10 



5.0 CONCLUSIONS 



From the results of UV-R field demonstration, the following conclusions can be 

 drawn: 



> The prototype field system shows good promise of detecting PAHs in water and 

 sediment in the range of 10-100 ppm and above. With further development, a rapid 

 means of assessing PAH distributions may be feasible in the next several years; 



> The optical/mechanical system was proven to be field-ready; 



> The system has detected station to station differences which may be related to PAH 

 concentration and composition; 



> The system has detected small sediment column spatial variations, which suggest 

 micro scale variation in PAH distribution within the top 0-15 cm of surface 

 sediments. 



The UV-R camera system developed by SAIC is still in its experimental stages. 

 Recommendations for improving the overall quality of the results include modifying the 

 camera system to reduce internal noise, stray reflections, and hot spots (replacing scratched 

 lenses, dirt on the first-surface reflecting mirror and higher quality RGB filters). The 

 second recommendation arises in the processing of the image data. A non-parametric 

 approach should be developed to characterize and isolate the internal noise to better permit 

 the separation of this signal from the sample data. Such noise reduction algorithms are 

 commonly employed in signal processing of acoustic data, for example, and could be 

 readily applied to the UV-R data sets. At present, the parametric method for signal 

 correction, while applicable to high PAH environments (e.g., oil spills) is too severe for 

 lower PAH environments created by non-point sources. This system was developed for 

 synchronous spectral analysis (i.e., emission intensity averaged over time). Future 

 configurations might consider application of this "time-resolved" spectrometry (i.e., 

 emission growth and decay rate) such as employed in other systems (Rudnick and Chen 

 1997). 



