At this time, insufficient data points preclude statistical analyses, but some preliminary observations 165can be made. It is evident that urban varnish is a widespread phenomenon and not just a local Washington problem. Second, on structures built with more than one type of stone, the varnish occurs only on the Triassic sandstone. This suggests that some property of this stone encourages the growth of the varnish. An exception to this rule is the identification of varnish on the Carboniferous sandstone of the Central Park’s Bethesda Fountain Plaza in New York City. Finally, the occurrence of varnish appeared to be an anomaly on the James Hill House in St. Paul, Minnesota, since it is located far from the Newark Supergroup region. However, records show that the building stone was shipped by railroad to St. Paul from Triassic red sandstone quarries at East Longmeadow, Massachusetts, which is within the Newark Supergroup.
Estimated growth rates on the Smithsonian Castle
At the time of writing the analysis of the most recent pXRF data from New York and Massachusetts sites has not been completed. However, the results from the 165-year-old Smithsonian Castle, which is located on the National Mall in Washington, can serve as an example of the method for calculating growth rates.
Three locations around the Castle were measured by pXRF. In addition to the southwest corner (Fig. 1), a patch at the east entrance was measured, which historic photographs show was free of varnish as late as 1985. A third patch was measured on a gate post of the Enid Haupt Garden, built in 1987 adjacent to the Castle using the same Seneca sandstone. Ten points each were measured on the varnish patch and a bare stone area at the SW corner, and five points each at the other locations. The pXRF counts data were converted to Mn layer thickness using the Fe/Fe ratio method for the Mn patch and bare stone.
The results are presented in Table 2 along with estimated ages. The calculated growth rates for the east entrance and the gatepost are reasonably close, on the order of 90 nm/yr. This is significantly higher than the maximum rate of 40 nm/yr observed for desert varnish (Liu & Broecker, 2008). However, the growth rate for the southwest corner is only a third of this, if its estimated age is based on the assumption that the layer began to grow as soon as the stone was put in place. Although historic photographic documentation is mainly in black and white and insufficiently detailed for conclusive determination, it appears that growth actually started much later. Dividing the thickness of the layer, 4 microns, by the rate of 90 nm/yr gives an age of 45 years or a start date of 1970. This is consistent with the period of great population growth and associated increase in automobile traffic around the Washington urban area.
Table 2: Varnish growth rates on the Smithsonian Castle, based on Fe/Fe ratios measured on Mn patches and bare stone using pXRF.
| Location | Thickness, µm | Time, years | Growth Rate, nm/yr |
| S.W. Corner | 4.0 ± 0.57 | 155 | 26 ± 3.7 |
| Gatepost | 2.41 ± 0.06 | 29 | 83 ± 2.0 |
| East Entrance | 2.38 ± 0.06 | 25 | 95 ± 2.4 |
Conclusions
Portable XRF provides a nondestructive method of acquiring data on its geographical distribution and rate of growth. The Mn/Fe counts ratio can be used to distinguish the Mn-rich varnish from other types of surface deposits. The direct Mn and indirect Fe/Fe ratio methods can be used to estimate the layer thickness and hence the growth rate. Patches of urban rock varnish have been identified by pXRF on buildings across the northern United States from Washington to New York City to Minneapolis. These patches have typically been observed on red Triassic sandstone. However, they have also been found growing on older Carboniferous sandstone in New York City’s Central Park. Growth rates estimated from datable patches on the Smithsonian Castle and nearby gate posts are in the range of 83 ± 2.0 to 95 ± 2.4 nm/yr. This is significantly higher than the maximum rate of 40 nm/yr observed for desert varnish.
166Acknowledgements
The authors would like to thank Bill Rebel of American Engineering Testing, Inc., for providing the XRF analysis of the James Hill House varnish sample.
References
Dorn R. I. 2007. Rock varnish. In: Geochemical Sediments and Landscapes. Nash D. J., McLaren S. J. (eds). London, Blackwell, pp. 246–297.
Grissom C., Aloiz E., Vicenzi E., Livingston R. A. 2018. Seneca sandstone: A heritage stone from the United States. In: Global Heritage Stone: Worldwide Examples of Heritage Stones, GSL Special Publication 486. London, Geological Society of London. https://doi.org/10.1144/SP486.4.
Gatuingt L. S. et al. 2016. Intrinsic parameters conditioning the formation of Mn-rich patinas on Luneville sandstones. In: Proceedings of the 13th international congress on the deterioration and conservation of stone. Hughes J. J., Howind T. (eds), vol 1. Paisley, Scotland, University of the West of Scotland, pp. 317–324.
Liu T., Broecker W. S. 2008. Rock varnish microexamination dating of late Quaternary geomorphic features in the drylands of western USA. Geomorphology 93:501–523.
Livingston R. A., Grissom C., Kavich G., Douglas J. 2020. Thickness measurement of thin manganese-rich layers on sandstone buildings using X-ray fluorescence attenuation, unpublished manuscript.
Livingston R. A. et al. 2016. Investigation of urban rock varnish on the sandstone of the Smithsonian. In: Proceedings of the 13th international congress on the deterioration and conservation of stone. Hughes J. J., Howind T. (eds), vol. 2. Paisley, Scotland, University of the West of Scotland, pp. 399–406.
Macholdt D. S. et al. 2017a. Black manganese-rich crusts on a Gothic cathedral, Atmospheric Environment 2017:205–220.
Macholdt D. S. et al. 2017b. Characterization and differentiation of rock varnish types from different environments by microanalytical techniques, Chemical Geology 459:91–118.
Matero F. G., Teutonico J. M. 1982. The use of architectural sandstone in New York City in the 19th century, APT Bulletin 14:11–17.
McNeil
