6.2 The preconditions for production of acceptable raw material
6.2.1 Possibilities to enhance yielding ability
Harvesting time
The harvesting system greatly influenced yield capacity of the species. Tall fescue was favoured by a harvesting system of two cuts (first cut at flowering and second in October) to a greater extent than reed canary grass. The superiority of the two-cut system compared with three or one cut systems was reported in earlier studies (Nissinen and Hakkola 1994, Pahkala 1997).
Two cuts of reed canary grass resulted in lower yields, especially on organic soil. In the years of experimentation, regrowth DM yield of reed canary grass comprised 5% to 32% and tall fes-cue 9% to 25 % of total harvested biomass. The regrowth ability was highly dependent on weath-er conditions during the post-harvest pweath-eriod. Pre-cipitation after the first cut contributed marked-ly to the regrowth. In the studies of Mason and Lachance (1983), performed in Quebec,
Cana-da, 43% of total biomass of tall fescue was from regrowth and for reed canary grass it was 32%.
The total DM yield of two cuts of reed canary grass and tall fescue however tended to decrease after the first year of harvest. Furthermore, for papermaking purposes the system of two har-vests, combined with drying the biomass, is like-ly to be too expensive even when the combined biomass from two cuts would be high during the first harvest years.
High DM yields of reed canary grass and tall fescue were obtained when the crops were har-vested in autumn at the seed ripening stage.
However, the DM content was less than 45% for both species, and the harvested biomass needed to be dried to reach a DM content of 85% before storage (Hemming et al. 1996). Reed canary grass gave the highest yields when harvested only once either in autumn or the following spring. However, the annual variation in yield was greater when harvested in autumn. Accord-ing to the studies of Mason and Lachance (1983), total annual yields of reed canary grass and tall fescue increased when the first harvest was de-layed. In their study, reed canary grass was su-perior in yielding capacity to timothy (Phleum Fig. 18. Flow-chart for the selection of fibre plant.
pratense L.), tall fescue and Kentucky bluegrass (Poa pratensis L.). Moreover, in Swedish stud-ies reed canary grass had the highest yield po-tential when it was compared with brome grass (Bromus inermis Leyss.), tall fescue, cocksfoot (Dactylis glomerata L.) and timothy (Wisur et al. 1993).
Of the crop species studied, reed canary grass was favoured most by a spring harvest. The first spring harvest was done two years after sowing and it resulted in 4–7 t DM ha-1. The following spring harvests gave yields of 6–8 t ha-1 in most years. DM yields increased with delayed harvest and increasing age of the ley compared with au-tumn harvests. This was also demonstrated in Swedish studies (Olsson 1993, Andersson 1994, Landström et al. 1996). Spring yields of reed canary grass were 6 to 10 t ha-1 annually and on organic soil even higher than 10 t: twice the an-nual yield of birch forest (4–7.5 t of DM ha-1), the maximum annual growth of which is 8–15 m3 (Ferm 1993). When reed canary grass was harvested in spring, the yield remained constant from the second year ley throughout eight years, but some variation caused by weather conditions was recorded in those years. The difference be-tween the average yield at autumn and spring harvests was not significant during the eight years because of the first harvest year, which was associated with the lowest yields in every ex-periment. Reed canary grass would benefit from spring harvesting with good persistence of the stand and with small variation in yield. Howev-er, its productivity in the UK was 7–12% less than that of Miscanthus and switchgrass (Chris-tian et al. 1999).
For tall fescue, delayed harvesting in spring resulted only in 37–54% of DM yields harvest-ed during the previous growth period. Low spring yields of tall fescue were associated with the growth habit of the species: a plant stand of tall fescue consisted mostly of leaves, and the crop was flattened tightly along the ground un-der snow cover. In spring, it was impossible to lift the lodged biomass with harvesters, and the plants were partly rotten. After spring harvest, tall fescue produced much fewer stems and
pani-cles than the plots harvested in summer or au-tumn. As a tussock grass, tall fescue may be more prone to damage during an early spring cut than reed canary grass. The reason for the enhanced formation of reproductive tillers could be also the lack of light in late summer of the previous year, when the tillers of tall fescue were initiat-ed or when the tillers were too young in autumn to respond to low temperature induction (Hare 1993). The enhanced effect of shading on tiller formation was reported for cocksfoot (Hare 1994). Because of poor biomass and straw yield, tall fescue was considered not to be suitable for spring harvesting.
The spring harvest duration of the present studies was about 10–15 days, when the mois-ture content of the grass was between 10% and 15%. The moisture content decreased to this lev-el even before ice had disappeared from soil, i.e.
in south Finland in late April. High moisture content of the soil or rain showers occasionally delayed harvesting for weeks. Hemming et al.
(1996) estimated that during the harvest period of two weeks in spring there would be 6–9 days when the weather conditions favour harvesting in Finland. It was also obvious that the harvest-ing has to be done before the new tillers are 15–
20 cm high. If the emerging shoots are taller at harvest, they may drastically reduce quality of the harvested biomass because of increasing moisture and mineral content of the biomass. In literature this is called the “harvest window”
problem. It is described for Miscanthus in the Netherlands (Huisman 1994, Venturi and Huis-man 1997). It is the period between the possible start of the harvest, defined by the decreasing moisture content of soil and biomass in spring time, and the end of the harvest period when the tillers grow too long.
Effect of nutrients
Increase in the supply of mineral nutrients from the deficiency range increases the growth of crop plants. The positive yield response to nitrogen application is well known for grasses (MacLeod 1969, Hiivola et al. 1974, Allinson et al. 1992, Gastal and Bélanger 1993). In this study, the
in-creased fertilizer application rates usually result-ed in increasresult-ed total yield of reresult-ed canary grass, when the biomass was harvested at the green stage in summer or in autumn. However, on a clay soil the yield increase with increasing fer-tilizer application rates was obvious also in spring up to 150 kg N ha-1. A rate of 200 kg N ha-1 did not improve yield beyond that promoted by 100 kg N ha-1. On organic soil, the spring yield response to increasing fertilizer rates was smaller than on clay soil and applications in excess of 50 kg N ha-1 did not result in higher DM yields.
The results show that growing reed canary grass on clay soil requires more N fertilizer to reach the same DM yield as growing the crop on or-ganic soil. Overuse of fertilizers may be uneco-nomic and cause environmental problems for farmers. Fertilizers represent the principal cost in cultivating reed canary grass when the rate of 70 kg N ha-1 is exceeded (Maunu and Järvenpää 1995). When reed canary grass was grown on clay soil and harvested in spring the economic optimum for the fertilizer application rates is likely to range from 50 to 100 kg N ha-1. Grow-ing reed canary grass on organic soil for paper-making might be advantageous as less fertilizer is required. However, more research is needed to have further long-term information on devel-opment and yield formation of reed canary grass on organic soil.
Yield of tall fescue was not enhanced at the highest fertilizer application rate of 150 kg N ha-1 during the two first years of harvest. On clay soil there were hardly any differences among the various treatments beyond 50 kg N ha-1 when tall fescue was harvested at the green stage. On or-ganic soil, the yield response for the highest rate was recorded only at the seed stage in 1995.
However, the results of Moyer et al. (1995) from young tall fescue swards showed yield increase by 53% as N application rate increased from 13 to 112 kg N ha-1, and by 69% as the rate increased from 13 to 168 kg N ha-1. At delayed harvesting in spring the differences in yield among the treat-ments seemed to be inconsistent and not statis-tically significant. If tall fescue is used as raw material for papermaking, fertilizer application
rates higher than 100 kg N ha-1 are not likely to improve yield.
Harvest losses
Stubble height markedly affected the harvested DM yield of reed canary grass. When grass was harvested at 5 cm instead of 10 cm, the DM yield was on an average more than 30% higher. The reasons for such a high yield difference may be several. When cut at the height of 10 cm versus 5 cm, the loss of the total biomass measured as the weight loss of 5 cm straw was 5.8%. The higher harvest losses at a stubble height of 10 cm caused by lodging were also evident, but were not measured in this study. In the study of Hor-rocks and Washko (1971), plants cut in spring leaving 10 cm stubble instead of 4 cm produced the same number of tillers, but the weight per tiller after a 4 cm cut was about 60% higher than that after a 10 cm cut. In the present study, the number of stems counted in autumn and spring was not affected by stubble height. However, the weight of individual tillers was not measured and the high yield, when cut at 5 cm, remained part-ly unexplained. Harvesting losses resulting from the harvester and baling would be high, but un-der favourable conditions were less than 15%
(Hemming et al. 1996). The summary of the fac-tors affecting harvested DM yield of reed canary grass is presented in Fig. 19.