Effect on nitrogen losses from a sheep grazing system by the randomised distribution of excreta

Abstract

Nitrogen excreted by ruminants in the form of dung and urine posses a potential source of pollution. The aim of this work was to assess the effect on the amount of nitrogen (N) that is lost from a sheep grazing system by leaching, volatilisation and denitrification when excreta is randomly rather than uniformly distributed over an area. The single heterogeneous paddock (SHP) option within the dynamic biophysical model EcoMod was used to allocate excreta to a grazed one hectare area. The SHP approach allows a paddock's soil nutrients, pasture production and species composition to be simulated independently within defined areas. Simulations used climate data for a 100 year period from the Ellinbank Research Farm in south-eastern Victoria, which has a temperate climate and Red Mesotrophic Haplic Ferrosol soil type. The pasture was perennial ryegrass (Lolium perenne)/white clover (Trifolium repens), which was grazed every 60 days by 45kg wethers (maintaining their liveweight). The average annual amount of N lost by leaching, volatisation or denitrification (kg N/ha) and their total amount per kg N applied and fixed by legumes were estimated for the most recent 40 years. It was assumed that one animal would produce 0.5m 2 of excreta each grazing day. Therefore, for a range of stocking densities from 200 to 2000 sheep/ha, excreta were allocated either uniformly to the whole 1 ha paddock or randomly to 1 to 10% of the grazed area (representing 200 to 2000 sheep/ha). The range of stocking densities was used to represent an increasing amount of excreta being applied within the grazing system. The model estimated that the average annual N losses from the system ranged from 91 to 67 kg N/ha and 92 to 95 kg N/ha from leaching, 3 to 21 kg N/ha and 3 to 19 kg N/ha from volatilisation, 5 kg N/ha and 5 to 4 kg N/ha from denitrification for low to high stocking densities for uniform and randomly distributed excreta respectively. This study showed that the random distribution of excreta significantly influences the annual N losses through leaching and denitrification from a grazing system. When the excreta from 1200 or more sheep/ha was distributed to 7% or more of the grazed area, the proportion of total N losses per year were on average significantly higher when excreta was distributed randomly compared to uniformly. Further development of the SHP approach to enable the random allocation of excreta in grazing systems with low stocking densities rather than 'mob' stocking would allow models to better apportion N losses, even though annual N losses may not be significantly different too if excreta was assumed to be uniformly distributed.