Scale-dependent effects of plant and animal functional characteristics on nutrient cycling in pastures
Final Report Abstract
Large herbivores play an important role in the nutrient cycle of natural and agriculturally used grasslands. The large majority of mineral nutrients these animals ingest are returned to the soil in form of urine and dung. This ‘herbivory cycle’ is usually faster than the alternative cycling of nutrients through decomposing plant litter. However, urine and dung return very high rates of nutrients to only small parts of the grazed area. In addition, the main plant nutrients, nitrogen (N), phosphorus (P) and potassium (K), are excreted to different proportions in either urine or dung, depending on their concentration in the animals’ forage. The stoichiometry of nutrient inputs at excreta patches therefore differs strongly from the average nutrient stoichiometry in the plant biomass. The efficiency with which plants can use nutrients from animal excreta should depend on the size of excreta patches, the degree of stoichiometric mismatch, and on the proportion of different plant functional groups. Whether the excreta-derived nutrients taken up by the plants are then circled through the herbivory or the litter-decomposition cycle will depend on the degree to which animals prefer or avoid areas affected by excreta patches for forage intake. As a consequence of the complex interactions between soil, plants and grazing animals, models and theories considering only average nutrient flows will fail to capture the effect of localized nutrient input in excreta patches on nutrient cycling and plant stoichiometry. Our study combined investigations at the scale of individual excreta patches with measurements of biomass and nutrient flows at the systems scale. We established a grazing experiment in which paddocks of 0.5 ha were grazed by either cattle or sheep, and where the grassland vegetation was either not manipulated (diverse) or was treated with a herbicide against dicots (grass-dominated). In this experiment, we measured biomass productivity, biomass and nutrient intake by animals and estimated nutrient return in either urine or dung over three years. We also measured the effect of cattle and sheep urine and dung patches on plant productivity and nutrient concentration as well on animal biomass intake directly in the grazing experiment. We complemented these measurements with more detailed ones on simulated excreta patches of the same animals in an adjacent grazing simulation experiment. We found surprisingly small plant responses to excreta patches: biomass productivity increased only at urine patches, and nutrient concentrations were largely unaffected by excreta patches. Apparent nutrient recovery exceeded 10% only on areas affected by cattle urine patches. In spite of lesser nutrient application rates at real and simulated sheep compared to cattle excreta, biomass responses differed little between excreta patches of the two animal species. Cattle avoided grazing both recent (c. 40 days) and older (c. 8 months) dung patches, sheep only recent ones, and these to a lesser degree. By contrast, neither more recent nor older urine patches were avoided by either animal species, leading to increased biomass and mineral nutrient intake at these patches. Although the diverse and grass-dominated vegetation differed considerably in the proportion of plant functional groups (proportion of grasses: 66 vs. 93% in the grazing and 51 vs. 91 in the simulation experiment), we found few differences between them in their general properties or in their reaction to excreta patches. Variation between replicates was surprisingly large for most variables, highlighting the difficulty of studying small-scale processes in a botanically diverse grassland with a past history of spatially heterogeneous nutrient input through excreta. At the paddock scale, only 20% of the area would by affected by animal excreta under cattle grazing, but 32% under sheep grazing. Scaling up from patch-scale results, excreta were calculated to increase biomass increase biomass productivity by 3% during the next stocking period and to decrease animal biomass intake by 3% under cattle grazing. Under sheep grazing, productivity and intake would increase by 2% and 8%, respectively. The unexpectedly small short-term effects of animal excreta we found contrasted with pronounced long-term effects in the form of large-scale nutrient redistribution which we found in our grazing experiment and adjacent cattle grazed pastures. In both experiment, extractable P and K concentrations differed, on average, by a factor of 1.8–5.5 between the paddock areas with the lowest and the highest density of excreta.
Publications
- (2015) Zusammenhang zwischen räumlichen Mustern des Weideverhaltens, Exkrementstellenverteilung und großräumiger Heterogenität der Bodennährstoffgehalte bei Rinder- und Schafbeweidung. 59. Jahrestagung der Arbeitsgemeinschaft Grünland und Futterbau, 27.–29.08.2015, Aulendorf. S. 204–207
Tonn, B., Stünke, R., Scheile, T., Isselstein, J.
- (2018) Effect of plant and animal functional traits on nutrient cycling in low-input pasture. Dissertation Georg-August-Universität Göttingen
Scheile T.
- (2018) Herbage biomass and uptake under low-input grazing as affected by cattle and sheep excrement patches. Nutrient Cycling in Agroecosystems 112: 277–289
Scheile T., Isselstein J., Tonn B.
(See online at https://doi.org/10.1007/s10705-018-9945-3) - (2018) Räumliche Präferenzen weidender Rinder verursachen Nährstoffumverteilung auf mehreren Maßstabsebenen. 62. Jahrestagung der Arbeitsgemeinschaft Grünland und Futterbau, 30.08.-01.09.2018, Kiel. S. 253–257
Tonn, B., Hüppe, C., Kunze, N., Raab, C., Isselstein, J.
- (2019) Sward patterns created by patch grazing are stable over more than a decade. Grass and Forage Science 74, 104-114
Tonn B., Raab C., Isselstein J.
(See online at https://doi.org/10.1111/gfs.12389)
