Groundnut is the main legume and oil seed crop with major economic importance, renowned for the high protein and oil content necessary for food as well as for industrial purposes. The breeding activities of yield enhancement and quality attribute are essential because the demand in the world concerning food supply production has been drastically increasing. These factors affect productivity, including variation due to genetics, environmental condition, and type of agronomy used. Pod yield (PY), grain yield (GY), seed per plant (SPP), and hundred seed weight (HSW) are some of the key parameters which are indicators of crop performance and have been well studied in crop development
(Patel et al., 1991; Anothai et al., 2008).
Even with unprecedented advances in agronomic management and genetic improvement, the interaction of these elements remains unpredictable and multifaceted, with extensive research required to identify the most suitable genotypes to employ in breeding. The objective of this study is to summarize the performance of 17 genotypes of groundnut on the most important agronomic traits in the hope of identifying their ability to enhance the yield and quality characteristics. The trait variation, for example, seed number per plant and seed weight, will facilitate the selection of suitable genotypes for different conditions and farming systems. The ANOVA table represented hereunder for the key traits (Table 2) related to genotype × environment interaction including genotype, environment and genotype × environment mean sum of squares.
F-values were high and significant as well (≤0.001) in the pooled ANOVA analysis for the four traits, Pod Yield (PY), Grain Yield (GY), Seeds per Plant (SPP), and Hundred Seed Weight (HSW), suggesting substantial genotypic effects on all traits. F-value of 8.05 was observed for Py with genotype mean square of 28,221.50 and matching error mean square of 3,507.65. This indicates that the variation in quantity of pods produced was mostly due to genetic variation. High genotypic effect and substantial impact of genotypic variation on grain yield were additionally substantiated by the significant F-value for Grain Yield (GY) of 6.05 with the genotypic mean square (41,612.29) was significantly higher than the error mean square (6,874.56). The SPP had a moderate genotypic effect, as shown by its F-value of 2.60 and a genotype mean square of 121,226.43, suggesting that there is a moderate contribution of genotype to variation in seed number per plant. It still suggested a significant genotypic effect, however. Hundred Seed Weight (HSW) also exhibited a high genotypic effect on seed size, as indicated by the F-value of 3.62. The genotype mean square was quite bigger (98.56) than the error mean square (27.24).
This discovery serves to emphasize the importance of genotype in the selection of elite breeding lines for drought tolerance. Selective breeding toward specific genotypic traits associated with the performance for yield and seed quality attributes would have a significant effect in enhancing performance under low rainfall (as practiced in the study area of the Pawe Agricultural Research Center), as the high significance levels for all traits indicated. In contrast, SPP had a relatively low F-value but considerable variation among genotypes, which implied its potential importance in the breeding strategies for improving the total production. The findings support the need for sustained breeding efforts to improve yield components for drought-prone conditions such as hundred seed weight, grain yield and pod yield.
LS Mean table (Table 3) gives least square means (adjusted against any variables) of different agronomic traits across genotypes, with statistical comparison to determine significant differences between treatments.
Mean values of the measured traits across all genotypes were significantly different, ranging from plant number per plot (PNPP) from 15.3 to 25.7, pod yield (PY, kg/ha) from 1508.0 to 2547.5 and grain yield (GY, kg/ha) from 1211.1 to 2044.4. Seed per plant (SPP) ranged from 21.2 to 41.2 and hundred seed weight (HSW, gm) ranged from 27.3 to 41.7.
Coefficient of variation (CV) showed there was high uniformity among traits under study and that SPP had the highest CV value at 21.52% with the highest seed number variability than other traits being studied. Least significant difference (LSD) values obtained were 183.24, 229.87, 456.2, and 12.13 for PY, GY, SPP and HSW, respectively, and allowed for determination of statistically significant differences among the genotypes.
The research also showed that genotype 104 (RDRGVT ICGV SM 01514) had the highest values of both PY (2547.5 kg/ha) and GY (2044.4 kg/ha), and the lowest values of both parameters were obtained by genotype 107 (RDRGVT ICGV SM 3530) (PY = 1508.0 kg/ha, GY = 1211.1 kg/ha). The highest average SPP was in genotype 113 (RDRGVT ICGV 8540) at 41.2, and the lowest seed number per plant was in genotype 107 at 21.2. For HSW, genotypes 110 and 112 had the highest seed weight of 41.0 g, while the lowest seed weight of 27.3 g was in genotype 103.
The data presented here demonstrate considerable variability in the agronomic characteristics of the genotypes tested, which is essential for choosing high-yielding varieties with desirable traits for breeding programs. The highest GY recorded in genotype 104, RDRGVT ICGV SM 01514, is in line with earlier studies where a higher yield potential was associated with enhanced pod and seed development
(Kebede et al., 2017). The performance of these genotypes suggest that it might be appropriate for high-input agricultural systems which are seeking to optimize grain yield.
In contrast, genotype 107 (RDRGVT ICGV SM 3530) had the lowest yields, which may be an indication of poor environmental adaptation or suboptimal genetic expression under the current field circumstances. This result is in line with the results of
Chaudhary et al., (2021), who found that genotypes with weaker adaptation to some settings have lower production potential.
Based on seed number per plant SPP, Genotype 113 or RDRGVT ICGV 8540 exhibited superiority over others and produced up to 41.2 seeds per plant. It is said that the crop yields are considerably dominated by quantity per plant through which seeds; on the positive aspect, shows good linkage towards the pod yields, as confirmed in
Baker et al., (2003). Probably, there can be possible causes of variance because of SPP genotypes attributed to genetics in seed-set parameters and reproductivity functions. According to
Patel et al., (2024), improving this feature is important to increase yields in leguminous crops.
The hundred seed weight (HSW) is another critical characteristic needed for production and quality. The highest HSW occurred in genotypes 110 and 112, with an average of 41.0 g. This characteristic, often linked to environmental factors such as nutrient availability and water stress
(Wang et al., 2023), suggests that these genotypes might be suitable for situations where seed size becomes the major factor in being marketable or processed.
The coefficient of variation for SPPl and HSW shows that these traits were relatively more variable between genotypes than PY or GY, which had lower CVs. This underlines the requirement for selection to be made for consistency in these attributes, especially seed number per plant, in a breeding programme aimed at enhancing yield potential. Moreover, the correlation between PY and GY further supports the hypothesis that higher pod yield could be directly translated to higher grain yield, as has been previously suggested by
Patel et al. (2024).
The LS Mean values and statistical significance on LSD, show success in the selection of genotype, since it assured yields above other genotypes, among them was genotype 104, RDRGVT ICGV SM 01514. Howsoever, environmental issues and management strategies combined with genotypic interactions require more research for practical application in an agricultural field setting
(Gulluoglu et al., 2017).
Genotypic variability and performance
All agronomic traits presented significant genotypic differences; hence, the groundnut lines analyzed exhibited a high level of genetic variation. Significant GEIs were reported for pod number per plant and seed weight. This implies that the environmental conditions prevailing in every cropping season impacted the performance of the traits under study
(Kalarani et al., 2023). This is justified for testing groundnut genotypes in different conditions in order to obtain the best stable lines for drought tolerance.
Some of the genotypes, such as RDRGVT ICGV SM 01514, RDRGVT ICGV 14788, and RDRGVT ICGV 8540, performed better than the control variety Werer-961 in pod and seed yield in both seasons. The high yield performance of these genotypes can be explained by the high number of pods per plant and seed weight
(Tadesse et al., 2018). For example, RDRGVT ICGV SM 01514 produced 3.2 tons per hectare, which was significantly higher than the control variety yield of 2.5 tons per hectare (Fig 1). Moreover, these lines had improved resistance to drought stress, as indicated by their ability to produce more pods and seeds even under reduced water availability.
Correlations among agronomic traits
The correlation matrix (Table 4) shows the Pearson correlation coefficients between agronomic traits, which express the direction and strength of the linear relationship that exists between each pair of traits among the genotypes tested.
The correlation analysis indicated high positive correlations between hundred seed weight and seed yield (r = 0.79) and between number of pods per plant and pod yield (r = 0.82). These findings indicate that selection for seed weight and number of pods can potentially increase pod and seed yield under drought stress
(Zhang et al., 2019). Seed yield and number of days to maturity also indicated a high negative correlation (r = -0.62), indicating that early-maturing genotypes perform better under drought stress conditions due to their shorter growth duration and lower water requirement
(Mubai et al., 2020). The growth-related characteristics included increased leaf area, greater dry matter accumulation, enhanced yield components and increased availability of nutrients during the growing season. (
Ananda and Sharanappa, 2017). In addition to having more space available to the groundnut plant, which increases photosynthetic activities and leads to higher growth and yield attributes, the presence of groundnut in the paired row system most likely had more synergistic effects than antagonistic ones. These factors ultimately reflected in the groundnut yield (
Hadiyal, 2020). Groundnut quality and productivity are influenced by environmental factors as well as enhanced agronomic techniques, particularly planting techniques, which are crucial to increasing groundnut crop yield
(Olayinka et al., 2020).
The association of the traits remained consistent between both study years, indicating that association among agronomic traits was maintained irrespective of varying environmental conditions
(Almeida et al., 2014). The findings highlight the requirement of selection for multiple traits such as pod number, seed weight, and maturation period during breeding for groundnut drought tolerance.
Normality and statistical analysis
Normality test and statistical analysis include testing the distribution of data to check whether it is normally distributed and then using statistical methods for appropriate analysis and interpretation. This graph uses a histogram drawn using blue bars to show the frequency distribution of PNPP values. An estimated probability distribution is marked by an overlaid density curve (smooth blue line). The histogram gives information on the spread and skewness in the data.
Distribution of agronomic characteristics was examined through histograms and Q-Q plots. The quantile-quantile (Q-Q) plot compares the ordered PNPP values (y-axis) with theoretical quantiles (x-axis) of a normal distribution. The red diagonal line represents the expected normal distribution, while the blue dots indicate actual data points. Deviations from the red line suggest departures from normality. Most of the characteristics were normally distributed, with slight deviations recorded from hundred seed weight and seed yield (Fig 1 and 2). These deviations would have been due to environmental factors to modify these characteristics, in this case, variations in rainfall and temperatures during the cropping season (
Reig-Valiente et al., 2018). Findings on pod number per plant and seed weight, however, showed very well standard distributions, with signs pointing towards the use of these characteristics being effective and indicative measures of drought resistance (Fig 1).
Fitted value plots are graphical representations of the model’s predictions of dependent agronomic traits for different genotypes or treatments (Fig 3). It was shown (as red dots) in comparison to the predicted values based on the regression model. A blue regression line, with a shaded confidence interval, represents the trend in the trait values. The slight decline indicates a weak the trend. The plots describe the relationship between predicted and observed values, thus enabling a determination of the model’s fit to the empirical data. Fitted values are the predicted outcomes of the statistical analysis, while the residuals being the difference between observed and fitted values indicate the accuracy of the predictions. Fitted value plots are crucial for trend detection, model assumption checking, and the identification of patterns that could reflect model inadequacy or unexplained variation.
