Physico-chemical and sensory evaluation of virgin olive oils from several Algerian olive-growing regions

– Olive cultivar diversity is rich in Algeria but most remain unexplored in terms of quality traits. This work aimed to evaluate the physicochemical and organoleptic quality of twenty olive oil samples belonging to four Algerian cultivars ( Chemlal , Sigoise , Ronde de Miliana and Rougette de Mitidja ) collected throughout the national territory. Physical-chemical and sensory results showed that 60% of the oils belong to the extra virgin category, while 40% were classi ﬁ ed as “ virgin olive oil ” . The results of the principal component analysis (PCA) revealed a great variability in fatty acids composition between the samples depending on the cultivar and origin. Oleic acid was the most abundant and varied between 64.84 and 80.14%. Extra virgin olive oils with quality attributes are eligible for a label. Rougette de Mitidja , Ronde de Miliana and Sigoise from Oran showed great potential. Résumé – Évaluation physico-chimique d ’ olive régions


Introduction
The olive tree (Olea europaea L.) is one of the most prominent fruit trees in Mediterranean countries, accounting for 97.9% of the total area of olive trees in the world (Rallo et al., 2018). More than 11 million ha of olive trees are grown in more than 47 countries, mainly in Spain (25%), Tunisia (13%), Italy (11%), Morocco (10%) and Greece (9%) (FAOSTAT, 2018). Algeria is one of the main producers of olives; olive trees have been ranked as the first fruit tree (Algerian Ministry of Agriculture). The area covered by olive trees has increased from 310 644 ha in 2010 (M. A.D.R, 2010) to 500 000 ha, in 2018 (M.A.D.R.P, 2018). The olive oil sector is increasingly considered as economic and social development of diverse regions, including Kabylia and Oranía. During the 2017/2018 campaign, Algeria produced 80 000 tons, ranking at the ninth position worldwide.
Virgin olive oil (VOO) extracted from fresh and undamaged fruits (Olea europaea L.), and appropriately processed, is a staple of the Mediterranean diet (Bedbabis et al., 2010). Olive oil is widely appreciated by consumers (Caporale et al., 2006) for its sensory properties (Angerosa et al., 2004;Bendini et al., 2007), nutritional and health benefits (Visioli and Galli, 1998). Several studies have highlighted the effectiveness of the bioactive compounds from olive oil in curing and prevention various chronic diseases, such as atherosclerosis, diabetes, cancer, neurodegenerative, and coronary heart diseases (Lou-Bonafonte et al., 2012;Visioli et al., 2018).
Olive oil quality is referred to international standards defined by several organizations (European Commission Regulation, 2016;Codex Alimentarius, 2017;IOC, 2018c;Conte et al., 2020). Olive oil quality assessment was performed by physicochemical parameters, including free acidity content, peroxide value, specific UV spectrophotometric absorptions (K 232 and K 270 ), and sensory evaluation (Gharbi et al., 2015).
The olive oil quality, nutritional and sensory characteristics are associated with the chemical composition, which is the result of a complex interaction between several environmental, agronomic, and technological factors (El Riachy et al., 2018). The cultivar, growing area, environmental conditions, soil, tree age, irrigation, fruit ripening, harvest time, fruit storage and processing system represent the important factors influencing the composition of olive oil (Gharbi et al., 2015;Ben Rached et al., 2017;Mele et al., 2018;Faci et al., 2021).
Algeria has an important potential to develop olive oil production, which could place it among the main producing and exporting countries. The quality of olive oil has not been consistently a priority, with only 15% of the quantity produced belonging to the category of virgin and extra virgin oils. Currently, there is a renewed interest in the production of quality olive oil.
The enhancement and improvement of olive oil require the study of its quality and chemical composition which must comply with international standards. This work has aimed to assess the quality of Algerian olive oil according to the cultivar and area of cultivation to valorize the best ones.

Plant material and sampling
The present work was carried on four cultivars of the local population, namely Chemlal, Sigoise, Ronde de Miliana and Rougette de Mitidja from different olive-growing regions of the country. Chemlal is the most widely grown cultivar in Algeria; it represents 40% of the Algerian olive grove. The fruits are small (about 1.7 g) and used for oil production; the oil yield varies between 14 and 22%. Sigoise occupies 25% of the Algerian olive grove and is located in the west plains. This cultivar is mainly used for the production of table olives which is characterized by medium-sized fruits (3.5 g). Ronde de Miliana occupies a limited olive-growing area. It is a dual-use cultivar (table olive and olive oil). The oil yield is estimated between 16 and 20%. Rougette de Mitidja is a cultivar intended for the production of oil with a yield varying from 18-20%.
The olive oil samples (twenty) produced during the 2018-2019 olive-growing season was collected from various producers during the months of November-December 2018.
Origin of the plant material and some agronomical conditions of each geographical area were shown in Table 1 and Figure 1. The olive trees have not undergone any chemical treatment.
Extraction of olive oil was performed immediately after harvest. The quantity of olives used for extraction is equal to or greater than 300 kg. The modern (automatic) three-phase system with a hammer crusher, malaxer and centrifugation of the olive paste is used for the extraction of the oil. All samples were packed in amber glass bottles without headspace and stored at 4°C until analysis. Different analyses were performed during the first ten of the month of January 2019 for all the samples.

Sensory analysis
A sensory analysis (median of defects, median of fruitiness, and panel classification test) of the samples was performed by 10 selected and trained panelists (IOC certified panel leader). The profile sheet was used according to the IOC Mario Solinas competition for extra virgin olive oils (T.30/ Doc. No.21, March 2018). The strengths of the positive (fruity, bitter, and pungent) and negative (fusty, winy, musty, muddy, rancid, metallic, and others) attributes were evaluated for each oil sample on a 10 scale. The sensorial evaluation results were estimated by the median, being valid when the coefficient of variation was less than 20 (IOC/T.20/Doc. No.15/Rev.10, 2018).

Colorimetric determination of total phenols
The total phenol content of the olive oil samples was determined by the Folin-Ciocalteu spectrophotometric method at 725 nm, using a gallic acid calibration curve (Gutfinger, 1981).

Fatty acid composition
The fatty acid composition of the oils was determined by gas chromatography (GC) as fatty acid methyl esters (FAMEs). FAMEs were prepared by cold transesterification with methanolic solution of potassium hydroxide (2N) according the method of European Commission implementing Regulation 2015/1833 amending Regulation (EEC) No. 2568/91. The solution containing the fatty acid methyl esters was injected in  , (Cyanopropyl at 50%) (USA) was used. The column temperature was isothermal at 200°C and the injector and detector temperatures were 250°C. Nitrogen as a carrier gas was used at flow rates of 1 mL/min. Fatty acids were identified by comparison of their retention times with those of standard reference compounds. Analyses were performed in duplicate.

Statistical analysis
All analyses were performed in triplicate, except for fatty acid composition, which is performed in duplicate; the results are presented as mean values. A simple analysis of variance was conducted according to the LSD test at the 5% threshold with GenStat Discovery Edition software. P ≥ 0.05 indicates that the effect is considered non-significant.
The correlation between the different analytical parameters is treated by principal component analysis (PCA), to highlight associations of individuals or links between variables. The data were statistically processed by the Statistica software version 12.5.

Oil content
The oil yield of the olives from the different samples is shown in Table 1. There are differences between the oil contents of the different cultivars. Chemlal from Djelfa (C5) has the lowest percentage (14%) and Sigoise from Oran (S3) has the highest percentage (22%). Al-Ruqaie et al. (2016) and Giuffrè (2017) showed that the cultivar significantly influences the oil content when samples come from the same harvest area.
The geographic origin of the samples also affects the oil content. For the same Chemlal cultivar, the yield varies from 14% for the sample from Djelfa (C5) to 22% for the sample from Bouira (C6). The same observation is noted for the cultivar Sigoise, whose yield varies from 15% (S1 from Blida) to 22% (S3 from Oran). The results are in agreement with those of Navas-López et al. (2020) who noted that genotype, environment and their interaction significantly affect the amount of oil.

Quality parameters
The parameters of the studied olive oils (Tab. 2) were within the limits of the European Commission implementing Regulation 2016/1227 amending Regulation (EEC) No. 2568/91 and IOC (2018b) and therefore the oils could be classified into the category of extra virgin olive oil, except samples of Chemlal from Tizi Ouzou (C2a and C2b), Sigoise from Ain-Temouchent (S2) and Djelfa (S4).
The region and cultivar had a highly significant effect (P 0.001) on the free acidity, the peroxide index, the K 270 , the water content, and significant effect (P 0.01) on the K 232 . Large variations in qualitative indexes according to the cultivar have shown by Piscopo et al., 2016). Significant differences were observed in most of the quality parameters between the extra virgin olive oil of "Sarıulak" olive cultivar produced in three different locations (Antalya, Karaman, Mersin) in the south region of Turkey (Arslan et al., 2013). The geographic area and environmental factors influence the characteristics of Arbequina oil (Borges et al., 2017).

Physical indices
The physical characteristics of all samples are presented in Table 2. The refractive index increases with the degree of unsaturation of the fatty acids. Indeed samples with higher unsaturation showed a higher refractive index. Our results are within the limits [1.468-1.470] set by Codex Alimentarius (2017) standards for extra virgin oils.
Density (MV) depends on the chemical composition of the oil and temperature. Therefore, our results are in the range [0.910-0.916] of Codex Alimentarius (2017). A highly significant difference between the different samples studied (P 0.001) according to refractive index and density.

Sensory analysis
The flavor and taste of olive oil are strongly correlated with its quality level and nutritional value. Sensory analysis of olive oils, based on median values of the positive and negative attributes carried out by a panel of trained tasters, is one of the most important evaluations used for virgin olive oil quality classification (Fernandes et al., 2018). First, sensorial features of olive oils were analyzed and the oils with defects were discarded. Twelve samples belonging to the extra virgin olive oil category were selected for sensory analyses (Tab. 3).
The results of the organoleptic analysis of the extra virgin oils, presented in Table 4, showed great diversity in its sensorial profiles. The variation in bitterness intensity ranging from 1 to 2.5 and pungency from 1 to 2.5. These attributes are scored on a scale of 1 to 3. The type of green fruitiness was the same for the different oils, but it was nuanced by the intensity which varies from 6 to 9.
The results (Tabs. 3 and 4) indicated that the organoleptic note varies from one cultivar to another; the oils of the Sigoise cultivar from Blida (S1) and Aïn-Temouchent (S2) were characterized by a fruity taste and a higher spicy median than other varieties.
The oil of Rougette de Mitidja cultivar from Blida (R1), with a very local diffusion, obtained a high score of 83.0, the best of all oils. It also showed a remarkable homogeneity as well as balanced attributes of intense fruity (9.0) and spicy (2.0) and bitter (2.0). The olive cultivar has been reported as one of the most crucial factors responsible for olive oil flavor variability (Tura et al., 2008;Mele et al., 2018). It is well established that the organoleptic properties of olive oil, which are strongly correlated to its geographical and varietal origins, are behind its wide commercialization and elevated market value (Rodney et al., 2010;Borges et al., 2017).
Samples of Chemlal cultivar, the most abundant olive cultivar in Algeria (40%), from different regions showed a remarkable difference in their sensory characteristics; it varies from ripe fruity with spicy and slightly bitter traits to medium green fruity with more pronounced fruit intensity, stronger spicy and bitter attributes. The same observation was noted for samples from the Sigoise cultivar where the oils have different sensory characteristics.
The cultivar factor has a very significant effect on sensory attributes. The aromatic notes depend on the volatile and phenolic compounds. The volatile composition of olive oil depends on several factors, such as the levels and activity of enzymes involved in different pathways (Angerosa, 2002). Results are expressed as mean ± SD of three sample replicates. Significance level: ***P < 0.001; **P < 0.01; *P < 0.05; NS = not significant. Free acidity: P < 0.001; Peroxide value: P < 0.001; K232: P < 0.05; K 270 : P < 0.001; H%: P < 0.001; Density: P < 0.001; n D 20 : P < 0.001; TP: P < 0.001. Fusty and musty oil S1 Medium green fruity R1 Intense green fruity C1 Ripe fruity C3 Ripe fruity C5 Medium green fruity Rm1 Medium green fruity S2 Intense green fruity SC1 Light green fruity SC2 Ripe fruity C7 Light green fruity C8 Medium green fruity C9 Medium green fruity

Total phenols
The quality of olive oil is significantly related to polyphenols content (Benlemlih and Ghanam, 2016). The total phenol content of the olive oil ranged from 46.29 to 351.45 mg/kg (Tab. 2). A high significant difference (P < 0.001) between the different samples from different regions and varieties was noted. These results are lower than those recorded on extra virgin olive oil of Algerian varieties (Douzane et al., 2013).
These low concentrations could be attributed to the olive oil extraction process in which most mills use the conventional three-phase centrifuge where a quantity of water has been added during the extraction process which has led to losses of phenolic compounds, vitamins, and aromatic components. It was reported that the two-phase decanter provides an oil rich in total polyphenols, ortho-diphenols, lower acidity, and better organoleptic quality compared to the three-phase press (Del Caro et al., 2006). Besides the cultivar, the degrees of ripening of the olives and the environmental conditions influence the concentration of phenolic compounds (Škevin et al., 2003). A negative correlation (r = À0.88) between drupe ripening degree and phenolic content of VOOs was found (Rotondi et al., 2004).
According to Borges et al. (2017), significant variations (P < 0.05) in the concentration of minor bioactive constituents were observed among oils from different growing areas, not only between Spanish and Brazilian samples but even within areas of the same country for Arbequina cultivar.

Fatty acid composition
The fatty acid composition is shown in Table 5. The most representative fatty acids are oleic, linoleic, palmitic, and stearic, which represent 98% of the composition of total fatty acids. Oleic acid was the most abundant and varied between 64.84 and 80.00%. The large-fruited varieties showed the highest amounts of oleic acid (Sigoise S1: 75%; Ronde de Miliana Rm: 73.88%; Rougette de Mitidja (R2: 80.14%) and lower levels in small-fruit varieties (Chemlal: 67%), a similar result as reported by Faci et al. (2021) who studied Chemlal of Tizi-Ouzou. These results corroborated with those obtained previously by Douzane (2012) for the same varieties: Sigoise (71.40%), Ronde de Miliana (80.18%) and Rougette de Mitidja (74.80%) collected from Olive production station (Sidi Aich, southern Bejaia). Lower percentages were noted for the Chemlal cultivar produced in several regions (Douzane, 2012). Algerian Chemlal cultivar presents a composition very close to the Tunisian cultivar "Chemlali" which contain low oleic acid and high concentrations of linoleic and palmitic acids (Manai et al., 2007).
Linoleic acid which was the dominant polyunsaturated fatty acid ranged from 6.80% to 11.70%, while linolenic acid (C18:3) was the minor and ranged from 0.56% to 0.77%. The levels of linoleic acid are inversely related to oleic acid. According to Sánchez Casas et al. (1999), for better preservation of oils' quality, their linoleic acid content shouldn't be upper than 10%. Lower levels of oxidative stability have been noted in oils containing high proportions of linoleic acid (Pardo et al., 2021). Also, Montaño et al. (2016) noted that oxidative stability was clearly correlated positively with oleic acid (R = 0.688) and negatively with linoleic acid (R = À0.710).
Palmitic acid, the major saturated fatty acids showed an important variation between samples, it ranged from 8.62% in Rougette de Mitidja cultivar (R2) to 17.57% in Chemlal cultivar (C5). All samples of Chemlal had high value (superior to 14%).The other varieties presented an intermediate value (13.24% for the Ronde de Miliana cultivar (Rm1), between 10.62 and 13.21% for the Sigoise from different orogins. Statistical analysis revealed a highly significant difference (P < 0.001) between the different samples of a different cultivar.
The values of fatty acids comply with the limits required by the European Commission implementing Regulation (2016/ 1227) and the International Olive Oil Council method (IOC, 2018a) on the characteristics of olive oils.
The percentages of saturated fatty acids (SFA), monounsaturated fatty acids (MUFA), and polyunsaturated fatty acids (PUFA) and the ratios of PUFA/SFA and oleic/ linoleic acids in all samples were also evaluated. It was observed that the oleic acid, the MUFA/ PUFA ratio, and the oleic/ linoleic ratio of the varieties with large medium fruits (Sigoise, Ronde de Miliana, and Rougette de Mitidja) are the highest and are clearly distinguished from the rates of "population cultivar" of small-fruit (Chemlal) (Tab. 5). Some results on the Chemlal cultivar are comparable to those observed by Louadj and Giuffrè (2010); however, our samples were characterized by a broad variation in the mentioned parameters above, due probably to the large geographical origins.
The results obtained in this work showed that the distribution of fatty acids is greatly influenced by olive cultivar. They agree those of others authors (Cicatelli et al., 2013;Fuentes et al., 2015;Cherfaoui et al., 2018). However, many researchers have shown that geographical location has a significant effect. Palmitic acid, oleic acid, linoleic acid, and wax content were found to be significantly affected by the growing region for some cultivars (Rodney et al., 2010). Noorali et al. (2014) noted that the fatty acid profiles could be used to separate olive samples according to their cultivar and growing area, respectively, the fatty acid profile was significantly affected by the growing area for almost all cultivars.

Multi dimensional analysis
The confrontation between the correlations circle of variables and the individual projections (Fig. 2) allowed for determining the characteristics of individuals according to their physicochemical characteristics, with the plane axis. Thus, axis 2 permitted us to distinguish that Chemlal cultivar has been regrouped for highlighting two subgroups, this is caused by the existence of subspecies within Chemlal cultivar to know as Chemlal of Tizi-Ouzou, precocious Chemlal of Tazmalt, Chemlal of Oued-Aissa, and Ali-Sharif's white Chemlal. The most abundant olive cultivar in Algeria, Chemlal, dominated Algerian olive orchards (40%), probably owing to its unique adaptation to various pedoclimatic conditions (Haddadi and Yakoub-Bougdal, 2010). Chemlal cultivar could be dispersed as a clone.
The medium-fruit cultivar Ronde de Miliana (Rm1) formed a single group, a sign of non-existence of a parental link with other varieties; the same results were noted previously by Douzane (2012). The large-fruited cultivar (Sigoise and Rougette de Mitidja) formed distinct groups. The heterogeneity of Sigoise could be connected to the existence of a mixture within the orchard of producers.
The individual projection allows us to distinguish the different groups (Fig. 2). This structure characterization was made according to the physicochemical analyzes.
The concept of quality in olives is complex, and numerous traits might be considered. Different definitions may apply according to the point of view and final goal of producers, traders, consumers, and/or nutritionists. The olive oil composition of cultivar results from a very complex multivariate interaction between the genotypic potential and the environmental, agronomic, and technological factors that characterize fruit growth and ripening as well as oil extraction and storage (Lavee and Wodner, 1991). Some olive varieties like Chemlal are diffused as a clone. No genetic variation exists, therefore between the individuals. All variations in the final product are consequently attributable to environmental and physical factors. The mechanisms underlying environmentally-caused variations in oil composition are for the most part unknown. The results of previous studies (Ben Temime et al., 2006;Borges et al., 2017) have highlighted the effect of cultivar-environment interaction in the expression of the quality of olive oil. The changes in oil composition of the Tunisian Chetoui cultivar according to the plantation origin were studied (Ben Temime et al., 2006). The results showed considerable variability in oil composition. The characteristics of the EVOO Arbequina from Brazil in comparison with Spanish Arbequina from different regions were recently studied (Borges et al., 2017). The findings reveal that geographic and climate aspects of producing areas may significantly influence the quality and composition of Arbequina olive oil.

Conclusion
The results showed that the cultivar plays an important role in the qualitative characteristics and sensory attributes of olive oils. The oil of Rougette de Mitidja cultivar from Mitidja (R1) showed the best sensory and physicochemical results. It also exhibited the best levels of oleic acid, qualifying it as the best extra virgin olive oil in Algeria for the 2018/2019 olive season. Ronde de Miliana (Rm1) also showed great potential.
Many other olive cultivars are unexplored in terms of quality traits, more effort should be made to assess them. On the other hand, olive growers must be encouraged to promote the local olive-growing heritage by cultivating varieties approved by the National Center for seeds and plants Control and Certification (CNCC).