Phytomass productivity of cutting and grazing grasslands with special reference to small-scale spatial variation in plant nutrient resources

Sown grasslands are generally managed via one of two methods: cutting or grazing. In cutting grassland (CG), fields are uniformly plowed, and subsequently, one or several mixed pasture plant species are uniformly seeded, managed, and cut. In grazing grassland (GG), several pasture plant species are uniformly seeded at the establishment stage of grassland, and livestock is introduced to the grassland after a certain volume of vegetation is achieved. In GG, nutrient cycling occurs via grassland vegetation, intake and excretion by animals in grassland, and grassland soil. The objective of this study was to build a mathematical model to investigate which of CG or GG produces more phytomass under an equal level of plant nutrient resources (PNR) such as mineral nitrogen, available phosphorous, potassium in soil and the derivatives in plant bodies. In the model, we assumed that the relationship between PNR and plant production per unit ground area (plot) in CG and GG would follow the Mitscherlich law; the frequency distributions of the PNR level in soil and vegetation per plot would follow a gamma distribution in GG, but a spatially uniform distribution in CG; PNR level in soil of CG would be equal to the mean PNR in soil of GG; and additional PNR in the form of fertilizer would be applied to achieve equal PNR levels between CG and GG ecosystems after cutting and grazing. The model outputs indicated that (1) under conditions in which mean PNR levels were equal between CG and GG, phytomass in GG after grazing was greater than that after cutting in CG if the spatial variation in PNR is not highly heterogeneous in GG, however, smaller than that just before cutting in CG for any spatial heterogeneity in GG; (2) resupply of PNR (i.e., fertilizer) after cutting or grazing practices (to maintain equivalent PNR levels between both grassland ecosystems) produce a higher phytomass in CG compared to GG. The reason is that CG receives much resupply of PNR to supplement losses by natural (such as runoff and infiltration) and artificial (cutting/carrying-out of phytomass) losses, but in GG only the natural loss is resupplied because there is no artificial loss; (3) the high spatial variations in PNR generally play a negative role to determine phytomass production (phytomass productivity, or phytomass yield) in GG; (4) the efficiency of PNR resupply to phytomass production (output/input ratio of PNR) is higher in GG, particularly under the condition with highly spatial variation in PNR, than that in CG.