13.3 Emissions from an Unknown Location




In this section we will use measurement data and the adjoint dispersion calculation (backward integration from the measurement location) to estimate the source location and magnitude of the emissions. Because of the uniform flow conditions during CAPTEX, for this and several of the following sections, we will use a hypothetical set of measurements at eight stations in the CAPTEX sampling region. Meteorological conditions varied sufficiently over a 60 h time period starting 1 September 0000 UTC to make an interesting example.

  1. We will only use the 3-hour duration sampling data that showed non-zero values from 00Z September 1st through 12Z September 3rd. The map shows the average concentration over the 60 h period at those stations. The object of this section is first to determine the source location and second to estimate the emission rate.

  2. The first step is to configure the menu in a way that can be used to predict the concentrations if the source location were known. In the Concentration / Setup Run menu, set the start time to 83 09 01 00 and run duration to 60 hours. The only meteorological file that contains data for 1-3 September is RP198309.gbl. In the Pollutant, Deposition and Grids menu set the emission rate to 0.0, it will be replaced in the pre-processing step, and center the concentration grid over the sampling domain at 41N 73W. A grid resolution of 0.25 should be sufficient. Define a unique output file name such as cnum. Because of the pre-processor, explicitly set the start and stop times for the emissions and sampling periods.

  3. Because we will be running the model with a coarse resolution meteorological data grid, fewer particles will be needed. Open the Advanced / Configuration Setup / Concentration / Menu #4 and reduce the particle release number to 5000 to speed up the calculations. Save to close and exit all menus.

  4. Now press the Special Runs / Geolocation menu tab to open another data input menu which is composed of four steps. In the first step you need to define the data file (DATEM format) that will be used to create the individual backward CONTROL file from each measurement. The hypothetical data for these examples, hypo_meas.txt, can be found in the CAPTEX directory. Use the browse button to find and select this file.

  5. In step 2 leave the defaults (measured value in the numerator of the source rate) and press the Execute button to create 29 CONTROL files in the working directory, one file per measurement. As an example, file CONTROL.023 contains the information from station 006, where the sample was collected from 21Z-00Z on the 2nd. Note this simulation, like all others, starts at 12Z on the 3rd, running for -60 hours (negative = backward), with the particle emissions occurring only from 00Z to 21Z. In this way all model simulation files contain the same number of output time periods. The emission mass equals the measurement. Zero measurements are assigned a nominally small emission rate (10-11). Execute Step 3 to run the dispersion simulations.

  6. When the calculations have completed, replace the Time aggregate default of one in Step 4 with 20 before pressing Execute so that the output will consist of one frame, the mean of all 29 simulations over all 20 three-hour sampling periods, representing the weighted source sensitivity function. The resulting source location pattern shows a maximum value near 43N 75W. These patterns are weighted by the measurements so that nearby samplers or those near the plume centerline, provide the greatest weight to the pattern.

  7. Before continuing, save the CONTROL and SETUP.CFG files to geol_control.txt and geol_setup.txt for future reference.

Now that the source location is known, the remaining question is what emission rate is required to match the measured concentrations?

  1. As noted in an earlier section, the model computes C = D Q, where C is concentration, Q is the emission value (mass) and D is the model dilution (m-3) factor. Although not precisely correct, we must assume that D forward, the dilution factor from the source to the receptor, is the same as D backward, the dilution factor from the receptor to the source. Then the source term can be estimated from measurements (M) as the ratio of M/D:
    • assume C=M
    • then Q=M/D.
    However, the model only computes D in the numerator, not its inverse. If we set the source term as 1/M for the calculation, the model will compute D/M which equals 1/Q.

  2. You may want to run Special Runs / Ensemble / Cleanup before starting this subsection. To avoid confusion, it may be best to rename the output file from cnum to cden (for denominator). To perform the 1/M calculation, it is necessary to repeat Steps 2 and 3, but this time selecting the Inverse radio-button. An example is shown in file CONTROL.023 for station 006, where the emission rate is the inverse of the measured value (1/846). Only non-zero measurements are used in this approach.

  3. Rerun the calculations and display. The smallest values on the graphic correspond to the larger emission rates, that is the further away you get from the actual source location, correspondingly larger emissions (inverse of the graphic values) are required to match the measured values. The exact value near the previously estimated source location (43N 75W) is difficult to determine from the graphic. Use Convert to Station on the file cmean to determine that the exact value is 0.584x10-15. The inverse would be 1.7x1015 pg/h or about 1700 g/h.

  4. The post-processing of the source attribution output files results in several different files (prob??, cmax??, cmean) in the working directory which are discussed in more detail in the hysplit help files. In this application, all the graphics are created from cmean, the mean value at each grid point for all the output files with the .000 suffix.

The results shown here used only measurement data in conjunction with the dilution factors computed by HYSPLIT to estimate the source location and magnitude. We assumed that the dilution factors are about the same in the forward or backward integration of the model. The results show a very close match with the actual hypothetical plume. However, the actual emission rate was 3000 g/h. Differences in the forward versus backward dispersion factors and the continuous 3-h emission assumptions used at the measurement locations, can result in lower correlations and a biased emission estimates.

3 m 8 s