The increasing spread of bioenergy crops offers significant opportunities for Europe to reduce greenhouse gas emission and ensure its energy supply. However, because of the low per unit-value, an adequate production under minimal external inputs is desirable for growers to profit from producing biomass (Knoll et al., 2012). Cynara cardunculus L. is native to the Mediterranean Basin, and includes the three taxa globe artichoke (var. scolymus L.), cultivated cardoon (var. altilis DC.) and their common ancestor wild cardoon (var. sylvestris Lam.) (Mauro et al., 2007; Portis et al., 2014). Both cultivated and wild cardoon have been recognised as promising energy crops for rainfed farmlands in Mediterranean climates. In particular, they produce lignocellulosic biomass and oil seeds (for solid biofuel and biodiesel) (Portis et al., 2012; Acquadro et al., 2013). The interest for these crops comes from their high biomass and energy yields 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 with 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 objective of our study was: i) to evaluate the long-term seasonal dynamic of biomass, achenes, oil and energy yield in cultivated and wild cardoon under minimal input; ii) to identify possible germplasm for breeding aimed at constituting specific genotypes for bioenergy purposes.
Materials and methods
Site and soil
A field experiment was conducted in Sicily (36° 47’ N, 14° 54’ E, 42 m asl), over seven years (from 2005-2006 to 2011-2012, hereafter referred as S1-S7) in a moderately deep calcic brown soil with sandy loam texture. The local climate is semi-arid-Mediterranean, with mild, wet winters and hot, dry summers. The mean seasonal rainfall over the period 1929-1998 was 481 mm, mostly concentrated from October to February. From March to August, temperature increases rapidly, with maximum temperatures ranging from 15.8 to 32.5°C.
Experimental design and crop management
The experiment was arranged as a randomized block design with four replications, 4 4 m each. Two genotypes were studied: a cultivated cardoon Bianco Avorio and a wild cardoon landrace collected from native stands in South-Eastern Sicily (36° 55’ N, 14° 43’ E, 502 m asl). Four-week-old seedlings were transplanted on November 5, 2005 (2 plants m–2), on a soil previously ploughed, harrowed and fertilised with 60 kg ha–1 N (as urea), 100 kg ha–1 P2O5 (as triple superphosphate) and 80 kg ha–1 K2O (as potassium sulphate). In late-February 2006, 60 kg ha–1 N (as urea) were applied. After transplanting an irrigation (30 m3 ha–1) was adopted. In the following years no external inputs were supplied. Crop regrowth from S2 onward was naturally allowed by rains.
Data collection and analysis
In each year, on late-August stand density was assessed, and the above-ground biomass was harvested by cutting the plants at ground level from the centre of each plot; plants’ subsamples were reweighed after desiccation at 105°C, to determine the dry biomass yield. Achenes were separated from the heads through a specific mini-thrasher. In each season the collected biomass was milled to 3.0 mm (MF10 IKA mill) to determine the lower heating value (C200-System-IKA calorimeter), while achenes oil yield was determined by standard procedure according to the UNI EN ISO 659:2009 norm. Collected data were submitted to analysis of variance, while means comparisons were performed through Fisher’s least significant difference test.
Results and discussion
On the whole S1-S7 period, a higher stand density was recorded in wild cardoon (1.97 plants m–2) than in cultivated cardoon (1.86 m–2) (Table 1). The former entry showed little variation over S1-S7 (from 1.99 to 1.94 plants m–2), while cultivated cardoon had a linear decrease over S1-S5 (from 1.98 to 1.83 plants m–2) and then a sharper decrease over S5-S7 (up to 1.71 plants m–2) (Table 1). Problems related to cultivated cardoon survival in long-term trials were pointed out as one of the limiting factors for its yield potentiality in the Mediterranean area (Gherbin et al., 2001). In our experiment, despite its lower yield, wild cardoon showed a greater ability to lend itself to long-term and zeroinput cultivation period, which well reproduce its natural cropping conditions in the Mediterranean area (Ierna et al., 2012). From a breeding point of view, this aspect should be taken into account, in order to increase the crop duration over time and improve its agronomical performances in Mediterranean marginal farmland.
Above-ground dry biomass yield
Over the period S1-S7, the biomass yield of cultivated cardoon was about 2 times higher than that of wild cardoon [1.46 vs 0.74 kg dry matter (DM) m–2 year–1] (Table 1), with the pattern of biomass yield across seasons significantly differing between genotypes. The above-ground dry biomass yield of cultivated cardoon significantly increased from S1 to S5 (from 0.97 to 2.20 kg DM m–2), then it dropped down to 1.06 kg DM m–2 at S7 (Table 1). By contrast, wild cardoon biomass yield increased from S1 (0.55 kg DM m–2) to S5 (0.88 kg DM m–2), afterwards it remained somewhat stable until S7 (0.81 kg DM m–2) (Table 1). This was a pivotal difference among the tested genotypes, since wild cardoon, though less yielding showed a more conservative profile in biomass yield over seasons. This reflects its stable percentage of plant survival over time and better adaptability to the xerothermic conditions of Mediterranean area. The latter trait might be taken into account in the context of possible breeding programs for this crop; indeed, a more stable biomass production over time could have in the near future a pivotal role for the construction of more robust and reliable feedstock production for the biomass supply chains.
Achenes yield and oil content
Cultivated cardoon yielded 139% more achenes than wild cardoon (55 vs 24 g m–2) (Table 1), but their yield trend across seasons was substantially different. In cultivated cardoon, the achenes yield was almost constant from S1 to S6 (on average, 60 g m–2), while it decreased in S7 (32 g m–2); the achenes yield of wild cardoon, on the contrary, increased by 24% at S6-S7 as compared to S1-S2 (Table 1). Achenes oil content did not significantly differ between genotypes, but their pattern significantly differed among seasons (Table 2). Indeed, in cultivated cardoon achenes oil yield tended to increase passing from S1 (25.0 g 100 g–1 DM) to S4 (26.1 g 100 g–1 DM), then it dropped down at S7 (24.0 g 100 g–1 DM) (Table 2), showing overall 27% coefficient of variation over the S1-S7 period. In contrast, wild cardoon showed a somewhat stable achenes oil content, ranging from 25.5 (S6) to 26.0 g 100 g–1 DM (S6-S7) (Table 2). Seed set in C. cardunculus is a delicate phase, whose success can be heavily influenced by both the climatic conditions during flowering and nutritional status of the plant. Our experiment demonstrates that, when compared to cultivated cardoon, wild cardoon has been slightly affected by these factors. This suggests a better photosynthetic efficiency during stressful climatic conditions occurring at the achenes filling stage (early summer), and under poor fertility soil conditions. All these traits should be taken into account in the light of a specific breeding program for bioenergy purposes, in order to constitute genotypes characterized by both adequate yield levels and adaptability to stressful conditions and marginal agroecosystems, and to maximize the profit from both biomass and biodiesel chains.
Lower heating value and energy yield
Lower heating value (LHV) of biomass including achenes was very similar in both Cynara accessions, but they showed a different pattern of variability along the S1-S7 period (Table 2). In cultivated cardoon LHV increased from S1 (17.4 kJ kg–1 DM) to S4 (19.8 kJ kg–1 DM), then showed a sharp decrease from S5 onward (up to 17.4 kJ kg–1 DM). On the contrary wild cardoon reported a more stable LHV along the whole experimental period, which ranged from 17.7 (S3) to 19.0 kJ kg–1 DM (S7) (Table 2). As regards the energy yield, cultivated cardoon doubled wild cardoon over the whole S1-S7 cropping period (27.5 vs 13.8 kJ ha–1 year–1); however, when comparing their energy yield pattern, both genotypes showed distinct trends across years (Table 2). Indeed, in cultivated cardoon the energy yield significantly increased from 18.6 (S1) to 40.3 kJ m–2 (S5), then drastically decreased from S6 onward (20.6 kJ m–2) (Table 2). Differently, in wild cardoon the energy yield continuously increased over the S1-S5 period (from 10.3 to 16.1 kJ m–2) and thereafter showing no significant variations (Table 2). Our experiment demonstrates the high ability of cultivated cardoon to valorise both the Mediterranean marginal lands and low inputs agronomical practices. This has important implications within the concerns about the antagonistic land use between food and energy crops, since high yield levels in marginal fertility farmlands could maximize the benefits from the large-scale implementation of this crop, minimizing, also, the environmental costs of producing bioenergy. However, it might be noted that such convenient energy yield in low fertility conditions was achieved by a genotype not specifically released for energy production, so we can argue that specific breeding programs could significantly improve the agronomic performances of the crop.
The data presented here indicate C. cardunculus L. (both cultivated and wild cardoon) as a very promising renewable energy crop in marginal soil managed under zero/minimal inputs. In this context, the long-period cropping of C. cardunculus L. reached 1.46 (cultivated cardoon) and 0.74 (wild cardoon) kg m–2 of dry biomass per year, of which 55 and 24 g m–2 of achenes, corresponding to an energy yield of 27.5 and 13.7 kJ m–2. The two botanical varieties displayed a different production pattern across seasons. Cultivated cardoon, due to its yield timing, appears to be more suitable for medium-period cropping systems.
Wild cardoon, on the contrary, revealed a continuous increasing over the seasons 1-5, followed by no significant yield variations from S5 onward. The conservative profile in biomass yield and in plant survival with crop ageing of wild cardoon together with its pests, pathogens and weeds resistance/tolerance should represent a good and additional opportunity for marginal Mediterranean areas to develop a new model of perennial bioenergy crops.