to minimize infiltration of visible light) compared with the intensities of the 

 sediment/interstitial water. From Figure 6, it is seen that ambient light leakage intensities 

 are similar to those found in the sediment for emission spectra of 350-450. However, 

 emission wavelengths of 500-600 (visible light wavelength) had much larger intensities. 

 This observation underscores the importance of having an efficient "shade" on the camera 

 to prevent visible light contamination. 



In order to exclude water column light noise in the analysis of sediment-associated 

 PAH spectra, the remainder of the analyses focus on the Middle and Bottom ROIs as these 

 areas encompass only sediment. Each of these ROI is presented separately by station in 

 Figures 7-10. As found for the whole sediment samples, the 365 nm excitation band 

 produces highest emission intensities for each of the locations. 



The same data presented by excitation wavelength in Figures 11-14 do reveal some 

 ROI-specific phenomena. For example, unique peaks are observed: 



1) at 299 excitation /430 emission for bottom left (BL) and bottom right (BR) ROI for 

 Sassafras and State Pier (Figure 11E,F); 



2) at 314/380, 314/420 and 314/450 nm combinations for Middle left ROI at State Pier 

 (Figure 12C); and 



3) at 335/440, 335/450 combination for Middle Left and Middle Right ROI at State 

 Pier and Sassafras Point, respectively (Figure 13 [C,D]). 



Finally, the 365 nm peak for the BR-ROI is higher for Shooters/Sassafras as found 

 previously for the whole sediment, but in the three other regions. Shooters emission was at 

 or below State Pier (Figure 14). Hence, the bottom right portion of the spectra was 

 dominating the return observed for the full image, possibly indicating a locally higher PAH 

 concentration. 



3.3. Intensity Analysis 



The data sets presented above have been processed using the fiill correction 

 algorithm (Equation 1) without regard to the intensity of the fluorescence emission. That 

 is, the data results represent average fluorescence emission over the entire ROI. In this 

 section, the variance of the fluorescence emission pattern are analyzed to discern whether 

 highly localized (i.e., pixel-scale) emissions occur which may relate to small amounts of 

 PAH materials (e.g., soot particles, oil droplets). The inspection of pixel returns within 

 ROIs reveal that relatively few of the pixels in each ROI actually contained detectable 

 emission data (Table 2). For example, the average number of non-zero pixels ranged 

 about 1-9% at Shooters 2; the other two sites were found to have similar statistical 

 distributions. It is also apparent that the number of non-zero pixels for the entire sediment 



