|
||||
OFR 01-0429: World Trade Center USGS Integration of Results and Conclusions |
About USGS / Science Topics / Maps, Products & Publications / Education / FAQ |
The results of analyses completed so far show a consistent picture: the samples are largely composed of gypsum, cellulose, and miscellaneous materials common in a building, with minor asbestiform minerals. However, one sample analyzed, the coating on a steel beam, indicates the presence of a significant abundance of chrysotile asbestos (as much as 20% by volume). The confirmed abundant chrysotile sample and the potential pockets of chrysotile indicated in the AVIRIS mineral maps indicates that asbestos can be found in localized concentrations. Thus, appropriate precautions should be taken when handling debris, especially coatings on metal beams.
Sample results are summarized in Table 1, below. To see the full resolution SEM image and description, click on the image in the table. In the spectroscopy column in the table below, CH indicates organic compounds, including paints and plastics. Fe2+ indicates minerals or materials containing ferrous iron. Amounts are qualitative and indicate (from low to high) trace as (tr), weak as (wk), and strong as (str). No indication indicates between weak and strong.
The composition of samples collected in the WTC area, as indicated by spectroscopy, XRD, SEM, and from the visual examination during splitting of the samples, show similarities, yet each sample shows differences. Thus, while the samples appear to be a "grey dust", the data indicate the dust was not well mixed. The sample analyses and the AVIRIS mapping results agree in this regard.
* Amorphous material is not identifiable by XRD, but its presence is detectable.
* "Possible trace chrysotile" means at or near the detection limit with XRD.
The question of asbestos distribution was investigated and the results show an asymmetric distribution pattern (Results Figure 1). More chrysotile was detected in an east-west direction than south. This pattern occurs in both the AVIRIS maps and from field samples (Results Figure 2). While there is a general trend, it is not exclusive, meaning that chrysotile was detected in all directions. It also should be noted that samples obtained next to each other (on the map this means a city block apart) can show different results: one has asbestos, another has no chrysotile above the detection limit.
Composition of samples on a centimeter scale was examined with a spectrometer. Small variations in chrysotile content throughout a sample were observed. Thus from scales of cm to tens of meters, chrysotile content varies. Such variability makes sampling and overall assessment of a site difficult.
The fact that some materials in the WTC debris were observed to contain higher levels of chrysotile (sample WTC01-08) on a steel beam, and that the coatings on the beams have largely been stripped, leads to the question of where did the coatings go and how well distributed/dispersed is the chrysotile? Because a patch of coating showed up to 20% chrysotile, and the field samples and the AVIRIS maps show varying levels of serpentine (chrysotile) leads to the possibility that other patches of chrysotile may exist in the debris.
The asymmetry in the AVIRIS iron-bearing materials map may be related to the asymmetry in the asbestiform minerals map. The AVIRIS data and the laboratory analyses of the field samples indicate a lower abundance of chrysotile in the the southern direction from the WTC, the same direction of the increase in iron-bearing materials. The one field sample, WTC01-08, from an iron beam, which had up to 20% chrysotile also contains a strong Fe2+ absorption. Thus one might expect a higher chrysotile content in iron-bearing materials. However, this is clearly not the case, at least in general. This may indicate other sources of the chrysotile besides the beam coatings.
AVIRIS imaging spectroscopy mapping provides a synoptic view that samples more area than possible with other methods. The AVIRIS maps shown here represent only a portion of the data collected, and effectively provide data for about 4.7 million sample locations, all obtained within a couple of hours. The sampling includes land, air and water.
The fact that the field sampling missed the highest concentrations of serpentines in the AVIRIS maps shows the limitations of limited sampling methodologies. Ideally, the field sampling team would have the AVIRIS materials maps to guide the field sampling. Unfortunately, this was not possible in this rapid response case (but we routinely employ such methods in geologic studies where the region does not change rapidly). Even so, the materials maps for this study were produced faster than any other imaging spectroscopy effort to our knowledge. The AVIRIS data were received within 24 hours of acquisition, and the data were initially calibrated to help the field team obtain the final calibration data with real time feedback via cell phone. In this case, scientists in Denver communicated composition of field calibration sites using initially calibrated AVIRIS data (of the parking lot structure) while the field team was investigating where the best portion of the parking lot was located. The real-time feedback resulted in avoidance of portions of the parking lot with strong absorption features, not visible to the human eye, that could have compromised the quality of the final calibration.
With further development of on-board solar calibration targets on the aircraft with the AVIRIS sensor, the refinement of analysis software, the development of more reference spectral libraries, and the use of faster computers, an even faster response is possible in the future. The challenge is formidable. To analyze the data for this study, we used approximately 300 gigabytes of disk space and performed over 50 trillion calculations. The results of the AVIRIS mapping are limited by knowledge of the spectral properties of materials and the detection levels are limited by the sensor signal-to-noise. The detection limits could be substantially improved with existing technology in a new sensor design.
The combination of field sampling with laboratory analysis and imaging spectroscopy remote sensing provide a powerful assessment combination. We estimate the analysis effort of this highly experienced team to be 1.8 person years to complete this study plus another 0.6 person-year for the AVIRIS data collection effort. This study includes analysis of 20% of the AVIRIS data from Sept 16, and 7% of the data from Sept 23 (thermal hot spot analysis only).
The scientific data from this study is presented with no assessment of health effects. It is beyond the scope of this study to assess health effects of a fraction of a percent chrysotile asbestos, for example.
Other Conclusions have already been presented in the Executive Summary.
Results Figure 1. Sample location map coded by asbestos detection.
Results Figure 2.WTC Sample analyses from Results Figure 1 are shown plotted on the serpentine/amphibole AVIRIS map. There is a loose correlation of chrysotile locations spread in an east-west direction in both the laboratory analyses and the AVIRIS data.
NEXT Section of Report: References
Back to document Table of Contents
AccessibilityFOIAPrivacyPolicies and Notices | |