Algeria is the largest country in Africa, covering more than two million square kilometers and 238 million ha but only 3.4% is arable land. Most agricultural activities are in the north of the country. The dominant crops are annual, and mainly field crops such as cereals, forages, food legumes and potatoes. Cereals are the predominant crops grown by Algerian farmers, covering  annually  3  to 3.5 million ha, nearly 40% of Algeria’s total agricultural land. However, the country depends on imports for 45% of its food because annual cereal production in Algeria is about 3.2 million tons (Benbelkacem, 1996) while the demand is around 8 million tons. During the last decade, wheat represents on average 67.1% of all cereal  production.  Yields  are low  due  to  several  abiotic  and biotic stresses. Scarcity and poor quality of underground water resources, low and erratic rainfall, drought recurrence, high and low temperatures and salinity are the key constraints to agricultural production.

 

Durum wheat (Triticum turgidum L. var. durum) is the traditional wheat crop in North Africa. It is used mainly for bread and couscous and freeke - made from immature green seed that is dried, grilled and broken and different sorts of traditional cookies (Abecassis et al., 1990; Desclaux et al., 2005; Kezih et al., 2013). In Algeria, durum wheat is sown annually over 1.2 million ha, with an average consumption of around 75 kg/person/year (Djermoun, 2009). Forty percent of the grain production is used for bread-making, mainly processed at the household level. The local genotypes of durum wheat, still grown by some farmers in the mountainous areas, are highly appreciated for their excellent bread-making and couscous quality. Improving durum wheat grain quality has become, in recent years, one of the main breeding goals in many Mediterranean countries, due to the increase in market demand for good quality durum wheat.

 

In Algeria, most of breeding efforts are concentrating on maximum yield potential under highly favorable environment but quality aspects have lower priority. A Multi-disciplinary approach has been adopted to promote newly selected varieties for different semi-arid targeted environments (durum and bread wheat, barley and triticale), this research resulted in cultivars able to withstand drought and heat but also responsive to improved moisture supply.

 

The released varieties during the late seventies have allowed a grain yield gain of at least 35% over the landraces, but their grain quality is highly affected by the environmental factors. The improvement of durum wheat quality to fit different end-uses (bread, couscous, pasta, local cakes, freeke) is an important objective of the research conducted by the National Agricultural Research Institute of Algeria (INRAA) and the collaborating International Centers, CIMMYT and ICARDA (Nachit et al., 1995).

 

The strong influence of environment and of genotype-environment interactions on grain quality hampered the improvement of durum wheat quality. Several approaches have been carried out on bread wheat to evaluate effects of genotype, environment and their interaction (Peterson et al., 1992; Annicchiarico et al., 2006). However, very little information is available on the relative importance of the effects of genotype, environment, and genotype-environment  interaction on the quality characteristics of durum wheat grown in the Mediterranean region. Michelena et al. (1995) studied the stability of 16 durum wheat varieties for different quality parameters under northern Spain conditions and found that spatial and temporal variation was very important for many traits. Juan-Aracil and Michelena  (1995) reported that the three order interaction latitude  x site x  year  was the main factor that affected quality parameters. In some other studies from Italy (Mariani et al., 1995; Nachit et al., 1995; Boggini et al., 1997; Novaro et al., 1997) a high influence of environment and genotype-environment interaction in determining durum wheat quality has been reported. This study was carried out in different wheat growing regions of Algeria that differ in climate, soil type and agronomic systems to overview the spatial variation in cereal quality traits of the four more widely cultivated durum wheat varieties conducted in three contrasting sites of Algeria during the 2010/2011. 

 

Statistical methods

 

In each trial, genotypes were sown in a randomized complete-block design with four replications. A one-way analysis of variance was done to compare more than two means from random independent samples (Dagnelie, 2003). This test was used to compare the different traits among varieties and sites. Mean comparisons between trials for each quality characteristic and a one way ANOVA were performed and least significant difference (LSD) values were calculated at the 5% probability level, groups were classified by the  Tukey’s test.

Growing conditions were different in each site; they were more favorable in the littoral area rather in the high plateau sites where it is more continental. According to Basso et al. (2012) the growing season vegetative rainfall occurs from December to February and the growing season reproductive rainfall occurs from March to May. Tiaret faced drought conditions with a total of 28.6 mm during booting stage, the most critical stage for spike constitution and early grain filling in this area; this drought period resulted in a reduced average yield (2.52 t/ha) due to water stress.  Guelma received the largest amount of reproductive rainfall with a total of 406.6 mm; from boot stage to physiological maturity it received 128.8 mm that made a quite good average grain yield (3.5 t/ha) superior to Tiaret site by one ton on average (Dagnelie, 2003; Basso et al., 2012; Juan-Aracil et al., 1995; Rizza, 2012). Rainfed conditions affected thus wheat yield and quality parameters; this has been reported by different authors in Mediterranean environments who showed that this effect is mainly due to weather conditions during grain filling and assimilates accumulation in the grain (Baenziger, 1985; Bahlouli et al, 2008; Diacono et al., 2012). The spatial variability of durum wheat varieties observed in all sites is probably mainly influenced by soil types (texture, bulk density and organic matter) (Mariani, 1995). During our trials, a similar relationship was expressed between rainfall with grain yield (RDT) and thousand kernel weight (PMG) at Setif, meaning that conditions favored the expression of those parameters simultaneously. As it is known, yield reduction observed at Tiaret was likely due to water stress but on the other hand it was associated with highest protein content measured by the near infra red (Rharrabti et al, 2003; Flagella, 2010; Akman et al, 2013; Mariani, 1995).

 

Yellow berry level was excellent (very low) in the high plateau sites. High yellow berry percent in the most humid areas on newly released genotypes was probably the result of the reduced uptake of water which has limited the N use efficiency of wheat varieties (Rharrabti, et al 2003; Desclaux, 2000). Yellow berry is generally related to low protein content (Fernandez and Conner, 2011), this was not the case in our study at Tiaret where this trait   was   lowest  for  yellow  berry  but  highest  for protein content. High percent of black point disease observed occurred as expected in wettest environment like in Guelma (Desclaux, 2000; Rharrabti et al., 2003; Fernandez and Conner, 2011; Akman et al., 2013).

Results deriving from different analysis (mean, anova, tukey test) show the high spatial variation existing between the different varieties for all quality parameters considered in this study. The differences were highly significant in all varieties for all traits except for grain yield at Guelma and Setif. This latter trait was lowest at Tiaret (2.5 Vs 4.5 t/ha at Setif). The newly released variety Beni Mestina (Lahn/Ch1 2003) out yielded all the other varieties in all sites. Humidity was in a good range at all sites averaging 12.8%. Protein content reached a high level (16.93%) at Tiaret site; the best overall level was on average in Waha with 14.3%.

 

It is important to notice that the best yielders did not have the best protein level meaning that the quantity of the product did not match with its quality. Our results showed the excellent results that appeared for protein content with almost no yellow berry in the grain and a high PMG in the high plateau semi arid sites implying clearly that rainfed conditions in the semi arid sites offer the best opportunity for production of good quality durum wheat. 

 

Under rainfed conditions, durum wheat cropping system in Algeria show sensitivity to spatial and climatic variability. Therefore, crops should be managed specifically according to the site where they are cultivated, especially by the use of techniques that improve protein content in their wheat varieties. Spatial analysis of yield and quality parameters allowed for the identification of Tiaret as the best site for grain quality because of its stable rainfall during grain filling period during our experiments. Varieties with shorter cycle and a good water use efficiency or drought tolerance released recently are better in littoral areas whereas varieties with long cycle are suitable in the high plateau sites. Our results suggest that, it is necessary to focus on each specific region with different environment types and select performing genotypes with the expression of best quality parameters for each variety in each site.

 

The authors have not declared any conflict of interest.