Yield Potential of Silage Sorghum: Cultivar Differences in Biomass Production, Plant Height, and Tillering Under Contrasting Soil Conditions in Central Europe

Abstract

We conducted a three-year field study to evaluate the above-ground biomass yield, plant height, and tillering capacity of eight Sorghum bicolor (L.) Moench varieties under two contrasting soil conditions (heavy clay soil and sandy soil) with different water retention. At the Field Experimental Station Žabčice of Mendel University in Brno, Czech Republic, we assessed yield performance and yield stability across years and environments. We applied standard agronomic practices and recorded detailed soil and climatic data. Significant differences were found among varieties and between locations in terms of plant height and tillering. KWS SOLE showed the most stable yield (11.80–15.63 t ha⁻¹), while LATTE, KWS TARZAN, and KWS HANNIBAL achieved the highest average yields (up to 20.16 t ha⁻¹). Plant height showed a strong positive correlation with biomass yield. This relationship underscores plant height as a valuable trait for selecting sorghum varieties with improved productivity and drought resilience. Variations in tillering capacity and environmental conditions also significantly influenced yield outcomes, highlighting the complex interaction between genotype and environment. These findings offer practical insights for cultivar selection and breeding strategies that aim to enhance the performance of sorghum varieties under the variable climatic conditions of Central Europe.

1. Introduction

Sorghum (Sorghum bicolor L. Moench), one of the oldest cultivated crops in the world [1], ranks among the most widely grown cereals globally, alongside maize, rice, and wheat [2]. Due to its subtropical origin, sorghum is traditionally cultivated in arid and semi-arid regions. However, recent climate changes in Central Europe, including in the Czech Republic, have sparked increased interest in sorghum as a supplementary crop, particularly due to the rising frequency of meteorological and soil droughts.

As documented by Trnka et al. [3] and Orság et al. [4], the number of days with average daily temperatures above 10 °C and the occurrence of tropical days have increased significantly over recent decades. Concurrently, suboptimal soil moisture conditions have become more prevalent, especially since the year 2000. These climatic shifts negatively impact the productivity of traditional crops such as maize and wheat.

Sorghum is considered a promising crop for dryland agriculture due to its morphological and physiological adaptations. It features a robust root system, low water demand per kilogram of dry matter [5], and several drought-tolerant traits, including a thick wax layer that protects leaves from UV radiation [6,7], high transpiration efficiency [8], and a pronounced stay-green effect [9,10]. Its C4 metabolism—shared with maize—contributes to enhanced productivity under high temperatures and moisture stress. In temperate climates, sorghum hybrids have demonstrated strong performance under water-limited conditions, while S. bicolor × S. sudanense hybrids perform well under more favorable moisture regimes [11]. Especially on light sandy soils, there are conditions for growing sorghum with greater yield certainty compared to corn, especially during more frequent periods of drought.

Despite these advantages, knowledge about the long-term yield performance and stability of silage sorghum cultivars under Central European field conditions remains limited. Previous studies have typically focused on tropical and subtropical environments, leaving a gap in understanding how different cultivars perform in temperate climates. In particular, we still lack evidence on how cultivar-specific traits—such as plant height, tillering capacity, and dry matter content—contribute to yield stability under contrasting soil water retention capacities. Addressing this gap is crucial for optimizing sorghum production as a sustainable forage source in regions increasingly affected by climate variability.

Therefore, the objectives of this study were (i) to evaluate the dry above-ground biomass yield and plant height of eight silage sorghum cultivars over three years under two contrasting soil conditions in the Czech Republic, and to assess tillering capacity during the two years when this trait was recorded; (ii) to determine the relationships between yield and key agronomic traits; and (iii) to assess yield stability in the context of increasing climate variability.

2. Materials and Methods

2.1. Site Description

Small-plot field experiments with selected sorghum varieties suitable for silage were conducted at the Field Experimental Station Žabčice, Mendel University in Brno, Czech Republic. The station is in the Dyje-Svratka Valley in South Moravia, approximately 25 km south of Brno (49.021800° N, 16.615600° E). The area experiences elevated aridity, with prevailing winds contributing to high evaporation of soil moisture. Lang’s Rainfall Index (ca. 52) classifies this site among the driest regions in the Czech Republic. The landscape is flat and erosion-resistant, situated at an elevation of ~190 m above sea level, with an average annual precipitation of 491 mm and a mean annual temperature of 10.3 °C.

Two sites with contrasting soil water retention capacities were selected: Obora and Písky.

• Obora (49.0255714° N, 16.6188814° E): Gleyic Fluvisol Clayic; heavy-textured, medium organic carbon content, neutral pH.
• Písky (49.0177722° N, 16.5917347° E): Arenic Chernozem; sandy-gravel substrate, lower water retention, weakly acidic, medium organic carbon content.

2.2. Cultural Practices, Treatments, and Varieties

At both experimental sites, primary tillage was carried out in autumn using disk harrowing followed by medium-depth plowing. In the spring (March–May), pre-sowing seedbed preparation was performed using a combination of disk and tine harrows to ensure optimal soil conditions and weed control.

A strip-plot layout without full randomization was employed, where sorghum varieties were sown side-by-side in adjacent strips. Three sampling sub-plots were established within each plot and were fully randomized to ensure objective data collection.

Eight sorghum varieties were evaluated in the experiment: RUZROK, NUTRI HONEY, PAMPA TRIUNFO XLT BMR, LATTE, KWS SOLE, KWS FREYA, KWS HANNIBAL, and KWS TARZAN, sourced from SEED SERVICE Ltd. (Vysoké Mýto, Czech Republic) and KWS OSIVA Ltd (Velké Meziříčí, Czech Republic).

Sowing was performed at the end of May in 2020 and 2022, and at the beginning of June in 2021, due to weather conditions affecting sowing timing. Seeds were sown at a depth of 3 cm with 0.45 m row spacing, using a HALDRUP SP-35 pneumatic precision seeder, at a rate of 245,000 seeds ha⁻¹. Immediately after sowing, the soil surface was compacted using Cambridge rollers.

In 2021 and 2022, a pre-emergent herbicide (312.5 g L⁻¹ S-metolachlor, 187.5 g L⁻¹ terbuthylazin; GARDOPRIM PLUS GOLD 500 SC) was applied at a rate of 2.0 L ha⁻¹, and in 2021, a post-emergent herbicide (dicamba 500 g kg⁻¹, tritosulfuron 250 g kg⁻¹; ARRAT) was used at 0.2 kg ha⁻¹.

2.3. Data Collection

2.3.1. Meteorological and Soil Conditions

Meteorological data were obtained from an automatic weather station operated by the Department of Agrosystems and Bioclimatology at Mendel University in Brno, located at the Field Experimental Station Žabčice. Technical specifications are detailed in Orság et al. [4]. The air temperature was measured using a Vaisala HMP155A sensor within a RAD14 radiation shield at a height of 2 m above ground level. Precipitation was monitored using a Met One 370 tipping-bucket rain gauge, positioned 1 m above ground level. Measurements were recorded every 30 s.

Meteorological data from the growing season (April–September) of each experimental year are presented in

Table 1. Mean air temperature (°C), years 2020–2022, locality Žabčice, Czech Republic.

Month	Year 2020	Year 2021	Year 2022	Period 1961–1990	Period 1991–2020
IV.	10.6	7.6	8.4	9.6	11.0
V.	13.0	12.7	16.1	14.6	15.6
VI.	18.3	20.9	20.3	17.7	19.2
VII.	19.9	21.2	20.7	19.3	20.9
VIII.	21.3	18.0	21.4	18.6	20.6
IX.	16.0	15.6	14.0	14.7	15.4
average	16.2	16.2	16.8	15.8	17.1

and

Table 2. Overall monthly precipitation (mm), years 2020–2022, locality Žabčice, Czech Republic.

Month	Year 2020	Year 2021	Year 2022	Period 1961–1990	Period 1991–2020
IV.	10.8	13.0	15.2	32.2	27.8
V.	81.8	84.8	43.0	62.8	52.2
VI.	128.6	59.8	54.6	68.6	61.7
VII.	44.0	45.6	98.8	57.1	68.9
VIII.	97.6	175.8	86.0	54.3	61.1
IX.	87.2	22.0	33.6	35.5	53.9
sum	444.6	401.0	331.2	311.5	325.7

.

According to the classification by Kožnarová and Klabzuba [12], the years were characterized as follows:

Year 2020: The average temperature during the growing season was 16.2 °C, which was 0.4 °C above the 1961–1990 long-term normal, classifying the year as a temperature normal year. Precipitation totaled 444.6 mm, representing 142.7% of the normal, and was therefore rated above normal.

Year 2021: The average growing season temperature was again 16.2 °C, which was 0.9 °C lower than the 1991–2020 long-term average, classifying it as a temperature normal year. Precipitation reached 401.0 mm, indicating above-normal precipitation.

Year 2022: The growing season temperature averaged 16.8 °C, matching the long-term average (1991–2020). Total precipitation was 331.2 mm, classifying the year as normal in both temperature and precipitation.

2.3.2. Soil Chemical Parameters

Soil samples were collected in 2022 from both experimental sites (Obora and Písky) at a depth of 0–0.3 m.

Chemical analyses included soil pH, macronutrient availability (P, K, Mg, Ca) using the Mehlich III extraction method, and organic carbon content (Cox). The results are presented in

Table 3. Soil parameters from the experiment localities Obora and Písky, Czech Republic.

Locality	Nutrient Content (mg kg−1)				pH	Cox
	P	K	Mg	Ca
Obora	88.2	154.0	316.0	4358	7.06	1.33
Písky	114.9	166.0	189.0	1967	6.47	1.42

.

2.3.3. Plant Trait Measurements

Plant Height and Number of Tillers

Plant height was measured at the optimal dry matter (DM) content in the aboveground biomass for all varieties and years based on three replications per variety before harvesting. Tiller count was evaluated only in 2021 and 2022 at both locations. For each variety, 30 plants were randomly selected for assessment.

Dry Matter Content and Dry Aboveground Biomass Yield

To ensure that the biomass was harvested at the optimal DM content (~28%), weekly monitoring began in mid-August. Harvesting occurred manually between late August and the end of September, varying by year and variety. Plants were cut approximately 0.1 m above the soil surface within three replicated plots (2.25 m² per plot; 5 m × 0.45 m).

Harvested biomass was weighed using a laboratory scale (precision ± 0.01 kg). Ten representative plants from each replicate were shredded using a Viking GE 450, homogenized, and sampled for DM analysis. The samples were pre-dried at 65 °C for 24 h and then dried at 105 °C for 4 h. Final average yield was expressed as t ha⁻¹ of dry aboveground biomass yield (DABY).

2.4. Statistical Analysis

Data on plant height and dry above-ground biomass yield were analyzed using analysis of variance (ANOVA) in Statistica 14 software [13]. Differences between means were tested using Tukey’s Honestly Significant Difference (HSD) test at a significance level of α = 0.05. Yield variability among sorghum varieties was visually represented using confidence interval (CI) plots (p < 0.05), highlighting the performance of each variety across to the sites over the three years.

Tiller number distributions were displayed using categorized histograms to assess varietal differences in tillering capacity.

3. Results

3.1. Plant Height

Average plant heights differed significantly among varieties (F = 401.6, p < 0.001) with significant main effects of Year (F = 71.6, p < 0.001) and Locality (F = 18.1, p < 0.001), and with significant two-way and three-way interactions: Year × Locality (F = 74.8, p < 0.001); Year × Variety (F = 14.2, p < 0.001); Locality × Variety (F = 3.7, p = 0.001); Year × Locality × Variety (F = 4.8, p < 0.001).

Complete results are shown in

Table 4. Analysis of variance for plant height.

Source	SS	DF	MS	F-Value	p-Value
Year	1.754	2	0.877	71.6	0.000
Locality	0.222	1	0.222	18.1	0.000
Variety	34.418	7	4.917	401.6	0.000
Year*Locality	1.831	2	0.916	74.8	0.000
Year*Variety	2.442	14	0.174	14.2	0.000
Locality*Variety	0.321	7	0.046	3.7	0.001
Year*Locality*Variety	0.818	14	0.058	4.8	0.000

Abbreviations: SS—sum of squares, DF—degrees of freedom, MS—mean square.

.

The RUZROK variety showed the most stable growth there, with plant height ranging narrowly between 2.28 and 2.34 m across all years. In contrast, the NUTRI HONEY variety exhibited significant interannual variability, especially at Obora, where its height decreased from 2.88 m in 2020 to 2.22 m in 2021. Both RUZROK and NUTRI HONEY belonged to the group of shorter sorghum varieties evaluated in this study. On the other hand, the LATTE variety showed the most uniform growth across sites and years, with plant heights consistently ranging from 3.02 to 3.38 m. Taller varieties included KWS TARZAN and KWS HANNIBAL, with KWS TARZAN reaching a maximum recorded height of 4.05 m at the Písky site in 2021. Similarly, KWS HANNIBAL achieved its highest average height of 4.38 m at the same site and year.

A comparative analysis of plant height at both experimental sites during the 2020–2022 period (Figure 1) revealed greater consistency in the plant height of individual varieties at the Obora site.

Overall, varieties tended to grow taller at the Obora site compared to Písky, while plant heights in 2022 were lower for most varieties relative to previous years. Inter-annual variation in plant height was strongly influenced by rainfall distribution and temperature. In 2020, the season featured the highest number of heavy rainfall days (17 ≥ 10 mm), which ensured a relatively even water supply. In 2021, frequent rainy days (69 in total) and several heavy and extreme rainfall events created highly favorable conditions, resulting in maximum growth of KWS HANNIBAL (4.38 m at Písky). In contrast, 2022 was the driest year (331 mm total precipitation, only eight heavy rainfall days), which restricted growth and reduced the height of KWS HANNIBAL to 3.12 m at Písky. NUTRI HONEY, as a shorter cultivar overall, exhibited particularly low stability, reaching only 2.22 m at Obora in 2021, highlighting its sensitivity to heat stress.

3.2. Number and Distribution of Tillers

The best tillering performance was observed in the NUTRI HONEY variety. In 2021, plants at both sites produced an average of 3 tillers, and 100% of plants formed at least one tiller. In 2022, the number of tillers decreased—plants at Obora developed on average one tiller (with 77% of plants tillering), while at Písky the average was two tillers (100% tillering). This variety showed a strong year effect on its tillering ability, particularly at the Obora site.

In contrast, the RUZROK variety exhibited relatively stable tillering across years and locations. In 2021, plants produced an average of 1 tiller, increasing to 2 tillers in 2022. Tillering occurred in 60–87% of plants, depending on the site and year, indicating minimal sensitivity to location-related conditions.

Table 5. Tillering frequency (% of plants with tillers) in sorghum varieties at the Obora and Písky sites in 2021 and 2022.

Location	Obora		Písky
Year	2021	2022	2021	2022
Variety	Distribution of Tillers Occurrence (%)
RUZROK	77	87	60	77
NUTRI HONEY	100	77	100	100
PAMPA TRIUNFO XLT BMR	70	60	77	67
LATTE	100	90	100	70
KWS SOLE	43	0	33	43
KWS FREYA	90	0	77	40
KWS HANNIBAL	80	0	60	0
KWS TARZAN	80	0	30	13

presents the frequency of tiller occurrence in individual sorghum varieties at the Obora and Písky sites.

A highly stable tillering pattern was observed in the PAMPA TRIUNFO XLT BMR variety, which consistently produced an average of 1 tiller per plant across both sites and years. This variety also exhibited the lowest variability in tiller distribution, with 60–77% of plants forming tillers.

A significant year effect was recorded at the Obora site for the group of varieties including KWS SOLE, KWS FREYA, KWS HANNIBAL, and KWS TARZAN. In 2021, between 70% and 80% of plants in these varieties produced 1 or 2 tillers, but in 2022, no tillering was observed in any of them at this location. The Písky site showed more heterogeneous tillering results, with only 24–50% of plants from these varieties forming tillers, and an average of 0–1 tiller per plant across the study period.

Figure 2 and Figure 3 illustrate the tillering performance at each location. The experiment showed that tillering ability was strongly affected by both year and variety, with a partially significant influence of location. These results indicate that tillering in sorghum varieties is highly dependent on genotype and annual climatic conditions, with some varieties exhibiting strong environmental sensitivity and others demonstrating greater stability.

3.3. Dry Aboveground Biomass Yield

Average DABY differed significantly among Year, Locality and Variety (Year: F = 29.09, p < 0.001; Locality: F = 21.10, p < 0.001; Variety: F = 55.96, p < 0.001), and there were significant two-way and three-way interactions: Variety × Year (F = 16.29, p < 0.001); Year × Locality (F = 20.91, p < 0.001); Variety × Locality (F = 3.06, p = 0.006); Variety × Locality × Year (F = 6.03, p < 0.001).

These results (

Table 6. Analysis of variance for the yield of above-ground dry matter.

Source	SS	DF	MS	F-Value	p-Value
Year	188.12	2	94.06	29.09	0.000
Locality	68.23	1	68.23	21.10	0.000
Variety	1266.83	7	180.98	55.96	0.000
Variety*Locality	69.31	7	9.90	3.06	0.006
Variety*Year	737.68	14	52.69	16.29	0.000
Locality*Year	135.23	2	67.2	20.91	0.000
Variety*Locality*Year	272.79	14	19.48	6.03	0.000

Abbreviations: SS—sum of squares, DF—degrees of freedom, MS—mean square.

) indicate that yields depend not only on genotype, but also firmly on environmental conditions. In particular, variation among years reflects the influence of precipitation and temperature distribution on plant performance.

The KWS SOLE variety demonstrated the most stable yield across both experimental sites during the monitored period, with an average DABY of 14.50 t ha⁻¹ at Obora and 13.38 t ha⁻¹ at Písky. An overview of DABY of sorghum varieties is presented graphically in Figure 4. The varieties KWS HANNIBAL and KWS TARZAN showed excellent average yields at both experimental localities. In 2021, KWS TARZAN achieved the highest average yield of 27.81 t ha⁻¹ at Obora and 24.33 t ha⁻¹ at Písky. In 2022, KWS HANNIBAL recorded the highest average yield at both sites, reaching 21.68 t ha⁻¹ at Obora and 27.14 t ha⁻¹ at Písky. Despite the limited precipitation between April and September in 2022, which negatively impacted plant development, KWS HANNIBAL achieved the highest yield of the entire observation period at the Písky site, highlighting its resilience under unfavorable conditions. This again illustrates that climatic effects largely drove inter-annual variation in yield, while genetic potential determined how strongly each cultivar responded.

The scatter plot of DABY and plant height (Figure 5) shows the relationship between plant growth and biomass production. Each point represents an observation, while the red regression line, along with its confidence interval, illustrates the overall trend. The statistically significant, strong positive correlation slope of the regression equation indicates that taller plants generally produce higher DABY (r = 0.622, p < 0.001). This suggests that plant height is a crucial trait contributing to biomass yield in the studied varieties.

To evaluate the impact of plant height on yield variability, the Spearman correlation coefficient was calculated. At Obora, plant height accounted for 17.6% of yield variability, whereas at Písky, its influence was slightly more pronounced at 19.2%. Although the overall impact of height on yield variability was moderate, a positive correlation between plant height and DABY was confirmed, especially in 2022. At both sites, the relationship was statistically significant (p < 0.05), with correlation coefficients of 0.786 at Obora and 0.810 at Písky. These results confirm the relevance of plant height as a breeding trait in sorghum genotypes, due to its strong positive correlation with above-ground biomass yield. This trait proves particularly valuable when selecting genotypes with improved drought resilience, as demonstrated by the performance of taller varieties under low-precipitation conditions in 2022.

Since plant height explained part of the variability in dry aboveground biomass yield, we further examined the influence of the number of tillers. The reason was that tillering capacity may represent an additional trait contributing to yield formation.

After analyzing plant height, the number of tillers was examined to explain the variation in DABY further. The regression (Figure 6) indicates a slightly negative trend (r = –0.12, p = 0.19), suggesting that higher tillering does not necessarily contribute to increased biomass and may even reduce yield efficiency.

Correlation analysis between precipitation and DABY (Figure 7) revealed a very weak negative relationship (r = –0.03, p = 0.86), which was not statistically significant. The regression line was nearly flat, indicating that precipitation had no clear or consistent effect on biomass production within the observed range. Site-specific analyses revealed opposite trends—weakly positive at Obora and moderately negative at Písky—yet these were inconsistent across years.

Given the limited explanatory power of rainfall, we further tested whether accumulated growing degree days (GDD) could serve as a more accurate predictor of yield. Across the three years and two experimental sites (Figure 8), the regression indicated a weak positive relationship (r = 0.21, p = 0.04) between DABY and GDD, suggesting that higher heat accumulation slightly increases biomass production. At the Obora site, which experienced less drought stress, a positive but non-significant correlation was observed (r = 0.37, p = 0.07). In contrast, at the Písky site, where drought stress was more pronounced, the correlation was weaker (r = 0.18, p = 0.27). These findings indicate that while temperature accumulation might contribute to yield formation under more favorable conditions, its explanatory power is limited under drought stress.

4. Discussion

4.1. Plant Height

In our experiment, plant height differed significantly among cultivars and showed strong year and site effects. Growth was more stable at Obora than at Písky: for example, the Czech-bred variety RUZROK varied by only 0.05 m across three years at Obora, whereas inter-site differences did not exceed 0.06 m. RUZROK typically reaches around 1.75 m under local conditions [14], but in our trials it reached 2.02–2.44 m even under reduced precipitation, indicating good environmental adaptability. The fertile soils at Obora may have contributed to this vigorous growth, consistent with reports by Kanbar et al. [15], who observed plant heights up to 2.8 m in southeastern Germany.

By contrast, NUTRI HONEY showed pronounced interannual and site-specific variability. Plants were approximately 0.6 m shorter at Obora in 2021 and at Písky in 2022. This confirms earlier observations that sorghum height depends not only on genetic background but is strongly influenced by agro-environmental conditions [16]. Van Oosterom et al. [17] further suggested that reduced stature may serve as an adaptive response to drought, which aligns with our findings. Since August corresponds to the rapid elongation phase in sorghum plants, lower temperatures during this period likely suppressed growth in temperature-sensitive varieties. Previous studies, such as those by Kamau et al. [18] and Maulana and Tesso [19], have shown that suboptimal temperatures limit nutrient uptake and photosynthetic activity, thereby reducing plant height. Despite these fluctuations, the average height of NUTRI HONEY was 2.60 m at Obora and 2.54 m at Písky, values comparable to those reported by Keskin et al. [20].

The LATTE hybrid showed high stability across years and sites, with interannual and inter-site differences not exceeding 0.3 m. Its average height was 3.18 m at Obora and 3.08 m at Písky, confirming earlier reports that LATTE is a drought-tolerant silage-type hybrid well adapted to Central European conditions [21].

Among the tall varieties, KWS TARZAN and KWS HANNIBAL consistently achieved the tallest plant heights. KWS TARZAN averaged 3.72 m at Obora (range 3.58–3.78 m) and 3.87 m at Písky (3.70–4.05 m), in line with Štěpánek [22], who reported maximum heights up to 5.5 m under optimal conditions. KWS HANNIBAL exhibited the highest variability, with heights ranging up to 1.27 m across environments. At Obora it averaged 3.82 m (3.58–4.03 m), while at Písky, the average plant height reached 3.64 m but ranged from 3.12 to 4.38 m. Interestingly, in this cultivar plant height did not always translate into higher biomass, and a negative association with dry above-ground biomass yield was observed (see Section 4.4).

4.2. Number of Tillers

Tillering capacity showed strong environmental sensitivity in our trials. In 2021, all varieties produced at least some tillers at both sites, but in 2022 several cultivars failed to tiller: at Obora four hybrids (KWS SOLE, KWS FREYA, KWS HANNIBAL, and KWS TARZAN) produced no tillers, while at Písky only KWS HANNIBAL showed complete absence. Despite these differences, tillering did not correlate significantly with dry biomass yield (Pearson r = –0.09, p = 0.64). This suggests that, under Central European conditions, the number of tillers alone is not a reliable predictor of yield performance.

The sensitivity of tillering to year and site is consistent with earlier reports, which indicate that precipitation timing and intensity during vegetative growth play a crucial role [23]. Field management of tiller numbers is therefore challenging due to their high variability across environments and seasons [24]. Physiological studies (e.g., Liu and Finlayson [25]) describe how sheath integrity and its eventual disruption influence axillary bud outgrowth, but our observations suggest that environmental stress can override these developmental cues. Genetic diversity in tillering tendency has also been documented [26], supporting the cultivar-specific differences observed here. More generally, tillering is recognized as a highly complex trait influenced by environmental conditions, hormonal signaling, and competition for assimilates [27,28,29,30]. While tillering contributes to canopy development and leaf area expansion [31,32], it can also influence biomass formation, depending on the context [33]. However, its role in driving dry biomass yield was limited in our experiment.

4.3. Dry Matter Content and Dry Aboveground Biomass Yield

Dry matter content at harvest was within the optimal range for silage production (30–35%), ensuring suitable ensiling quality [34,35,36]. However, in our trials, the DM content did not show any consistent relationship with dry above-ground biomass yield across years or sites. This suggests that DM content, although critical for harvest timing and silage quality, was not a primary factor driving yield variability.

In contrast, apparent differences emerged among cultivars in terms of DABY. Variety KWS SOLE showed the most stable performance, with average yields of 14.50 t ha⁻¹ at Obora and 13.38 t ha⁻¹ at Písky, similar to values reported by Kubeš [37]. Varieties KWS TARZAN and KWS HANNIBAL achieved the highest yields, but with more substantial year-to-year fluctuations. For example, in 2021, variety KWS TARZAN achieved yields of 27.81 t ha⁻¹ at Obora and 24.33 t ha⁻¹ at Písky, which is consistent with the high yields reported by Povolný and Hampl [38]. In 2022, variety KWS HANNIBAL recorded 21.68 t ha⁻¹ at Obora and 27.14 t ha⁻¹ at Písky despite below-average precipitation. This suggests that tall hybrids have high yield potential but are more sensitive to environmental variation, in line with broader evidence linking plant height and biomass to yield performance across environments [39].

Variety NUTRI HONEY displayed site-specific responses: in 2020, it yielded 19.03 t ha⁻¹ at Obora versus 13.48 t ha⁻¹ at Písky, confirming earlier reports that soil type influences its productivity [20]. In other years, however, yields were higher at Písky, indicating strong genotype × environment interactions. Similar resilience of tall hybrids under water-limited conditions has also been reported by Islam and Sedgley [40] and Hammer et al. [41].

4.4. Climatic Drivers and Genotype × Environment Interactions

Interannual yield variability was partly associated with climatic conditions, but no single parameter consistently explained the differences. When tested across the entire dataset, climatic variables had only limited explanatory power: precipitation showed no apparent effect, and neither did mean growing-season temperature. Similar conclusions were reported by Guerra et al. [42], who demonstrated that climatic variability strongly influenced the performance of sorghum biomass across Brazilian regions, but could not be reduced to a single driver.

The strong Year × Variety and Year × Locality interactions detected by ANOVA (Table 6) confirm that cultivar performance was highly dependent on specific environmental conditions. For example, variety KWS HANNIBAL achieved the highest recorded yields in 2022 despite limited precipitation, highlighting its resilience, whereas other cultivars performed better under more favorable rainfall patterns in 2020. These results are consistent with research by Atim et al. [43], who also observed that multi-environment trials of forage sorghum hybrids revealed significant genotype × environment effects on biomass yield and stability. Dolapčev Rakić et al. [44] further highlighted that no single environmental parameter explained yield variability, underlining the complexity of sorghum adaptation.

Our findings confirm that the yield stability of sorghum varieties in Central Europe is not solely explained by climate, but by genotype × environment interactions, in which cultivars exploit or tolerate variable conditions under different environmental conditions.

4.5. Agronomic Traits: No Single Driver of Yield Instability

Beyond climatic influences, correlation analyses with plant traits further clarified the sources of yield variability. Plant height accounted for approximately 17–19% of yield variability and showed a positive correlation with DABY, particularly in 2022, but its explanatory power was limited. These interpretations are consistent with international research on plant height and biomass yield performance across diverse environments [45], as well as with multi-environment forage sorghum varieties evaluations, which show that plant morphology contributes to, but does not solely determine, yield stability [43,45].

Dry matter content and tillering capacity (evaluated in 2021 and 2022) did not show a significant correlation with yield. This aligns with findings by Dolapčev Rakić et al. [44], who reported that yield variability in forage sorghum could not be predicted by a single agronomic trait but instead reflected complex trait × environment interactions.

Several other traits not measured in detail here—including the number of tillers per plant [33], leaf area and number of leaves [46], and stomatal density and relative water content [47]—are known to affect biomass accumulation and could contribute to differences among genotypes. Likewise, agronomic practices [48] interact with genetic predisposition to shape performance. These findings underscore that no single trait is sufficient to explain yield variability; rather, a combination of morphological and physiological factors must be considered when evaluating sorghum cultivars for resilience under Central European conditions.

4.6. Practical Implications for Cultivar Selection

From a practical perspective, our results highlight key differences among cultivars. Variety KWS SOLE exhibited the most stable yields across environments, making it suitable for systems where consistency is critical. In contrast, varieties KWS TARZAN and KWS HANNIBAL achieved the highest yields but also showed greater interannual variability, indicating high potential under favorable conditions but reduced predictability under stress. Variety NUTRI HONEY displayed strong site-specific responses, performing better on heavy soils in 2020 but yielding more on sandy soils in later years. Such differences underscore the importance of considering both average yield and yield stability when recommending cultivars for silage production in temperate, drought-prone regions.

4.7. Limitations and Future Directions

This study has some limitations. Tillering data were collected only in 2021 and 2022, which restricts the ability to fully assess its contribution across variable years. Future work should therefore focus on (i) integrating daily weather data and crop stage-specific responses, (ii) applying multi-environment stability models such as AMMI and GGE biplot analysis to quantify better genotype × environment interactions [49,50], and (iii) linking field observations with high-performance phenotyping and genomic tools. Recent advances in drone- and satellite-based monitoring [51], combined with genome-wide association studies [52], already show promise in identifying traits associated with drought tolerance and yield stability. At the molecular level, several drought-responsive regulatory networks, including dehydration responsive element binding proteins (DREB), LEA proteins, and ABA signaling pathways, have been identified as critical for sorghum varieties adaptation to water deficit [53]. Incorporating such approaches will enhance our understanding of the mechanisms underlying resilience and support the breeding of sorghum cultivars adapted for increasingly variable Central European conditions.

5. Conclusions

Our study highlights the substantial influence of cultivar selection, environmental conditions, and site-specific soil characteristics on dry matter yield in forage sorghum varieties. While medium-heavy soils are generally considered more favorable, our results suggest that certain hybrids can outperform expectations even on lighter soils, particularly under water-limited conditions. The demonstrated yield stability and drought resilience of varieties such as KWS HANNIBAL and KWS TARZAN underscore the potential of sorghum as a reliable forage crop in regions vulnerable to climate change.

Given its adaptability, high biomass productivity, and tolerance to drought stress, sorghum represents a promising supplementary forage crop for the Czech Republic, especially in areas increasingly affected by soil moisture deficits and rising temperatures. These findings support the integration of sorghum varieties into local crop production systems, sustainability and resilience under changing climatic conditions.