Biomass partition and productive aptitude of wild and cultivated cardoon genotypes (Cynara cardunculus L.) in a marginal land of Central Italy

Industrial Crops and Products 95 (2017) 191–201

Contents lists available at ScienceDirect

Industrial Crops and Products journal homepage: www.elsevier.com/locate/indcrop

Biomass partition and productive aptitude of wild and cultivated cardoon genotypes (Cynara cardunculus L.) in a marginal land of Central Italy Ulderico Neri, Bruno Pennelli, Giampiero Simonetti, Rosa Francaviglia ∗ Consiglio per la ricerca in agricoltura e l’analisi dell’economia agraria, Centro di Ricerca per lo Studio delle Relazioni tra Pianta e Suolo (CREA-RPS), Via della Navicella, 2-4, 00184 Roma, Italy

a r t i c l e

i n f o

Article history: Received 13 June 2016 Received in revised form 14 September 2016 Accepted 21 October 2016 Keywords: Cardoon Genotypes Yield Biomass partition Mediterranean crops

a b s t r a c t A very limited number of field studies have addressed the suitability of Cynara cardunculus L. genotypes to local environmental conditions in terms of productive aptitude. Four genotypes of Cardoon, under low input conditions (reduced nitrogen fertilization and rainfed conditions) were compared in a marginal land of Central Italy during the 2012–2014 period: two cultivated (CDL07 and Gigante) and two wild cardoon genotypes (RCT10 and Tolfa). At the ripening stage of capitula, all plant components were weighed separately, yields were calculated, and biometric traits were measured. Genotypes showed different productive aptitudes and suitability to the pedo-climatic conditions of growth. The total aboveground dry biomass per plant ranged between 114.4 and 353.6 g with variable values within genotypes and years, whereas the partition of the aboveground biomass was strongly affected by genotypes only. Analysis of variance (ANOVA) showed a general prevalent influence of the genotype factor on crop yields. Capitula components were of greatest importance in cultivated cardoons, but strongly affected by the annual climatic trends. The maximum achenes yield was in 2012 with 3.2 t ha−1 as a mean of the cultivated cardoons, whereas the wild genotype Tolfa reached a mean production of 2.1 t ha−1 . In general the wild genotype RCT10 demonstrated a poor adaptability to the conditions of the experimental site. On the other hand, wild genotype Tolfa might be incorporated into the local cropping systems as an industrial or bioenergy crop, due the low management inputs required, its adaptability to the local conditions, and the fairy good aboveground biomass and achenes production. © 2016 Elsevier B.V. All rights reserved.

1. Introduction Growing energy crops such as cardoon is a nontraditional land use option able to convert solar energy into stored biomass relatively efficiently, and leading to positive input/output energy balances for the overall system. In addition, under specific soil and climatic site conditions, bioenergy crops can be optimized since low input systems requiring limited nutrients and chemical inputs are needed (Sims et al., 2006). When combining production and sustainability, cardoon helps to reduce soil degradation due to the protection of its dense canopy against the erosion caused by the intense precipitations that occur in Mediterranean areas (Grammelis et al., 2008; Lag-Brotons et al., 2014).

∗ Corresponding author. E-mail address: [email protected] (R. Francaviglia). http://dx.doi.org/10.1016/j.indcrop.2016.10.029 0926-6690/© 2016 Elsevier B.V. All rights reserved.

Over the last few years a renewed and growing interest in the cardoon (Cynara cardunculus L.), an old plant with new uses in bioenergy, bioindustry and functional foods, has been observed. Cardoon is a perennial plant native to the Mediterranean Basin belonging to the Asteraceae family, and includes two botanical varieties (Gatto et al., 2013): cultivated cardoon (var. altilis (DC)) and the ancestor wild cardoon or wild artichoke (var. sylvestris (Lamk) Fiori). The cultivated cardoon has been cultivated as a vegetable since ancient times, but the land area devoted to this crop has never been large (about 2000–3000 ha), and mainly localized in Spain, Italy, France and Greece (Ierna and Mauromicale, 2010). The wild cardoon is a robust thistle with a characteristic rosette of large spiny leaves and branched flowering stalks. It is distributed over the west and central part of the Mediterranean basin (Portugal to west Turkey), as well as Madeira and Canary Islands; in postColumbian time it colonized some parts of the New World and has

192

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

spread as a weed in parts of Argentina and California (Marushia and Holt, 2006). Both cultivated and wild cardoons produce lignocellulosic biomass and oil seeds for solid biofuel and biodiesel (Portis et al., 2012; Acquadro et al., 2013), and have been recognized as promising energy crops for rainfed farmlands in Mediterranean climates under low external management energy supplies (Ierna et al., 2012). This peculiarity is supported by the excellent adaptation of Cynara spp. to the Mediterranean climate, owing to the positive balance between the phases of the growth and development cycle under Mediterranean climatic trends, the capacity of photosynthesizing during winter time, as well as the capacity of nutrient uptake from deep soil layers (Fernández et al., 2006). The deep root system of cardoon annually produce high below-ground biomass, and its active root pools and soil microorganisms are important to stock soil carbon and as a nutrient cycling reservoir (Mauromicale et al., 2014). The economic analysis of cardoon in comparison with other herbaceous annual crops, demonstrated the low cultivation costs, the higher total revenues, and its suitability for the inclusion in arable cropping systems in marginal lands (Papazoglou and Rozakis, 2011; Francaviglia et al., 2016; Mehmooda et al., 2016). In recent years, Cynara cardunculus L. has been considered for different industrial applications. Achenes can be utilized for oil production for human consumption, and after oil extraction, the cake could be used for animal feed. The evaluation of whole cardoon achenes for feeding ruminants has been conducted by Cajarville et al. (2000). Roots could be used for extraction of inulin (Raccuia and Melilli, 2004, 2010; Raccuia et al., 2005), a fructose polysaccharide of interest for food and not-food applications (Ritsema and Smeekens, 2003). The use of fresh biomass as forage for livestock feeding is another possible application of the crop (Fernández et al., 2006). C. cardunculus L. has also been used for medicinal purposes (Kraft, 1997). Leaves, rich in polyphenols, are used because of the pharmacological properties of their constituents and extracts (Clifford, 1992; Grancai et al., 1994; Valentao et al., 2002; Curt et al., 2005; Fernández et al., 2006; Pinelli et al., 2007; Ciancolini et al., 2013). Recently, there has been an increase in the use of these polyphenolic compounds also in cosmetics (Lupo, 2001; Peschel et al., 2006). In terms of dry biomass, yields are very variable in relation to the pedo-climatic conditions, the cropping techniques and the genotypes compared. Studies from Italy, Portugal and Spain report dry biomass yield ranging from 10 to 15 to 30 t ha−1 due to differences in irrigation inputs and fertilizer applications (Gonzáles et al., 2004; Raccuia and Melilli, 2007; Gominho et al., 2014). With reference to biomass partition, Raccuia and Melilli (2007) reported that 38% were heads and 62% were stalks + leaves. The aims of the research were to evaluate the biomass yield and partition in two cultivated and two wild cardoon genotypes under a low input rainfed cultivation system, to evaluate their response and adaptability to the local environment, and to assess their potential use for food or bioenergy purposes. An advanced statistically based interpretation of the data was adopted to assess the suitability of the different genotypes to the local conditions and possible uses.

(Francaviglia et al., 2016), but yields were not economically significant in the first year, that can be considered as a stabilization stage (Angelini et al., 2009), thus only the last three years are included in this study. Soils were Eutric Cambisols (WRB, 2014), and more details about the site characteristics are given in Francaviglia et al. (2016). Marginal area in the context of the paper refers firstly to a low soil fertility level as evidenced by the low organic C (0.81%) and total N (0.105%) values, and the consequent unfavourable C/N ratio (7.7) indicating an accelerated decomposition of soil organic matter. In addition, soil erosion rates measured in the same farm during other field experiments were higher than 12 t ha−1 in the period September–November (Bazzoffi et al., 2015). The long term mean climate (Fig. 1) has a mean annual temperature of 15.2 ◦ C (24 ◦ C in July–August, 7 ◦ C in January), and 800 mm total rainfall (28 mm minimum in July). According to the updated Köppen-Geiger climate classification (Kottek et al., 2006), the climate is warm temperate with hot summers (Cfa). In 2012, total rainfall was 713 mm of which 421 mm from September to December, with a long dry period in June and July; temperature was higher in comparison with the long-term values in the period June–August. In 2013, total rainfall was considerably higher than the mean value (1130 mm), particularly in the spring, summer and winter months, and temperature was lower (14.6 ◦ C). During 2014 total rainfall was 908 mm, and temperature was higher than the long-term value (16.0 ◦ C). 2.2. Experiment set-up and crop management The characteristics of cultivated (Cynara cardunculus L. var. altilis DC) and wild (Cynara cardunculus L. var. sylvestris Lam) cardoon genotypes were studied, both in terms of biomass yield and its partition. Four genotypes were compared: two cultivated genotypes (CDL07 and Gigante) and two wild cardoons, one from Sicily (RCT10) and the other from Latium (Tolfa Mountains). The research is part of a field experiment set up in autumn 2010 (Francaviglia et al., 2016) following a split-split plot scheme with 3 replicates (12 plots, plot size 28 m2 , planting density 8 plants m−2 ), where the crop genotype was the main factor. A basal dressing fertilisation with 300 kg ha−1 of triple superphosphate (P2 O5 46%) was applied in October 2010. The experiment included two transplanting periods at the beginning of 2011 (second factor) due to the low seeds germination, with a total of 24 sub plots. Nitrogen fertilization was added as third factor (0–50 kg N ha−1 from urea) in the second and third year (November 2011 and 2012), with a final total of 48 sub-sub plots (Francaviglia et al., 2016). Yearly samplings on each sub-sub plot (4 plants randomly selected but avoiding the borders) were taken in late August at stage 8N9 of the BBCH scale proposed for C. cardunculus (Archontoulis et al., 2010), to determine the yield (fresh and dry biomass of stalks, leaves, capitula, achenes, receptacles, bracts and pappi), the biomass partition, and the biometric parameters (plant height, basal diameter of stalks, no and weight of capitula per plant, weight of receptacles, bracts and pappi, weight of achenes per capitulum, 1000 seeds weight). Fresh samples of the biomass components (200 g of plant material) were oven dried at 105 ◦ C until constant weight was reached.

2. Materials and methods 2.3. Statistical analyses 2.1. Study area The field experiment was carried out in a marginal area of the hills of Latium Region (Central Italy) at the CREA-RPS research farm of Tor Mancina near Rome (lat. 42◦ 06 N, long. 12◦ 40 E, alt. 43 m a.s.l.) during a 3-year period (2012–2014) and under rainfed conditions. The study is the follow-up of an experiment started in 2011

Chi-Square test was used to check normality of data distribution, and Levene’s test to check homogeneity of variances; Box and Cox and logarithmic transformations of data, before ANOVA, were used when necessary (untransformed data are reported and discussed). Differences among treatments were determined by analysis of variance (ANOVA), related to the experimental design adopted in the

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

193

Fig. 1. Monthly climate (2012–2014) for rainfall (R) and temperature (T) and long term values in the study area.

field (split-split-plot design). The year effect was tested in the ANOVA, and analyzed as further subplot in a split-split-split plot design, according with some authors that suggest this approach for perennial crops (Petersen, 1994). General Linear Model procedure of the Statistica 8.0 software package (Statsoft, Tulsa, USA) was used to perform the ANOVA, considering the block as random factor to calculate the appropriate error terms for the F test. Additionally, multiple means comparisons were performed through the Tukey HSD test for the Genotypes × Year interaction and Fisher’s protected least significant difference test (LSD test) for the Genotypes and Year averages, both at least at P ≤ 0,05. Linear regression analyses were performed to estimate the relationship between the dependent variable total biomass as dry matter per plant and the independent variables stalks, leaves and capitula weights to evaluate the productive aptitude of the four genotypes to the local environmental conditions. 3. Results 3.1. ANOVA results Results of the analysis of variance for most parameters showed significant differences mainly among genotypes (G), years (Y) and their interaction (G × Y) (Table 1). The G factor showed a general prevalent influence on biometric and yield parameters during the three years of research, and the nitrogen factor (N) never showed significant effects. The transplanting factor (TP), introduced in the experimental design due to problems in the seed germination and maintained in the statistical analyses to detect the possible variance induced, showed significant differences only in the stalks diameter. The interactions of the three factors (G × TP, G × N, and TP × N) never showed significant effects during the experiment (Table 1). 3.2. Biometric parameters The initial planting density (8 plants m−2 ), allowed the crop to maintain a constant plant density in 2012 and 2013, and to prevent the tendency to form lateral branches. However, the gradual fungal blight causing root rot due to Sclerotium rolfsii Sacc., was compensated by the formation of lateral branches in Tolfa and RCT10 wild genotypes; conversely, Gigante and CDL07 cultivated genotypes showed a high sensitivity to the pathogen, and in the last year (2014) the plant density decreased to 4.7 and 1.7 plants per square meter, with important negative consequences on yields.

All biometric parameters (Table 2) were strongly affected by the genotype and the yearly climatic patterns. Gigante showed the best vegetative vigor with 182 cm of plant height and 2.5 cm of basal stalk diameter; Tolfa, CDL07 and RCT10 showed significantly lower plant heights among the three genotypes and in comparison with Gigante (131, 110 and 96 cm respectively), while basal stalk diameters were not statistically different among the three genotypes (2.0, 2.1 and 1.8 cm respectively) but significantly different in comparison with Gigante. Considering the number of capitula per plant, 2013 showed a significant and general decrease in comparison with the 3-year average value (7.4), with an average of 5.7 capitula per plant. Conversely, in 2012 the average value was higher (9.2), and close to the average in 2014 (7.3). Wild genotypes showed a clear and significantly higher number of capitula per plant (9.2 as average) in comparison with the cultivated genotypes (5.6 as average); in particular CDL07 showed the lowest number of capitula per plant (4.6 as average). Conversely, the cultivated genotypes showed the highest average weight of capitula, with 40.4 g CDL07 statistically different in comparison with Gigante (29.3). Wild genotypes Tolfa and RCT10 showed a lower and significantly different weight of capitula (14.4 and 9.5 g respectively). The mean weights of achenes per capitulum and 1000 seeds were strongly dependent on the genotype, and significantly higher in the cultivated rather than in the wild genotypes, indicating the importance of these traits in determining the grain yields. In particular, the mean weight of achenes per capitulum was 7.2 and 5.2 g in the cultivated genotypes CDL07 and Gigante respectively, 1.4 and 2.6 g in the wild genotypes RCT10 and Tolfa respectively. This parameter was significantly influenced by the climatic pattern, and in 2013 showed a general decrease (3.3 g as mean value) in comparison with 2012 and 2014 (4.6 and 4.4 g respectively). The weight of 1000 seeds was rather stable during the experiment, and the mean yearly values were not significantly different from the general mean (28.7 g). However, the mean values of the cultivated genotypes were significantly different in comparison with the wild genotypes, but no significant differences were shown within the two cultivated and wild genotypes. As average, this parameter was 32.8 g in CDL07 and Gigante, 24.7 g in RCT10 and Tolfa. 3.3. Biomass partition The highest aboveground biomass per plant expressed as dry matter (Table 3) were observed in 2012 and 2014, and were statistically different in the two years (277.7 and 243.8 g respectively);

194

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

Table 1 Biometric and yield parameters ANOVA results. Probability level (P) of F test for biometric and yield (dry matter) parameters during the 3 years of cultivation. Factors: genotype (G), transplanting period (TP), nitrogen (N) and year (Y). Parameters

G

TP

N

Y

(G × TP)

(G × N)

(G × Y)

Plant height (cm) Stalks diameter (cm) Capitula per plant (no) Capitula m−2 (no) 1000 seeds weight (g) Achenes per capitula (g) Average weight of capitula (g) Average weight of receptacles, bracts and pappi (g) Aboveground per plant (g) Stalks per plant (g) Leaves per plant (g) Capitula per plant (g) Achenes per plant (g) Receptacles, bracts and pappi per plant (g)

0.000 0.003 0.002 0.001 0.000 0.000 0.000 0.000 0.015 0.013 0.001 0.009 0.006 0.009

ns 0.039 ns ns ns ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns ns ns ns ns ns

0.024 ns 0.000 0.000 ns 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000

ns ns ns ns ns ns ns ns ns ns ns ns ns ns

ns ns ns ns ns ns ns ns ns ns ns ns ns ns

0.001 0.004 0.000 0.000 ns 0.000 0.001 0.001 0.001 0.000 0.000 0.001 ns 0.000

ns = not significant.

Table 2 Biometric parameters during the 3 years of cultivation. Year

CDL07

Gigante

RCT10

Tolfa

Meana

Plant height (cm) 2012 2013 2014 Meanb

114 bc 114 bc 103 ab 110 B

188 e 183 e 175.58 e 182 D

92 a 100 a 96 a 96 A

125 cd 133 d 134 d 131C

130 AB 132 B 127 B 130

Stalks diameter (cm) 2012 2013 2014 Meanb

2.3 cdef 2.0 abcd 2.0abc 2.1 A

2.4 def 2.5 ef 2.6 f 2.5 B

1.6 a 2.0 abc 1.9 abc 1.8 A

1.8 ab 1.9 abc 2.2 bcde 2.0 A

2.0 ns 2.1 ns 2.2 ns 2.1

Capitula per plant (no) (ln) 2012 2013 2014 Meanb

6.1 c 3.9 a 3.9 ab 4.6 A

7.5 cd 6.4 cd 5.7 c 6.5 AB

11.6 e 6.4 c 10.7 e 9.6 B

11.5 e 6.0 bc 9.0 de 8.8 B

9.2 C 5.7 A 7.3 B 7.4

Capitula m−2 (no) (B-C) 2012 2013 2014 Meanb

49 abd 31 df 7g 29C

60 abe 51 ab 27 f 46 B

93 c 51 ab 79 ce 74 A

92 c 48 ad 71 bce 70 A

74 B 45 A 46 A 55

Average weight of capitulum (g) 2012 2013 2014 Meanb

44.2 e 38.7 d 38.3 d 40.4 D

25.6 c 24.1 c 26.1 c 25.3C

11.3 ab 9.5 a 7.7 a 9.5 A

14.0 14.9 b 14.3 b 14.4 B

23.8 B 21.8 A 21.6 A 22.4

Average weight of receptacles, bracts and pappi (g) (B-C) 2012 2013 2014 Meanb

36.4 f 32.8 e 30.4 e 33.2 D

19.8 d 20.5 d 19.9 d 20.1C

9.2 abc 8.4 ab 6.8 a 8.1 A

11.3 bc 12.5 c 11.6 bc 11.8 B

19.2 B 18.5 B 17.2 A 18.3

Achenes per capitulum (g) (B-C) 2012 2013 2014 Meanb

7.8 c 5.9 b 7.8 c 7.2 D

5.8 b 3.6 e 6.2 bc 5.2C

2.1 a 1.1 d 1.0 d 1.4 A

2.7 a 2.4 a 2.7 a 2.6 B

4.6 C 3.3 A 4.4 B 4.1

1000 seeds weight (g) 2012 2013 2014 Meanb

34.7 32.9 34.4 34.4 B

31.0 32.2 31.3 31.3 B

26.3 23.3 25.7 25.7 A

24.5 20.2 23.7 23.7 A

29.1 ns 27.1 ns 28.7 ns 28.7

Averages of genotypes × years interaction (in italics) with different lowercase letters are significant with the Tukey HSD test (P ≤ 0.05). (ln) or (B-C) = natural logarithm or Box and Cox transformation before ANOVA. Untransformed data are shown. ns = test F not significant. a Averages of years (right column) with different uppercase letters are significant with the Fisher LSD test (P ≤ 0.05). b Averages of genotypes (bottom row for each trait) with different uppercase letters are significant with the Fisher LSD test (P ≤ 0.05).

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

195

Table 3 Aboveground biomass during the 3 years of cultivation as dry matter per plant. Year

CDL07

Gigante

RCT10

Tolfa

Meana

Aboveground biomass (g) (B-C) 2012 2013 2014 Meanb

351.8 a 173.3 bc 224.3 cd 249.8 AB

353.6 a 280.8 ad 333.5 a 322.7 B

175.1 bc 114.4 e 150.2 be 146.6C

230.3 cd 173.2 be 267.2ad 223.6 A

277.7 C 185.5 A 243.8 B 235.7

Stalks biomass (g) (B-C) 2012 2013 2014 Meanb

92.4b 16.7 d 57.0 a 55.3 A

150.4 e 55.5 a 112.8 b 106.2 B

58.1 a 47.2 a 57.8 a 54.4 A

87.8 bc 66.2 c 94.9 b 82.9 B

97.2 C 46.4 B 80.6 B 74.7

Leaves biomass (g) (ln) 2012 2013 2014 Meanb

17.8 21.4 30.9 23.4 B

38.3 93.4 87.1 72.9C

c

c

11.8 18.3 10.0 A

24.7 58.5 27.7 B

14.0 A 37.8 B 48.7 C 33.5

Capitula (g) (B-C) 2012 2013 2014 Meanb

241.6 d 135.3 ab 136.5 ab 171.1C

165.0 b 132.0 ab 133.6 ab 143.5 BC

117.0 ab 55.3 c 74.1 c 82.2 A

142.5 ab 82.3 c 113.9 a 112.9 AB

166.5 C 101.2 A 114.5 B 127.4

Achenes (g) (B-C) 2012 2013 2014 Meanb

41.1 20.1 27.6 29.6 A

38.7 20.4 31.8 30.3 A

20.6 6.4 9.3 12.1 B

26.7 13.1 21.5 20.4C

31.8 C 15.0 A 22.5 B 23.1

Receptacles, bracts and pappi (g) (ln) 2012 2013 2014 Meanb

200.5 a 115.2 a 108.9 a 141.5 A

126.3 d 111.5 a 101.8 a 113.2 BC

96.4 a 48.9 b 64.8 bc 70.1 A

115.9 c 69.2 bc 92.4 ac 92.5 AB

134.8C 86.2 A 92.0 B 104.3

Averages of genotypes × years interaction (in italics) with different lowercase letters are significant with the Tukey HSD test (P ≤ 0.05). (ln) or (B-C) = natural logarithm or Box and Cox transformation before ANOVA. Untransformed data are shown. a Averages of years (right column) with different uppercase letters are significant with the Fisher LSD test (P ≤ 0.05). b Averages of genotypes (bottom row for each trait) with different uppercase letters are significant with the Fisher LSD test (P ≤ 0.05). c Leaves weight was not determined in 2012 due to the total fall before the harvest. Tukey test not performed.

in 2013 the productivity was the lowest (185.5 g) and statistically different. The highest mean values during the three years were observed in the cultivated genotypes Gigante and CDL07 (322.7 and 249.8 g per plant respectively); conversely, the lowest productivity was observed in the RCT10 wild genotype from Sicily (146.6 g per plant), in agreement with the biometric traits, whereas the wild genotype Tolfa showed a mean productivity of 223.6 g per plant during the three years, not significantly different in comparison with the cultivated genotype CDL07. The productive pattern of the Tolfa genotype was of particular interest since in 2013, which was the most unfavorable year from the climatic point of view, the production was fairly good (173.2 g per plant), and an increase to 267.2 g was observed in 2014, not significantly different in comparison with the cultivated genotypes. The partition of dry biomass as absolute and percentage values (Fig. 2) has shown marked differences among the genotypes. Considering the stalks, the most productive genotype was Gigante (106.2 g per plant and about 33% of the total biomass), followed by Tolfa (82.9 g per plant and about 37% of the total biomass). The two genotypes from Sicily (CDL07 and RCT10) showed a stalk biomass of 54–55 g per plant, equal to about 22 and 37% respectively. Considering the leaves production, 2012 stands out due to the fully senescence and premature detachment before harvesting in the two wild genotypes, so that the sampling was not carried out. Conversely, the highest leaves production was obtained in 2014 for most genotypes (Table 3). The highest mean leaves production was observed in Gigante (72.9 g per plant and about 23% of the total biomass), which confirms the peculiarity of this genotype which was originally selected for the enlarged bleached petiole to be used

as food. Leaves in Tolfa genotype were 27.7 g per plant and about 12% of the total biomass, whereas lower values were observed in RCT10 and CDL07 from Sicily, 10.0 and 23.4 g per plant (about 7 and 9% respectively). The CDL07 genotype from Sicily stands out for the biomass production of the floral apparatus where the dry biomass of the capitula components including the achenes was 68% of the total as average value, whereas values ranging from 45 to 56% were observed in the other genotypes (Fig. 2). Achenes production in CDL07 was about 12% of the total, and the highest yield was observed in 2012 (41.1 g per plant) whilst the 3-year average was 29.6 g per plant (Table 3), confirming the high productive potential of achenes of this genotype. Gigante showed a mean production of achenes not statistically different in comparison with CDL07 (30.3 g per plant), but the partition of this trait was lower (about 9%) due to the very high total aboveground biomass of this genotype. The wild genotype Tolfa showed an interesting production of achenes, with a mean value equal to 20.4 g per plant representing about 9% of the total biomass. Conversely, the other wild genotype RCT10 showed the lowest productivity (12.1 g per plant equal to about 8% of the total biomass). A particular attention must be paid to the partition of the other components of the capitula (receptacles, bracts and pappi) representing a very high amount of the total biomass, about 35–41% in Gigante and Tolfa, 48% in RCT10 and about 57% in CDL07. The last two genotypes, both from Sicily, showed a mean production of this trait equal to 70.1 and 141.5 g per plant respectively. The regression analyses of the total aboveground biomass and the different plant components (stalks, leaves, capitula, achenes

196

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

Fig. 2. Absolute and percentage partition of the plant components in the different genotypes. Cultivated genotypes (Gigante, CDL07), wild genotypes (RCT10, Tolfa).

240 200

y = -34.238 + 0.4353x r2 = 0.6491

y = 7.1361 + 0.3224x r2 = 0.8216

160 120

Stalks biomass (g)

80

RCT10

40

Gigante

0 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

240

y = -30.0106 + 0.3417x r2 = 0.8929

200 160

y = 10.894 + 0.3222x r2 = 0.9578

120 80

CDL07

40

Tolfa

0 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

Aboveground biomass (g) Fig. 3. Regression of above ground vs. stalks biomass as dry matter per plant (g). Cultivated genotypes (Gigante, CDL07), wild genotypes (RCT10, Tolfa). Regressions are significant at P < 0.001. n = 36 for each genotype.

and the remaining floral components) allowed to evaluate the stability and the productive aptitude of the different genotypes (Fig. 3–7). Despite an excellent productive capacity in terms of total aboveground biomass, Gigante showed a strong instability in the partition of leaves (r2 = 0.0934) and stalks (r2 = 0.6491); conversely, the partition of capitula (r2 = 0.8640), achenes (r2 = 0.6687) and floral components (r2 = 0.8045) was rather stable. CDL07 showed a strong stability in the production of capitula (r2 = 0.9424) and floral components (r2 = 0.9135), and a fairy good productive stability of achenes (r2 = 0.8108), indicating the excellent capacity of this genotype to maximize the achenes yield rather than the total aboveground biomass. The wild genotype from Sicily RCT10, which showed a limited productive potential due to the mentioned difficulties to adapt to the new pedoclimatic environment, showed a good stability in the biomass partition of the different plant components, with the exception of the variability of leaves (r2 = 0.2976). The maximum stability in the partition of all the biomass components was observed in the wild genotype Tolfa from Latium, both in the reproductive organs (achenes: r2 = 0.7497; floral components:

r 2 = 0.7946) and more markedly in the vegetative organs (leaves: r2 = 0.9002; stalks: r2 = 0.9578). The stability of Tolfa yields indicates a strong adaptation of this genotype to the areas of the Latium region, where the selective pressure has probably determined a plant genetic heritage with very stable vegetative and reproductive traits. 3.4. Biomass and grain yield The main productive results are shown in Table 4. The highest total dry biomass yields were obtained with the Gigante genotype, with a 3-year average value equal to 22.3 t ha−1 statistically not significant in comparison with the Tolfa genotype with 17.7 t ha−1 . The maximum annual yield was obtained in 2012 with 22.2 t ha−1 , and a general decreasing trend was observed in 2013 (14.8 t ha−1 ). In 2014 annual yield was also low (13.0 t ha−1 ) due to the already mentioned fungal blight (Sclerotium rolfsii Sacc.) causing root rot to which both cultivated genotypes showed a high sensitivity. Conversely, the wild genotypes showed a productive increase in 2014

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

197

160

y = 37.2158 + 0.1107x r2 = 0.0934

y = 5.1918 + 0.0745x r2 = 0.2976

120 80

Leaves (g)

40

Gigante

RCT10

0 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

160

y = 16.4239 + 0.0277x r2 = 0.0793

120

y = -13.8191 + 0.2517x r2 = 0.9002

80

CDL07 Tolfa

40 0 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

Aboveground biomass (g) Fig. 4. Regression of above ground vs. leaves biomass as dry matter per plant (g). Cultivated genotypes (Gigante, CDL07), wild genotypes (RCT10, Tolfa). Regression is significant at P < 0.001 for Tolfa, not significant for the other genotypes. n = 36 for cultivated genotypes, n = 24 for wild genotypes.

450

y = -20.4938 + 0.7003x r2 = 0.8608

400 350 300 250

y = -2.9778 + 0.454x r2 = 0.8640

Capitula (g)

200 150 100 50 0

Gigante

RCT10 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

450 400 350

y = 6.7205 + 0.4749x r2 = 0.8013

y = 13.5867 + 0.6306x r2 = 0.9424

300 250 200 150 100

CDL07

50 0 0

100

200

300

400

500

600

Tolfa

700 0

100

200

300

400

500

600

700

Aboveground biomass (g) Fig. 5. Regression of above ground vs. capitula biomass as dry matter per plant (g). Cultivated genotypes (Gigante, CDL07), wild genotypes (RCT10, Tolfa). Regressions are significant at P < 0.001. n = 36 for each genotype.

due to their higher tolerance to the pathogen and the favorable climatic pattern. In particular, the highest yield was observed in Tolfa (20.9 t ha−1 ) confirming the higher stability, adaptability and strength of this genotype. The highest achenes yields (Table 4) were observed in 2012 with a mean value equal to 2.6 t ha−1 , and the maximum in CDL07 and Gigante (3.3 and 3.1 t ha−1 respectively). Lower and statistically different yields were observed in the wild genotypes, even if the yield of Tolfa was fairly good (2.1 t ha−1 ). The decrease of achenes yields was consistent starting from 2013, in coherence with the total biomass, due to the unfavorable climatic conditions with lower temperatures in the summer period where the development of capitulum and seed ripening take place, while in 2014 the decrease

was particularly evident in the cultivated genotypes due to the Sclerotium fungal blight. In particular, CDL07 showed yields lower by about 50% in 2013 (1.6 t ha−1 ) and 85% in 2014 (0.5 t ha−1 ) due to the higher sensitivity to the fungus. The 3-year average value of the achenes yields showed the highest productivity in Gigante (2.1 t ha−1 ), followed by CDL07 (1.8 t ha−1 ) and Tolfa (1.6 t ha−1 ), while RCT10 was the genotype with the lower and statistically different yield (1.0 t ha−1 ). 4. Discussion Biomass yield results were generally in good agreement with the scientific literature available from other Mediterranean countries.

198

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

80 70

y = -7.2584 + 0.1163x r2 = 0.6687

y = -5.8311 + 0.1223x r2 = 0.6280

60 50 40 30 20

RCT10

Achenes (g)

10

Gigante

0 0

100

200

300

400

500

600

700 0

80 60

200

300

400

500

600

700

y = 0.3304 + 0.0899x r2 = 0.7497

y = 3.812 + 0.1032x r2 = 0.8108

70

100

50 40 30 20

CDL07

10

Tolfa

0 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

Aboveground biomass (g) Fig. 6. Regression of above ground vs. achenes biomass as dry matter per plant (g). Cultivated genotypes (Gigante, CDL07), wild genotypes (RCT10, Tolfa). Regressions are significant at P < 0.001. n = 36 for each genotype.

400 350

y = 4.2807 + 0.3376x r2 = 0.8045

y = -14.6627 + 0.5779x r2 = 0.8956

300 250

Receptacles, bracts and pappi (g)

200 150 100

Gigante

RCT10

50 0 0 400

100

200

300

400

500

600

700 0

300

200

300

400

500

600

700

y = 6.3901 + 0.3851x r2 = 0.7946

y = 9.7747 + 0.5274x r2 = 0.9135

350

100

250 200 150 100

Tolfa

CDL07

50 0 0

100

200

300

400

500

600

700 0

100

200

300

400

500

600

700

Aboveground biomass (g) Fig. 7. Regression of above ground vs. receptacles, bracts and achenes biomass as dry matter per plant (g). Cultivated genotypes (Gigante, CDL07), wild genotypes (RCT10, Tolfa). Regressions are significant at P < 0.001. n = 36 for each genotype.

In a study performed in the Catania Plain (Sicily, southern Italy) with a low irrigation input (100 mm), Foti et al. (1999) reported about 30.5 t ha−1 in cultivated cardoon, 18.8 t ha−1 in wild cardoon as 3-year average dry biomass value. Lower yields but in rainfed conditions were obtained in Basilicata (Piscioneri et al., 2000) with dry biomass yields for cultivated cardoon genotypes ranging from 10 to 15 t ha−1 during the third year of cultivation. The latter results were in agreement with the dry biomass yields obtained in Spain (Gonzáles et al., 2004; Ochoa and Fandos, 2004), but were lower than the yield of cultivated cardoon, equal to 20.6 t ha−1 as a 4-

year average, reported by Mauromicale and Ierna (2004) in Sicily, with a very low irrigation input (50 mm). Raccuia and Melilli (2007) reported a 3-year average dry biomass values in Sicily of 12.2 and 23.0 t ha−1 for wild and cultivated cardoon respectively, and with 50 mm irrigation. In Sardinia and under Mediterranean climate, Ledda et al. (2013) observed a 3-year average dry biomass yield of 10.2 t ha−1 for cultivated cardoon (range 6.8–11.6 t ha−1 ). In a large scale field experiment in Portugal, Gominho et al. (2014) reported a total dry biomass yield of 9.7 t ha−1 , (range 4.4–18.4 t ha−1 ); Vasilakoglou and Dhima (2014) observed total dry biomass yields

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

199

Table 4 Yield parameters in dry matter during the 3 years of cultivation. CDL07

Gigante

RCT10

Tolfa

Meana

28.1 d 13.9 abc 3.9 g 15.3 A

28.3 d 22.5 df 16.1 bce 22.3 B

14.0 abc 9.2 a 11.1 ab 11.4 A

18.4 cef 13.9 abc 20.9 def 17.7 AB

22.2 C 14.8 B 13.0 A 16.7

Achenes (t ha−1 ) (B-C) 2012 2013 2014 Meanb

3.3 f 1.6 a 0.5 b 1.8 A

3.1 ef 1.6 a 1.5 ad 2.1 A

1.7 a 0.5 b 0.7 bc 1.0 B

2.1 ae 1.1 cd 1.7 a 1.6 A

2.6 B 1.2 A 1.1 A 1.6

Stalks (t ha−1 ) (B-C) 2012 2013 2014 Meanb

7.4 b 1.3 c 1.0 c 3.2 A

12.0 d 4.4 a 5.4 ab 7.3C

4.7 a 3.8 a 4.3 a 4.2 AB

7.0 b 5.3 ab 7.4 b 6.6 BC

7.8C 3.7 A 4.5 B 5.3

Leaves (t ha−1 ) (ln) 2012 2013 2014 Meanb

1.4 1.7 0.5 1.2 ns

3.1 7.5 4.2 4.9 ns

c

c

1.0 1.4 0.8 ns

2.0 4.6 2.2 ns

1.1 3.0 ns 2.7 ns 2.3

Capitula (t ha−1 ) (B-C) 2012 2013 2014 Meanb

19.3 f 10.8 ab 2.4 e 10.8 ns

13.2 b 10.6 ab 6.4 cd 10.1 ns

9.4 ab 4.4 c 5.5 c 6.4 ns

11.4 ab 6.6 cd 8.9 ad 9.0 ns

13.3 C 8.1 B 5.8 A 9.1

Receptacles, bracts and pappi (t ha−1 ) (ln) 2012 2013 2014 Meanb

16.0 f 9.2 a 1.9 e 9.0 ns

10.1 a 8.9 a 4.9 b 8.0 ns

7.7 ad 3.9 b 4.8 bc 5.5 ns

9.3 a 5.5 bcd 7.2 acd 7.3 ns

10.8 C 6.9 B 4.7 A 7.5

Year −1

Aboveground biomass (t ha 2012 2013 2014 Meanb

) (B-C)

Averages of genotypes × years interaction (in italics) with different lowercase letters are significant with the Tukey HSD test (P ≤ 0.05). (ln) or (B-C) = natural logarithm or Box and Cox transformation before ANOVA. Untransformed data are shown. ns = test F not significant. a Averages of years (right column) with different uppercase letters are significant with the Fisher LSD test (P ≤ 0.05). b Averages of genotypes (bottom row for each trait) with different uppercase letters are significant with the Fisher LSD test (P ≤ 0.05). c Leaves weight was not determined in 2012 due to the total fall before the harvest. Tukey test not performed.

equal to 11.0 and 21.8 t ha−1 in Northern and Central Greece respectively. Dry matter achenes yields in Italy commonly range from 1.3 t ha−1 (Curt et al., 2002) to 2.6 t ha−1 (Foti et al., 1999), and up to 3.0 t ha−1 (Raccuia and Melilli, 2007). Achenes yields reported by Gominho et al. (2014) in Portugal were 2.2 t ha−1 as average (range 0.5–3.4 t ha−1 ); Vasilakoglou and Dhima (2014) measured achenes yields equal to 1.3 and 2.2 t ha−1 in Northern and Central Greece respectively. Significant differences among genotypes for aboveground biomass yield and its partition were observed by Cravero et al. (2012) in three botanical varieties of Cynara cardunculus, and by Raccuia and Melilli (2007) in a 3-year trial performed with wild and cultivated cardoons accessions. In particular, Raccuia and Melilli reported that 38% were heads and 62% were stalks + leaves. Mauromicale et al. (2014) reported that cultivated cardoon showed a higher leaves, stalk and achenes incidence (39.2, 45.7 and 3.9%, respectively) as compared to wild cardoon (34.4, 35.3 and 3.2%, respectively). Scarce information is available regarding the partition of the capitula biomass. Fernández and Curt (2005) showed that the dry capitula biomass in cardoon is partitioned on average as: receptacle 18%, bracts 25%, fruits 32% and light material 25% (hairs, pappi, and remains of corolla, stamens and styles). Similar values were recorded by Piscioneri et al. (2000). Gominho et al. (2009) found that the capitula biomass partition is a function of capitulum size. Results from Cravero et al. (2012) for cardoon showed that the

receptacle was the most important, and its weight accounted for 50% of the total aboveground biomass. In relation to the oil content of achenes, Francaviglia et al. (2016) showed quite constant average values of 18.5% for the two cultivated genotypes (CDL07 and Gigante), and lower contents of 17.4 and 16.7% in the wild genotypes (Tolfa and RCT10 respectively). Average values were not significantly different in the four studied genotypes. Average oil yields were 0.45 t ha−1 in cultivated genotypes, and 0.23 t ha−1 for wild cardoons (0.28 and 0.18 t ha−1 for Tolfa and RCT10 respectively). 5. Conclusions Aboveground biomass production and partition were mainly driven by genotype and the year of cultivation. Wild genotypes (Tolfa and RCT10) were less productive than cultivated genotypes (CDL07 and Gigante) in terms of total dry biomass. Similarly, grain weight was significantly lower in the wild genotypes in comparison with the cultivated species. As average, both aboveground total biomass and achenes followed the order Gigante > CDL07 > Tolfa > RCT10. The stability and the productive aptitude of the four genotypes was evaluated with an advanced statistically based analysis which showed different results among the genotypes, and could allow to assess the suitability of the different genotypes to the local conditions. In particular, the capacity of CDL07 genotype to maximize the grain yield rather than the total aboveground biomass was shown.

200

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201

The wild genotype RCT10 showed a limited productive potential due to the difficulties to adapt to the new pedoclimatic environment. The maximum stability in the partition of all the biomass components and a high tolerance to pathogens was observed in the wild genotype Tolfa, while cultivated genotypes were more sensitive to the pathogen infection which reduced the plant density and decreased the yields. The tested cultivated cardoon genotypes can be grown with different purposes, e.g. food (Gigante) since this genotype was originally selected for the enlarged bleached petiole, biodiesel (CDL07) and alternative energy production (both genotypes). The adaptability of wild Tolfa genotype to the local conditions shown by the more stable biomass partition, coupled with the fairy good aboveground biomass and achenes production, suggests that this variety might be easily incorporated into the local cropping systems as an industry crop to produce oil for human consumption if capitula are collected separately and threshed, biodiesel or bioenergy purposes with the utilisation of the residual lignocellulosic biomass for solid biofuel production. Acknowledgements The research was part of the BIOSEGEN Project (BIOcarburanti di SEconda GENerazione), funded by the Italian Ministry of Agricultural Food and Forestry Policies (MiPAAF Decree 17532/7303/10), and coordinated by Dr. Vito Pignatelli, Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA). We acknowledge the contribution of CNR-ISAFOM (Istituto per i Sistemi Agricoli e Forestali del Mediterraneo), U.O.S. Catania (Dr. Salvatore A. Raccuia) for providing the seeds of CDL07 and RCT10 genotypes and his valuable suggestions during the research. References Acquadro, A., Portis, E., Scaglione, D., Mauro, R.P., Campion, B., Falavigna, A., Zaccardelli, R., Ronga, D., Mauromicale, G., Lanteri, S., 2013. CYNERGIA project: exploitability of Cynara cardunculus L. as energy crop. Acta Hortic. 983, 109–150. Angelini, L.G., Ceccarini, L., Nassi o Di Nasso, N., Bonari, E., 2009. Long-term evaluation of biomass production and quality of two cardoon (Cynara cardunculus L.) cultivars for energy use. Biomass Bioenergy 33, 810–816. Archontoulis, S.V., Struik, P.C., Vos, J., Danalatos, N.G., 2010. Phenological growth stages of Cynara cardunculus: codification and description according to the BBCH scale. Ann. Appl. Biol. 156, 253–270. Bazzoffi, P., Francaviglia, R., Neri, U., Napoli, R., Marchetti, A., Falcucci, M., Pennelli, B., Simonetti, G., Barchetti, A., Migliore, M., Fedrizzi, M., Guerrieri, M., Pagano, M., Puri, D., Sperandio, G., Ventrella, D., 2015. Environmental effectiveness of GAEC cross-compliance Standard 1.1a (temporary ditches) and 1.2 g (permanent grass cover of set-aside) in reducing soil erosion and economic evaluation of the competitiveness gap for farmers. Ital. J. Agron. 10 (Suppl. 1), http://dx.doi.org/10.4081/ija.2015.10.s1.710. Cajarville, C., Gonzalez, J., Repetto, J.L., Alvir, M.R., Rodriguez, C.A., 2000. Nutritional evaluation of cardoon (Cynara cardunculus) seed for ruminants. Anim. Feed. Sci. Technol. 87, 203–213. Ciancolini, A., Alignan, M., Pagnotta, M.A., Vilarem, G., Crinò, P., 2013. Selection of Italian cardoon genotypes as industrial crop for biomass and polyphenol production. Ind. Crops Prod. 51, 145–151. Clifford, M.N., 1992. Sensory and dietary properties of phenols. Bull. Liaison Groupe Polyphen. 16, 19–31. Cravero, V., Martin, E., Crippa, I., López Anido, F., García, S.M., Cointry, E., 2012. Fresh biomass production and partition of aboveground growth in the three botanical varieties of Cynara cardunculus L. Ind. Crops Prod. 37, 253–258. Curt, M.D., Sánchez, G., Fernández, J., 2002. The potential of Cynara cardunculus L. for seed oil production in a perennial cultivation system. Biomass Bioenergy 23, 33–46. Curt, M.D., Sánchez, G., Fernández, J., 2005. Cynara cardunculus a source of silymarin. Acta Hortic. 681, 461–467. Fernández, J., Curt, M.D., 2005. State of the art of Cynara cardunculus L. as an energy crop. In: Proceedings 14th. European Biomass Conference & Exhibition. Biomass for Energy, Industry and Climate Protection, Paris, France, pp. 22–27. Fernández, J., Curt, M.D., Aguado, P.L., 2006. Industrial applications of Cynara cardunculus L. for energy and other uses. Ind. Crops Prod. 24, 222–229. Foti, S., Mauromicale, G., Raccuia, S.A., Fallico, B., Fanella, F., Maccarone, E., 1999. Possible alternative utilization of Cynara spp. Part I. Biomass, grain yield and chemical composition of grain. Ind. Crops Prod. 10, 219–228.

Francaviglia, R., Bruno, A., Falcucci, M., Farina, R., Renzi, G., Russo, D.E., Sepe, L., Neri, U., 2016. Yields and quality of Cynara cardunculus L. wild and cultivated cardoon genotypes. A case study from a marginal land in Central Italy. Eur. J. Agron. 72, 10–19. Gatto, A., De Paola, D., Bagnoli, F., Vendramin, G.G., Sonnante, G., 2013. Population structure of Cynara cardunculus complex and the origin of the conspecific crops artichoke and cardoon. Ann. Bot. 112, 855–865. Gominho, J., Lourenc¸o, A., Curt, M.D., Fernández, J., Pereira, H., 2009. Characterization of hairs and pappi from Cynara cardunculus capitula and their suitability for paper production. Ind. Crop Prod. 29, 116–125. Gominho, J., Lourenc¸o, A., Curt, M.D., Fernández, J., Pereira, H., 2014. Cynara cardunculus in large scale cultivation: a case study in Portugal. Chem. Eng. Trans. 37, 529–534. Gonzáles, J., Pérez, F., Fernández, J., Lezaun, J.A., Rodriguez, D., Perea, F., Romero, C., Ochoa, M.J., García, M., 2004. Study of Cynara cardunculus L: lignocellulosic biomass production in dry conditions. Acta Hortic. 660, 221–227. Grammelis, P., Malliopoulou, A., Basinas, P., Danalatos, N.G., 2008. Cultivation and characterization of Cynara cardunculus for solid biofuels production in the Mediterranean region. Int. J. Mol. Sci. 9, 1241–1258. Grancai, D., Nagy, M., Suchy, V., Novomesky, P., 1994. Cynarin from the fresh flower buds of Cynara cardunculus. Fitoterapia 65 (3), p282. Ierna, A., Mauromicale, G., 2010. Cynara cardunculus L. genotypes as a crop for energy purposes in a Mediterranean environment. Biomass Bioenergy 34, 754–760. Ierna, A., Mauro, R.P., Mauromicale, G., 2012. Biomass, grain and energy yield in Cynara cardunculus L. as affected by fertilization: genotype and harvest time. Biomass Bioenergy 36, 404–410. Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World Map of the Köppen-Geiger climate classification updated. Meteorol. Z. 15, 259–263. Kraft, K., 1997. Artichoke leaf extract: recent findings reflecting effects on lipid metabolism, liver and gastrointestinal tracts. Phytomedicine 4, 369–378. ˜ J., Bartual-Martos, J., 2014. Effects of Lag-Brotons, A., Gómez, I., Navarro-Pedreno, sewage sludge compost on Cynara cardunculus L. cultivation in a mediterranean soil. Compost Sci. Util. 22, 33–39. Ledda, L., Deligios, P.A., Farci, R., Sulas, L., 2013. Biomass supply for energetic purposes from some Cardueae species grown in Mediterranean farming systems. Ind. Crop Prod. 47, 218–226. Lupo, M.P., 2001. Antioxidant and vitamins in cosmetics. Clin. Dermatol. 19, 467–473. Marushia, R.G., Holt, J.S., 2006. The effects of habitat on dispersal patterns of an invasive thistle, Cynara cardunculus. Biol. Invasions 8, 577–593. Mauromicale, G., Ierna, A., 2004. Biomass and grain yield in Cynara cardunculus L. genotypes grown in a permanent crop with low input. Acta Hortic. 660, 593–597. Mauromicale, G., Sortino, O., Pesce, G.R., Agnello, M., Mauro, R.P., 2014. Suitability of cultivated and wild cardoon as a sustainable bioenergy crop for low input cultivation in low quality Mediterranean soils. Ind. Crops Prod. 57, 82–89. Mehmooda, M.A., Ibrahim, M., Rashid, U., Nawaz, M., Ali, S., Hussain, A., Gull, M., 2016. Biomass production for bioenergy using marginal lands. Sustain. Prod. Consumpt., http://dx.doi.org/10.1016/j.spc.2016.08.003. Ochoa, M.J., Fandos, A., 2004. Evaluation of vegetable cardoon (Cynara cardunculus L.) populations for biomass production under rainfed conditions. Acta Hortic. 660, 235–239. Papazoglou, E.G., Rozakis, S., 2011. Cardoon cultivation for combined bioenergy production and cadmium phytoextraction: an economic evaluation. In: Proceedings of the 3rd International CEMEPE & SECOTOX Conference, Skiathos, June 19–24 (ISBN 978-960-6865-43-5). Peschel, W., Sanchez-Rabaneda, F., Diekmann, W., Plescher, A., Gartzia, I., Jimenez, D., Lamuela-Raventos, R., Buxaderas, S., Codina, C., 2006. An industrial approach in the search of natural antioxidants from vegetable and fruit wastes. Food Chem. 97, 137–150. Petersen, R.G., 1994. Agricultural Field Experiments − Design and Analysis. Marcel Dekker, Inc, New York, USA. Pinelli, P., Agostini, F., Comino, C., Lanteri, S., Portis, E., Romani, A., 2007. Simultaneous quantification of caffeoyl esters and flavonoids in wild and cultivated cardoon leaves. Food Chem. 105 (4), 1695–1701. Piscioneri, I., Sharma, N., Baviello, G., Orlandini, S., 2000. Promising industrial energy crop, Cynara cardunculus: a potential source for biomass production and alternative energy. Energy Convers. Manage. 41, 1091–1105. Portis, E., Scaglione, D., Acquadro, A., Mauromicale, G., Mauro, R., Knapp, S.J., Lanteri, S., 2012. Genetic mapping and identification of QTL for earliness in the globe artichoke/cultivated cardoon complex. BMC Rese. Notes 5, 252. Raccuia, S.A., Melilli, M.G., 2004. Cynara cardunculus L., a potential source of inulin in Mediterranean environment: screening of genetic variability. Aust. J. Agric. Res. 55, 693–698. Raccuia, S.A., Melilli, M.G., 2007. Biomass and grain oil yields in Cynara cardunculus L: genotypes grown in a Mediterranean environment. Field Crops Res. 101, 187–197. Raccuia, S.A., Melilli, M.G., 2010. Seasonal dynamics of biomass inulin, and water-soluble sugars in roots of Cynara cardunculus L. Field Crops Res. 116, 147–153. Raccuia, S.A., Patanè, C., Melilli, M.G., 2005. Multiple utilisation of plants in Cynara cardunculus L. var. sylvestris Lam.: inulin yield. Acta Hortic. 681, 475–482. Ritsema, T., Smeekens, S., 2003. Fructans: beneficial for plants and humans. Curr. Opin. Plant Biol. 6, 223–230.

U. Neri et al. / Industrial Crops and Products 95 (2017) 191–201 Sims, R.E.H., Hastings, A., Schlamadinger, B., Taylor, G., Smith, P., 2006. Energy crops: current status and future prospects. Global Change Biol. 12, 2054–2076. Valentao, P., Fernandes, E., Carvalho, F., Andrade, P.B., Seabra, R.M., Bastos, M.L., 2002. Antioxidative properties of cardoon (Cynara cardunculus L.) infusion against superoxide radical: hydroxyl radical and hypochlorous acid. J. Agric. Food Chem. 50, 4989–4993.

Vasilakoglou, I., Dhima, K., 2014. Potential of two cardoon varieties to produce biomass and oil under reduced irrigation and weed control inputs. Biomass Bioenergy 63, 177–186. WRB, 2014. World Reference Base for Soil Resources 2014. International Soil Classification System for Naming Soils and Creating Legends for Soil Maps World Soil Resources Reports No. 106. FAO, Rome, Italy.

201