Reach Us +3225889658
Performance Evaluation of Sorghum (Sorghum bicolor (L.) Moench) Genotypes for Grain Yield and Yield Related Traits in Drought Prone Areas of Ethiopia
ISSN: 2329-8863

Advances in Crop Science and Technology
Open Access

Our Group organises 3000+ Global Conferenceseries Events every year across USA, Europe & Asia with support from 1000 more scientific Societies and Publishes 700+ Open Access Journals which contains over 50000 eminent personalities, reputed scientists as editorial board members.

Open Access Journals gaining more Readers and Citations
700 Journals and 15,000,000 Readers Each Journal is getting 25,000+ Readers

This Readership is 10 times more when compared to other Subscription Journals (Source: Google Analytics)
All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.
  • Research Article   
  • Adv Crop Sci Tech 2019, Vol 7(2): 423

Performance Evaluation of Sorghum (Sorghum bicolor (L.) Moench) Genotypes for Grain Yield and Yield Related Traits in Drought Prone Areas of Ethiopia

Amare Seyoum*, Adane Gebreyohannes, Amare Nega, Habte Nida, Taye Tadesse, Alemu Tirfessa and Tamirat Bejiga
Melkassa Agricultural Research Centre, National Sorghum Research Program, Ethiopian Institute of Agricultural Research, Adama, Ethiopia
*Corresponding Author: Amare Seyoum, Melkassa Agricultural Research Centre, National Sorghum Research Program, Ethiopian Institute of Agricultural Research, Adama, Ethiopia, Tel: +251962069454, Email: [email protected]

Received Date: Feb 06, 2019 / Accepted Date: Mar 20, 2019 / Published Date: Mar 29, 2019


Sorghum is an indigenous crop of Ethiopia grown in highly diverse environments of having water stress, soil fertility and temperature conditions. Developments of sorghum varieties resistant to drought and producing better grain yield while addressing the plant biomass requirement is one of the strategies in the sorghum breeding program to the dry lowland environment. A total of 22 early maturing sorghum genotypes were evaluated by including one standard check Melkam to estimate the grain yield and stability of performance across the test environments. The trial was conducted using RCBD in row and column arrangement 4. The analysis of variance using additive main effect and multiplicative interaction (AMMI) model revealed significant difference (p<0.01) for genotypes, locations and genotype by location interactions. The two principal components explained 60.4 % further variation due to genotype by location interaction. Based on the overall performance of grain yield, flowering time and plant height and the grain yield stability the variety 2005MI5064 was identified as a potential candidate for production to the target environment. The result also showed that the varieties IESV92084-DL and 2001MS7015 had better grain yield performance and stability across environment, which could be used as potential parental lines for genetic improvement in the sorghum improvement program.

Keywords: Variety; GEI; AMMI; Interaction principal component; Regression coefficient


Sorghum (Sorghum bicolor (L.) Moench) is a diploid C4 cereal crop which was domesticated in Africa. It has 2n=20 chromosome and genome size of 750 Mb [1]. Sorghum is predominantly self-pollinated short day plant with the degree of spontaneous cross pollination, in some cases, reaching up to 30%, depending on the shape and type of panicles. It is an indigenous crop of Ethiopia grown in highly diverse environments of having water stress, soil fertility and temperature conditions. In Ethiopia, sorghum grows from the lowland areas which receive lower amount of rainfall and has high temperature to the highland characterized by low temperature and higher amount of rainfall [2]. Sorghum is the fifth major cereal crop in the world and third in Ethiopia which is the most important dry land crop grown by 6 million small holder farmers in more than 1.9 million hectare of land.

Sorghum is considered as major food security crop in Ethiopia which is contributing 18% of the total grain production [3]. Although sorghum is primarily grown in Ethiopia as grain crop, farmers uses its biomass for different purposes which is equally valued as grain. The crop has a wider agro-ecological adaptation, however, is best suited and widely grown in the dry land areas, where water scarcity is limiting for crop production. More than 70% of the area for sorghum production is in Africa and India. USA, India, Mexico, Nigeria, Sudan, and Ethiopia are among the top ranked sorghum producing countries [4]. In Ethiopia sorghum is grown in 1.9 million ha of land and the share of the dry lowland agro ecology is 66%-70% of the total sorghum growing areas. Sorghum in Ethiopian has multiple uses. As the grain is preferred for food the biomass is equally important for farmers to use it as animal feed. Farmers feed their cattle while the plant is the field and store the biomass for the drier season. Hence, variety development considered dual purpose interest both grain and biomass yield.

Sorghum grain has a wide range of nutritional compositions and constituted about 11% water, 340 k/cal of energy, 11.6% protein, 73% carbohydrate and 3% fat by weight [5]. Ethiopia has a wealth of sorghum genetic resources that could be used for increasing productivity and nutritional quality of sorghum. In fact, sorghum is traditionally grown with low adoption rate of improved varieties and production techniques, it has low productivity. The recurrently occurring mainly in the drought prone sorghum growing areas crop failure is becoming eminent every two or three years. The recurrent drought coupling with the lower usage of improved sorghum production technologies increases the vulnerability of sorghum growing farmers in the dry lowlands of Ethiopia. In a recent study it was estimated that poverty level is about 29% of the sorghum growing households [6].

In Ethiopia, drought is usually occurring due to delay in onset, dry spell after sowing, drought during critical crop stage (flowering and grain filling stage) and too early cessation of rain. These situations can be overcome by developing sorghum genotypes which are resistance to moisture stress [7]. Developments of early maturing and drought tolerance varieties containing stay green trait that assist the plant to delay leaf senescence are the major strategies in the national sorghum breeding program of Ethiopia. Genotypes specifically developed for the target environment through crossing of elite materials likely to be tested for their suitability to the moisture stress environments.

Exploitation of genetic variability is the most important tool in plant breeding, and this has to be inferred by phenotypic expression. The consequences of the phenotypic variation depend largely on the environment. This variation is further complicated by the fact that not all genotypes react similar ways to the change in environment. If relative performance of genotypes is different in different environments, then G × E interaction becomes a major challenge to crop improvement. Genotype by environment interaction is the variation, arising from the lack of correspondence between the genetic and non-genetic effects in multi-location trials.

Interactions may involve change in rank order for genotypes between environments and change in the absolute and relative magnitude of the genetic, environmental and phenotypic variances between environments [8]. In general, genotype by environment interactions happen when two or more genotypes perform differently in several environments and are selected as differential genotypic sensitivities to environments. The different response of genotypes across the testing environment is considered as a hindrance in selecting and recommending of crops and cause yield fluctuation [9]. Genotype by environment interaction may offer opportunity for selection and adaptation of genotypes that showed positive interaction with the specific location which helps in the effective utilization of specifically adapted genotypes [10].

Different statistical models are being used to quantify the genotype by environment interactions including the classical analysis of variance (ANVOA), stability and multivariate analysis. The additive, multiplicative interaction component (AMMI) model is a hybrid analysis that incorporates both the additive and multiplicative components of the two-way data structure. In this model the additive portion is separated from interaction by analysis of variance (ANOVA) and the multiplicative component is further decomposed by interaction principal component [11]. Therefore, this study was initiated to know and quantify the magnitude of genotype by environment interaction and identify high yielding, high biomass and stable sorghum genotypes.

Materials and Methods

Description of test environment

The field testing was conducted during the main cropping season of at seven environments of four locations (Kobo, Mieso, Sheraro and Shaorobit ), representing the moister stressed dry lowland areas of Ethiopia located in the altitude range of 1297-1513 m.a.s.l, where sorghum is predominantly frown by small holder farmers crop grown (Table 1).

Location Longitude Latitude Altitude in m.a.s.l Soil type Rain fall in mm Minimum T° Maximum T°
Kobo 39°38'E 12°09'N 1513 Vertisol 678 14.8 32
Miesso 39°21'E 8°30'N 1470 Vertisol 571 16 31
Sheraro 39°9'E 14°6'N 1179 Vertisol 615 20.4 34
Shoarbite 39°93'E 10°35'N 1500 Vertisol 713 17.7 33

Table 1: The agro-ecological characteristics of the experimental locations.

Description of test genotypes

A total of 22 genotypes including one popular released variety Melkam as a standard check evaluated from 2012 to 2014 (Table 2). The genotypes were derived from the pedigree breeding program through crossing of landraces with early maturing elite breeding lines and improved varieties. These genotypes were selected based on their overall field performance and adaptation to the targeted environments. In addition, five genotypes previously introduced from ICRISAT were included in the evaluation.

Entry # Genotype Pedigree Seed source
1 2001MS7003 LocalBulk(white)/SRN-39 Developed by pedigree breeding
2 2001MS7013 PGRC/E#222880/ICSV-1/KAT369-1 Developed by pedigree breeding
3 2001MS7015 PGRC/E#222880/ICSV-1/KAT369-1 Developed by pedigree breeding
4 2001MS7037 PGRC/E#222878/ICSV708 Developed by pedigree breeding
5 IESV92084-DL IESV92084-DL Introduced from ICRISAT
6 IESV92168-DL IESV92168-DL Introduced from ICRISAT
7 IESV92199-DL IESV92199-DL Introduced from ICRISAT
8 IESV92057-DL IESV92057-DL Introduced from ICRISAT
9 IESV9027-DL IESV9027-DL Introduced from ICRISAT
10 2001MS7007 CR:35:5/DJ1195/N13 Developed by pedigree breeding
11 2005MI5060 WSV387/P9403 Developed by pedigree breeding
12 2005MI5064 WSV387/P9404 Developed by pedigree breeding
13 2005MI5065 WSV387/P9405 Developed by pedigree breeding
14 2005MI5066 M36121/P9401 Developed by pedigree breeding
15 2005MI5069 M36121/P9402 Developed by pedigree breeding
16 2005MI5070 M36121/P9403 Developed by pedigree breeding
17 2005MI5075 3443-2-OP/P9401 Developed by pedigree breeding
18 2005MI5079 3443-2-OP/P9401 Developed by pedigree breeding
19 2005MI5081 3443-2-OP/P9403 Developed by pedigree breeding
20 2005MI5082 3443-2-OP/P9403 Developed by pedigree breeding
21 ICSR24005 ICSR24005 Developed by pedigree breeding
22 Melkam WSV387 Check popular released variites

Table 2: List of sorghum genotypes used for this study and their origin.

Experimental design

The experiment was conducted at Mieso, Sheraro, Shoaroit and. Kobo from 2012 to 2014. Randomized Complete Block Design (RCBD) was used to lay down the genotypes with three replications in a row column arrangement to minimize the special variability in estimating the genetic value. Each plot contained three rows of 5 m length separated by 0.75 m. At all locations sowing was done in between last week of June to first week of July when sufficient rain was received. Plantation was done manually, and population was adjusted by thinning considering 0.15 m as spacing between plants. DAP fertilizer was applied at planting time with the rate of 100 kg/ha and Urea was side dressed when the plant reached at knee height 50 kg/ha basis. Weeding was conducted at least three times during the growing period in each of the test sites depending on the level of weed infection. The following data were collected and analyzed to identify stable and superior genotypes compared the standard check variety.

Days to 50% flowering: The time between days to emergence to 50% of the plants in a plot reached half-bloom stage.

Plant height (cm): The length from the base of the plant to the tip of the panicle.

Grain yield per plot (GY): Grain yield in kilogram of plants from the three rows and adjusted to 10% moisture level and converted to kg ha-1.

Days to 90% physiological maturity (DTM): The number of days from emergence to the stage when 90% of the plants in a plot reached at physiological maturity, i.e., the stage at which when the panicle lose their pigmentation and begin to dry.

Plant aspect (PAS): Over all agronomic desirability score (drought tolerance, earliness, head exertion and compactness, grain size and shape, thresh ability, disease and insect resistance, etc) was measured using 1-5 score where 1=excellent and 5=poor; data scored by the same individual across sites and year were preferably used for this analysis.

Analysis of variance (ANOVA)

The collected grain yield and related data was subjected to Mixed model analysis of variances (ANOVA). R software version 3.4.0 was used.

The locations were considered as random and genotypes as fixed effects. The sources variation were partitioned into blocks, treatments and error terms. The treatment effect further decomposed into three components: G, E and GEI effects. According to Ding et al. [12] the following linear model was used for this experiment.

Yijr=μ+αi+βj+αβij+bj+εijr …………….(1)

Where yijr, is the average value of the dependent variable of genotype i in environment j and block r, μ is a grand mean, αi is the effect of the ith genotype. βj, is the effect of the jth environment, αβij is the effect of the ith genotype by the jth environment, bj is the block effect at the jth environment and ijr ε, is the residual error.

Stability analysis: In this study the following well-known analysis model were used for the stability of analysis.

Eberhart and Russell’s joint regression model: According to Eberhart and Russell [13] model the use of joint linear regression where the yield of each genotype is regressed on the environmental index. The behavior of the genotype was determined by the model:


Where: Yij=the mean performance of the ith genotype in the jth environment, μi=the grand mean of the ith genotype over all the environments, βi=the regression coefficient which measures the response of the ith genotype on environmental index, Ij=the environmental index obtained by the difference between the mean of each environment and the grand mean and δij=the deviation from regression of ith variety in the jth environment.

The additive main effects and multiplicative interaction (AMMI) method: The following AMMI model equation were computed:


Where, Yij=the mean grain yield of genotype i in environment j, μ=the grand mean, αi=the deviation of the genotype mean from the grand mean, βj=the deviation of the environment mean from the grand mean, λn=the singular value for the IPCA n, N=the number of PCA axis retained in the model, γin=the PCA score of a genotype for PCA axis n, δjn=the environmental PCA score for PCA axis n, θij=the AMMI residual and Eij=the residuals. According to Zobel et al. [11] the degrees of freedom (DF) for the IPCA axis were calculated as DF=G+E-1-2n; Where: G=the number of genotypes, E=the number of environments and n=the nth axis of IPCA.

GGE Biplot Model: GGE biplot were used to have a complete and visual estimation of all features of the data by forming a biplot that simultaneously represents both mean performance and stability, optimized environments for specific genotypes and identifies megaenvironments. According to Yan [14] GGE analysis is partitioned of G +GE into principal components through singular value decomposition of environmentally centered yield data. The following researchers like Tesfaye et al. [15] and Gasura et al. [16] have been used GGE biplot to conduct multi-environment trials and in their finding, they found that the existence of a good testing environment for selecting widely adaptable and high yielding cultivar.

Results and Discussion

Data collected on 21 sorghum genotypes developed by the Ethiopian institute of agricultural research, National sorghum research program including one popular check variety were subjected to analysis of variance (ANOVA) for key characters in order to check the presence of significant difference among genotypes and the results are presented in Tables 3-5. Genotypes and check varieties were tested across seven environments for grain yield and other related traits.

Source of Variation  Mean Square
Genotype (Gen) 21 2435941** 2299.4** 59.92** 1.5727**
Environment (Env) 6 85824967** 9769.5** 869.5** 20.3579**
Gen × Env 126 1128764** 303.4** 19.84** 0.5718**
Env × Block 14 1969891** 558.8* 61.5** 0.2132NS
Residuals 294 236943447 274.7 13.74 0.1501

Table 3: Mean squares of combined analysis of variance (ANOVA) of sorghum genotypes evaluated at Miesso, Erer, Kobo, Shoarobite during the main cropping seasons of 2012 to 2014.

S No Genotype Pedigree Erer Shewarobit Kobo Mieso Mean
2013 2014 2013 2014 2014 2012 2014
1 2001MS7003 LocalBulk(white)/SRN-39 3028 2836 4432 4898 3317 1961 2306 3254
2 2001MS7013 PGRC/E#222880/ICSV-1/KAT369-1 2904 3071 4633 4906 3802 1264 1647 3175
3 2001MS7015 PGRC/E#222880/ICSV-1/KAT369-1 3658 2933 6282 6004 2963 2199 2399 3777
4 2001MS7037 PGRC/E#222878/ICSV708 2924 2169 3810 4148 1866 1194 1558 2524
5 IESV92084-DL IESV92084-DL 3197 3230 6358 6821 2768 2087 2577 3863
6 IESV92168-DL IESV92168-DL 2849 2148 5865 6128 1849 1972 2398 3316
7 IESV92199-DL IESV92199-DL 2785 3277 4091 3481 2382 2547 2360 2989
8 IESV92057-DL IESV92057-DL 2993 3378 3940 4615 1977 3529 2477 3273
9 IESV9027-DL IESV9027-DL 3146 1973 5230 4521 3061 2779 1886 3228
10 2001MS7007 CR:35:5/DJ1195/N13 3690 3117 6157 5739 2419 2095 2633 3693
11 2005MI5060 WSV387/P9403 3040 3647 4769 4745 2497 2345 2582 3375
12 2005MI5064 WSV387/P9404 2630 3520 5717 5180 3669 3416 2345 3782
13 2005MI5065 WSV387/P9405 3258 3381 4839 6060 3135 2631 2310 3659
14 2005MI5066 M36121/P9401 2997 3313 4606 4913 2514 2224 2522 3298
15 2005MI5069 M36121/P9402 2489 2581 4612 4452 2757 3750 2583 3318
16 2005MI5070 M36121/P9403 2625 4047 3488 5893 1423 2421 2052 3136
17 2005MI5075 3443-2-OP/P9401 3656 3257 4104 4832 2766 2019 1524 3165
18 2005MI5079 3443-2-OP/P9401 2370 2756 4548 4300 2845 1979 1904 2957
19 2005MI5081 3443-2-OP/P9403 3508 2424 3791 3678 2113 2093 2676 2898
20 2005MI5082 3443-2-OP/P9403 2967 4000 4702 4680 1358 1828 2463 3143
21 ICSR24005 ICSR24005 2336 2785 4558 4233 2137 1551 2681 2898
22 Melkam WSV387 2744 1911 5339 5496 3356 2939 3212 3571
 Mean 2991 2989 4812 4987 2590 2310 2322 3286
 CV (%) 16 13 13 8 14 14 11  
LSD 765 638 998 694 592 547 403  

Table 4: The combined and mean grain yield (Kg/ha) of early maturing sorghum varieties evaluated from 2012 to 2014 in four different sites.

S No Genotype Pedigree Erer Shoarobit Kobo Mieso Mean
2013 2014 2013 2014 2014 2012 2014
1 2001MS7003 LocalBulk(white)/SRN-39 154 174 195 181 187 195 165 179
2 2001MS7013 PGRC/E#222880/ICSV-1/KAT369-1 138 164 186 189 176 198 170 174
3 2001MS7015 PGRC/E#222880/ICSV-1/KAT369-1 142 170 180 182 181 192 187 176
4 2001MS7037 PGRC/E#222878/ICSV708 165 176 215 179 196 193 190 188
5 IESV92084-DL IESV92084-DL 149 182 184 192 160 192 185 178
6 IESV92168-DL IESV92168-DL 152 193 215 206 182 217 198 195
7 IESV92199-DL IESV92199-DL 160 169 207 179 173 193 193 182
8 IESV92057-DL IESV92057-DL 148 189 192 188 178 210 185 184
9 IESV9027-DL IESV9027-DL 137 160 194 187 170 198 170 174
10 2001MS7007 CR:35:5/DJ1195/N13 155 168 204 170 186 197 170 178
11 2005MI5060 WSV387/P9403 160 194 186 168 196 183 207 185
12 2005MI5064 WSV387/P9404 178 187 209 179 184 200 202 191
13 2005MI5065 WSV387/P9405 195 208 214 197 187 218 200 203
14 2005MI5066 M36121/P9401 180 180 219 198 187 197 198 194
15 2005MI5069 M36121/P9402 164 185 182 178 188 217 195 187
16 2005MI5070 M36121/P9403 167 183 205 189 184 198 195 189
17 2005MI5075 3443-2-OP/P9401 214 193 233 185 197 242 193 208
18 2005MI5079 3443-2-OP/P9401 137 182 167 183 187 170 187 173
19 2005MI5081 3443-2-OP/P9403 152 174 184 175 178 180 188 176
20 2005MI5082 3443-2-OP/P9403 140 183 188 179 180 178 188 177
21 ICSR24005 ICSR24005 159 183 184 188 181 208 192 185
22 Melkam WSV387 117 144 184 174 150 188 170 161
Mean 157 179 197 184 181 198 188 183
CV (%) 8 8 13 7 8 10 9  
LSD 27 23 ns ns 23 34 ns  

Table 5: Combined and mean plant height of the test genotypes evaluated in each of the test sites from 2012 to 2014.

Analysis of variance across test environment

Combined analysis of variance (ANOVA) across locations for grain yield (GY), days to flowering (DTF), plant height (PTH) and overall agronomic desirability (PAS) revealed highly significant variability among the genotypes with significant environment, and genotype by environment interaction effect (Table 3). The high mean square values for locations across the test environments indicating that the test genotypes responded differently across environments. Hence, decision on selecting genotypes need to consider the target environments. Similar findings have been reported by Assefaw [17] and Gasura et al. [16] on multi-environment evaluation of sorghum genotypes.

Mean performance of advanced early maturing sorghum varieties

The mean performance of the 22 genotypes including one standard check (Melkam) for grain yield, days to flowering, plant height, days to maturity and overall agronomic desirability performance at dry lowland representative testing sites (Miesso, Erer, Shoarobite and Kobo) are given in Table 4. In the current research finding high significant effect of genotypes for Grain yield (kg/ha), Plant height (cm) and Days to flowering, Days to maturity and overall agronomic desirability was observed which is in agreement with the findings of many researchers [18-23]. The mean grain yield of the test genotypes across environment was 3286 kg/ha with the range between 2524 to 3863 kg/ha. In comparison to the improved standard check variety (Melkam) six of the tested varieties gave the highest mean grain yield advantage ranged between 2 to 8% (Table 4).

The variety IESV92084-DL gave the highest mean grain yield (3863 kg/ha) across all the environments. The highest mean grain yield performance was obtained at Shoarobit which 4987 kg/ha for 2014 and 4812 kg/ha for 2013 crop season, while the lowest mean grain yield (2310 kg/ha) was obtained in 2012 at Mieso. This showed the seasonal variability which has been affected the variation in genetic performance of the test genotypes.

Based on over all agronomic desirability score (PAS) two promising genotypes 2005MI5064 and 2005MI5065 with the pedigree of WSV387XP9404 and WSV387XP9405 respectively have been identified for verification trial and these genotypes super imposed on three on station and 6 on farm and later, after national variety release committee (NVRC) evaluated their field performance the variety 2005MI5064 has been released for commercial production. The released variety 2005MI5064 is mainly preferred for tall plant height (which is the number one criterion of farmers in dry lowland sorghum growing areas), excellent head exertion (the line has an exertion of about 7 cm between ligule of flag leaf to head base) compared to head base almost entirely covered by leaf sheath in the recent released varieties (Dekeba and Melkam), semi compact head, high and stable yield.

The released variety 2005MI5065 (WSV387XP9405) is preferred mainly for tall plant height having a plant height advantage of 23.7%, head exertion and compactness but this candidate genotype was not accepted by the national variety release committee. Generally, the released candidate variety 2005MI5064 is preferred mainly for tall plant height having a plant height (190.79 cm) advantage of 16.7%, head exertion and compactness but it also has grain yield (3810.96 kg/ha) advantage of 7% over the standard check Melkam.

As plant biomass is an important attribute to sorghum growing farmers, the test genotypes were evaluated for plant height. Plant height had the highest heritability across environments (87%) indicating the repeatability of the genotype’s performance in plant height across the testing sites. The mean plant height across environment was 183 cm with the range of 161 to 208 cm (Tables 4 and 6). All the test genotypes had the highest mean plant height, however, two of the test varieties (2005MI5075 and 2005MI5065) had mean plant height greater than 200 cm. As a measure of adaptation to the target environment the test genotypes were compared for flowering time with the check variety. Hence, seven of the tested genotypes had similarity in flowering time with the check variety Melkam. This result indicating that these varieties could adapted to the environment where can adapt and perform.

2001MS7003 77.88 78.2 121.24 121.22 179.77 179.36 3265.73 3253.97 3.28 3.29
2001MS7007 76.22 75.73 121.9 122.11 178.95 178.41 3498.66 3688.04 3.56 3.71
2001MS7013 78.24 78.73 121.81 122 174.99 173.85 3223.46 3175.21 3.39 3.46
2001MS7015 77.93 78.27 123.05 123.67 174.64 173.45 3546.24 3776.71 3.25 3.25
2001MS7037 79.59 80.73 123.37 124.11 187.15 187.84 2874.08 2524.13 3.53 3.67
2005MI5060 78.06 78.47 121.07 121 182.42 182.4 3330.77 3375.18 3.28 3.29
2005MI5064 77.88 78.2 121.32 121.33 189.71 190.79 3564.62 3810.96 2.79 2.58
2005MI5065 75.99 75.4 121.57 121.67 200.9 203.66 3426.91 3554.34 2.73 2.5
2005MI5066 77.61 77.8 120.5 120.22 193.27 194.88 3325.3 3364.99 2.9 2.75
2005MI5069 77.61 77.8 121.32 121.33 184.46 184.75 3299.94 3317.73 2.84 2.67
2005MI5070 77.88 78.2 122.96 123.56 186.65 187.27 3202.27 3135.73 3.19 3.17
2005MI5075 78.24 78.73 123.21 123.89 204.21 207.46 3215.67 3160.69 3.39 3.46
2005MI5079 75.05 74 119.84 119.33 173.91 172.61 3121.91 2985.97 3.59 3.75
2005MI5081 79.73 80.93 122.06 122.33 175.46 174.41 3059.21 2869.13 3.28 3.29
2005MI5082 78.65 79.33 123.13 123.78 177.09 176.27 3205.99 3142.64 3.42 3.5
ICSR24005 76.22 75.73 120.99 120.89 183.47 183.6 3074.48 2897.58 3.73 3.96
IESV9027-DL 74.83 73.67 118.94 118.11 175.45 174.38 3264.57 3251.81 3.25 3.25
IESV92057-DL 74.96 73.87 118.53 117.56 183.92 184.12 3265.55 3253.65 3.45 3.54
IESV92084-DL 77.43 77.53 120.25 119.89 176.62 175.74 3592.29 3862.54 3.05 2.96
IESV92168-DL 76.17 75.67 119.84 119.33 189.95 191.06 3301.39 3320.42 3.33 3.38
IESV92199-DL 77.57 77.73 121.81 122 179.71 179.28 3050.59 2853.06 3.13 3.08
Melkam 75.18 74.2 119.51 118.89 163.15 160.24 3435.84 3570.98 2.9 2.75

Table 6: Combined Meta-R (Multi Environment trial analysis) of sorghum genotypes evaluated across seven environment.

Combined ANOVA determines if GEI is a significant source of variation or not and estimates it but does not provide insight into the patterns of genotypes or environments that give rise to the interaction [24]. Therefore, the combined data was also analyzed using AMMI model that further partitions GEI into interaction principal component axis (IPCA) components. Hence, the AMMI model analysis had partitioned the GEI into the first two significant IPCAs. AMMI ANOVA given in Table 7.

Source of variation SS DF MS % of sum of square
Environment 514949802.3 6 85824967.05** 72.7
Genotype 51154766.16 21 2435941.25** 7.22
Interaction 142224244 126 1128763.85** 20.08
IPC1 48723428.72 26 1873978.03** 34.3
IPC2 37083643.86 24 1545151.83** 26.1
IPC3 22908132.9 22 1041278.77ns 16.11
IPC4 18189841.35 20 909492.07ns 12.79
IPC5 8621549.17 18 478974.95ns 6.06
IPC6 6697648 16 418603.00ns 4.71
Residuals 264521915.1 308 858837.39 -

Table 7: AMMI model analysis of variance for grain yield of sorghum genotypes in seven environments of four locations.

Additionally, Eberhart and Russel regression model used to see the stability performance of the tested genotypes (Table 8).

Genotypes Mean Sd Bi S2di R2
2001MS7003 3253.97 1071.59 0.89 -125205.68 0.9
2001MS7007 3688.04 1631.71 1.4 -139595.78 0.96
2001MS7013 3175.21 1389.89 1.05 328825.84 0.74
2001MS7015 3776.71 1683.87 1.46 -184471.93 0.98
2001MS7037 2524.13 1133.71 0.95 -137321.8 0.91
2005MI5060 3375.18 1038.17 0.88 -177247.55 0.93
2005MI5064 3810.96 1237.84 0.95 0.16899 0.76
2005MI5065 3554.34 1407.08 1.19 -0.87684 0.92
2005MI5066 3364.99 1038.37 0.89 -212348.91 0.96
2005MI5069 3317.73 933.97 0.65 115588.72 0.63
2005MI5070 3135.73 1496.56 1.03 776775.09 0.61
2005MI5075 3160.69 1159.68 0.91 58601.15 0.8
2005MI5079 2985.97 1063.47 0.88 -120758.17 0.89
2005MI5081 2869.13 781.47 0.56 -22779.32 0.66
2005MI5082 3142.64 1349.84 1.02 303822.83 0.74
ICSR24005 2897.58 1103.66 0.91 -86996.54 0.88
IESV9027-DL 3251.81 1244.11 0.96 141069.9 0.78
IESV92057-DL 3253.65 889.63 0.61 91525.29 0.62
IESV92084-DL 3862.54 1907.27 1.67 -232439.09 0.99
IESV92168-DL 3320.42 1868.99 1.6 -74349.5 0.95
IESV92199-DL 2853.06 679.83 0.54 -175548.31 0.83
Melkam 3570.98 1344.39 1.01 299244.77 0.74

Table 8: Regression coefficient (bi) and standard deviation from linearity of regression (S2di) for the test genotypes using Eberhart and Russell model.

Stability analysis across the testing sites

The combined analysis showed the significant genotype by environment interaction, which would have effect in the ranking of genotypes in each of the testing sites. AMMI model has been used to further partitions GEI into interaction principal component axis (IPCA) components and assess stability of the test genotype across locations. The AMMI analysis of variance for the (additive main effect) showed significant difference (p ≤ 0.01) for the testing locations, genotypes and genotype by location interaction (Table 7).

The result showed that the environment accounted the maximum sum of square 72.70% followed by the genotype by environment interaction sum of square (20.08%) and the genotype sum of square was found to be the lowest (7.22%). The larger proportion of variance caused by the significant variation of the test environment in soil characteristics and availability of moisture. The first principal component (IPCA1) captured 34.3% of the total variation while the second principal component explained 26.1% of the variation, which add up to 60.4 of the total variation. The most reliable model for AMMI can be estimated using the first two IPCAs and illustrated that most of the interaction occurs in the first few axes.

The graphical presentation of grain yield plotted against the IPC 1 grouped the test environments and genotypes based on the effect to the environment interaction (Figure 1). From the seven test environments the trial conducted at Sherarao (SR) on both 2013 and 2014 were distinctly grouped while the other five environments closely grouped together (Figure 1). This study have shown that the closeness between pairs of locations or pairs of genotypes in the biplot is proportional to their similarity for genotype by location interaction effects. This suggested that environments that have similar genotype by environment interaction effect can be combined to identify genotypes suitable to the targeted environment.


Figure 1: AMMI biplot for grain yield of 22 sorghum genotypes evaluated in seven environments. GYLD (Grain yield)- Y-axis plotted against the first IPCA1; The roman number represented genotype number and environments designated by MS12=Miesso in 2012, ER13=Erer in 2013,SR13=Shoaarobit in 2013, MS14=Miesso in 2014, KB14=Kobo in 2014, ER14=Erer in 2014 and SR14= Shoaarobit in 2014.

The interaction principal component 1 (IPCA1) plotted in the x-axis and the interaction principal component two (IPCA2) plotted in the yaxis (Figure 2) showed that the first Interaction Principal Component (IPC1) explained 34.3% while the second interaction principal component explained 26.1% the two interaction principal components with a cumulative effect of 64.4% of the genotype by environment interaction effect.


Figure 2: AMMI biplot for grain yield of 22 sorghum genotypes evaluated in seven environments. This Biplot showing genotypes grain yield stability and preferential adaptation environment using the first two IPCAs. The roman number represented tested genotype and environments designated by MS12=Miesso in 2012, ER13=Erer in 2013, SR13=Shoarobite in 2013, MS14=Miesso in 2014, KB14= Kobo in 2014, ER14=Erer in 2014 and SR14=in 2014.

Genotypes plotted to the center are considered to be stable across the test environments. Hence, genotypes 2001MS7003, 2001MS7037 and ICSR24005 were found to be the most stable across environments. In addition, the best performing sorghum genotypes across environments (2005MI5064 and 2005MI5065) were also had better stability.

Analysis based on Eberhart and Russel regression model

The Ebrehart and Russell [13] analysis model revealed that two of the test genotypes (2005MI5064 and 2005MI5065) that had better grain yield performance from the standard check (3811 and 3554) respectively, and they had regression coefficients approaching one (0.9476 and 1.186) in the same order and acceptable deviation from regression (-0.168, and -0.876) respectively, suggesting that these genotypes are stable and widely adaptable than the other genotypes (Table 7). An ideal genotype has the highest average grain yield, a regression coefficient (bi) value of approximately one and a mean square deviation from regression (s2di) value close to zero [13,25,26].


Combined analysis of variance displayed highly significant GXE interaction, which is a major challenge in the course of variety development and release activities. Besides, variations were significant among the test sites. Selecting genotypes in diversified testing locations and assessing yield stability of early maturing sorghum genotypes is quite vital. The AMMI analysis for the additive main effect and multiplicative interaction effect reveled significant difference for Genotype, testing location and genotype by testing location interaction. The first interaction principal component (IPCA 1) captured major part of the interaction 34.3 and the second interaction principal component explained additional 26.1 and both together explained 60.4% of the genotype by environment.

Based on the all agronomic desirability score two promising candidate genotypes 2005MI5064 and 2005MI5065 with the pedigree of WSV387XP9404 and WSV387XP9405 respectively have been identified for verification trial and these genotypes super imposed on three on station and 6 on farm and then their performance evaluated by the national variety release committee (NVRC) and the first candidate genotype called 2005MI5064 verified in 2016 for commercial production for dry lowland sorghum growing areas.

The candidate variety 2005MI5065 (WSV387XP9405) is preferred mainly for tall plant height having a plant height advantage of 23.7%, head exertion and compactness but this candidate genotype was not accepted by the national variety release committee. The variety 2005MI5064 is preferred mainly for tall plant height having a plant height (190.79 cm) advantage of 16.7%, head exertion and compactness but it also has grain yield (3810.96 kg/ha) advantage of 7% over the standard check Melkam.


Citation: Seyoum A, Gebreyohannes A, Nega A, Nida H, Tadesse T, et al. (2019) Performance Evaluation of Sorghum (Sorghum bicolor (L.) Moench) Genotypes for Grain Yield and Yield Related Traits in Drought Prone Areas of Ethiopia. Adv Crop Sci Tech 7: 423.

Copyright: © 2019 Seyoum A, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.