Sorghum (Sorghum bicolor (L.) Moench) is a major cereal crop cultivated both for its grain, used for human consumption and for its residues, used as animal feed [1]. Because of its high ecological plasticity, sorghum is able to adapt to a wide range of agro-climatic conditions, both in tropical and temperate regions, making it a strategic crop face of climate risks.

Africa is recognized to be a center of domestication of several cereals, including sorghum, which is the second most important cereal after maize in terms of cultivated area (22% of global cereal acreage) according to [2]. In 2022, sorghum FAO global production was estimated at 57.58 million tons with a harvest area of 40.76 million hectares, placing it fifth among the world’s most produced cereals, after maize, wheat, rice, and barley [3]. In developed countries, sorghum is primarily used for animal feed, while in Africa and India, it is a staple food for millions of people [4].

Despite significant progress in sorghum varietal improvement, both through conventional methods (mass selection, pedigree selection, mutagenesis) and the introduction of new improved varieties, productivity levels remain relatively low in smallholder farming areas. Yields rarely exceed 1.5 t ha⁻¹ in many production zones, which does not adequately meet the growing food needs of the population [5].

In Niger, sorghum plays a dominant role in agricultural production systems. In 2019, it was cultivated on approximately 3.8 million hectares, representing 20% of the cultivated area ranking as the second most widely cultivated cereal after millet [1]. However, the national average yield which is very low (506 kg ha⁻¹ in 2019) remains below the estimated global average of 1,445 kg ha⁻¹ for the same year [1]. According to [3], national production of sorghum was approximately 2.10 million tons from a harvested area of 3.78 million hectares. This low productivity is mainly due to several major constraints, including low soil fertility, pest and disease pressure, and the variability and scarcity of rainfall.

Since 1970s, significant efforts have been made by breeding programs in West Africa, with a particular focus on improving intermediate sorghum varieties, especially the Guinea-Caudatum type, to increase their yield potential and their adaptation to local agro-ecological conditions [6]. In this context, the agro-morphological and agronomic evaluation of sorghum varieties from different agro-ecological zones appears to be a relevant approach for identifying high-performing genotypes well-suited to the production conditions of Niger, particularly in Dosso region (13˚02′46ʺN, 3˚11′50ʺE) in the Southwest, which is highly exposed to the effects of climate change.

The overall objective of this study is to evaluate the agronomic performance of various sorghum varieties originating from different West African countries in order to identify varieties that are best adapted to the agro-ecological conditions of Niger. Specific objectives were to identify high-performing sorghum varieties in terms of grain yield, which are likely to contribute to improved food security and to evaluate the response of different varieties to major production constraints, including low soil fertility, pest pressure, diseases and low rainfall.

The study was conducted at the INRAN experimental station in Bengou (Dosso Region, Gaya Department; 3.45˚E longitude and 11.88˚N latitude). Bengou is located at 3˚33′25″E longitude and 11˚59′44″N latitude, with an average altitude of 172 m. Bengou lies in the Sudanian zone with a unimodal rainfall regime (June–October), annual rainfall of 800 mm - 900 mm, peak precipitation in August, and mean temperatures of 24˚C - 35˚C. Soils are ferruginous tropical soils, sandy-loam, well-drained, slightly acidic to neutral, and low to moderately fertile. The station was established in 1974 as a research site dedicated to the Sudanian zone, characterized by an average annual rainfall ranging from 800 to 900 mm. During the period of the experiment, total cumulative rainfall recorded at Gaya was approximately 820 mm - 830 mm, which is within the normal range for the agro-ecological zone. The plant material used consists of twenty-four (24) sorghum varieties coming from different NARs and international research institutions, including one local variety used as control (Table 1).

Table 1. List of varieties used.

The trial was conducted using a randomized incomplete block design with three replications. Each replication comprised five subblocks of five varieties. Sowing was carried out in hills, with a spacing of 0.4 m between hills and 0.8 m between rows. As one of the varieties originating from Togo did not germinate, the evaluation ultimately included 24 varieties.

The phytosanitary and agronomic assessment was carried out by assigning severity scores based on the estimated percentage of leaf area affected by diseases and pests.

The severity of disease attacks was rated on a scale of 0 to 5, where 0 corresponds to the absence of symptoms (0% of leaf area infected), 1 to a highly resistant variety (1% - 10% of leaf area infected), 2 to a resistant variety (11% - 25% of leaf area infected), 3 to a moderately susceptible variety (26% - 50% of leaf area infected), 4 to a susceptible variety (51% - 75% of leaf area infected), and 5 to a highly susceptible variety (>75% of leaf area infected).

Data collection involved randomly selected samples from the center rows of each plot to minimize edge effects and inter-varietal competition. A total of thirteen (13) agro-morphological and phytosanitary variables were measured to assess the agronomic performance of the sorghum varieties. These variables were selected as part of the process of identifying the highest-performing varieties best adapted to the agro-ecological conditions of the study area.

The collected data were entered into an Excel spreadsheet. This software was also used to create tables and graphs using means and standard deviations. The different variations were evaluated by determining, for each variable, the mean, the coefficient of variation, and the observed probability. One variety originating from Togo failed to germinate in all experimental units. Consequently, this genotype was excluded from data collection and statistical analysis. The trial was therefore analyzed on the basis of twenty-four (24) varieties instead of the initially planned twenty-five (25). Because the missing variety occurred systematically across all replications, it did not introduce imbalance among treatments within blocks. The randomized incomplete block design was thus maintained, and the analysis of variance (ANOVA) was performed considering only the effectively established varieties. Mean comparisons were conducted on the reduced dataset using the Bonferroni test at the 5% significance level.

The significant variability observed among sorghum varieties for phenological, morphological, phytosanitary, and yield-related traits reflect strong genetic diversity among the tested germplasm. Such diversity is essential for identifying varieties adapted to specific agro-ecological conditions, particularly in rainfed systems characterized by climatic uncertainty such as the Sudanian zone of southwestern Niger.

Late-maturing varieties generally exhibited greater plant height and biomass production, confirming that longer vegetative cycles allow extended photosynthetic activity and greater assimilate accumulation. This trend is consistent with previous studies reporting a positive association between cycle length, plant height, and biomass production in sorghum grown under favorable rainfall conditions [7][8]. In Bengou, where rainfall during the experimental period was within the normal range (820 mm - 830 mm), late-maturing varieties were able to fully exploit available soil moisture, resulting in enhanced vegetative growth.

Plant height plays an important role in lodging susceptibility, productivity and intercropping potential [9]. A short stem height, less than 2 m, is often preferred for the intensification of rainfed crops [10]. The observed variation in plant height can be explained by the genetics of the varieties and their adaptability to the environment. The number of mature green leaves ranged from 3 to 7.5, with variety ICSB 176008 exhibiting the maximum and ISRA-S-756-34 the minimum. Green leaf area is essential for photosynthesis, light capture, and CO₂ fixation, directly contributing to biomass production [11]. A positive correlation between the number of green leaves and dry biomass confirms the importance of foliage in dry matter accumulation [12].

The strong performance of the MDK variety can be attributed to its local adaptation to the Sudanian agro-ecological conditions of Gaya. Its tall stature, high dry biomass, and superior 1000-seed weight indicate efficient assimilate partitioning toward both vegetative and reproductive organs. Despite producing fewer panicles than some varieties, MDK achieved the highest grain yield, demonstrating that grain yield in sorghum is more strongly influenced by grain filling efficiency and seed weight than by panicle number alone. This finding is supported by the strong positive correlations observed between grain yield, seed weight per plot, and 1000-seed weight.

Disease pressure varied significantly among varieties, highlighting differences in genetic resistance. The high resistance observed in most varieties to anthracnose and zoned leaf spot likely contributed to maintaining green leaf area during grain filling, thereby supporting sustained photosynthesis and yield stability [13]. Disease tolerance is particularly important in smallholder systems where chemical control options are limited. The negative correlation between seed mold severity and flowering time suggests that late-maturing and taller varieties may escape late-season humidity conditions that favor mold development, as previously reported [14].

A negative correlation between the date of 50% flowering and the number of harvested panicles suggests that delayed flowering can reduce the number of productive panicles, particularly in Sahelian regions where rainfall can cease early [6]. Furthermore, the positive correlation between the number of green leaves and cycle length shows that, in C4 plants such as sorghum, the number of leaves increases with cycle length [15], confirming previous observations.

Although the grain yields obtained under experimental conditions were substantially higher than the national average, this difference can be explained by optimal crop management practices, including timely sowing, appropriate plant density, and effective weed control. Nevertheless, the strong performance of locally developed varieties such as MDK suggests that significant yield gains are achievable under farmer conditions if basic agronomic practices are improved. This confirms that productivity gains in Niger can be achieved not only through exotic germplasm but also through the optimization and dissemination of well-adapted local varieties.

MDK’s high grain yield is mainly attributed to its local adaptation, tall stature and high biomass, superior grain filling (high thousand-seed weight), and good tolerance to major diseases, combined with optimal experimental management conditions. Although yields exceeded the national average, MDK’s local origin supports its potential for scaling under improved agronomic practices in Niger.

This study demonstrated significant agronomic variability among sorghum varieties evaluated at INRAN Bengou research station. Several varieties showed superior grain yield, early maturity, and strong resistance to anthracnose and zoned leaf spot. The locally adapted variety MDK exhibited the highest grain yield, associated with high biomass production and superior grain weight, highlighting its strong potential for scaling under improved agronomic management. These results confirm that yield improvement in Niger can be achieved through the promotion of locally adapted varieties combined with appropriate crop management practices.