Content uploaded by Robert G. Bednarik
Author content
All content in this area was uploaded by Robert G. Bednarik on Jul 23, 2021
Content may be subject to copyright.
43
Rock Art Research 2019 - Volume 36, Number 1, pp. 43-48. R. G. BEDNARIK
KEYWORDS: Rock art – Direct dating – Microerosion analysis – Recalibration – ICRAD
ADVANCES IN MICROEROSION ANALYSIS
Robert G. Bednarik
Abstract. This paper reviews current developments in microerosion analysis, including the
testing of a universal calibration curve based on regional precipitation. This enables the use
of the method in regions that are unlikely to provide suitable calibration surfaces. The paper
also considers the creation of the archive of the International Centre of Rock Art Dating, and
the need for it to facilitate the testing of rock art age estimates.
Introduction
Microerosion analysis was deliberately designed
to provide a relatively simple method for the direct
dating of petroglyphs with a high degree of reliability,
even if its precision was always regarded as relatively
low. Bearing in mind that ‘indirect’ (archaeological)
methods of age estimation (e.g. by excavation, stylistic
reasoning) of rock art created by reductive processes
(petroglyphs) have proved to be quite unreliable,
and have led to estimates that have been wrong by as
much as 250 times (e.g at Siega Verde, Spain; Bednarik
The concept of direct rock art dating refers to the
direct physical relationship between the rock art and
propositions concerning this relationship (IFRAO Rock
Art Glossary).
About thirty years ago we developed the meth-
applied it at Lake Onega in Russia (Bednarik 1992,
1993). This microscopic approach to estimating the age
of petroglyphs comprises not a single method, but a
cluster of methods focused on the idea that a variety
approximate ages of the rock art motifs they pertain to.
These chemical weathering processes apply especially
in unsheltered locations and they result in cumulative
results that are a function of time. So far, only two
such methods have been applied: the measurement of
micro-wanes on crystals that were fractured during the
production of petroglyphs; and the selective, alveolar
retreat of certain rock types of component minerals
that weather much faster than others.
-
tensively since the early 1990s and now forms a core
approach of the International Centre for Rock Art
Dating at Hebei Normal University, China. It has been
applied in all continents except Antarctica, and in some
of them with considerable success. The crucial variable
utilised in this method derives from the primarily
chemical weathering of a surface feature created when
a petroglyph was produced. In fracturing mineral crys-
tals, edges are formed that are perfectly sharp at that
surface as a function of time, but it does so selectively.
face. Therefore, the edge becomes progressively more
rounded, a process following a fundamental law of
nature. The same law applies to the geometry of heat
-
trates a cube of sugar; or the way macro-wanes develop
on rock, such as basalts, sandstones or granites. Mac-
ro-wanes in geology have long been recognised as a
measure of surface age, and
claim credibly to estimate the ages of such wanes to
within 10–20% accuracy on two rock types (cf. Trendall
law is, however, more complex than they envisaged,
because the rock surface retreats not only at the edge
(as they assumed), but also on the surfaces forming it.
The governing geometry (Bednarik 1992, 1993a, 2007)
was eventually explained, which opened the way to
quantifying the process variables.
The second microerosion method so far applied
typically of two minerals co-occurring in a rock. For
the cement of sandstone is of colloid silica and dissolves
In this method the retreat of the cement is determined
microscopically, and if the rate of retreat can be estab-
lished from a surface of known age nearby, such cali-
bration can lead to securing age estimates of petroglyph
surfaces of unknown antiquities. Despite its evident
Rock Art Research 2019 - Volume 36, Number 1, pp. 43-48. R. G. BEDNARIK
44
validity and procedural simplicity, this method has
only been applied very rarely so far (Bednarik 1995).
The analysis of micro-wanes, by contrast, has be-
come a standardised and much used method. That
employment over many years and by several analysts
has resulted in various collaborative improvements,
in the establishment of routine procedures and in
-
tion. Essentially, there are two complications. First, in
not impossible, to secure calibration surfaces. These
are rock surfaces of known antiquities that contain
and feldspar have been utilised). They include rock
inscriptions, monuments, ancient bridges and other
structures, glacial striations; such features of known
ages are simply lacking in many regions of the world.
Second, it became increasingly evident that the vari-
ations in the microerosion process across countries,
especially large countries, mean that any calibration
curves obtained were likely to apply only on regional
scales. Such climatically uniform regions tended to be
small and, obviously, independent of national borders.
This paper endeavours to address these issues and
method as it has developed over the years.
Calibrating microerosion coecients
While applying microerosion analysis in the
suitable calibration site for this arid region. Up to that
time, the practice in such cases was to substitute a mi-
apparently similar climatic region. In the Kalahari
experiment we considered the use of the Spear Hill
calibration from north-western Australia (Beaumont
and Bednarik 2015: 170). The climate seemed broadly
similar to the target region, although apparently more
arid, but in trying to test the assumption made we de-
cided to compare the precipitation rates of Spear Hill
and southern Kalahari. In the process we happened to
experimentally
-
they formed an almost perfect alignment (op. cit.:
Saudi Arabia (Bednarik and Khan 2005), Spear Hill in
Australia (Bednarik 2002), Deyunshan in China (Tang
et al. 2017), Grosio in Italy (Bednarik 2001) and Vila
Real in Portugal (Bednarik 2003).
-
engraved date, was considered inadequate because of
its low age. This was despite the close match between
μm-
μm. In that case, a regional calibration
μm was adopted as a compromise
In another important development, microerosion
that she has discovered a rock inscription on schist
‘Yan shi hou san shi li’ (‘The stone is thirty miles far
Shu style, a calligraphic style established during the
in one of the characters yielded an average micro-wane
width of 11.5 μm (spectrum 9–15 μm), which would
to the Deyunshan calibration curve. However, if the
μm.
suggested to be superior.
The question now arising is, which way should
microerosion methodology be developed further?
Should we focus on collecting more calibration values
or simply rely on the universal curve?
Our answer is that there needs to be a combination
of both approaches. The universal calibration curve
is not yet as well developed as we would like it to be
before we can rely on it exclusively. The acquisition
of more calibration references must continue where
this is possible, but it needs to be recognised that
in many parts of the world it is unlikely to succeed.
There are simply no weathered fracture surfaces of
known ages available. The strategy therefore needs to
be two-pronged: to secure more calibrations wherever
Site Age Coecient Precipitation
1150–1200 μm/ka
b. Spear Hill, Australia Average 7 values μm/ka
c. Deyunshan, China μm/ka
d. Lianyungang, China μm/ka
e. Grosio, Italy 12 000 μm/ka 920 mm
f. Vila Real, Portugal 1900 12.0 μm/ka
Table 1. Calibration points, their ages, their coecients and average annual precipitation.
45
Rock Art Research 2019 - Volume 36, Number 1, pp. 43-48. R. G. BEDNARIK
possible, but apply the UCC exper-
imentally when it seems unlikely to
secure suitable calibration surfaces
and to always convey the original
wane-widths data set in case age
the future. That is precisely the pur-
pose of full publication of data sets.
To show where we are at this point
in time, Figure 1 presents the most
detailed UCC established so far. It
is based on the annual precipitation
values and known ages of six sites as
listed in Table 1.
Table 2 adjusts the estimates
previously secured in two Chi-
Provinces, calibrated with the only
reference then available, the Deyun-
shan inscription. For Ningxia, the
adjustment to the microerosion coef-
calibration value determined by
revision takes into account the sig-
averages across the large country.
Another set of microerosion-based
age estimates in need of review are
the six determinations procured from
two of the twenty-seven sites of the
Kalatrancani petroglyph complex
near Cochabamba in central Bolivia.
Region Number of sample Prev. age esti-
mate
Precipitation
& coecient
Universal curve
age estimate
Ningxia
Province
μm
120
E2330 + 90 / - 210
Lianyungang calibration
Province
μm
E2150 + 300 / - 110
E2210 + 210 / - 90
Table 2. Corrections of age estimates obtained in China before allowing for regional dierences in precipitation.
Corrections are derived from the universal calibration curve.
Figure 1. Improved and updated universal calibration curve for microerosion
analysis of quar fracture micro-wanes.
Rock Art Research 2019 - Volume 36, Number 1, pp. 43-48. R. G. BEDNARIK
46
These were provisionally based on the Grosio calibration
in the light of the region’s precipitation data and the
yields the results shown in Table 3.
There is no need to adjust other age estimates de-
rived from microerosion analysis. Although the result
from Al Usayla near Riyadh in Saudi Arabia is based
on the Umm Sinman calibration (Bednarik and Khan
cannot be checked as there are no rainfall data available
from Shuwaymis (we requested the installation of a
weather station only recently, as part of our nomina-
tion for World Heritage listing). It is, however, likely
(Bednarik and Khan 2017). The only
age estimate from the country’s far south, from Ta’ar,
has its own calibration from nearby c
other published microerosion estimates have either
been furnished with their own local calibrations or,
in the case of the most recent determinations, have
already been referenced to the UCC, or at least partially
Bednarik in press).
Developing microerosion analysis
These new developments show that microerosion
-
liably estimating the approximate age of petroglyphs
currently at our disposal. The method is still evolving
available from around the world. However, it has al-
ready become a routine procedure whose weaknesses
and strengths are well appreciated, and which is being
applied successfully by various teams.
It needs to be emphasised that another essential
development in microerosion analysis is the need to ac-
quire much more data for the alternative method men-
solution rates of two minerals co-occurring in a rock,
particularly crystalline grains and colloidal cement.
Measuring the retreat of the more soluble component is
relatively simple and is a direct reference to the amount
of solution a surface has been subjected to. That mea-
surement can be a function of time if it can be calibrated
against surfaces of known ages of the same rock type in
to apply than the more widely used measurements of
only by the process of kinetic energy metamorphosis
(Bednarik 2015a, 2015b), the results of which are read-
ily detectable by a specialist using optical or scanning
electron microscopy (Bednarik 2019).
However, more elementary developments in the
practices of microerosion analysis refer to the routine
procedures of collecting and archiving data. The most
is the repeatability of experiments to render data
testable. This prerequisite determines the standing
art age estimates, most of the methods so far applied
fail to meet this criterion, for a variety of reasons. For
example, samples removed for destructive analysis
obviously cannot be re-analysed. If there is adequate
material sampled, sample splits may be available, but
with various methods of destructive sampling the
quantities of sample available are minute. There may
also be variations in the composition of the substanc-
es constituting the dating criterion that are of very
repeat analyses. For instance, the dating criterion in
uranium-thorium analysis is the ratio of 230Th and U,
but that relationship can be (and apparently often is)
distorted in carbonate speleothems, the material most
relevant in the context of cave art dating. The relevance
of this ratio depends entirely on the system being a
closed system, which in the case of speleothems is
rarely the case. In the C analysis of speleothems it has
been known since the method’s earliest applications
are very densely crystalline stalagmites. The same is
likely to apply in uranium-series dating, yet nearly
all the results so far reported derive from porous or
extremely thin coatings. The solubility of U in water
and the potential presence of detrital Th have to be ad-
C and U–Th age determinations of split samples need
Site Number of sample Prev. age
estimate
Precipitation
& coecient
Universal curve
age estimate
Isay Rumi
South, Kala-
trancani
Bolivia-Kala1-EQ-25/7/2007
μm
Bolivia-Kala2-EQ-25/7/2007 E530 + 50 / - 90
Bolivia-Kala3-EQ-25/7/2007
E390 + 50 / - 100
Ph’alta Rumi,
Kalatrancani
E790 + 90 / - 210
Table 3. Corrections of age estimates for the Kalatrancani petroglyph complex, central Bolivia, based on the Grosio
calibration curve, and the corrections suggested by the universal calibration curve.
47
Rock Art Research 2019 - Volume 36, Number 1, pp. 43-48. R. G. BEDNARIK
al. 2003). It is incumbent upon the U–Th analysts to
explain why in most of the cases when the method
was used in tandem with C, its results were three to
the direct dating of rock art: they are all burdened with
and repeatability is usually impaired.
Whereas the adversities with microerosion analysis
are well understood and appreciated, any mention
of complications with other rock art dating methods
tends to elicit antagonistic responses from their advo-
cates (e.g. Pike et al. 2017), which limits constructive
dialogue severely. In contrast to destructive methods,
microerosion study is fully repeatable, and remains
of re-measuring micro-wanes in centuries from now,
-
cients in the very long term. The prospects of achieving
that kind of facility with any other rock art dating meth-
od are not promising. The opposition to destructive
sampling of rock art is expressed, for instance, in the
recent decision of the French Commission of Historical
Monuments to demand that all applications to regional
authorities for permits to sample Palaeolithic cave art
be referred to its central agency (Sauvet et al. 2015). In
view of the recent proliferation of such activities and
the sensationalist reporting of the results, such caution
seems pertinent. At this stage in the development of
direct dating of rock art it is certainly premature to
overuse destructive sampling methods and to reject
calls for checking their results against those of alter-
native methods (Pike et al. 2017). The rock art in ques-
tion has survived for millennia and will survive for
much longer, while the methodology in question will
no doubt be continuously developed and improved.
Rock art dating is likely to be discredited by precipitate
claims that are intended to augment academic careers
rather than a sound knowledge base.
This raises the question of how to develop micro-
erosion analysis to facilitate its maximal potential. In
order to be able to re-measure micro-wanes centuries
from now, two prerequisites are indispensable: every
measured wane must be re-locatable and the original
wane width measurements must be preserved. The
-
scribed in Bednarik (2017). Each measured wane is
the individual petroglyph, and the location of the
measured micro-wane within that petroglyph. The
last factor is only needed when there is a possibility
that the petroglyph may have experienced renewal at
some time, because if all fractures in one motif derive
from one single event of production, all wane-widths
in it should yield similar measurements. The second
prerequisite is that the wane-widths originally deter-
mined must remain available for all future. Both these
requirements can be met by the recent establishment of
the International Centre of Rock Art Dating (ICRAD)
which will maintain a register of all such results.
petroglyphs unlimited numbers of times in the future
is illustrated by considering the prediction that a motif
estimated to be 300 years old today should in 300 years
from now yield wane widths twice as great as today.
This absolute testability of predictions is the hallmark
of solid science and contrasts with the opportunistic
use of methods known to be subject to numerous fac-
tors of uncertainty that remain inadequately explored.
The only weakness of microerosion analysis is the
issue of precipitation rates having varied in the past.
Obviously rainfall does not remain the same through
time, but the long-term trends over the most recent
millennia, which have yielded the largest number of
results, are not so dramatic that they are likely to exceed
proceeds into the Metal Ages and beyond, variations
-
accelerated or decelerated rates of erosion for limited
periods of time, but still preserve relative parity of
results. Secondly, over very long timespans, these
variations are even likely to cancel themselves out and
especially then remain within the parameters of stated
tolerance limits. The issues of this source of imprecision
-
At this stage in the method’s development reliability
of results is simply considered more important than
their precision.
Conclusion
of rock art (and had it rejected by archaeologists who
believed that rock art needs to be dated by archaeology,
not by science) have been marked by lively activity in
some areas, relative neglect in others; by wide vari-
ations in credibility of methods and results; and by
over-enthusiastic and fervently defended notions. All
a repository of all ‘direct dating’ results, even those
that appear to have been refuted. This rock art dating
archive needs to provide, for each entry, the possibility
of testing it, how it was acquired, and whether there
was destructive sampling involved in its procurement.
Rather than securing more new data it may be more
productive at this stage to end the probationary phase
of direct rock art dating, instead focusing on consolidat-
ing the data of recent decades and creating an archive
designed to be fully transparent about the derivation
of age estimates. In all probability new methods will
be developed in the future and applied to data secured
Rock Art Research 2019 - Volume 36, Number 1, pp. 43-48. R. G. BEDNARIK
48
by current means. Testing of previous results needs
to be encouraged and facilitated, e.g. by providing
information required for re-analysis.
Acknowledgments
the Lianyungang calibration curve and microerosion coef-
Tang Huisheng. I am also grateful to the three RAR referees
of this paper.
Robert G. Bednarik
International Centre for Rock Art Dating
Hebei Normal University
robertbednarik@hotmail.com
REFERENCES
, P. B. and R. G. 2015. Concerning a
cupule sequence on the edge of the Kalahari Desert in
South Africa. Rock Art Research
Fingerlinientradition. Anthropologie 23: 73–79.
, R. G. 1992. A new method to date petroglyphs.
Archaeometry
, R. G. 1993. Geoarchaeological dating of petro-
glyphs at Lake Onega, Russia. Geoarchaeology
, R. G. 1995. The age of the Côa valley petroglyphs
in Portugal. Rock Art Research
, R. G. 1997. Direct dating results from rock art: a
global review. AURA Newsleer
, R. G. 2001. Petroglyphs in Italian Alps dated. Acta
Archaeologica
, R. G. 2002. First dating of Pilbara petroglyphs.
Records of the Western Australian Museum
, R. G. 2003. First microerosion calibration curve
for Iberia. ARKEOS – Perspectivas em Diálogo
, R. G. 2007. Rock art science: the scientic study of
palaeoart, second edition. Aryan Books International,
New Delhi.
, R. G. 2009. Fluvial erosion of inscriptions and
petroglyphs at Siega Verde, Spain. Journal of Archaeological
Science
, R. G. 2012. U-Th analysis and rock art: a response
to Pike et al. Rock Art Research
, R. G. 2015a. The tribology of cupules. Geological
Magazine
, R. G. 2015b. Kinetic energy metamorphosis of
, Horizons in Earth
Science Research 13, pp. 119–
New York.
, R. G. 2017. Developing ICRAD. Rock Art Research
–115.
, R. G. 2019. Tribology in geology and archaeology.
Nova Science Publishers, New York.
progress report. Purakala (in press).
, R. G. and M.
Arabian rock art. Rock Art Research
, R. G. and M. 2017. New rock art complex in
Saudi Arabia. Rock Art Research
weathered basaltic crust in absolute chronology. Nature
-
derthal. International Newsleer on Rock Art
, H. W. 1951. Altersbestimmung an Sinter mit radio-
Die Höhle
Kalksinter, ihre Bedeutung für die C-Altersbestimmung
Geologisches Jahrbuch
why U–Th is the way to go. Quaternary International
and 2003. Cross dating (Th/U-C) of
calcite covering prehistoric paintings in Borneo. Quater-
-
and
Uranium-series dating of carbonate formations overlying
Palaeolithic art: interest and limitations. Bulletin de la
Société Préhistorique Francaise
and R. G. 2015.
The Kalatrancani petroglyph complex, central Bolivia.
Rock Art Research 32(2): 219–230.
and R. G.
Rock Art Research 35(1):
and
2017. Uranium-thorium dating method and Palaeolithic
rock art. Quaternary International
W. and R. G.
microerosion dating project in China. Rock Art Research
W. and R. G.
Rock Art Re-
search
engravings. Appendix in W. D. L. Ride and A. Neumann
(eds), Depuch Island
Western Australian Museum, Perth.
RAR 36-1280
