https://www.sciencedirect.com/science/article/abs/pii/S0926669099000485
P.C. Struik a, S. Amaducci ab, M.J. Bullard c, N.C. Stutterheim a, G. Venturi b, H.T.H. Cromack d
Abstract
Fibre hemp may yield up to 25 t above ground dry matter per hectare (20 t stem dry matter ha−1) which may contain as much as 12 t ha−1 cellulose, depending on environmental conditions and agronomy. Its performance is affected by the onset of flowering and seed development. Effects of cultivar and management on yield and quality were tested at three contrasting sites in Italy, the Netherlands and the UK in three years, making use of standardised protocols for experimental design and research methodology. Highest yields (up to 22.5 t dry matter ha−1) were obtained in Italy when later cultivars were used. Attainable yields proved slightly lower in the Netherlands and much lower in the UK. The quality of the cellulose was relatively stable over the growing season, but lignification may proceed rapidly some time after flowering. Crop development was very rapid and crops maintained green leaf area for a long time, thus radiation interception was considerable. The radiation use efficiency changed during development. It was lower after flowering (about 1.0 g MJ−1 PAR) than before (about 2.2 g MJ−1 PAR). Growing earlier cultivars to obtain some seed set advanced the reduction in radiation use efficiency. Nitrogen proved to affect yield only slightly. A relatively small amount of fertiliser will be adequate to cover the crop’s needs. Plant density declined during growth in a site-specific manner when it was high initially. Very low plant densities may not show this self-thinning but reduced yield and (especially) quality. Final plant densities were proven to depend more on initial plant stands than expected from literature. This was true at all three contrasting sites and in the different years. Nitrogen and plant density hardly interacted within one site. Results suggest that hemp can yield large quantities of useful cellulose when ecologically adapted cultivars are sown in proper plant densities. The cultivation is environmentally friendly with little harmful accumulation or emission of chemical inputs. More research on ideotyping is required and breeding efforts should be broadened.
Introduction
Fibre hemp (Cannabis sativa L.) is grown for a multitude of end products derived from the cannabinoids, seed, fibre and wooden core. It may also be grown as a shelter crop. As a fibre crop, it is one of the oldest non-food crops world-wide (Schultes, 1970). In Europe, flax and hemp were the most important fibre crops from the 16th to the 18th century. In the 19th century the cultivation of hemp in Europe declined but recently interest has been renewed, for example in Germany, France, the Netherlands, the UK, Spain and Italy, but also elsewhere in the world.
This renewed interest is triggered by several developments. Firstly, agricultural overproduction of some commodities within the EU has stimulated the search for novel uses of land, and hemp has been rediscovered as an interesting ‘new’ crop with a large plasticity, which allows it to be grown under a wide variety of agro-ecological conditions. Moreover, it is very high yielding compared with many other crops (Van der Werf et al., 1996). Due to this plasticity it may contribute to the (economic, environmental, agronomic and social) sustainability of arable farming. Secondly, hemp is an attractive non-food crop, which produces a wide variety of renewable resources, in a way that is much more efficient than with other non-food crops. Thirdly, because it produces many components that may be useful to mankind, it is an excellent model crop for the development of multi-output systems through stepwise breakdown of biomass into several useful components. The latter characteristic puts it ahead of many other non-food crops, which usually produce only one type of raw material.
Hemp also has one major disadvantage: it is associated with use of illegal narcotics and it is impossible without detailed costly analysis to discriminate before seed set between fibre types and drug types in a hemp crop. Consequently, only cultivars which set seed in a given member state are permitted to be grown in the EU.
Growing hemp is not difficult: the crop requires little or no biocide, suppresses weeds efficiently and has limited demands with respect to fertiliser usage or crop rotation (Van der Werf, 1994). The main problem might be crop establishment: hemp is very sensitive to poor soil structure and shortage or excess of water during early stages of growth (authors’ personal experiences). The crop is high yielding, partly because it already shows full ground cover after a thermal time of about 400–450°C d (see e.g. Van der Werf, 1994, Van der Werf et al., 1995a), whereas a crop such as sugar beet requires 600–700°C d (see e.g. Smit and Struik, 1995).
Several physiological features, however, require special attention in breeding and crop management, because they are determinant for crop yield and quality. Firstly, hemp is a short day plant. This behaviour affects the crop production, because once the flowering starts the efficiency with which intercepted radiation is converted to dry matter drops rapidly (Van der Werf, 1994). It is therefore attractive to prolong the growing season by selecting later cultivars for higher latitudes (Van der Werf, 1994, Stutterheim et al., 1999). In southern Europe, genotypes should be selected with a longer critical photoperiod to profit optimally from the available growing season (S. Amaducci et al., unpublished data). The cultivars that are grown in Europe are usually of French origin, and have a critical photoperiod between 14 and 15.5 h. Their behaviour is therefore different in different eco-regions of Europe and they may not be able to make full use of the potential of the seasons all over Europe.
Secondly, there is a large degree of heterogeneity in the crop (Van der Werf et al., 1995b). This is partly caused by sexual dimorphism: the differences in rate of growth and development between male and female plants are large (Van der Werf and Van den Berg, 1995). The male plants tend to flower and senesce earlier. Also within the same sex, large plants are suppressing smaller ones and thus plant-to-plant variation can become considerable and may even result in self-thinning (Van der Werf et al., 1995c). Particularly in dense stands intra-specific competition generates a size hierarchy and thus increases variability (Van der Werf et al., 1995b). This variation may limit yields, may reduce the efficiency of resource use, and may result in variable quality.
Thirdly, growers should aim at a high cortical surface area compared to the crop volume because this will realise a high bark:core ratio. The fibre quality in the bark (mainly primary bast fibres and some secondary bast fibres) is much better than in the core (mainly high-lignin, libriform fibres) (Bedetti and Ciareli, 1976, Bosia, 1976). This can be realised by aiming at high densities (Van der Werf et al., 1995b). The bast fibre content of the stem increases with plant density (Jakobey, 1965). Moreover, a dense crop causes strong elongation of the primary bast, producing long low-lignin fibres (F. Engels, personal communication). This means that despite the self-thinning, large quantities of hemp seeds are sown to establish a very dense crop. Farmers thus have to cope with the difficulties caused by the fierce plant-to-plant competition.
Finally, the development of the individual internodes over time is important. Individual internodes produce increasing amounts of cellulose until the lumen of the cells is completely filled. There is an asynchrony in this event along the stem (F. Engels, personal communication), but for the crop as a whole there may be a clear (but difficult to assess) moment at which the cellulose yield is maximal and after which the main process that continues is the encrustation of lignin in the cell walls. Timing of the harvest in relation to environmental conditions may therefore be crucial.
These four aspects of growing hemp in Europe mean that the genotype×environment×crop management interaction is relevant and should be studied in more detail. This paper evaluates the yield formation and the effects of cultivar and crop management thereon for three years and three contrasting sites as recorded in field experiments based on the same protocol and methodology.
