emerging to the harvest were calculated (McMaster and Wilhelm 1997). Tbase was 3 °C (Werker and Jaggard 1997).

The precipitation pattern between years was different. The all-temperature parameters were lower in the second year, so GDD accumulation was reduced in the second year. In addition, the number of days of low temperatures below 3 °C limiting sugar beet growth processes was 9 days higher in the second year, and the number of days between 3 and 12 °C promoting the bolting disposition was 22 days higher in the second year than in the first year (Table 1).

Table 1 Environmental variables of the experiment area.
Full size table

Data Collection

Twenty representative beets from each plot were pulled out to determine yield components, i.e. root length from the crown to the tip with 2 cm diameter (RL), maximum root diameter (RD), fresh root weight (RW), root fork number (RF), fresh leaf weight separated from the crown (LW) and sugar content (SC) per plant. SC was determined by the polarimetric method with a portable refractometer. After removing the border effects, 80 sampled plants in all plots were hand-harvested to determine root and leaf yield. Leaf and root yields were summed to calculate biological yield (BY). The harvest index (HI) was determined by the ratio of root yield to leaf yield. Consequently, sugar yield was determined by multiplying root yield by sugar content.

Physiological parameters like SPAD, which is used to detect the chlorophyll content of leaves, and chlorophyll fluorescence (Fv/Fm), which is used to determine the maximum quantum yield of photosystem II (PSII), were measured from the uppermost fully developed leaves of ten plants of each genotype at BBCH19 (nine and more leaves fully developed) and BBCH35 (leaves cover 50% of the ground) stages (Meier et al. 1993). SPAD 502 Plus Chlorophyll Meter (Konica Minolta Co., Japan) and Pocket PEA (Hansatech Instruments, England) portable instruments were used for physiological observations, respectively. The physiological measurements were obtained during a specific time window of the day, from 11:00 to 14:00 h, when the solar radiation intensity was relatively stable on cloudless days, and the wind speed was less than 5 km h−1 (Ergo et al. 2018). These conditions were selected to minimize the potential influence of environmental factors on the physiological parameters, allowing for more accurate and reliable data acquisition.

Data Analysis

Firstly, after Shapiro–Wilk’s test for the normality of residuals, a three-way analysis of variance (ANOVA) was performed to determine if there were any treatment-by-year interactions for the traits studied (Table S1). Year, sowing method and cultivar were fixed effects, and replication was a random effect. Differences between years were significant for all traits except sugar content, SPAD (BBCH19) and Fv/Fm (BBCH19) (Table S1). Thus, the data collected were analysed separately each year according to the split-plot design in RCBD. A two-way mixed effects ANOVA was performed within each year, where the sowing method and cultivar were regarded as fixed effects and (sowing method × replication) was a random effect (Table S2). For the traits where differences between years were found to be insignificant, the results of three-way ANOVA were considered (Table S1). When a main effect or interaction effect was significant (p < 0.05), means were compared using the LSD test. We used JMP® Pro 16.0.0 (SAS Institute Inc.) to perform all statistical analyses.

Results

In the first year of the study, we observed that the performance of the sowing method varied according to the cultivars. Terranova in flat sowing and Aranka in ridge sowing attracted attention for LW, RD, RW and RL (Fig. 1). In the second year, Terranova in flat sowing and Aranka cultivar in ridge sowing produced more leaves, while RD and RW increased significantly in ridge sowing. We found that RF increased significantly only in the 2nd year in flat sowing; Terranova was prone to forking among the cultivars.

Fig. 1
figure 1

Mean values of yield components regarding year, sowing method and cultivars. When the interaction was significant, means not followed by the same letter(s) are significantly different. Upper- and lowercase letters were used to compare sowing methods and cultivars among themselves, respectively

Full size image

All of the traits mentioned above were strongly affected by the year (Table S1). The mean LW and RW of the first year were 38% and 22% higher than that of the second year. Similarly, we observed that root diameters narrowed, roots shortened, and RF increased (Fig. 1).

The sowing method × cultivar interaction, which was significant in terms of BY and RY in the first year, indicated the performance of Terranova in flat sowing and Aranka in ridge sowing. However, Terranova produced significantly higher BY in the second year in both sowing methods. Moreover, we determined that the RY was significantly higher in the ridge sowing. The relationship between BY and RY was reflected in the harvest index, and the harvest index of the Terranova cultivar in ridge sowing was significantly higher than that of flat sowing in both years. Aranka followed Terranova in ridge sowing. In both years, we found that the SC values of the Dionetta cultivar were significantly higher in both sowing methods. The performance of the Terranova cultivar in the ridge and flat sowing, respectively, remained in second place. In SY, which is a function of root yield and sugar content, the superior performance of ridge sowing was clearly evident. Among the cultivars, we determined that Terranova had significantly higher SY, which was significant in the 2nd year (Figs. 2 and 3).

Fig. 2
figure 2

Mean values of yield, sugar content and harvest index regarding year, sowing method and cultivars. When the interaction was significant, means not followed by the same letter(s) are significantly different. Upper- and lowercase letters were used to compare sowing methods and cultivars among themselves, respectively

Full size image
Fig. 3
figure 3

Variations in root yield and sugar content regarding year, sowing method and cultivars

Full size image

In the BCCH 19 growing stage, we found that SPAD was affected by year × sowing method, year × cultivar and sowing method × cultivar interaction. In the first year, the SPAD value was found to be high in ridge sowing, while there was no significant difference in the second year. The Terranova cultivar performed well in both years, followed by the Aranka cultivar. While Terranova and Aranka had similar SPAD values in flat sowing, the superiority of Terranova became evident in ridge sowing. In BCCH 35, Aranka in ridge sowing had the highest values in the first year, whereas in the second year, Terranova and Aranka had significantly the highest values in both sowing methods. Fv/Fm in early-stage values showed that Aranka was affected by stress in the second year of flat sowing. In the second stage, the Terranova cultivar for the first year and ridge sowing for the second showed significant performance (Fig. 4).

Fig. 4
figure 4

Mean values for physiological traits. When the interaction was significant, means not followed by the same letter(s) are significantly different. Upper- and lowercase letters were used to compare sowing methods and cultivars among themselves, respectively

Full size image

Figure 5 presents the correlation coefficients among observed traits. Significant and positive correlations between RY and RW, RD, RL, LW, BY, SY, SPAD and Fv/Fm values at BCCH 35 indicated that yield components, chlorophyll content and Fv/Fm contribute significantly to yields. The effect of RD on SC was significant and negative. Another important correlation was that high Fv/Fm values at BCCH 35 positively decreased the RF value.

Fig. 5
figure 5

Correlogram showing the relationships between studied traits

Full size image

Discussion

According to two years of results, the average root yield in our study was lower than that of Ertürk and Ağır (2022), whereas it was similar to Garcia and Bellido (1986). The sugar content of the cultivars tested in this study was higher than those mentioned in both study years. In a 3-year study conducted in southern Italy, Rinaldi and Vonella (2006) recorded yields similar to our study (56.80 t ha−1) and lower sugar content (14.9%). The point that should not be ignored is that all of these researchers irrigated sugar beet many times (2–7 times) during the winter growing season. Therefore, higher root yield and lower sugar content than in our study could be attributed to irrigation during the winter growing season. These results may also be evidenced by the fact that although the total water consumption of sugar beet during the growing season is about 900–1200 mm (Dunham 1993; Hills et al. 1990), the total rainfall during the winter season in our study was 374.45 mm on a two-year average. Water availability is the most important factor limiting sugar beet yield (Jaggard et al. 1998). In addition to water availability, temperature is one of the important environmental factors affecting sugar beet yield (Milford et al. 1985; Bojović et al. 2014, 2019, 2022). The fact that the average temperature was 17.6 °C in the study of Ertürk and Ağır (2022) and the average temperature was 14.29 °C in this study confirms the assumption that temperature affects yield. However, this temperature assumption cannot fully explain the high root yield and low sugar content found by Rinaldi and Vonella (2006). Because of the mean winter growing season (from emergence to harvest, December–August; October–July; and November–July), temperature in their study was 14.94 °C. Although the temperatures were similar to those in our study, the root yield and sugar content differences can be attributed to the total vegetation period. In their study, autumn-sown sugar beet was harvested at 206, 262 and 254 days in different years. Therefore, a longer vegetation period increases growing degree days, which has a direct and considerable relationship with yield and quality. As a result, the two-year average growing degree days value during the winter growing season in our study was 1726 °C, while the average growing degree days value in their study was 3443 °C, which can explain the assumption of the length of the vegetation period. In addition, the low sugar content in their study may have been caused by the prolonged vegetation period, especially in July and August. In a study conducted in Georgia, the sugar beet sown earliest in autumn reached the highest yields (Webster et al. 2016). In their study, the lowest yield of 42 t ha−1 was obtained in sugar beet sown at the end of December 2012 and harvested at the end of May 2013. Even this lowest yield is higher than our lowest yield value (36.12 t ha−1 in 2022). These results may be because of the cumulative growing degree days at which their lowest yield value was obtained was about 2400 °C, while ours remained at 1497 °C. In addition, they also stated that the sugar content was 16.1%, which is the general average of the year, location and cultivars. They underlined that variety and sowing time did not affect sugar content.

The differences between years (56.72 vs. 46.45 t ha−1) indicated that the effects of climatic factors on yield formation were significant (Fig. 3). Firstly, the minimum temperature is lower in all months except December in the first year than in the second year. At the same time, total rainfall in all months except January and March is lower in the first year. In contrast, the number of days below 3 °C, the minimum temperature for sugar beet growth (Milford et al. 1985), is lower in the first year, especially in February and March. While the number of days below 3 °C in the first year was 45, it was 54 days in the second year. While 10 days in January and 15 days in March were below 3 °C in the first year, these numbers were recorded as 16 and 21 days in the same months in the second year (Table 1). The low yields observed in ridge sowing plots in the second year indicate that ridge sowing is insufficient to eliminate the negative effect of temperatures below 3 °C. Similarly, minimum temperature has been reported to be the most limiting factor for the growth of sugar beet in winter (Loel and Hoffman 2014).

Hoffmann and Kluge-Severin (2011) reported that the yield of sugar beet sown in autumn can be well explained by GDD, indicating that sugar beet growth is controlled by temperature and day length. Accordingly, the yield difference between years in this study can be reasonably attributed to GDD. This is because the yield difference between the years was 22%, and the GDD difference was 31%. The closeness between the ratios may confirm the assumptions mentioned above for the relationship between GDD and yield.

It is also reported that under proper agricultural management practices, sugar beet yield potential is related to intercepted solar radiation during the growing period (Eggleston et al. 2010; Finkenstadt 2014), and this climatic variable is crucial for winter sugar beet cultivation. Therefore, one climate variable that differs between years and should be considered is the amount of solar radiation intercepted. It is worth noting that the total solar radiation amount between emergence and harvest decreased from 757.30 kWh m−2 in the first year to 673.80 kWh m−2 in the second year. However, it is noteworthy that the solar radiation amount in other months except December, January and May is higher in the second year. Each year, cumulative GDD and solar radiation had a significant linear relationship. Hoffmann and Kluge-Severin (2011) claimed that solar radiation and temperature, and thus GDD, are closely correlated during the winter season and that these variables cannot be evaluated separately. As a result, it can be considered that another explanatory climatic variable for the yield difference between years in this study is the amount of solar radiation.

Another climate variable that varies between years in this study was rainfall. Our study was completely conducted under rainfall conditions without irrigation during all growth stages, even germination and emergence. Although the yield was higher in the first year, total rainfall was about 13% less than in the second year. In addition, the rainfall in December, the month of sowing and emergence, was 44% higher in the second year (158.80 vs. 110.50 mm). While 50% of field emergence was reached in 10 days in the first year, this period was 21 days in the second year. For the first year, the cumulative GDD between sowing and emergence was 104.90 °C, while this value was 174.65 °C in the second year. On the contrary, Hoffmann and Kluge-Severin (2011) reported that sugar beets need a GDD of about 100 °C to reach 50% field emergence, whether sown in spring or autumn, and GDD can well explain that field emergence. They also emphasized that for this close relationship between field emergence and GDD to be meaningful, other germination and growth factors, such as soil moisture and oxygen supply, must be near the optimum. The fact that the relationship between field emergence and GDD could not explain the emergence times in our study confirms the importance of other growth factors, which are indispensable for germination and emergence. Perry and Harrison (1974) suggest that the inhibition of monogerm sugar beet seed germination by excess water in the soil could be explained by the restriction of oxygen flow through the basal pore of the seed. They also noted that the pores are a pathway of oxygen to the seed and that excess water prevents the flow of oxygen to the seed by closing the pores. As a result, this impairment of oxygen diffusion (hypoxia) is not lethal for the seed but delays germination (Dasberg and Mendel 1971; Perry and Harrison 1974; Richard et al. 1989). Moreover, Håkansson et al. (2012) reported that at high soil moisture content, the disruption of oxygen supply can occur when the percentage of water-filled pores increases more than the percentage of air-filled pores. However, they emphasized that the effect of oxygen supply limitation on germination was lower than the considerable negative effect of waterlogging. It has also been stated that delayed field emergence was caused by a water level above the seeding depth and poor soil drainage (Durrant et al. 1988). These hypotheses were confirmed by the fact that the total rainfall between sowing and emergence in the second year in our study was 151.20 mm, while it was only 37.20 mm in the first year. As a result, 38% of the total rainfall in the growing season in the second year occurred between sowing and emergence, compared to only 11% for the same period in the first year. Accordingly, the total rainfall between field emergence, when active growth begins, and harvest differed in the two years. In the first year, the total rainfall during the active growth period was 315.20 mm, while this amount remained at 245.30 mm in the second year. In addition to the difference in rainfall distribution between the years, the GDD between field emergence and harvest was 1955.70 °C in the first year and 1496.65 °C in the second year, which is about 31% lower than the first year for the same period (Table 1). As a result, excessive rainfall in the second year prolonged the emergence period; therefore, the sum of temperature and solar radiation, the most critical factors for sugar beet growth, decreased between field emergence and harvest compared to the first year. The yield decrease in the second year can be attributed to said reasons.

In both the first and second years, root yield was higher in ridge sowing compared to flat sowing (Fig. 2). Considering the average of two years, 27% more root yield was obtained in ridge sowing compared to flat sowing (57.75 vs 45.43 t ha−1). While ridge sowing produced 19% more root yield than flat sowing in the first year, this ratio increased to 38% in the second year. As can be seen from Fig. 3, the variation in root yield was less in ridge sowing under all treatments (year + cultivar). Ahmad et al. (2019) found the yield of sugar beet to be 62 t ha−1 in ridge sowing in autumn-sown, which was about 8.40% higher than in flat sowing (57.20 t ha−1), and reported that ridge sowing is superior to other sowing methods (flat or bed sowing) to obtain higher yield and sugar content. In addition, Ahmad et al. (2010) expressed that in autumn-sown, ridge sowing sugar beet reached 43% higher root weight than flat-sown sugar beet. As in autumn-sown, it was emphasized that in spring sowing, yield and sugar content were higher in ridge sowing compared to flat sowing (Krause et al. 2009). Bhullar et al. (2009), Krause et al. (2009), and Ahmad et al. (2019) reported that the fact that the soil on the ridge tends to warm up faster than the soil at ground level may be a possible reason for the higher yields obtained in ridge sowing, which hypothesis was in agreement with the results of Hoffmann and Jungk (1996) who declared that increased soil temperature increases the growth of young sugar beet seedlings, and the results of this study also confirm assumptions. Besides, the very loose soil around the root in ridge sowing may be a factor that improves plant growth by improving soil aeration and deeper root penetration, which allows the utilization of water in deeper soil layers during the rainless period, thus increasing yields (Khan et al. 2012; Stephan et al. 2020). This assumption was verified in our study with longer root length, wider root diameter and higher root weight in ridge sowing. In this study, the root length, root diameter and weight were found to be higher in ridge sowing than in flat sowing, which is similar to the findings of Ahmad et al. (2019). Furthermore, in this study, the number of forked roots, which is another soil compaction indicator, was generally lower in the ridge sowing. Root length (0.78**), root diameter (0.69*) and root weight (0.91***) were significantly and positively correlated with yield, confirming the significant association between root development and yield (Fig. 4). Lower yields in flat sowing can be attributed to soil compaction and hypoxia, which can be caused by excessive rainfall, especially during the early growth period (Awad et al. 2012), and soil compaction has been reported to affect emergence and root growth (Gemtos and Lellis 1997; Gemtos et al. 2000; Mahmoud et al. 2012). In our study, excessive rainfall in December and February in the second year compared to the first year confirmed the hypothesis of yield decrease due to soil compaction and insufficient oxygen supply during the early development period. Despite this yield decrease, the yield in the second year was still higher in ridge sowing. This is because, as emphasized by Starbuck (2005), the soil on the ridge warms up faster and, more importantly, dries out more quickly than the soil at the ground level, which can help prevent soil compaction and hypoxia.

Khan et al. (2012) showed that the number of leaves and leaf area were higher in ridge sowing. This circumstance is probably because, as Krause et al. (2009) pointed out, the enlarged soil surface in ridge sowing probably increased the absorption of incident heat radiation due to the more favourable angle of incidence of sunlight on the ridge. Accordingly, better light absorption will directly affect early leaf development (Stephan et al. 2020) and improve the performance of photosynthesis. This assumption was consistent with the leaf weight found to be higher in ridge sowing compared to flat sowing in our study (384.45 vs. 309.77 g). Leaf weight was significantly and positively correlated with yield (0.66*), supporting the significant relationship between photosynthetic performance and yield. The significant and positive relationship between sugar yield and high Fv/Fm values at BBCH 35, which is an indicator of the absence of stress, exhibited that this trait could be used as an evaluation criterion under stressful conditions (Fig. 5).

Differences in cultivars, sowing method, population density, location and year affect the quality of sugar beet (Märländer 1991; Rother 1998; Sogut and Arioglu 2004; Ahmad et al. 2012; Awad et al. 2012; Bojović et al. 2019). Among these factors, it is well known that the cultivation method affects sugar beet quality by directly and indirectly affecting the micro-environment in which sugar beets grow (Seadh et al. 2013). In support of this information, it has been reported that sugar content was better in ridge sowing than in flat sowing (El-Sarag 2009; Ahmad et al. 2019). However, in our study, although the difference between years in terms of sugar content was insignificant, the difference between years in terms of sugar yield was significant due to differences in root yield between years. In addition, year × sowing method × cultivar interaction was significant for sugar content. Therefore, it can be said that the cultivar effect is more significant on sugar content than the year and sowing method, because while there was inconsistency between years and sowing methods in terms of sugar content, the performance of cultivars was consistent across years and sowing methods. In line with this inference, Hoffmann and Kluge-Severin (2011) reported that sugar beet sown in February or March as early sowing did not form more cambium rings in which sugar is stored compared to the normal sowing in April. They emphasized that this indicates that the number of cambium rings is determined by genotypic effects and that this number cannot change much, even under favourable environmental conditions. There is a physiological limit to sugar-storing cells, which is specific to the genetic background of commercial cultivars, regardless of environmental variables, management practices and growing period extension (Schnepel and Hoffman 2016). In contrast to our findings, Webster et al. (2016) reported that cultivars and sowing dates had no significant effect on sugar content in sugar beet sown at different dates in autumn. In addition, the average sugar content was 16.10% in the mentioned research, and it was determined to be 17.12% in our research. One of the most important reasons for the difference in sugar content between the studies may be the bolting observed in their study. This is because the transition from the vegetative to the reproductive stage induces the transport of dry matter from the root to the stem; therefore, the sugar content may decrease (Hoffmann and Kluge-Severin 2011).

Conclusion

Two-year results indicated that autumn-sown sugar beet without irrigation could be successfully cultivated in temperate and rainy climate zones without bolting and frost damage. The yield and quality differences between the two years could well be described with climatic factors such as rainfall, temperature and solar radiation. Although waterlogging during germination negatively affected the emergence duration and yield, the stress-reducing effect of ridge sowing was considerable. As in this study, the number of days below 3 °C was essential for the adaptation of sugar beet in autumn sowing areas. We strongly recommend ridge sowing for high yield and sugar content in the winter growing season, especially under waterlogging conditions that may occur during the seedling stage. This study has proven that the sugar content can be higher without any loss in yield compared to the classically grown summer sugar beet.

Autumn-sown sugar beet cultivation, harvested in the first half of June, allows second-crop cultivation of crops such as cotton, maize, soybean and sunflower. In locations similar to the climate of the Aegean Coastal Zone, in the crop rotation dominated by winter wheat, summer cotton and maize, autumn-sown sugar beet can be included in the crop pattern in terms of yield, quality and economics. Further studies are needed in different soil types and sugar beet cultivars to elucidate better the performance of sugar beet yield and quality under the Mediterranean Climate Zone.