Search

The study was conducted with the aim of inventorying and assessing the level of sorghum landraces richness, estimating the extent and patterns of phenotypic diversity, identifying race type, and predicting suitable areas of production using future climate scenarios in Tigray, Northern Ethiopia. Overall, 358 landraces from 125 independent farmers’ fields in 20 districts belonging to four zones of Tigray were surveyed. Landraces richness, frequency of character states, and diversity levels were estimated via Margalef’s, Menhinick’s and Shannon–Weaver (H′) indices for each zone and altitude classes. Suitable areas for sorghum cultivation in the future were predicted using DIVA-GIS. Altogether, 140 distinctly named landraces were identified and collected. Southern zone has the highest richness (DMg = 10.74, DMn = 5.0) followed by central (DMg = 8.54, DMn = 3.80). The estimated H′ for the phenotypic character states ranged from 0.24 (seed form) to 0.95 (rachis branch length). Highest diversity estimate was found in northwest (H′ = 0.71) followed by central zone (H′ = 0.64). Respectively, H′ was 0.70, 0.68, and 0.61 in lowland, intermediate, and highland elevations. In general, lowlands of northwest, central, and southern zones are identified as potential sites for germplasm conservation and utilization. The races durra, bicolor, caudatum, and durra-bicolor were found with durra as the dominant race (79%). All the currently surveyed areas and many other sites in Tigray are predicted to be suitable for future sorghum production under the changing climate scenario. Copies of all the 358 sorghum landraces are deposited in the national gene bank for use in future sorghum breeding work. The wealth of sorghum landraces diversity could be used in sorghum improvement programs.

A set of 192 barley genotypes of wide eco-geographic origin were evaluated for flowering time (FT), the major adaptive trait in Ethiopian barleys. Morphological, functional and DArT markers were used to explore trait-marker associations, population structure, genetic diversity and extent of pair-wise linkage disequilibrium (LD). A wide magnitude of variation in FT was detected within and across environments and spanned at least 50–60 days with clear G × E interactions (GEI). The average DArT polymorphism estimated by gene diversity and polymorphic information content (PIC) was 0.33 and 0.27 respectively. Four distinct subpopulations, with one outlying group were identified which indicated that the populations were structured. The genome-wide LD extended up to 25 cM, with extensive variation at chromosome level. Markers associated with FT were found close to known QTLs for FT such as VRN-H1, VRN-H3, Ppd-H1, Ppd-H2 and earliness per se (Eps-2S), but also in regions known for FT-like and CO-like genes working under short days. In some QTLs the magnitudes and directions of the effects varied with environments. It is proposed that genes conditioning tolerance to drought and/or waterlogging interact with FT genes and modify the phenotype. A non-random association of alleles on different chromosomes—vrs1 (2H), BYDV resistance (3H) and srh (5H) associated with altitude and FT indicated a co-adapted gene complex. A set of 95 genotypes were selected to represent the Ethiopian section of the Barley Core Collection.

Tigray (region) is one of the major finger millet growing regions in Ethiopia and an important site from an archeobotanical point of view. Three zones of Tigray (east, central and west) were identified as representative sites in the region and a total of 14 districts/ ‘Woreda’ were surveyed. Thirty-seven landraces/farmers’ varieties of finger millet were identified/recorded. Farmers in Tigray undertake pre and post harvest selection in finger millet and sometimes they also select seeds from storage based on a number of attributes. Farmers maintain diversity as a way to ensure harvest security or stability of production, to promote diversity of diet and income sources, minimize crop failure risk, reduce insect and disease incidences and ensure efficient use of labour. The traditional management of finger millet in the entire study area is generally found to be demand driven, showing the existence of potential sites for on-farm conservation. The high morphological diversity (H =0.76 ± 0.09) found in the gene bank collections of Tigrayan origin also reveals the importance of linking ex situ with in situ conservation activities. Furthermore, the enhancement and conservation significance of the crop is discussed.

Twenty two accessions of barley landrace/farmers' varieties collected from Bale and North Shewa in situ conservation zones were characterised using 18 qualitative and quantitative morphological traits. Phenotypic frequencies for individual qualitative characters across in situ conservation zones, districts, and strategic sites (localities) have shown mixed and some peculiar patterns. Varieties from Bale conservation zone are predominantly white-yellow seeded whereas varieties from North Shewa are purple-black seeded. White-yellow seeded varieties are more frequent in the lower altitude examined, and the purple-black seeded varieties are more frequent at the highest altitude (≥ 2650 ). While six row types occur at higher frequencies in almost all the sites in North Shewa, the irregular types are dominant at Bale. The highest frequency of six row types is found at an altitude ≥ 2650 , whereas the two rowed and irregular types are found below 2650 . Two row types occur at low frequency at both sites. At Bale, glume awn shorter than glume is the most frequent character whereas glume awn longer than glume is more frequent at North Shewa. Glume awn longer than glume appears to increase with increasing altitude. Caryopsis cover is relatively monomorphic with respect to distribution across zones. The covered types are most frequent in almost all altitudinal ranges. Estimates of diversity index (H′) for individual qualitative characters suggest that polymorphism is common in varying degrees for most characters, which indicates the existence of wide range of variation. On the basis of cross validation using discriminant function among the quantitative characters, the landrace varieties from North Shewa seem to be more diverse than the ones from Bale. From the results of multiple regression analysis, the zonal variation is significantly associated with all the quantitative morpho-agronomic characters except plant height. The great majority of the varieties from Bale were perfectly identified and named by farmers. The reliability analysis confirmed that there was a remarkable positive degree of consistency between farmers naming of landrace varieties.