The concentrations of many atmospheric trace gases are increasing, and the implications of the increases in some of these gases, such as carbon dioxide, are widely appreciated. There is not as wide a concern, however, about increasing concentrations of tropospheric ozone and the reactive nitric oxides, despite clear evidence that these gases are important to atmospheric chemistry, terrestrial ecology and human health.
Tropospheric ozone is both a potent greenhouse gas and one of the major atmospheric pollutants. Ozone is produced though a series of reactions involving both hydrocarbons and NOx (NO + NO2). In many non-urban settings, NOx concentrations control whether ozone is produced or destroyed during these reactions. At NOx concentrations below approximately 50 parts per trillion by volume (pptv), the series of reactions involving NOx, atmospheric oxygen and hydrocarbons result in net ozone destruction. When NOx concentrations rise above this "critical value," the reactions lead to net ozone production. Measurements of the relationships between ozone and NOx concentrations in the atmosphere generally result in positive correlations between the two gases, until very high levels of NOx are reached (greater than ~0.5 ppb) when all of the available OH is consumed and net ozone production declines. The interdependency between NOx concentrations and ozone production results in a correlation between NOy (all N oxides except for N2O) deposition and ozone concentrations (figure 8.1). High levels of NOy deposition are generally found in industrial/urban areas with high NOx emissions and high ozone concentrations.
NOx is emitted by a variety of pathways including combustion, lightning, aircraft, soil microbiological reactions and biomass burning. The relative importance of these different production pathways varies both spatially and seasonally. In urban areas the combustion pathway generally dominates while in rural areas, soil and biomass burning emissions of NOx can be the major source. Natural sources of NOx, lightning, biomass burning, and soil emissions are the most difficult to quantify and thus have the greatest uncertainty associated with them. Comparisons of emissions amongst 3-D chemical transport models shows that they vary by 2 to 3 fold (table 8.1).
In addition to controlling concentrations of ozone, NOx emissions represent a major vector for the transport of reactive nitrogen between ecosystems. Because nitrogen is a limiting nutrient in many ecosystems, the increased deposition of nitrogen may result in increased plant growth and thus increased sequestration of atmospheric CO2 in the biosphere. This effect has been modeled by Townsend, Holland and Braswell using a simple terrestrial ecosystem model known as N-Dep.
The N-Dep model is a perturbation model that is run in conjunction with a full ecosystem model (CENTURY). N-Dep calculates the effects of increased nitrogen availability on plant and microbial dynamics. Using stoichometric relationships between carbon and nitrogen and assumptions about the allocation of carbon in plants, N-Dep can be used to estimate the amount of carbon that is fixed through photosynthesis as the result of nitrogen deposition to the terrestrial biosphere. In addition, the model can be used to track the residence time of the carbon in the biosphere following fixation and the relative importance of changes in plant allocation patterns to the retention of carbon in terrestrial ecosystems.
N deposition fields to drive N-Dep were generated using five different 3-dimensional (3 -D) chemical transport models. Previous work had demonstrated the importance of both the quantity and spatial distribution of the deposited nitrogen. The distinct spatial distributions of the N deposition reflected the different modeling approaches including differences in sources, transport, chemistry, and deposition parameterizations. Thus the results reflect uncertainties in our understanding of all the processes involved. The five 3 -D chemistry transport models used were: ECHAM, GCTM, GRANTOUR, IMAGES, and MOGUNTIA.
In addition to the spatial distribution of deposition, it is important to consider the spatial distribution of the ecosystems receiving the nitrogen. Only natural ecosystems are capable of responding to the added nitrogen by storing additional carbon. Forests, which store carbon in wood at a C:N ratio greater than 150, can store more than an order of magnitude more carbon per mole of nitrogen deposited than grasslands, which store carbon in soil organic matter at a C:N ratio of 12:16. The difference between agricultural and natural systems is due to the large amounts of nitrogen fertilizer already added during agricultural activities. Because there is already a great deal of nitrogen in agricultural systems, it is unlikely that any additional nitrogen in deposition will significantly increase carbon storage.
The deposition of N on land, oceans and natural vegetation was different between the five atmospheric chemistry models compared in this experiment. While all the models estimated maximum deposition rates near 50 degrees north latitude, there was considerable North and South spread. In addition, the partitioning of deposition between oceans, land and natural vegetation varied considerably between the models. As a result, the estimated carbon uptake due to nitrogen deposition for the five atmospheric chemistry models ranged from 0.45 to 0.6 Pg C yr-1 (Table 8.2).
It is also important to note that none of these models included a consideration of ammonia deposition which could add significantly to the total amount of nitrogen deposited on terrestrial ecosystems. Total ammonia fluxes are estimated to be currently 45 Tg N yr-1 by Dentener and Crutzen for the MOGUNTIA model. In comparison, total annual NO X fluxes are estimated to be 35 Tg N yr-1 (Table 8.3). Inclusion of ammonia in the carbon storage calculation increased the globally integrated sink to 1.1-1.6 Pg C yr -1 depending on the assumption used.
The inclusion of ammonia deposition into annual estimates of nitrogen deposition could significantly alter estimates of carbon storage in natural ecosystems. In addition, ammonia fluxes represent another form of biosphere/atmosphere nitrogen exchange that occurs in addition to the changes in the global fluxes of the nitrogen oxides. Taken together, these changes in the global nitrogen cycle represent a dramatic human perturbation to one of the most important global biogeochemical cycles. Improved understanding of the ecological controls and atmospheric dynamics of these changes in the nitrogen cycle will be critical to assessments of the ecological, atmospheric and human implications of an altered nitrogen cycle.