Different K sources combined with MBM gave an increased yield ofbarley compared with the use of MBM alone, whereas supply of K in addition to MBM did not significantly influence the yield of wheat. The soils used in this experiment had higher concentrations of readily available K than planned. The soils used in this experiment were within the group of sandy soils with a low content of acid-soluble K (acid-soluble K minus readily available K, 8-24 mg 100 g-1) as described by 0gaard

et al. (2002). 0gaard et al. (2002) found no yield response to applied K in perennial grass leys when the amount of readily available K exceeded 10 mg (100 cm3)-1 (8 mg 100 g-1 and bulk density 1.25 mg m3). In the present experiment the readily available K level was not significantly lowered owing to only a small negative K balance in the MBM treatment (Table 3.6), and was clearly above the expected minimum readily available K level based on texture (0gaard et al. 2002). Sufficient K in the soil is therefore the most probable reason for not finding a significant effect of K supply on K uptake in wheat grain. Using sand of the same origin as the Elverum sand in this study, Haraldsen et al. (2010) found that application of 80 and 160 kg N ha-1 in MBM alone and other organic N fertilizers with low K content gave signifi­cant K deficiency and reduced barley yield compared with the use of the same amounts of mineral NPK and a liquid anaerobic digestate based on source-separated household waste. That batch of sand had a considerably lower content of readily available K than the sand used in this study, and had a readily available K level close to the expected minimum level as described by 0gaard et al. (2002).

Although the effect of K supply was not large in this study, the mixture of MBM and BWA gave at least the same yield as mineral NPK or supply of K, Mg and S in addition to MBM. The concentrations of potential plant-available K in the crushed rock powder types was too low for making commercial fertilizer products, and the amounts used in this experiment did not influence K uptake and did not cause any change in the level of readily available K or nonexchangeable K in the soils. Combining N-rich waste (human urine) and wood ash gave more biomass of red beets than mineral fertilizer (Pradhan et al. 2010), and Kuba et al. (2008) as well as Bougnom et al. (2010, 2011; see Chap. 7) found positive effects on growth of mixing wood ash with compost. These examples indicate a potential of combining

N-rich and K-rich waste streams as fertilizer or soil amendments. Especially for use in organic cropping there is demand for recycled NPK fertilizers, which have predictable effects.

The Ca-to-K ratio of the BWA used in the present experiment was not optimal in the mixture with MBM, and the ash also had a higher concentration of Ni than allowed in materials that can be used as fertilizers for agricultural crops in Norway (maximum quality class II according to the Norwegian Ministry of Food and Agriculture 2003). BWA Akershus (Table 3.8) has a low Ca-to-K ratio and can be categorized as quality class I according to the Norwegian Ministry of Food and Agriculture (2003). Because a smaller amount of BWA Akershus than of the ash used in this study is needed to obtain the same NPK ratio, it is expected that a mixture of MBM and BWA Akershus will cause a smaller pH increase than the mixture of MBM and BWA in this study. The pH increase of 0.5 after a single application of 1,200 kg BWA ha-1 represented an estimated addition of about 700 kg CaO equiv ha-1, and caused a significant increase in the amount of readily available Ca in the soil (Table 3.7). According to Franzefoss (2007) natural acidification (by leaching and acid precipitation) represents an annual demand of lime of 100-200 kg CaO ha-1 in eastern Norway and 200-400 kg CaO ha-1 in coastal areas in western and central Norway. A liming effect of 200-300 kg CaO equiv ha-1 will be suitable for a fertilizer that is to be used annually for cereals. The ash of cereal waste from a milling plant had interesting properties as a PK fertilizer, as the levels of heavy metals were lower than the limits for use on agricultural land (Norwegian Ministry of Food and Agriculture 2003). For further development of organic mixtures based on MBM plus BWA, ash of similar quality as BWA Akershus (Table 3.8) will be selected. The challenge is to find sufficient quantities of BWA with similar properties, in order to establish commercial production of organic NPK fertilizer by combination of waste streams.

2.3 Conclusions

A mixture of MBM and BWA with a low Ca-to-K ratio and lower concentrations of heavy metals than the limits in the governmental regulations for organic fertilizers may give a NPK fertilizer with reliable effects of all three major plant nutrients. Such a fertilizer, based on combining waste streams from society, will be more complete fertilizer product than the raw materials represent and is of special relevance and interest for organic cropping. The BWA used in the mixture with MBM in the present experiment did not have optimal chemical properties, and gave too high an increase in pH for annual use for cereals. Further investigations on optimization of mixtures of MBM and BWA are needed before such fertilizer is ready for commercial production.

Acknowledgements The pot experiment was a part of a project supported by the programme “ORIO-Organic Waste Products and Recycling of Resources” and Norsk Protein AS (grant no.

403). Ellen Zakariassen is thanked for skilful technical assistance and experimental work. Additional analyses of ash and further work on the fertilizer concept were carried out as a part of WP 1.4 “Residues upgrading and use” in CenBio, Bioenergy Innovation Centre (http://www. cenbio. no), which is supported by the Research Council of Norway, Norsk Protein AS and Akershus Energi AS.