In summary, results of this study show that the average reduction in GWP due to CO2 sequestration by European forests is set off by N2O emissions for ~10%, whereas the net uptake of CH4 is negligible compared to CO2 sequestration. On average, the effect of nitrogen deposition on increasing N2O emissions is estimated at ~10-15% of the increased CO2 sequestration, implying that the positive effect on increasing the vegetation CO2 sink is also much larger than the effect on increasing the N2O emissions. The effect of nitrogen deposition on reduced CH4 uptake is very small and highly uncertain. In this approach, the impacts of indirect N2O emissions and additional N2O emissions following forest disturbances are not accounted for, but the same holds for off-site carbon sequestration. Both aspects might nearly double the real emissions or sequestration, leading to a similar percentage in which N2O emissions set off CO2 sequestration.
Although the general conclusions are robust, the complexity of the processes involved and the large scale and diversity of the forests indicate that the impact of environmental factors on the emissions and sinks of CO2, especially on N2O and CH4, remains highly uncertain. The research to improve the estimates of GHG emissions and the nitrogen deposition impacts on those emissions is an integrated interdisciplinary approach consisting of: (i) field measurements (including satellite observations above large forest areas) in extensive continental networks including measurements of N2O and CH4; (ii) process studies on the interaction of carbon and nitrogen in the laboratory; (iii) further development (using results of process studies) and testing (using field measurements) of detailed process-oriented biogeochemical models at plot scale; and (iv) upscaling of results by developing and using empirical models and simplified process-based models, in connection with results from field measurements and detailed model studies. Such an approach is foreseen in the Nitro Europe Project, which will be carried out during 2006-2011. Using such an integrated approach enables identification of knowledge gaps and/or ecosystems that are underrepresented, and the approach towards these gaps should be prioritized to maximize the effectiveness of the research investments.
Regarding field measurements, it is important to perform year-round measure ments, preferably for a longer time period. By increasing the number of measurements to cover at least 1 year, and with data on weather and environmental conditions, the empirical models can be significantly improved. A better understanding of the basics of N2O production as a function of soil moisture, temperature and nitrogen availability is needed to improve plot-scale models and estimates of the European greenhouse budget, and subsequently to devise strategies to minimize emissions. Such information can be derived from process studies using modern stable isotope techniques and advanced soil incubation techniques.
Apart from field and laboratory studies, the uncertainty in GHG emissions and nitrogen deposition impacts on them may be significantly reduced by the further development and testing of process-oriented models at a plot level. Recently several biogeochemical models such as the DNDC (Li et al., 1992) and CENTURY (Parton et al., 1996), which are able to mimic the complexity of processes observed in field and laboratory studies, have been developed not only to simulate GHG exchange but also for nitrate leaching or NH3 volatilization. They are already capable of simulating a wide variety of ecosystems, and such models can be used to extrapolate results to other combinations of weather and climate, land use, soil type and (forest) management practices by running them for longer time periods (e.g. 10-30 years). However, it should be realized that these simulation models are able to 'explain' at best only 50% of the measured daily variation in N2O fluxes over time.
One of the most challenging aspects, also attempted in this chapter, is to extrapolate results from the plot scale to the regional, continental and even global scale. Application of a detailed model, such as DNDC, implies the use of many assumptions and generic estimates regarding inputs, management and model parameters. This may completely offset the advantage of using validated detailed plot-scale models on a large regional scale. The data uncertainties (model inputs and model parameters) involved in this process can be ascertained by applying such a model on a plot with detailed input and output data. Comparison of the model results using these input and output data on a national or continental scale gives information on the loss of reliability caused by upscaling because of less detailed data information.
Considering the problem of data availability in detailed models, it might be just as reliable to use results of simulations of process models to generate artificial datasets from which simple 'empirical' relationships are derived, such as Eq. 17.7. These simple relationships can then be used to derive large-scale soil emissions, as done for N2O. In this context, a relatively simple process-based model (e.g. INITIATOR2; De Vries et al., 2005a) can also be a compromise between a simple empirical model and a detailed process-oriented model approach (e.g. DNDC). This implies the need for further development and application of both detailed and simple process-oriented models assessing the loss of reliability in emissions estimated caused by upscaling because of less detailed data information by comparing the results of both types of models using the approach mentioned earlier.
Finally, it is important to realize that for IPCC estimates only the anthropogenic CO 2, N2O and CH4 emissions are relevant. So the natural CO2, N2O and CH4 source strength itself is not relevant, except for the deviation from the natural base line. IPCC estimates of biogenic GHG emissions thus focus on sources that are directly influenced by human activities, such as fertilizer-induced N2O emissions from agricultural soils while N2O emissions from forests are accounted for as indirect emissions (currently only related to effects of agricultural NH3 emissions and not to NOX emissions). In this context, it is important to establish an improved linkage between anthropogenic nitrogen deposition and atmospheric nitrogen emissions by using improved large-scale atmospheric dispersion models. This aspect is also foreseen in the Nitro Europe project. Only when the information about the impact per unit of nitrogen deposited on the CO2, N2O and CH4 emissions - positive or negative (sink) - is properly linked to nitrogen emitted by anthropogenic sources can the potential impact of changes in nitrogen emissions on GHG emissions be quantified.
Was this article helpful?