ArticlePDF Available

The Impact of Industrial Pollution on the Rock Art of Murujuga, Western Australia

Authors:
  • WA Australian Coastal Society

Abstract and Figures

MacLeod and Fish have recently suggested that there is no adverse impact on the engraved rock art of Murujuga (the Burrup Peninsula) from industrial pollution. This highly controversial conclusion demands examination because it could influence future government decision-making concerning ongoing applications to expand industrial activity on Murujuga. We, therefore, review the data and arguments underpinning that conclusion. We find the conclusion unsubstantiated, misleading and potentially damaging for the long-term preservation of the Murujuga rock art. Evidence suggests that the petroglyphs are already actively degraded by industrial pollution.
Content may be subject to copyright.
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
3
KEYWORDS: Petroglyph – Industry emissions – Rock varnish – Acid rain – Murujuga – Scientic integrity
THE IMPACT OF INDUSTRIAL POLLUTION ON THE
ROCK ART OF MURUJUGA, WESTERN AUSTRALIA
Benjamin W. Smith, John L. Black, Stéphane Hœrlé, Marie A. Ferland,
Simon M. Diey, Jolam T. Neumann and Thorsten Geisler
Abstract. MacLeod and Fish have recently suggested that there is no adverse impact on the
engraved rock art of Murujuga (the Burrup Peninsula) from industrial pollution. This highly

decision-making concerning ongoing applications to expand industrial activity on Murujuga.
           
conclusion unsubstantiated, misleading and potentially damaging for the long-term preser-
vation of the Murujuga rock art. Evidence suggests that the petroglyphs are already actively
degraded by industrial pollution.
Introduction
Murujuga and its rock art are of exceptional global

incorporating Murujuga, has over a million rock art
engravings or petroglyphs said to capture more than
50 000 years of Indigenous knowledge and spiritual
beliefs. The Murujuga rock art is among Australia’s
most extraordinary Aboriginal heritage sites and is
of Outstanding Universal Value (Australian Heritage
Council 2012; DBCA 2021). The rock art is thought to
contain some of the earliest known representations
in the world of the ‘human face’. It includes some of
the world’s oldest complex geometric designs, extinct
animals including the ‘fat-tailed kangaroo’ and ‘thy-
lacine’, as well as recording the evolving fauna of the
area through global climate change during and after
the last Ice Age (McDonald and Veth 2009; Mulvaney
   
both to current Aboriginal Custodians and to the global
community.
From the 1960s, sections of the Burrup Peninsula
have been increasingly industrialised, initially with
an iron ore port and salt production facility and, from
the late 1970s, with large petrochemical industries,
including natural gas plants and others that use the
by-products from gas liquefaction. The area is now

in the Southern Hemisphere’ (Mulvaney 2011b: 17).
Aside from the unconscionable direct destruction of
petroglyphs during the construction of this industry

upon the rock art of polluting emissions have been pre-
sented and discussed in many reports and publications
(e.g. MacLeod 2000, 2011; Bednarik 2002, 2007a, 2007b,
2009; Lau et al. 2008; Mulvaney 2011b; Moodie 2016;
Parliament
of Australia 2018; González Zarandona 2020). Conclu-
sions from governmental reports have been shown
to understate the evidence of damage to the rock art

Black et al. 2017a).
Ian MacLeod and Warren Fish have recently pub-
lished a conference paper (MacLeod and Fish 2021) in

to the factors controlling the decay mechanisms on
engraved rocks in the Pilbara region of Western Aus-
tralia’. Given the massive volume of published and
grey literature on this topic, this is a big statement.
The paper concludes that ‘the present monitoring data
shows that there is presently no adverse impact on the
rock engravings from industrial pollution …’. This is

expectations of decades of studies (Bednarik 1994, 2002,
2007b, 2009; MacLeod 2005, 2011; Mulvaney 2011b;

Moodie 2016; Black et al. 2017b; Ramanaidou et al.
2017; Data Analysis Australia et al. 2018). The basis of
this claim is a set of pH, nitrate, chloride ion and redox
measurements on rock surfaces at eight sites across
Murujuga, as well as some rock surface colour mea-
surements. We review the current understanding of
factors controlling the synthesis and decay of Murujuga
rock surfaces to assess whether the extraordinary claim
made by MacLeod and Fish (2021) can be considered

Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
4
The red-brown rocks of Murujuga – the living rocks
The rich red-brown colour that characterises the
visible faces of the granophyre and gabbro rocks of
Murujuga is not the original natural colour of the rocks,
which is blue-grey. Tectonic forces and temperature
   

new rock faces degrade extremely slowly (Pillans and
   
weathering rind, or leach zone, comprising a mixture
        
5–10 mm in 30 000 years depending on the rock type
(Bednarik 1979, 2007b; Donaldson 2011; Ramanaidou
and Fonteneau 2019). The weathering rind is covered
by the hard, red-brown surface outer layer that is
only 1 to 200 microns in thickness (Liu and Broecker
2000; Ramanaidou and Fonteneau 2019). It comprises
a ferromanganeous crust, known as the patina or rock
varnish. The rock art was made by breaking
through that patina and into the weathering rind
to provide a colour and contour contrast (Fig. 1).
Evidence from studies into the growth of
the ferromanganeous patina on rocks in simi-
lar environments to Murujuga shows that it is
formed by biomineralisation processes, where
budding bacteria and micro-fungi concentrate
manganese and iron compounds and form the
hard, cement-like components of the patina
(Miller et al. 2012; Dorn 2020; Lingappa et al.
2021). These microbes have evolved in extremely
harsh, low moisture, high-temperature rock
surface environments and have developed
strategies to protect themselves against radiation
and oxidative stress.
Dust is the primary source of manganese
and iron oxide on rock surfaces (Bednarik 2002;
Macholdt et al. 2019). Manganese oxide arriving
on the rock surface is reduced to H-Mn++ ions by pho-
tochemical or biological processes when moisture from
dew is present (Lingappa et al. 2021). The H-Mn++ ions

cyanobacteria, Chroococcidiopsis, with manganese being
concentrated up to 50 times or more from that in the
local environment. Chroococcidiopsis is the dominant
organism living on rock surfaces in extremely dry and
harsh desert environments, whether in the Antarctic or
hot deserts (Mishra 2020). This bacterium has evolved
to tolerate extreme desiccation, ionising radiation and
ultraviolet light and can maintain normal metabolic
processes following the loss through desiccation of
more than 50% of its weight (Kvíderová et al. 2011).
Lingappa and colleagues (2021) report that the
manganese compound in the patina is predominantly
in the Mn4+
mixtures of Mn4+and Mn3+. These authors consider that
this mixing of manganese
ion types is consistent with
the oxidation-reduction
recycling process occurring
within the patina, where
it is part of an ecosystem.
Figure 2 shows examples of
electron microscope images
of budding bacteria in rock
varnish and the concen-
tration of manganese and
iron within the bacteria.
The rock varnish-forming
microbes are thought to lie
dormant for much of their
lives, becoming active only
when moisture concen-
trations are sufficient for
active metabolism. When
the organisms die, their
biomass is oxidised, and
the magnesium and iron-
Figure 1. An example of a rock art panel showing part of the
Murujuga landscape (photograph: BWS).
Figure 2. Electron microscope images of rock varnish showing the bacteria sheaths
(hyphae) concentrating manganese-iron-rich material in rock patina (left) and a
budding hypha is emerging from a cocci bacterial form (right). The centre boom
image is an elemental spectral analysis of the hyphae (left) compared with the adjacent
patina (centre top), showing marked concentration of manganese (Mn) and iron (Fe)
in the bacteria (right image adapted from Dorn 2020: Fig. 13.3(a); left and centre
images from Dorn 2020: Fig. 13.2 originally from Krinsley et al. 2009: Fig. 5(B).
Images with permission from R. I. Dorn – The American Geographical Union grants
permission to use gures in academic works).
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
5
rich components are incorporated
with clays into the rock patina (Dorn
2020; Lingappa et al. 2021). Bacterial
remains and fossilised fungi have
been observed within the patina of
rocks collected from the Pilbara in
Western Australia, with Murujuga
being part of this region (Flood et al.
2003). Krinsley et al. (2017) suggest
that as few as one bacterium every
~400 years may be incorporated into
rock varnish under warm desert con-
ditions. Consequently, the rock patina
grows at extraordinarily slow rates of
1 to 10 microns per 1000 years (Dorn
and Meek 1995; Liu and Broecker
2000; Dorn 2009).
The manganese and iron oxides
and hydroxides in the patina are
formed only under near-neutral
and alkaline conditions (Goldsmith et al. 2014). The
ratio of manganese to iron compounds varies with

local acidity-alkalinity (Dorn 1990; Broecker and Liu
2001). Proportionately more iron compounds form in
drier conditions with a more alkaline environment
(Dorn 1990). The colour of the patina varies with the
proportion of darker manganese compounds relative
to the proportion of redder ferrous oxide compounds.
When suitable conditions exist, the patina is generally
formed in layers, but it can also be largely amorphous
and then is known as ‘rubbly varnish’ as illustrated in
Figure 3 (Garvie et al. 2008).
Over periods of 10 000 to 30 000
years, the outer patina can grow over
the weathering rind of the original
petroglyph to reduce the colour con-
trast and show primarily a contour or
texture contrast (Fig. 4). Critically, the
petroglyphs are lost if the outer patina
is removed either by being dissolved
or detached from the underlying
weathering rind (Fig. 4).
The inuence of pH
on Murujuga rock art decay
Published studies agree that the
rich red-brown patina of the Muruju-
ga rocks, as with other forms of rock
varnish, is dissolved with increasing
acidity (Bednarik 2002, 2007b; Ma-
cLeod 2005, 2011; Black et al. 2017b;
cf. Dorn 2020). Even a ‘[s]ubtle change
[…] in surface pH of only 0.3’ (Ma-

future survival of the rock art.
The impact of increasing acidity on
the chemical changes occurring within
the patina is most clearly illustrated by
consideration of the Pourbaix diagrams for manganese
and iron compounds (Black et al. 2017b). Figure 5
shows the Pourbaix diagram for manganese. The
diagram shows the three equations from MacLeod
and Fish (2021) for the synthesis of Mn3O4 (equation 1,
circled 16), MnO2 (equation 2, circled 20) and Mn(OH)22+
(equation 3, circled MF3). The range of rock surface pH
values reported in MacLeod (2005), Black et al. (2017b),
and MacLeod and Fish (2021) are also shown on the
diagram, as well as the domain of stability of water
where water is thermodynamically stable.
As acidity increases (i.e., pH falls), the manganese
Figure 3. Optical micrographs (A and B) of ultrathin sections of rock varnish
showing layering (LV) with dierent colours and more amorphous ‘rubbly
varnish’ (RV) overlaying a quar (Q) rock surface (from Garvie et al.
2008, Fig. 1. ‘Fair use’ permission presumed from The Geological Society
of America).
Figure 4. An ancient petroglyph where the patina has grown over the weath-
ering rind to provide only a contour contrast from the background rock
(centre-left) and where the patina has detached from the weathering rind
(centre-right) and the portion of the petroglyph removed (from Pillans
and Field 2013; with permission from B. Pillans, and through Elsevier
License 5163861131736).
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
6
compounds in the patina are altered through a series
of chemical transformations accompanied by colour
changes (brown-black Mn3O4, black Mn2O3, brown or
black MnO2
as pale pink) to Mn++,+++ soluble ions. The soluble
manganese ions may then be washed from the rocks
in rainfall events.
There is ample evidence that a high concentration
of natural organic acids will dissolve the manganese
and iron compounds that harden rock patina. The
outer patina is completely removed from rock surfaces
at Murujuga with natural acid from trees (Bednarik

2017) and also from areas of rock surfaces under lichens
(Dorn 1990; Dragovich 1987) and microcolonial fungi
(Dragovich 1993) at other desert locations in Australia
and southern United States of America (U.S.A.). For
this reason, it is not a surprise that acidic industrial
emissions have been found to degrade or destroy rock
surfaces (Dorn 2020; Giesen et al. 2014).
Industrial NOx emissions also cause an increase in
nitrite and nitrate concentrations on Murujuga rock
surfaces. MacLeod (2005) showed that there was a log
(10-fold) increase in the growth of bacteria, yeasts,
fungi, and lichens for each unit increase in rock surface
nitrate. These organisms release organic acids that
reduce the pH of the rock surface. There is a strong
relationship between nitrate and pH on Murujuga
rock surfaces (MacLeod 2005; MacLeod and Fish 2021).
Gleeson and colleagues (2018) speculate that changes
to the microenvironment of the rock patina on Muru-

species and metabolism of the microorganisms on the
rock surface. Large numbers of invasive opportunistic

      -
cient quantity will dissolve the manganese and iron
compounds in the patina (Dorn 1990). Second, these
organisms will outcompete the specialised patina form-
ing Chroococcidiopsis strains, which evolved to survive
in extremely low nitrogen rock surface environments

2019).
Bednarik (2009) states that ferromanganeous patina
will not occur in areas where the pH of the rain is 5.6
or less, which is consistent with the observation that
manganese ions leach from dust or rock varnish when
pH falls to 5.7 (Goldsmith et al. 2014;.
Osborn and colleagues (1981) measured the pH of rain
in the Arizona rangelands, where rock varnish is nota-
Figure 5. Pourbaix diagram of Mn (manganese) compounds. With increasing acidity, the dierent Mn oxides
progressively become unstable (from Mn3O4 to Mn2O3 to MnO2) until only dissolved Mn ions can exist.
Any change from the natural pH (blue dashed line) results in surface loss or fresh mineral deposits, which
changes the chemical composition of the rock surface and, thus, the appearance of the rock art (MacLeod and
Fish 2021: 4). Equation 3 (MacLeod and Fish 2021:5) is at about MF3; however, it does not normally appear
on the Pourbaix diagram (Hem 1963). Diagram adapted from Pourbaix (1974) and Black et al. (2017b) with
additional data from Black et al. (2017b); and MacLeod and Fish (2021).
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
7
bly widespread, to have a mean of 6.8 in places where
no polluting industries were operating. Bednarik (2002:
36) states that the pH of natural rain in the Pilbara re-
gion of Western Australia was ‘in the order of 6.8 prior
to industrial development, and reached about pH 7.0
to 7.2 at Murujuga’.
   -
nation of how pH changes on rock surfaces following
a rain event with a lower pH due to the dissolution of
atmospheric carbon dioxide. They propose that rain,
-
ly low to dissolve manganese from dust, but when it
lands on rocks, reactions occur that remove protons
from the carbonic acid (occurring naturally in rain) to
form alkaline minerals. As the sun dries the rocks, the
carbon dioxide from the carbonic acid is released, while
the alkaline metals remain and further increase pH
(i.e., raise alkalinity) with manganese hydroxides from
Mn2+ to Mn4+ continuing to form until pH 8 is reached.
Similarly, Goldsmith and colleagues (2014) report that
moist dust from the Negev desert in Israel has a pH of
about 8 and an EH of about 0.6 V. Dorn (1990) found a
strong correlation between soil pH and the pH of rock
varnish at 67 sites across Arizona and California in the
U.S.A. The pH of the rock varnish was predominantly
in the range 6.7 to 9, with values lower than these being


produce organic acids.
MacLeod and Fish (2021: 4) assert that ‘the natu-
ral pH of local weathered gabbro and granophyre is
5.5±0.2’. This stands in striking contrast to MacLeod’s
own recordings (Black et al. 2017b) and the known
pH of rocks upon which ferromanganeous patinas
survive in other parts of Australia and the world. Such
an allegedly low pre-industrial rock surface pH cannot
be reconciled with the observations made by Bednarik
(1979), who found that the mean pH was 5.9 across 30
sites on Murujuga where bird droppings had dissolved
the rock patina. Bednarik’s evidence suggests that,
contrary to MacLeod and Fish (2021: 4), rock art will
be completely degraded over time on any rock that

With this broader knowledge of the natural for-
mation processes of rock patinas and the deleterious

       
MacLeod and Fish (2021: 6). From the eight sites that
they monitored between 2017 and 2019, all of the rocks
       
event on 6 June 2018, average pH levels had reached
an acidity in which manganese ions are known to be
leached (all falling to between pH 5.2 and 4.4). Whilst
MacLeod and Fish place much emphasis on the impact

they present shows that rock surface acidity returns to
unacceptable levels soon after even a major rain event.
The pH of rocks taken from Murujuga to the Western
Australian Museum before industrialisation was found
to be 6.8±0.2 (Black et al. 2017b), which corresponds

by organic acids in the U.S.A. (Dorn 1990). By contrast,
the mean surface pH of the eight rock sites monitored
by MacLeod and Fish was 4.71 in 2019. This change in
average pH from pre-industrial times is dramatic and,
according to the Pourbaix diagram and other evidence

cause the patina to dissolve.
A rain event does not truly ‘reset’ the decay clock,
as MacLeod and Fish (2021) suggest. It dissolves some
of the dry, acidic dust pollution from industry that has
been deposited on the rock surfaces. As the dust mixes
with water, the resulting solution bathes the surfaces
in highly corrosive acids that will dissolve the rock
patina in the same way that acid rain eats away at stone
buildings and monuments in polluted cities. Very light
rains, the most common type at Murujuga, as well as
morning dews (which can be heavy during winter
months), react with the dry SO2 and NOX pollution
particles, oxygen and other chemicals to form sulphuric
(pKa, or acidity, -3.0) and nitric (pKa -1.37) acid. These
acids can reach very low pH and be highly corrosive to
ferromanganeous patinas. In most years, the amount
   
remove the acidic dust from most surfaces (Clark et al.
2018). It is only during heavy rains, typically cyclonic
events, when this acidity is truly washed away. Even
this is context-dependent. Smooth granophyre rocks
may be washed clean, whilst the rough gabbros may
   

pores. Major cyclonic events have been recorded on
average twice a decade over the past forty years at
Murujuga (Sudmeyer 2016; Holmes 2021). Whilst this
cyclonic removal of acidic particles assists the survival
of the rock surface patinas, such events happen too

   
away dissolved manganese and iron ions from the
pores in the rock surface and diminishes the integrity
of the patina (Andreae et al. 2020).
As a result, for the bulk of the time, the rocks at
Murujuga are currently subjected to a level of acidity
known to be corrosive to ferromanganeous patinas.
Our own commissioned pH testing in 2017, conducted
by MacLeod and reported in MacLeod and Fish (2021),
shows that rocks at site 4, adjacent to the Woodside
petrochemical plants, have an average pH of below 4.

to cause loss of iron and manganese particles from the
types of patinas that exist at Murujuga and that this
damage becomes extreme below a pH of 5.7 (Bednarik

follows that the average rock surface measurements of
pH 5.29 (2017), pH 5.51 (2018), pH 4.71 (2019) reported
by MacLeod and Fish (2021), even though these years
included a cyclonic event and a massive industrial
ammonia leak (alkaline) (MacLeod and Fish 2021: 4),
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
8

rock surfaces on the majority of days in the majority of

leads to surface loss’ (2021: 4).
Colour changes on the rocks
Colour change should be an important proxy for
patina decay, since any leaching of manganese and
iron compounds from the rocks will cause the surface
to become lighter in colour (Black et al. 2017b). In prac-
tice, the measurement of colour change on the rocks
of Murujuga has proven to be challenging. CSIRO at-
tempted it as part of a Western Australian government
monitoring program at Murujuga from 2004 until 2014.
The highly variable results were heavily criticised, both
for the method by which they were generated and
for how they were analysed and interpreted (Black et
al. 2017a; Data Analysis Australia 2016, 2017). There
were repeated inconsistencies in results from year to
year that arose from changing the colour measuring
-
ditions under which the readings were taken
(Black et al. 2017a). Black and colleagues
(2017a) also demonstrated that precise colour
change monitoring is almost impossible;
they provided the example of a close-up
photograph of a Murujuga rock surface and
demonstrated that the displacement of the
colour instrument measuring head by only
0.5 mm from the previous measurement
location could result in a substantial change
in measured colour (Fig. 6 and Table 1).

(2016) reanalysed the CSIRO 2004 to 2014
data (Markley et al. 2015) and found that
70% of all spots measured (engraving and

(L* increased) and that none were darker
over the decade of measurement. CSIRO
themselves had claimed that there was no
consistent trend in colour change, but these
claims were made without statistical analysis

the method, the extent of colour degradation
  
enough over a decade for it to be evident,
even in a colour measuring program that has
been shown to be imprecise.
   
one of the colourimeters formerly used by CSIRO and
that they measured average colour across the surface,
    
the same rock. Therefore, their work is subject to the

(Black et al. 2017a). MacLeod and Fish (2021: 7, Fig 8)
show colour contrast between background rock and
engraved areas for 2019 but present results for only
four of the eight sites because 
the regression (site 7), were claimed to be statistically
-


Figure 8 does 

pH of the surface. Indeed, they conclude that ‘[w]ithout
pH monitoring data, it is unwise to base conservation
   
Figure 6. Three square areas (A, B, C) with the same area as a 3 mm
diameter aperture positioned on a photograph of the surface patina
of a Murujuga rock (Bednarik 2007b: 224, Figure 19) and moved
0.5 mm to the right in the lower panels (reproduced from Black et
al. 2017a: Fig. 6).
Table 1. L*, a* and b* colour space values predicted for each square with the same area as a 3 mm diameter aperture
using the Image Color Summarizer (Krzywinski 2017) (reproduced from Black et al. 2017a: Table 6)
Square Position L* a* b*
Avg Median Min Max Avg Median Min Max Avg Median Min Max
A 0, 0 51 51 29 85 26 26 10 44 22 22 6 50
B 0, 0 46 46 17 80 25 25 1 47 19 17 -8 47
C 0, 0 59 60 22 79 34 35 16 52 29 29 1 49
A 0.5, 0 50 51 14 75 24 24 8 41 19 19 -2 46
B 0.5, 0 48 48 17 78 25 25 1 49 19 18 -8 51
C 0.5, 0 55 56 22 76 30 31 8 52 24 26 0 46
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
9
between the engraved and background areas of
the images.’ It is clearly not on an alleged colour
change that MacLeod and Fish base their extraor-
dinary conclusion that industrial pollution does
not impact the rock art.
Acidic emissions
MacLeod and Fish (2021) assert that ‘Wood-
side Petroleum introduced low NOx burners
between 2005 and 2015, which resulted in great-
ly reduced nitrate levels’, and this seems to be
one reason for their optimism that ‘the present
monitoring data shows that there is presently
no adverse impact on the rock engravings from
industrial pollution …’. However, they provide
no NOx     
reduction. The claim appears to be based on the
supposition that rock surface nitrate concentra-
tions are lower on average in 2017–19 than measured by
MacLeod (2005) in 2003–04. However, this supposition
is invalid because:
i)    

2017–19;
ii) MacLeod and Fish (2021) chose the highest value of
6.53±5.1 ppm recorded in 2003–04 for comparison;
iii) MacLeod and Fish (2021) report only a mean val-
ue of 0.7±0.5 ppm, when there was a wide range
in nitrate concentration sufficient to develop a
regression equation relating nitrate concentration
to pH; and
iv) MacLeod and Fish (2021) did not mention that ni-
trate measured pre-industrialisation was 0.3 ppm
(MacLeod 2005).
The emissions data reported by the industry at
Murujuga gives less room for optimism. Woodside

8900 tonnes of NOx annually. On top of the
Woodside and ongoing shipping emissions, since
2017, there have also been new and increasing
NOx emissions from Yara Pilbara Nitrates. Yara
Pilbara reported recently that their fertiliser plant
     
x
annually (Yara Pilbara 2019).
Contradictory to statements by MacLeod and
Fish (2021), the industry emissions data reported
to the National Pollution Inventory (2020) illus-
trate that NOx emissions rose steadily from 2000
until 2017 (Fig. 7). Woodside explained the appar-
ent drop in 2014 of Karratha Onshore Gas Plant
emissions as a change in the way emissions were
calculated rather than a real change in emissions
(Woodside 2019b). The reported tenfold drop in
Pluto onshore gas treatment emissions from 2018
to 2020 has not been explained. Given that there
were no reported major adjustments to any of
the main industrial processes or facilities in the
year of this ten-fold reduction (National Pollution
Inventory 2020), the reduction needs explanation.
Even if the NOx emissions have declined over the
last three years, the emission rate across Murujuga re-
mains high enough to damage the rock art, as seen by
the continuing low rock surface pH. A reduction in NOx
concentrations from earlier years is not the criterion for
assessing whether the rock art is being damaged. The
criterion is the quantities of acidic compounds from
industrial emissions deposited on the rock surfaces,
which will dissolve the outer rock patina. Clearly,
from the discussion above, when rock surface pH falls
below 5.9, the long-term survival of the rock art is in
jeopardy (Bednarik 1979).
The distribution of atmospheric nitrogen dioxide
across Murujuga is clearly illustrated in the data
from the Copernicus satellite for a week in 2021 (Fig.
8). Although the highest concentration is located
directly over the Woodside facilities and notably in
Figure 7. NOx emissions as reported annually by industry at
Murujuga (from National Pollution Inventory 2020).
Figure 8. Mean values of NO2 concentration over seven days
(26/4/2021 – 2/5/2021) derived from satellite data (Copernicus
data hub 2021). Analysis and map by Christopher Swain.
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
10
the immediate vicinity of various sampling sites, the
impact is evident across a wide area of Murujuga. A
similar distribution of NOx deposition due to industry
across Murujuga and adjacent islands was reported
by Parsons et al. (2021) in their study for the Western
Australian Government into the impact of cumulative
air emissions across Murujuga.
As we have noted above, nitrogen oxides form acid
rain when removed from the atmosphere during a rain-
fall event, and they also undergo dry deposition as dust
falling on the rocks at other times. This nitrous-rich
dust then forms nitric acid during a period of light
rainfall and dew. Results from chemical deposition
monitoring in 2012–2014 provided by Woodside show
     
deposited on the rocks (Woodside 2019a, Appendix
E – Air Quality, Table 3-8).
We have presented evidence on NOx emissions, but
emissions of SOx, particularly from shipping, are also
extremely high (Black et al. 2017a), and these will be
having an equally profound impact on the rock patinas
of Murujuga, forming sulphuric acid during periods
of light rainfall and dew.
MacLeod and Fish’s (2021) claim that these massive
and highly acidic industrial emissions are having no
impact on Murujuga rock art is, therefore, simply not
    -
dence presented. What then is the basis of the claim?
We are genuinely unsure, but it seems to relate to Ma-
cLeod and Fish’s (2021) term ‘precipitation’.
‘Precipitation’
There is a suggestion by MacLeod and Fish (2021)
that somehow rock patinas, whilst currently etched by
acidity during the bulk of the time, are quickly recuper-
ating during the brief interludes (presently only a few

level above 6.5. Unlike previous studies, MacLeod and
Fish (2021) emphasise that ‘alkalisation (from sea salt
or ammonia leaks) causes fresh minerals to deposit and
thereby change the appearance of the engravings’ (p. 4).
They report observing from photographs ‘an increased
amount of the purple-black patina’ on the sample
rock art from Deep George (Nganjarli) after a massive
ammonia leak event from Yara Fertilisers that had
turned rock surfaces in this area unusually alkaline.
The following year, they found that the reverse side
of the rock ‘had a deep-purple patina associated with
the presence of Mn3O4
redox (voltage) measurements. Unfortunately, no L*,
a* and b* data to quantify the colour measurements are
presented. They explain the colour changes in terms of

that this new ‘patina’ was dissolving (p. 5), presumably
as the rock acidity returned to below a pH of 6.5. They
describe having seen the same process operating at
two other sites among the eight they were monitoring.
They conclude that ‘dry deposition of ammonia and
wind-borne sea salts works to mitigate the release of
iron and manganese minerals from the Murujuga rock
engravings’ (p. 8). This claim seems to lie at the heart
of MacLeod and Fish (2021) assertion that ‘there is
presently no adverse impact on the rock engravings
from industrial pollution …’.
As we have described in this paper, the ferroman-
ganeous rock patinas of Murujuga were formed by bio-
mineralisation processes where microbes concentrate
manganese and, to a lesser extent, iron compounds
so as to form the hard cement-like components of the
patina (Miller et al. 2012; Dorn 2020; Lingappa et al.
2021). They do this only under alkaline conditions,
and it happens at extraordinarily slow rates of 1 to 10
microns per 1000 years (Dorn and Meek 1995; Liu and
Broecker 2000; Dorn 2009). Dorn and Krinsley (1991)
showed clearly that reprecipitation of manganese and
ferric compounds following acidic dissolution removes
the natural layering of the rock varnish, increasing
porosity and redistributing the manganese and ferric
deposits along rock wall fractures and not as a new
form of varnish. Thus, even if reprecipitation of man-
ganese and iron compounds occurred, it would not be
within the hyphae of dead microbes, and the structure
and morphology of the patina would be changed with
a detrimental impact on the long-term survival of the
rock art. We doubt that MacLeod and Fish (2021) have
observed examples of this process, and they certainly
do not provide evidence of it. Whatever the cause of the
colour changes observed by MacLeod and Fish (2021),
their data shows it was for a short period, and the
re-deposition was already being dissolved by acidity
within just a year.
Therefore, the bigger picture is what is important
here, namely that the industrial emissions at Murujuga
   and thereby
pose a constant threat to the integrity and survival
of the rock art. Occasional industrial ammonia leaks
and extreme weather events that temporarily alleviate

     
and cannot reverse the ongoing deterioration of the
rock art. The cause for genuine concern is real and
cannot be wished away, particularly as new industrial
developments are currently proposed that will result
in increased acidic emissions.
Conclusion
A detailed understanding of the rock decay pro-
cesses at Murujuga has evolved over more than 20
years from painstaking local research, and this has
    
the formation processes of ferromanganeous patinas in
low rainfall environments. The evidence we have pre-
sented clearly shows that, with the currently recorded
acidity levels, the rock patina and associated art will
degrade and disappear over time, as has occurred on
rocks with continual bird droppings (Bednarik 1979).

because the Western Australian State Government is
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
11
considering development applications for a further
50 years of natural gas processing and from two new
industries seeking to construct plants at Murujuga.
We caution in the strongest possible terms that the
       
not acceptable as an excuse to allow the placement of
further polluting industry at Murujuga. The statement
is not supported by current science.
 
is of such concern that existing industry should be
subjected to far more stringent emission controls to
bring Western Australia into line with other global
leaders. Excellent technology is available and utilised
successfully elsewhere that substantially reduces
emissions by industries of the type located at Muru-
juga. One of the parent companies of Murujuga’s
own Yara Pilbara is a market leader in manufac-
turing emissions control technology. The Western
Australian Government should heed the warnings of
contemporary science and ensure that new industry
is placed away from Murujuga, in one of the many

strict emissions control requirements are enacted to
reduce pollution at Murujuga and bring the State in

pollution and climate change. In direct contrast, the
Norwegian Government placed a tax on emissions
of NOx in 2007 (recently 23.48 NOK or AUD 3.40 per
kg NOx   
x in
forming acid rain and damaging human health. A tax
of this magnitude would raise approximately $30 m
annually from Woodside operations alone and would
undoubtedly change behaviour and help to ensure the
long-term preservation and protection of the Indige-
nous cultural heritage contained in the irreplaceable
Murujuga rock art.
Given that the bigger claims of MacLeod and Fish
(2021) are at odds with existing science, the onus is on

upon which they base their claims (including their
statistical methodology) and secondly, they should
make their data publicly available. MacLeod and Fish
acknowledge in their paper (2021: 3) that they both
work as contractors to the petrochemical industry
producing the acidic emissions at Murujuga. The tide
is turning against those that seek to play down the
impacts of industrial pollution upon the world. Even
within the industry, big vested interests are changing
tack and, for example, supporting the shift to sustain-
able energy and net zero-carbon emissions (Murray
2020; Samios and Harris 2021).
Acknowledgments
 -
original Corporation for their support. We are grateful for
the institutional support that we receive from the Centre for
Rock Art Research and Management in the School of Social
Sciences at the University of Western Australia and the

publication is produced as a part of the Murujuga Rock Art

of industrial pollution on Murujuga rock art, was funded by
ordinary citizens who gave an amount in excess of $285 000
through crowdfunding. This extraordinary level of public
support shows the extent of public concern that exists on this
topic. We thank each of the many hundreds of people that
supported this research. The views expressed in this paper


insightful comments of Dr Ilona Box. We thank Christopher
Swain for analysing the Copernicus satellite data and the
production of the NO2 map in Figure 8.
Conicts of interest
JLB and MAF are members of Friends of Australian Rock
-
lished in 2006 to protect, preserve and promote Australia’s
Aboriginal rock art, particularly the ancient petroglyphs of
the Dampier Archipelago, including Murujuga. We are all
scientists from a range of disciplines, determined to uphold

paper were voluntary.
Prof. Benjamin W. Smith
Centre for Rock Art Research and Management
University of Western Australia
Perth, WA 6009
and Rock Art Research Institute, Wits University, South
Africa
Benjamin.Smith@uwa.edu.au
Dr John L. Black
Centre for Rock Art Research and Management
University of Western Australia
Perth, WA 6009
Australia
John.Black@uwa.edu.au
Dr Stéphane Hœrlé
CNRS UMR 5199 PACEA
Université de Bordeaux
CS 50023, 33615
Pessac
France
and Rock Art Research Institute
School of Geography
Archaeology and Environmental Studies
University of the Witwatersrand
Johannesburg
South Africa
stephane.hoerle@sfr.fr
Dr Marie A. Ferland
P.O. Box 3309, East Perth, WA 6892
ferlandmarie@gmail.com

Apex Biometry Pty Ltd,
South Fremantle, WA 6162
Australia
simon@apexbiometry.com
Jolam Neumann and Prof. Dr Thorsten Geisler
Institut für Geowissenschaften (ehemals Steinmann-Institut)

Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
12
Universität Bonn
Meckenheimer Allee 169
D-53115 Bonn
Germany
jolam@uni-bonn.de, tgeisler@uni-bonn.de
REFERENCES

       and   2020.
Geochemical studies on rock varnish and petroglyphs in
the Owens and Rose Valleys, California. Plos One E 15(8):
e0235421. hps://doi. org/10.1371/journal.pone.0235421.
Australian Heritage Council 2012. The potential Outstanding
Universal Value of the Dampier Archipelago site and threats
to that site: a report by the Australian Heritage Council to the
Minister for Sustainability, Environment, Water, Population
and Communities, Australian Heritage Council, hps://
www.environment.gov.au/system/les/pages/5b14f51b-b7e1-
432f-8049-1e653713607d/les/outstanding-universal-values-
may2012.pdf, accessed 20 August 2021.
R. G. 1979. The potential of rock patination analysis
in Australian archaeology — part 1. The Artefact 4: 1438.
1994. Dampier rock art under siege. AURA
Newsleer 11(2): 14–15.
R. G. 2002. The survival of the Murujuga (Burrup)
petroglyphs. Rock Art Research 19(1): 29–40.
R. G. 2007a. Dampier rainwater as acidic as beer:
CSIRO. Rock Art Research 24(1): 139–141.
R. G. 2007b. The science of Dampier rock art — part
1. Rock Art Research 24(2): 209–246.
R. G. 2009. Experimental colorimetric analysis of
petroglyphs. Rock Art Research 26(1): 55–64.
 J. L. and S.  2016. Reanalysis of the colour
changes from 2004 to 2014 on Burrup Peninsula rock
     
Government, Department of Environment Regulation.
J. L., I.  and S.  2017a. Inadequacies of
research used to monitor change to rock art and regulate
industry on Murujuga (‘Burrup Peninsula’), Australia.
Rock Art Research 34(2): 130–148.
and 2017b. Theoretical

art sites on Burrup Peninsula, Western Australia. Journal
of Archaeological Science: Reports 12: 457–462.
H. 2019. The Cyanobacterium Chroococcidiopsis and
its potential for life on Mars. Journal of Astrobiology and
Space Science Reviews 2: 398–412.
W. S. and T.  2001. Rock varnish: recorder of
desert wetness? GSA TODAY 11(8): 4–10.
and 2018. Rainfall regimes
over northwestern Australia. Quarterly Journal of the Royal
Meteorological Society 144(711): 458–467.
Copernicus data hub 2021. Copernicus data hub. Downloaded
data in netCDF (*.nc) format, extracted in ASCII format

by Christopher Swain, hps://scihub.copernicus.eu/.
Data Analysis Australia, J.  and K. 
2017. Review of CSIRO report of Burrup Peninsula rock art
monitoring, Department of Environment Regulation,
Western Australia. https://www.der.wa.gov.au/images/
documents/our-work/consultation/Burrup-Rock-Art/DAA-
independent-review-report---May-2017.pdf, accessed 23
August 2021.
Data Analysis Australia, J. , K.  and
M.  2016. Review of statistical aspects of Burrup
Peninsula rock art monitoring, Department of Environment
Regulation, Western Australia. hps://www.wa.gov.au/sites/
default/les/2020-10/DWER-Murujuga-DAA-independent-
review-report-Nov-2016.pdf, accessed 23 August 2021.
Data Analysis Australia, J. , A. , M. 
and G.  2018. Analysis of Burrup Peninsula rock art,
Department of Water and Environment Regulation,
Western Australia. hps://www.yara.com.au/siteassets/about-
yara/reports/rock-art-monitoring-reports/analysis-of-burrup-
peninsula-rock-art-2017-daa.pdf/, accessed 23 August 2021.
DBCA 2021. Murujuga World Heritage nomination update.

Government of Western Australia. https://www.dpaw.
wa.gov.au/images/news/newsleers/Murujuga%20News%20
February%202021.pdf?goal=0_079762b204-4d8596aa86-
78172569&mc_cid=4d8596aa86&mc_eid=be71443e0d,
accessed 26 August 2021.
D, M. 2011. Understanding the rocks: rock art and
the geology of Murujuga (Burrup Peninsula). Rock Art
Research 28: 35–43.

in rock varnish microlaminations on western U.S.A.
volcanics. Palaeogeography, Palaeoclimatology, Palaeoecology
76: 291–310.
R. I. 2009. Desert rock coatings. In A. J. Parsons and A.
D. Abrahams (eds), Geomorphology of desert environments,
pp. 153–186. Springer Netherlands, Dordrecht.
R. I. 2020. Anthropogenic interactions with rock
varnish. In K. Dontsova, Z. Balogh-Brunstad and G.
L. Roux (eds), Biogeochemical cycles: ecological drivers
and environmental impact, pp. 267–283. Geophysical
Monograph 251, 1st edn, American Geophysical Union
and John Wiley and Sons, Inc.
D, R. I. and D. H. K 1991. Cation-leaching sites in
rock varnish. Geology 19: 1077–1080.
 R. I. and N.  1995. Rapid formation of rock
varnish and other rock coatings on slag deposits near
Fontana, California. Earth Surface Processes and Landforms
20(6): 547–560.
D. 1987. Weathering of desert varnish by lichens.
Readings in Australian Geography. Proceedings of the 21st
IAG Conference, Perth, pp. 407–412.
D. 1993. Distribution and chemical composition
of microcolonial fungi and rock coatings from arid
Australia. Physical Geography 14(4): 323–341.
N., E. , D.  and D.  2017.
Burrup Peninsula Aboriginal petroglyphs: colour change &
spectral mineralogy 20042016, No. EP161761, CSIRO.
hps://www.wa.gov.au/sites/default/les/2020-10/DWER-
Murujuga-Burrup-Peninsula-Aboriginal-Petroglyphs-Colour-
Change-Spectral-Miner.pdf, accessed 15 January 2017
B., C.  and T.  2003. Microbial fossils
detected in desert varnish. 34th Lunar and Planetary Science
Conference: Abstract #1633. Lunar and Planetary Institute,
Houston, TX.
L. A., D. M.  and P. R.  2008. Nanometer-
scale complexity, growth, and diagenesis in desert
varnish. Geology 36(3): 215–218.
 M. J., A. , P. A. , B. , A. D.
 and D. W.  2014. Condition assessment
and preservation of open-air rock art panels during
environmental change. Journal of Cultural Heritage 15(1):
49–56.
D. B., M. , B.  and J. L. 
2018. Rock-art microbiome: influences on long term
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
13
preservation of historic and culturally important
engravings. Microbiology Australia 39(1): 33–36.
Y., M.  and Y.  2014. From dust
to varnish: geochemical constraints on rock varnish
formation in the Negev Desert, Israel. Geochimica et
Cosmochimica Acta 126: 97–111.
J. A. 2011. The destruction of heritage:
rock art in the Burrup Peninsula. International Journal of
the Humanities 9(1): 325–341.
J. A. 2020. Murujuga: rock art, heritage,
and landscape iconoclasm. University of Pennsylvania Press,
Philadelphia.
 J. D. 1963. Chemical equilibria and rates of manganese
oxidation: chemistry of manganese in natural water. US

, J. 2021. Tropical cyclone impacts on the Western
Australian coast and extreme wind speeds in Region D.
Australian Journal of Structural Engineering 22(2): 110–119.
 D., R. I.  and B.  2009.
Astrobiological implications of rock varnish in Tibet.
Astrobiology 9(6): 551–562.
D., B.  R. I.    and
R. F 2017. Mn-Fe-enhancing budding bacteria in
century-old rock varnish, Erie Barge Canal, New York.
The Journal of Geology 125: 317–336.
M. 2017. Image Color Summarizer. Version 0.76,
hp://mkweb.bcgsc.ca/color-summarizer/?analyze, accessed
5 March 2017.
J., J.  and M. Šimek 2011. In situ response
of Nostoc commune s.l. colonies to desiccation in central
Svalbard, Norwegian High Arctic. Foea 11(1): 87–97.
D., E. , S. , A. , M.
, M.  and B.  2008. Burrup
Peninsula Aboriginal petroglyphs: colour change and spectral
mineralogy 2004–2007, CSIRO. hps://www.der.wa.gov.au/
images/documents/our-work/programs/Colour_Change_and_
Spectral_Mineralogy_2007_report_Sept2008.pdf, accessed
15 January 2017
U. F., C. M. , A. , N. L. , D.
P. , G. , A. D. , G. L. , D. R.
 and J. S.  2021. An ecophysiological
explanation for manganese enrichment in rock
varnish. Proceedings of the National Academy of Sciences
U.S.A. 118(25): e2025188118; https://doi.org/10.1073/
pnas.2025188118.
T. and W. S.  2000. How fast does rock varnish
grow? Geology 28(2): 183–186.
 D. S., K. P. , A.  and M. O.
 2019. Rock varnish on petroglyphs from the
Hima region, southwestern Saudi Arabia: chemical
composition, growth rates, and tentative ages. The
Holocene 29(8): 1377–1395.
I. 2000. Rock art conservation and management: the
past, present and future options. Studies in Conservation,
45(sup3): 32–45.

and microbiological activity on the surface pH of rocks
in the Burrup Peninsula. Triennial meeting (14th), The
Hague, 1216 September 2005, pp. 386–393. James & James,
London.
 I. 2011. Consuming corrosion – A journey into
the realm of materials conservation. 18th International
Corrosion Congress 2011, Perth. hps://www.researchgate.
net/publication/286294741_Consuming_corrosion_-_A_
journey_into_the_realm_of_materials_conservation, accessed
6 September 2021.
I. and W.  2021. Determining decay mecha-
nisms on engraved rock art sites using pH, chloride ion
and redox measurements with an assessment of the
impact of cyclones, sea salt and nitrate ions on acidity.
Paper presented to the ICOM-CC 19th Triennial Conference
Preprints, Beijing, 17–21 May 2021.
, T., M. , E. , D.  and
D.  2015. Burrup Peninsula Aboriginal
petroglyphs: colour change and spectral mineralogy
2004-2014. CSIRO, Australia. Report EP1410003. hps://
www.der.wa.gov.au/our-work/programs/36-burrup-rock-art-
monitoring-program.
J. and P.  2009. Dampier Archipelago

Heritage Listing. Archaeology in Oceania 44(S1): 49–69.
A., A. , M. S. , M. 
, M. , V. , L. , J. ,
A.  and S.  2012. Biogenic
Mn oxide minerals coating in a subsurface granite
environment. Chemical Geology 322: 181–191.
S. 2020. Cyanobacterial imprints in diversity and
phenology. In P. K. Singh, A. Kumar, V. K. Singh and A.
K. Shrivistava (eds), Advances in Cyanobacterial Biology,
pp. 1–15. Academic Press, New York.
C. 2016. Rock art decay ‘sure as night follows day’,
says retired Museum boss, ABC News, 10 December 2016.
hps://www.abc.net.au/news/2016-12-10/burrup-peninsula-
rock-art-concerns/8107114, accessed 18 August 2021,
K. 2011a. About time: toward a sequencing of the
Dampier Archipelago petroglyphs of the Pilbara region,
Western Australia. In C. Bird and E. R. Webb (eds), ‘Fire
and hearth’ forty years on: essays in honour of Sylvia J. Hallam,
pp. 30–49. Records of the Western Australian Museum,
Supplement 79, Perth.
K. 2011b. Dampier Archipelago: decades of
development and destruction. Rock Art Research 28(1):
17–25.
K. 2013. Burrup and beyond: a short guide to the area’s
cultural heritage and history. Rio Tinto Mining Company
of Australia, Perth.
 2020. Which major oil companies have set net-zero
emissions targets? NS Energy 16th December 2020, hps://
www.nsenergybusiness.com/features/oil-companies-net-zero/,
accessed 3 September 2021.
National Pollution Inventory 2020. Results — individual
facility detail. http://www.npi.gov.au/npidata/action/
load/individual-facility-detail/criteria/jurisdiction-facility/
WA0196/state/WA/year/2018, accessed 3 September 2021.
Norway Tax Administration 2021. NOx Tax. hps://www.
skaeetaten.no/en/business-and-organisation/vat-and-duties/
excise-duties/about-the-excise-duties/nox/, accessed 3
September 2021.
O, H. B., L. R. C and J. 1981. Rainwater
quality in southeastern Arizona rangelands. USDA – AEA
Southwest Rangeland Watershed Research Centre. hps://re-
pository.arizona.edu/bitstream/handle/10150/301257/hwr_11-
155-160.pdf?sequence=1, accessed 17 September 2021.
L. M., D. S. , K. P. , B. , U.
, B. , D. , G. H. , A. M. 
and M. O.  2020. Geochemical insights into the
relationship of rock varnish and adjacent mineral dust
fractions. Chemical Geology 551: 119775.
Parliament of Australia 2018. Protection of Aboriginal rock
art of the Burrup Peninsula. Senate Environment and
Communications References Committee, Parliament of
Australia, hps://www.aph.gov.au/Parliamentary_Business/
Rock Art Research 2022 - Volume 39, Number 1, pp. 3-14. B. W. SMITH et al.
14
Committees/Senate/Environment_and_Communications/
BurrupPeninusla/Report, accessed 1 March 2017.
P, M., J. J, J. G and B. B 2021. Study
of the cumulative impacts of air emissions in the Muru-
juga airshed. Ramboll Australia Pty Ltd., 41 St Georges
Terrace, Perth, Western Australia 6000, Australia.
P, B. and L. K.  2013. Erosion rates and weath-
ering history of rock surfaces associated with Aboriginal
rock art engravings (petroglyphs) on Burrup Peninsula,
Western Australia, from cosmogenic nuclide measure-
ments. Quaternary Science Reviews 69: 98–106.
M. 1974. Atlas of electrochemical equilibria in aqueous
solutions - 19. Manganese, technetium, rhenium. Vol. 307,
English transl., National Association of Corrosion
Engineers, Houston, TX, U.S.A.
 and L. C F2019. Rocky relation-
ships: the petroglyphs of the Murujuga (Burrup Peninsula
and Dampier Archipelago) in Western Australia. Austra-
lian Journal of Earth Sciences 66: 761698.
 E., G.  and D.  2017.
Extreme weathering experiments on the Burrup Peninsula/
Murujuga weathered gabbros and granophyres, CSIRO,
https://www.der.wa.gov.au/images/documents/our-work/
consultation/Burrup-Rock-Art/Extreme-Weathering-Burrup-
Report-2017-.pdf, accessed 28 August 2021.
and2021. Rupert Murdoch’s newspapers,
24-hour news channel to champion net zero emissions.
Sydney Morning Herald 6th September 2021. hps://www.
smh.com.au/business/companies/rupert-murdoch-newspapers-
24-hour-news-channel-to-champion-net-zero-emissions-
20210905-p58oyx.html, accessed 6th September 2021.
 R. 2016. Climate in the Pilbara. Bulletin 4873,
Department of Agriculture and Food, Western Australia,
Perth.
Woodside 2019a. North West Shelf Project Extension:
Environmental Review Document EPA Assessment No.
2186, EPBC 2018/8335, Revision 1 - December 2019,
Woodside. hps://les.woodside/docs/default-source/current-
consultation-activities/australian-activties/north-west-
shelf-project-extension---environmental-review-document.
pdf?sfvrsn=a8b10277_4, accessed 22 August 2021.
Woodside 2019b. Karratha gas plant annual environment
report. July 2018 – June 2019 Reporting period. hps://
parliament.wa.gov.au/publications/tabledpapers.nsf/display
paper/4013422cb6d1449dcd8dbb70482584ba0005bf78/$le/
tp-3422.pdf, accessed 3 September 2021.
Yara Pilbara 2019. Yara Pilbara Nitrates 2019 Annual Compli-
ance Report EPBC 2008/4546 Technical Ammonium Nitrate
Plant      hps://www.
yara.com.au/siteassets/about-yara/reports/annual-environ-
mental-compliance-reports-epbc-20084546/650-200-acr-y
pn-0007.pdf/, accessed 28 August 2021.
RAR 39-1370
... The nitrogenous substances stimulate adventitious microbial growth, which produce organic acids that can dissolve the rock surfaces upon which the rock art is engraved (MacLeod 2005). Nitrogen and sulphur oxides also combine with rain and dew to form highly caustic acids that are similarly able to dissolve the rock surfaces (Bednarik 2009;Black, MacLeod, and Smith 2017;Gadd 2017;Smith et al. 2022). ...
... Murujuga has become one of the world's most contested landscapes; a test-case as to whether Australia will take the hard actions needed to deliver on its commitments to emissions reductions and a sustainable future. We have demonstrated elsewhere that pollution will damage the rock surfaces and the petroglyphs of Murujuga (Black, MacLeod, and Smith 2017;Smith et al. 2022). ...
... The now exposed large flat surfaces on which the rock art sits were produced by tectonic forces and temperature changes over aeons. The weathering process of the rocks is described by Smith et al. (2022). Newly broken blue/grey rock faces degrade extremely slowly, producing an orange/yellow coloured weathering rind with a thickness of 5-10 mm over around 30,000 years (Pillans and Fifield 2013). ...
Article
Full-text available
This paper considers the effects of industrialisation upon one of the world’s most significant rock art sites, Murujuga (Burrup Peninsula), located in north-west Western Australia. Photographs of 26 petroglyphs taken prior to or early in the industrialisation of the area were compared with recent photographs to assess whether the presence of industry is accelerating degradation. Fifty per cent of the petroglyphs showed indications of changes, and two showed substantial damage. The bulk of the changes can be attributed directly to industrial activity in the area which commenced in the 1960s. All changed petroglyphs, with two exceptions, were in relative proximity to industry. A reduction in industrial emissions is considered essential if damage to the rock art is to be limited and this iconic cultural place is to remain largely intact for future generations.
... The observation of accelerated degradation would not be surprising, considering that in its National Pollutant Inventory the Australian Government lists at least 320 t of ammonia, nearly 9000 of nitrogen oxides, and more than 300 t of sulphur dioxide that were released into the air by the industrial complexes on the peninsula in 2019 alone [23]. Smith et al. showed, using observations from the Copernicus satellites, that nitrogen dioxide was widely distributed across the archipelago, and highly concentrated over the industrial area [24]. It is known that nitrogen oxides released into the air lead to acid rain [25] which could result in accelerated dissolution of the rock varnish on Murujuga and thus a faster decay of the petroglyphs [24]. ...
... Smith et al. showed, using observations from the Copernicus satellites, that nitrogen dioxide was widely distributed across the archipelago, and highly concentrated over the industrial area [24]. It is known that nitrogen oxides released into the air lead to acid rain [25] which could result in accelerated dissolution of the rock varnish on Murujuga and thus a faster decay of the petroglyphs [24]. Bednarik reported that the pH of natural rain in the Pilbara region of Western Australia was in the order of 6.8 prior to industrial development and reached values between 7.0 and 7.2 at Murujuga [18], which is essentially consistent with observed neutral pH values for rain in the Arizona rangelands, where desert varnish is notably widespread and no polluting industries were operating [26]. ...
... In particular, a recent publication by MacLeod and Fish concludes that industrial emissions have no adverse impact on the petroglyphs [32]. Smith et al. claim that this conclusion is based on unsubstantiated, misleading and scientifically unsound data [24]. Therefore, there is an urgent need to design and perform studies that can ultimately answer the question as to whether industrial emissions are leading to accelerated weathering of the rock varnish. ...
Article
Full-text available
Murujuga in Western Australia has the largest concentration of ancient rock engravings (petroglyphs) in the world. However, the Murujuga rock art is potentially threatened by local industrial air pollution, in particular by acid rain, but unambiguous scientific evidence is still missing. Here, we report on results of an accelerated weathering experiment, simulating Murujuga weather and climate conditions that was designed and performed to test whether the expected small changes in chemical, mineralogical, and physical characteristics of the rock surface can be detected and reliably quantified by various analytical means. Locally acquired Murujuga granophyre and gabbro samples with natural varnish were artificially weathered for up to four months in a climate chamber under conditions that simulated 2 years of natural weathering. Mineralogical, chemical, and physical changes were qualitatively monitored by X-ray diffraction and confocal Raman spectroscopy, and quantified by colorimetry, portable X-ray fluorescence spectrometry, and micro-computed tomography. In addition, artificial rainwater that was sprinkled over the rock samples was collected and analysed by inductively-coupled plasma mass spectrometry. The results show significant chemical and physical changes of the surfaces of the rock varnish after 1 month of artificial weathering. The analytical results demonstrate that it is possible to quantitatively monitor small changes caused by the weathering of gabbro and granophyre. Therefore, such a semi-actualistic experimental approach, when carefully designed, potentially allows testing the hypothesis that the weathering rate of the Murujuga petroglyphs is increased by local industrial air pollution. Further experimental work is currently under way.
... It is known that nitrogen oxides released into the air lead to acid rain which could result in accelerated dissolution of the rock varnish on Murujuga and thus a faster decay of the petroglyphs. 24 Measured pH of the rainwater in the township of Dampier dropped from about 7.0 in the 1960s 18 to a mean value of about 4.6 for sites monitored on Murujuga in 2007. Because both the solubility of Fe and Mn oxides and their dissolution rate increase dramatically when pH is lowered 18,, , it is evident that ongoing industrial development could erase the petroglyphs if industrial emissions were to continue at the levels currently reported. ...
... 31 Smith et al. claim that this conclusion is based on unsubstantiated, misleading and scienti cally unsound data. 24 Therefore, there is an urgent need to design and perform studies that can ultimately answer the question as to whether industrial emissions are leading to accelerated weathering of the rock varnish. Such studies must deliver robust and reliable data that can be used to support governmental decisions about further industrial developments on Murujuga and assist in the development of strategies to protect the rock art. ...
Preprint
Full-text available
Murujuga in Western Australia has the largest concentration of ancient rock engravings (petroglyphs) in the world. However, the Murujuga rock art is potentially threatened by local industrial air pollution, in particular by acid rain, but unambiguous scientific evidence is still missing. Here, we report on results of an accelerated weathering experiment, simulating Murujuga weather and climate conditions that was designed and performed to test whether the expected small changes in chemical, mineralogical, and physical characteristics of the rock surface can be detected and reliably quantified by various analytical means. Locally acquired Murujuga granophyre and gabbro samples with natural varnish were artificially weathered for up to four months in a climate chamber under conditions that simulated two years of natural weathering. Mineralogical, chemical, and physical changes were qualitatively monitored by X-ray diffraction and confocal Raman spectroscopy, and quantified by colorimetry, portable X-ray fluorescence spectrometry, and micro-computed tomography. In addition, artificial rainwater that was sprinkled over the rock samples was collected and analysed by inductively-coupled plasma mass spectrometry. The results show significant chemical and physical changes of the surfaces of the rock varnish after one month of artificial weathering. The analytical results demonstrate that it is possible to quantitatively monitor small changes caused by the weathering of gabbro and granophyre. Therefore, such a semi-actualistic experimental approach, when carefully designed, potentially allows testing the hypothesis that the weathering rate of the Murujuga petroglyphs is increased by local industrial air pollution. Further experimental work is currently under way.
... In such cases, an approach of benign neglect by relevant individuals or organisations needs to be replaced by active management. In other situations, damage to rock art can occur through apparently indirect mechanisms such as establishing polluting industries close to rock art sites, thus leading to increased rates of rock surface weathering [30,31]. Since the World Heritage nomination of the Murajuga Cultural Landscape, the Australian government has limited future industrial approvals to one new plant [32,33] and instigated a rock art monitoring program [34,35]. ...
Article
Full-text available
Using rock art conservation as a focus, this paper outlines the levels of legislated protection afforded to designated natural and cultural areas/sites in Australia and describes the co-management approach adopted in 1998 in relation to Mutawintji National Park in western New South Wales. The park encompasses four different protection categories: a Historic Site, a Nature Reserve, a National Park, and a State Conservation Area. Known for more than a century, the Historic Site is a major area of rock art containing Aboriginal engravings, paintings and stencils. Management of the Historic Site is a key concern, given the tourist interest and associated potential for accelerated deterioration of cultural heritage. The Mutawintji Plan of Management pointed to the importance of Mutawintji for Aboriginal people to connect with the country, and the co-management model encouraged tourism development as a means of providing employment opportunities as Aboriginal guides. No special legislative requirements in relation to rock art conservation, beyond those already in existence, were applied to the co-management system. Using field knowledge involving rock art research and early guide training programs at Mutawintji and literature sources, this paper suggests possible future approaches to rock art conservation in the Mutawintji Lands.
... While it's possible that the fires contributed to the accumulation of these particles, there are other sources of amorphous carbon that could be responsible. For example, industrial pollution could also produce these particles and deposit them on the object's surface [61,62]. However, considering the location of this open-air rock art shelter, which experiences absence of industrial interference, it is logical to conclude that the most probable source of amorphous carbon is related to fires. ...
Article
Full-text available
This paper investigates Prehistoric rock art painted figures attributed to the post-palaeolithic schematic style. Through SEM-EDS, Raman spectroscopy and X-microfluorescence analysis, a chemical-mineralogical characterisation is made, and natural alteration products (biological, by lichen and chemical, through the formation of patina) are investigated to assess the impact of lichen colonization on the open-air rock-art sites. The eventual aim is to develop a control strategy for the Juanita rock-shelter in Extremadura, Spain. The main mineral component in the red pigment (hematite) was identified using micro-Raman spectroscopy. Energy-Dispersive X-ray microfluorescence (ED-µXRF) registered iron, aluminium, and potassium as main elements in the pigments, indicating red ochre or earth ochre were used to produce them. The presence of calcium and phosphorus is linked to alteration, as both elements can be attributed to the substrate or to mineral accretion over the pigment itself. SEM observations have revealed the presence of early lichen colonization on the painted surface. Furthermore, amorphous carbon particles were identified on this surface. From these observations it is hypothesized that the state of conservation of the rock art is related to external factors (organic materials, patina, fire) more than to alteration of the substrate.
... 2017). Although recently it has been suggested that ammonia emissions, sea-spray and periodic rainfall events decrease the acidity (MacLeod and Fish 2021; see Smith et al. 2022 for contra view). terra australis 55 Once a member of the CSIRO team tasked by the state's BRAMMC with conducting the scientific studies into emissions and rock art, Eric Ramanaidou has continued his research involvement into Murujuga's petroglyphs. ...
... 2017). Although recently it has been suggested that ammonia emissions, sea-spray and periodic rainfall events decrease the acidity (MacLeod and Fish 2021; see Smith et al. 2022 for contra view). terra australis 55 Once a member of the CSIRO team tasked by the state's BRAMMC with conducting the scientific studies into emissions and rock art, Eric Ramanaidou has continued his research involvement into Murujuga's petroglyphs. ...
Book
Full-text available
Australia has one of the largest inventories of rock art in the world with pictographs and petroglyphs found almost anywhere that has suitable rock surfaces – in rock shelters and caves, on boulders and rock platforms. First Nations people have been marking these places with figurative imagery, abstract designs, stencils and prints for tens of thousands of years, often engaging with earlier rock markings. The art reflects and expresses changing experiences within landscapes over time, spirituality, history, law and lore, as well as relationships between individuals and groups of people, plants, animals, land and Ancestral Beings that are said to have created the world, including some rock art. Since the late 1700s, people arriving in Australia have been fascinated with the rock art they encountered, with detailed studies commencing in the late 1800s. Through the 1900s an impressive body of research on Australian rock art was undertaken, with dedicated academic study using archaeological methods employed since the late 1940s. Since then, Australian rock art has been researched from various perspectives, including that of Traditional Owners, custodians and other community members. Through the 1900s, there was also growing interest in Australian rock art from researchers across the globe, leading many to visit or migrate to Australia to undertake rock art research. In this volume, the varied histories of Australian rock art research from different parts of the country are explored not only in terms of key researchers, developments and changes over time, but also the crucial role of First Nations people themselves in investigations of this key component of their living heritage. ** DOWNLOAD FOR FREE: http://doi.org/10.22459/TA55.2022 **
... Rock art, in the form of petroglyphs and pictograms, is found worldwide and has an undoubtedly immense value as it is considered one of the first forms of expression of ancient societies and the prehistoric precursor to art [1][2][3][4]. As part of the natural landscape, petroglyphs are constantly exposed to anthropogenic and natural weathering processes [5][6][7], but despite this, knowledge regarding preservation and conservation of this valuable cultural heritage is limited. ...
Article
Full-text available
Petroglyph sites exist all over the world. They are one of the earliest forms of mankind’s expression and a precursor to art. Despite their outstanding value, comprehensive research on conservation and preservation of rock art is minimal, especially as related to biodeterioration. For this reason, the main objective of this study was to explore the factors involved in the degradation of petroglyph sites in the Negev desert of Israel, with a focus on biodegradation processes. Through the use of culture-independent microbiological methods (metagenomics), we characterized the microbiomes of the samples, finding they were dominated by bacterial communities, in particular taxa of Actinobacteria and Cyanobacteria, with resistance to radiation and desiccation. By means of XRF and Raman spectroscopies, we defined the composition of the stone (calcite and quartz) and the dark crust (clay minerals with Mn and Fe oxides), unveiling the presence of carotenoids, indicative of biological colonization. Optical microscopy and SEM–EDX analyses on thin sections highlighted patterns of weathering, possibly connected to the presence of biodeteriorative microorganisms that leach the calcareous matrix from the bedrock and mobilize metal cations from the black varnish for metabolic processes, slowly weathering it.
Chapter
Full-text available
There is no doubt that the past decades have brought exciting and novel understandings about geographic distributions, chronologies and analytical methods to the studies of rock art. Even from the lurch into the twenty-first century, this has been a fast forward: increasing confirmations of early image-making in Australia and other places; successful application of a new dating method to reveal previously unimagined figurative images in very deep time in Borneo; a proliferation of rock art knowledge and research; and expanded and inter-connected communities of researchers are just a few among many examples of fast-breaking news for the field. But at the same time, some of the practices that are decried by the arena of “slow science” are still with us and have, perhaps, precisely as part of the “globalization” of rock art research, become more entrenched by those who consider the field to be more competitive than collaborative, still motivated by the pull of “origins” research and claims, and the lack of retractions when, indeed, a need for such is at hand and for the betterment of the field. Slow science promotes time to think, rather than haste to get out the big next “scoop”; it promotes the reminder that we are enmeshed more than ever in broader social interests, human experiences and human needs, and for a more lasting and even an ethical science, racing ahead is deeply problematic. This chapter will explore the issues implicated by the fast-moving world with its dampening of local knowledges and alienations of non-experts as is situated in rock art research and the benefits/mandates of what slow science can bring to the field. In fact, I will suggest that rock art research is an ideal field for advancing the benefits and the power of slow science.
Chapter
Microorganisms found in various climatic conditions are able to effect the surface of rock art monuments. In this work we consider the influence of bacteria on biomineralization on the unique rock art monument “Tomskaya Pisanitsa” (Western Siberia, Russia). The biomineralization products obtained in vitro and in vivo were investigated by X-ray powder diffraction, binocular and scanning electron microscopy as well as energy-dispersive spectroscopy. Cultures of cyanobacteria and organotrophic bacteria were identified on the basic of the nucleotide sequences of the 16S rRNA gene. The manifestation of bacterial carbonate biomineralization on rock art monument “Tomskaya Pisanitsa” was found. Using experimental modeling it was shown that the formation of biogenic calcite (in some cases together with calcium oxalates) can occur by action cyanobacteria and organotrophic bacteria living on the bedrock. The results of the study can be used for conservation of rock monuments.
Article
Full-text available
Significance Rock varnish is a prominent feature of desert landscapes and the canvas for many prehistoric petroglyphs. How it forms—and, in particular, the basis for its extremely high manganese content—has been an enduring mystery. The work presented here establishes a biological mechanism for this manganese enrichment, underpinned by an apparent antioxidant strategy that enables microbes to survive in the harsh environments where varnish forms. The understanding that varnish is the residue of life using manganese to thrive in the desert illustrates that, even in extremely stark environments, the imprint of life is omnipresent on the landscape.
Article
Full-text available
We investigated rock varnish, a thin, manganese- and iron-rich, dark surface crust, on basaltic lava flows and petroglyphs in the Owens and Rose Valleys (California) by portable X-ray fluorescence (pXRF) and femtosecond laser-ablation inductively-coupled-plasma mass spectrometry (fs-LA-ICPMS). The major element composition of the varnish was consistent with a mixture of Mn-Fe oxyhydroxides and clay minerals. As expected, it contained elevated concentrations of elements that are typically enriched in rock varnish, e.g., Mn, Pb, Ba, Ce, and Co, but also showed unusually high enrichments in U, Cu, and Th. The rare earth and yttrium (REY) enrichment pattern revealed a very strong positive cerium (Ce) anomaly and distinct negative europium (Eu) and Y anomalies. The light rare earth elements (REE) were much more strongly enriched than the heavy REY. These enrichment patterns are consistent with a formation mechanism by leaching of Mn and trace elements from aeolian dust, reprecipitation of Mn and Fe as oxyhydroxides, and scavenging of trace elements by these oxyhydroxides. We inferred accumulation rates of Mn and Fe in the varnish from their areal densities measured by pXRF and the known ages of some of the lava flow surfaces. The areal densities of Mn and Fe, as well as their accumulation rates, were comparable to our previous results from the desert of Saudi Arabia. There was a moderate dependence of the Mn areal density on the inclination of the rock surfaces, but no relationship to its cardinal orientation. We attempted to use the degree of varnish regrowth on the rock art surfaces as an estimate of their age. While an absolute dating of the petroglyphs was not possible because of the lack of suitable calibration surfaces and a considerable amount of variability, the measured degree of varnish regrowth on the various petroglyphs was consistent with chronologies based on archeological and other archaeometric techniques. In particular, our results suggest that rock art creation in the study area continued over an extended period of time, possibly starting around the Pleistocene/Holocene transition and extending into the last few centuries.
Article
Full-text available
The unicellular Chroococcidiopsis is extremely resistant to desiccation, UV-irradiation, salt toxicity and high temperature. It requires only low light intensities for growth, and is protected against destruction from high energy UV-light by pigments such as carotenoids and scytonemin. Cyanobacteria, such as Chroococcidiopsis, have minimal nutrient requirements and can meet their nitrogen demand by means of the conversion of the atmospheric dinitrogen molecule to ammonium ions (nitrogen fixation). Water is the electron donor for photosynthesis, but can possibly be replaced by molecular hydrogen gas. A member of the genus Chroococcidiopsis is likely the best organism for exploring the potential for life on Mars. For this goal to be achieved, however, many strains exist on Earth that still must be explored to find the best suited one.
Article
Full-text available
We investigated rock varnish formed on sandstone and petroglyphs in the Hima area, southwestern Saudi Arabia. To characterize the rock varnish, we made in-situ measurements by portable x-ray fluorescence (pXRF) and analyzed samples by femtosecond laser-ablation inductively coupled–plasma mass spectrometry (fs LA-ICP-MS). Detailed chemical analysis of the rock varnish samples and adjacent soil or aeolian dust yielded information about the varnish’s geochemical context and formation mechanism. Untypically low positive Ce anomalies in the rock varnish samples correlated with negative Ce anomalies in the dust, supporting the hypothesis that the dust is the source of the varnish material. To study the varnish development, we made use of the fact that engraving the petroglyphs exposes a fresh bare sandstone surface without varnish, on which varnish regrows subsequently. We determined by pXRF the areal density of manganese (Mn) and iron (Fe) that had been deposited as rock varnish since the creation of the rock art. The rates of Mn deposition in the newly formed varnish were then estimated by correlating the areal density of Mn in Ancient Arabian and Old Arabic inscriptions with their known age ranges. The observed deposition rates showed substantial variability resulting from differences in exposure conditions of the rock surface, but were in a range comparable with that of our previous measurements in northwestern Arabia. This variability could be reduced significantly by referencing the measurements to the intact varnish adjacent to the individual petroglyphs. This normalization provided a much clearer relationship between varnish deposition and age, and enabled tentative ages to be assigned to rock art motifs without previously known ages. These tentative ages spanned most of the Holocene period and were consistent with the culturally or ecologically derived ages of the animal and human figures depicted in the rock art and the styles of scripts used in different periods.
Article
Full-text available
Murujuga ('Burrup Peninsula') in northwest Western Australia contains the largest concentration of petroglyphs in the world. A substantial petrochemical industrial complex has been placed in close proximity to this cultural treasure. A review of the publically available research reports used by governments to justify establishment of industry and the levels of acceptable emissions reveals the research is not scientifically credible in design, method-ology, analysis or interpretation. Each report has serious errors, which renders most results to be of little value. There remains a knowledge deficit on the condition of petroglyphs. No credible decisions about the effects of industry on the rock art can be made using the reports.
Article
This paper reviews the numbers of tropical cyclones in the southern Indian Ocean, and the land-falling cyclones that have impacted the Pilbara and Gascoyne coastlines of Western Australia since 1970, with particular emphasis on those of Category 4 strength and above. It shows reductions in impacts of severe cyclones on that coastline in the most recent two decades. Nearly as many severe cyclones have impacted the coast outside the current northern boundary of Region D in AS/NZS 1170.2 as have occurred within that boundary, suggesting the limits of Region D may need reviewing. Recorded and corrected extreme wind gusts exceeding 22 m/s from eight coastal stations have been processed as a group, and individually for 3 stations with long records. This indicates that the extreme value distribution in the draft standard DR AS/NZS 1170.2:2020 is quite adequate, without any additional factors, but the predicted extreme wind speeds for Carnarvon are well under the Region D specifications. The calculated wind direction multipliers for Region D show higher values from east and northeast; this can be explained by the wind directions generated by the clockwise rotations created by the cyclonic vortices, as the storms cross the coastline, or as they pass along the coastline at near full strength.
Chapter
This chapter focuses on the ubiquitous biogeochemical coating known as rock varnish (sometimes called desert varnish in an arid climate) that forms naturally as a result of interaction between budding bacteria and clay minerals. This chapter interfaces with this book through exploring natural processes that generate this coating and the sensitivity of rock varnish to both natural climatic changes and anthropogenic forcings. High‐resolution electron microscopy reveals that budding bacteria concentrate both manganese (Mn) and iron (Fe) on cell surfaces. Postdeposition processes mobilize nanoscale Mn and Fe that cements clay minerals to the underlying rock or preexisting varnish. Although other hypotheses exist to explain varnish formation, this model is the only proposed formation process that has a rate‐limiting step and does not fall pray to the “varnish rate paradox.” Despite rates of varnish accretion of microns per millennia in warm deserts and microns per century in more mesic settings, anthropogenic processes have altered its biogeochemistry in a variety of ways. Globally, lead fallout has contaminated the surface‐most micrometer. Regionally, acid fog and anthropogenic dust generation alters varnish textures. Locally, ash from wildfire combines with graffiti such as chalk to coat varnishes. Humans have also applied “artificial varnish” to minimize the aesthetic impact of road construction in wealthier communities.
Article
The petroglyphs of the Murujuga (Burrup Peninsula and Dampier Archipelago) in Western Australia are of exceptional cultural value for the nation. It is Australia’s largest and most significant collection of aboriginal rock art essentially as petroglyphs, with the number estimated at more than one million engravings. The Murujuga is a textbook example of ancient and modern times colliding as it also hosts potentially polluting, major industrial complexes such as iron ore and salt ports; liquefied natural gas, liquid ammonia and ammonium nitrate plants; railway lines, pipelines and rock quarries. The work presented in this paper is underpinned by an annual monitoring study of 10 selected sites including two control sites located on Dolphin and Gidley islands and eight test sites located closer to the industrial areas. The main rock types of the Murujuga, gabbro and granophyre, have been affected by weathering consisting of a cm-thick leached zone capped by a skin of orange and red iron and manganese oxides. The petroglyphs were engraved by removing the few top millimetres of the iron oxide-rich layer and the contrast with the paler leached zone clearly showing the carved motifs. Phosphorus is abnormally enriched in the leached zone and the surface coating, with manganese exclusively on the surface coating, demonstrating the impact of bird droppings and ‘desert varnish’ on the rocks. The colour difference and the hematite–goethite ratio, measured with a field portable reflectance spectrometer, between the background and engraving provides a tool to estimate the relative age of the petroglyphs and the selection of the backgrounds. As the fine-grained granophyres were preferentially chosen from the coarser-grained gabbros to host the petroglyphs, the goethitic (or yellower) backgrounds were prepared as canvas before being engraved. Many petroglyphs register the food items of the area, which changed as the last ice age ended and sea levels rose. We hypothesise that based on their locations and colour difference the constraint for the oldest age of the petroglyphs exceeds 17 000 years BP.