Blackened Limestone Pebbles: Fire at Subaerial Unconformities

Abstract

Introduction

Field Observations & Some Experimental Results

Discussion

Submarine Blackening & Accumulation of Salt-and-Pepper Sands

Conclusions

Acknowledgements

References

Discussion

The material which causes blackening is difficult to discern in thin section. Figure 5.7 is a thin section of artificially blackened soilstone crust. It is identical in appearance to its unblackened counterpart except that what were darker brown areas have turned dark grey to black. Finely disseminated organic matter which causes the brown color in soilstone crust (Multer and Hoffmeister 1968) apparently chars to dark grey or black. Thin sections of naturally blackened soilstone crust and pebbles look identical to the artificially blackened samples. Even in thin sections of heat-blackened coral, discrete black particles were impossible to discern. Scanning electron microscopy might reveal individual charred particles; however, its use was beyond the scope of this study.

Figures 5.3B and C are photomicrographs of two different black pebbles from the breccia shown in Figure 5.3A. Note in Figure 5.3B that blackened ooid grains surrounded by clear calcite cement give the pebble its overall black color. The darkened pebble in Figure 5.3C, on the other hand, is a fragment of soilstone crust. We think it significant that the black pebbles in the breccia shown in Figures 5.3A, B, and C occur in late Pleistocene eolian deposits situated approximately 20 m above sealevel. These 20- to 30-m-thick cross-bedded eolian dunes were deposited shortly before or during the last Pleistocene glacial lowering of sealevel, and there is no evidence of a higher sealevel stand in this area since sealevel rose again during the Holocene. The age and position of these black pebbles therefore are compatible with our forest fire hypothesis, and blackening mechanisms involving subtidal or hypersaline conditions can be ruled out.

The data from LECO and Rock-EVAL analyses of experimentally blackened material are interesting but equivocal (Table 5.2). The control sample contained 0.53 wt % organic carbon by LECO and 0.89 wt % organic carbon by Rock-EVAL. The heat-blackened sample, however, contained slightly less, 0.46 wt % and 0.19 wt %, respectively. Surprisingly, the naturally blackened Pleistocene pebbles from the breccia shown in Figure 5.2B contained more carbon than its unblackened counterpart, although the values are less than for both the control and experimentally blackened Holocene crusts. It should be pointed out, however, that these Pleistocene pebbles are not of soilstone crust but instead are coralline grainstone, so a direct comparison cannot be made.

Surprisingly, the data for unblackened subtidal skeletal grains collected in Florida Bay show significantly more total organic carbon than the blackened grains analyzed from the same sediment sample. The Marquesas Keys samples, which were older (buried beneath 8 m of lime mud and silt) and at a depth of 30 m below sealevel, showed more carbon in the black sample than in their unblackened counterpart. We do not know how to explain these few data adequately but do not find it surprising that heated crust samples apparently contain less carbon than unheated samples. Carbon was probably lost by vaporization during heating.

We do not know why darkening was obtained so quickly at only 400°C in the later experiments, when the previous work listed in Table 5.1 indicated that 510°C for 6 h was required to obtain maximum blackening. There are at least two possibilities: (1) All samples listed in Table 5.1 were small sawed samples approximately 1 cc in size. The later experiments (see Figs. 5.5A, B, C) utilized larger hand specimens. (2) The thermometer and heat control devices, which were later changed, were defective. At best, temperature control was no better than 5% and likely as low as 10%. Regardless of the precision, it seems likely that temperature is more critical than time.

Although all our experiments and observations support the fire hypothesis, we nevertheless feel that it would be fruitful to determine more precisely the temperature-time relationships. To obtain the data, more expensive and sophisticated temperature control, not now available to us, will be required. In addition, more geochemical analyses should be aimed at identifying the blackening agent. If there is a unique chemical signature, then it might lead to development of a simple geochemical test that could distinguish subaerial from submarine blackened pebbles and grains.

All our observations, simple experiments, and carbon analyses support the fire hypothesis. We thus find it interesting that Barthel (1977), who also noted anthropogenic blackening of limestone in the Florida Keys, thought fire-blackening a hindrance to discovering the true cause of blackening. Certainly there has been much burning associated with land-development projects. One of us (EAS) observed many forest fires in the Florida Keys during the late 1940s and early 1950s, when that was a common land-clearing technique. There is evidence, however, of natural forest fires. Figure 5.8 is a photomicrograph of a charred twig or wood fragment imbedded within soilstone crusts from Ramrod Key, Florida. The C¹⁴ work of Robbin and Stipp (1977) showed that these crusts range in age from about 5000 y to the present. Inclusion of charcoal throughout the crusts therefore indicates that natural fires were occurring in the area a few thousand years ago, before influx of modern man. We are concerned, however, that we often see charred wood fragments in unblackened laminated soilstone crust. If our hypothesis is correct, then this material is probably the result of minor grass fires which did not generate sufficient heat to blacken thoroughly the crust in which they became incorporated. Some crusts in this area, however, do contain blackened laminae, which we think were caused by fire before deposition of subsequent unblackened layers (Fig. 5.4).

We consider the experiment shown in Figures 5.6A and B to be particularly significant. The experiment shows that even though breccia components experienced the same temperature conditions, only some of the previously unblackened pebbles became blackened. Soilstone crust lining the breccia accumulation also turned black. The fractured black pebble in Figure 5.6A is consistent with the in situ forest fireheating hypothesis. The pebble was clearly fractured in place, possibly during the same heating event that blackened it. It is probable, however, that it was blackened before emplacement and was subsequently fractured by pedogenic processes. If this was not the case, the pebble next to it which blackened experimentally would have already been blackened by the heat that colored the fractured pebble. Although probably transported after blackening, its angularity argues against significant transport.

Figure 5.1A shows two pebbles which display a pronounced color gradient. The one at the upper left indicates a heat source from below, whereas the one to the right of center indicates a source from above. We believe the gradient was caused by heat, more or less in situ, but that the upper left pebble has been rotated by pedogenic processes, possibly roots, or burrowers, or some other process of which we are unaware.

Ward et al. (1970) described blackened soilstone crust pebbles mixed with unblackened crust pebbles in and around a hypersaline pond on Isla Mujeres off Mexico, and concluded that the color was caused by organic matter, probably algae, which had been darkened by reducing hypersaline waters. Analyses showed that the blackened fragments contained no more iron, manganese, or sulfur than nonblackened fragments (Ward et al. 1970). Analyses by LEGO carbon analyzer showed a slightly higher (2000 to 3000 ppm) concentration of organic carbon (versus 1000 to 2000 ppm) in unblackened soilstone crust pebbles. Ward et al. (1970) also concluded that there is no evidence of crusts being blackened today but that blackening occurred during the early Holocene, and the crusts were subsequently fractured by desiccation to form angular pebbles. We suggest that the blackening was caused by forest fires and that the recent relative rise in sealevel created the ponds which inundated both blackened and unblackened crusts and pebbles with hypersaline waters. Our analyses in Table 5.2 would suggest little relationship between total organic carbon and color. The total organic carbon data of Ward et al. (1970) for blackened and unblackened crusts overlap and are not considered sufficient evidence that blackening was caused by reducing hypersaline waters. Although reducing waters can cause blackening, our simple experiments and occurrences of black pebbles in young eolian deposits demonstrate that hypersaline water is not a necessary condition.

Although we may be the first to propose an "instantaneous" fire blackening origin, we are not the first to implicate fire as a cause of limestone blackening. Strasser (1984) noted the possible role of fire in Florida and the Bahamas but concluded that "the blackening substances are transported by percolating waters" (p. 1103). He concluded from his analyses that traces of organic matter in the 0. 1 to 0.5% range caused blackening. Although Strasser also listed blackening by anoxic bottom water, he concluded that in ancient limestones blackened pebbles indicate subaerial exposure and the former presence of islands and coastlines. We would like to stress that indeed there are grains blackened under subtidal conditions, and they should not be confused with subaerially blackened pebbles.

« Field Observations & Some Experimental Results | Submarine Blackening & Accumulation of Salt-and-Pepper Sands »