Beschreibung:
Abstract: Livestock production systems are dominant greenhouse gas sources in sub-Sahara African (SSA), including emissions of nitrous oxide (N2O) from excreta deposited onto pasture. However, a limited number of studies are available to date. Using the default emission factor (EF) from the Intergovernmental Panel on Climate Change (IPCC) guidelines, which represents a global mean value mainly based on observations in OECD countries, may cause strong uncertainties due to differences in the management practices, climate and soils between OECD countries and countries and livestock systems in SSA. To narrow uncertainties of national greenhouse gas inventories in SSA it is thus needed to develop region-specific N2O emission factors for excreta dropped on tropical pasture which do consider regional characteristics. In my thesis the calculation of N2O emission factors for excreta dropped on tropical pastures was based on a series of field trials in which I varied dung quantity, quality and urine amount. These field trials were carried out during the dry and the rainy season to also investigate how weather conditions and hydric seasons (dry versus rainy) are affecting emissions. In a final experiment I investigated the importance of soil properties on N2O emissions due to excreta deposition. Measurements of N2O fluxes were done using an automated chamber systems as well as manual chambers to measure N2O. The use of automated chambers allowed to obtain a high resolution dataset and to quantify short-term emission peaks. However, as the number of automated chambers was limited also manual chambers were used, specifically while investigating the effect of soil types on N2O emissions from excreta dropped on tropical pastures. My results show that differences in dung mass, dung quality and urine volume did not cause significant differences in terms of dung or urine N2O EF. Surprisingly, dung application only marginally stimulated the soil N2O fluxes regardless of the mass or quality of dung patches, with EFs being a magnitude lower as found for dung applications to temperate grasslands. This significant difference in N2O EFs was due to the low nitrogen (N) concentration, high carbon (C) to N ratio, as well as the rapid crust formation of dung under tropical weather and soil conditions. In contrast to my observations for dung, application of urine to tropical pastures resulted in rapid increases in N2O fluxes, though the stimulatory effect was higher in the rainy as compared to the dry seasons. Again, EFs for urine were significantly lower in my experiments as compared to experiments carried out in temperate regions. The influence of soil type on N2O emissions after excreta deposited onto pasture was also tested during my study using intact soil cores from five soil types that are wide-spread in Kenya. Through two field trials, I could show that soil properties were significantly affecting soil N2O emissions as triggered by urine but not dung additions to pastures, with highest N2O emission responses found for soils with pH values close to neutral and rich in clay. Based on my findings I calculated a specific cattle excreta N2O EF (0.13%) as well as disaggregated EFs for dung (0.05%) and urine (0.29%) for Kenya, thereby considering excreta properties, soil properties and climate (rainy and dry seasons). The cattle excreta N2O EF was calculated assuming N partitioning between urine and dung as 34:66 based on the feed diet in SSA. This N excreta partitioning is different from the IPCC 2019 refinement, which assumes a N partitioning between urine and dung as 66:34 and causes an overestimation of N2O emissions from excreta on tropical rangelands in dry climates (IPCC 2019: 0.20%; my study: 0.13% of excreta N deposited onto rangeland). In the frame of manure decomposition experiments, with dung from cattle, sheep and goat being investigated and exposed at four sites differing in climate conditions, I could show that cattle manure is decomposing faster than manures from sheep or goats. This difference in manure decomposition could be linked to manure cellulose concentration, with higher values supporting higher decomposition rates. Moreover, I found that manure decomposition rates were higher under wetter conditions, but that even after more than one year (378 days) approx. 30% of cattle manure and 50% of goat and sheep manure remained. My studies show that distinct, so far not sufficiently accounted differences in N2O EFs and manure decomposition dynamics exist for excreta dropped on tropical pastures if compared to existing studies for temperate regions. These differences are mainly due to lower quality of excreta in terms of N concentrations and composition of the manure with regard to e.g. lignin or cellulose, i.e. factors linked to feed supply and feed quality. In addition, environmental factors such as pronounced dry periods affect N2O emissions from excreta and the decomposition of manures. These pronounced differences in decomposition dynamics and N2O EFs might be addressed in further studies in order to narrow uncertainties in N2O emissions from livestock systems in the tropics and subtropics