Assumptions and Limitations of the Approach and Model

Every model and modeling approach has its assumptions and limitations. EKMA/OZIP is no different:

Assumptions:

VOC, NHMC, and NMHC Over the years a number of seemingly ambiguous methods to group volatile hydrocarbons have been developed. In OZIP any organic compound, including methane, may be entered as a species by including it in at least one reaction in the MECHANISM option. In addition, OZIP allows grouping of a number of species into one unique set of organics, the group named VOC (volatile organic compounds, which is usually used to describe those classes of anthropogenic emitted organics that are subject to control).
The group of individual organic species that comprise VOC is defined within the MECHANISM option; the species are named and the number of carbons in each is given. The BOUNDARY option allows the user to list the fractions of each species included in the VOC group that constitute emitted VOCs, surface transported VOCs, and aloft transported VOCs. These three sets of information allow OZIP to account for emissions and transport of the mass in the group VOC.
More important, however, is the fact that this specific method of grouping facilitates the EKMA and ISOPLETH options. By lumping all organic species subject to emission control as VOC, it is a simple function for OZIP to diminish VOC emissions in ozone attainment calculations. (Similarly, NO_x and CO are controlled in EKMA calculations, where NO_x is a combination of NO and NO₂ and CO represents carbon monoxide.)
Therefore, VOC is a specialized OZIP subset of often-used organic compound groupings called non-methane organic compounds (NMOC) and non-methane hydrocarbon compounds (NMHC). VOC is reported in ppmC. The ratio of VOC to NO_xis not necessarily the ratio of all NMOC to NOx. No effort is made in any OZIP version to combine and count all organic species as NMOC.
A number of scenario-specific assumptions are being provided to you, including composite emissions inventories, scenario-specific cost estimates, and other datasets.

Limitations:

The EKMA approach and the OZIP model are typically not used for showing regulatory compliance in today's environment. Both the approach and the model were developed in the late 1970's, and have been replaced by newer models and techniques, such as the Urban Airshed Model (UAM) and the soon-to-be-published Models-3 suite of air quality modeling tools. This scenario uses EKMA and OZIP for several reasons:

The "good" news:

EKMA and OZIP are relatively easy to use, yet comprehensive enough to still provide a realistic learning environment.
EKMA and OZIP incorporate the same chemical reaction mechanisms (Carbon-Bond IV and RADM) as are found in newer air quality models
Data from OZIP can easily be "visualized" using public-domain or commonly available visualization software, such as gnuplot or newer spreadsheets. UAM and newer models require high-end visualization software such as AVS. These packages are expensive and require high-performance workstations or supercomputers to run.

The "bad" news:

EKMA is limited to single day episodes. Newer models, such as UAM, allow for multiple-day scenarios. EKAM/OZIP require multiple single-runs, thus introducing additional errors and biases into the calculations
EKMA only predicts a single region-wide peak ozone value, and as such cannot be used to evaluate spatial (grid-based) and temporally-varying emission control measures
EKMA cannot handle complicated meteorological datasets, including topographical influences on meteorological conditions

Pitts and Finlayson-Pitts (Atmospheric Chemistry, 1986) also described some problems and limitations of the EKMA modeling approach, reproduced here:

The effects of [the EKMA] strategy will clearly depend on the particular NHMC/NO_x ratio (i.e., on which side of the ridge line a particular location is found) and thus will vary from region to region.
The isopleths are really only applicable to urban areas and ignore the impacts on downwind urban and rural areas. Thus increased NO_x will lead to increased NO₂ and hence O₃ when integrated over an entire basin. Areas somewhere downwind of the urban centers will therefore be subjected to increased O₃ levels.
This strategy focuses only on O₃ production. It negelects the fact taht increase NO_x is expected to lead to increased concentrations of species such as HONO, which we have seen, can act as photoinitiators in multi-day smog episodes. Additionally, smog chamber and modeling studies show that increased NO_x leads to increased formation of other pollutants such as nitric acid, which is a key component of acid fogs and rain. Increased formation of other secondary pollutants such as nitroarenes, which may be mutagens and/or carcinogens, may also occur. However, while these other secondary pollutants are of concern, their production cannot be quantitatively related to ozone yields.
Increased concentrations of NO₂ are expected in the urban area, and NO₂ itself has significant health and other effects.
The most severe air pollution episodes generally occur in multi-day episodes. In these episodes, characterized by very stable meteorological conditions, the peak concentrations of O₃ and other secondary pollutants generally increase from one day to the next. This is attributed to the buildup of reactive species such as HONO, HCHO, and so on, which act as initiators when the sun rises and thus increase the rate of photooxidations. Neither smog chamber studies nor modeling efforts have been very successful at reproducing such multi-day episodes. In addition, few experimental data from ambient air studies on the concentrations of such photoinitiators are available to test model predictions for these important species.
In summary then, while isopleths can be used as a semi-quantitative guide in formulating control strategies, one must exercise caution in extrapolating them to ambient air in large urban areas. In particular, the results not only in urban areas but also in downwind regions must be considered, and emphasis should be placed not solely on O₃ but on other associated co-pollutants such as HNO₃ as well. Furthermore, the impacts of proposed controls on multiday episodes must be taken into account.
Finally, the isopleths are based on data from conventional NMOC-NO_x photooxidations. In some settings, e.g. in some industrial areas, an unusual pollutant mix may exist, whose reactivity depends not only on the NMOC-NOx chemistry, but also in part on the chemistry associated with the industrial emissions. For example, where there are sources of molecular chlorine, Cl₂, a significant portion of the hydrocarbon loss may be due to attack by Cl atoms formed on the photolysis of the Cl₂, in addition to attack by OH. In such situations, the products formed may also be somewhat unusual; for example, in laboratory studies, a variety of chlorinated organics have been identified in irridated mixtures of ethene in air containing Cl₂. In such cases, isopleths developed for NMOC-NO_x mixtures may not be applicable.

Developed by