9439Ma: Drought and the Maya Collapse
- Jan 24, 2003
>Jeff Baker said:Martin Peach said:
>"There is one major problem with both Gill's book and the O-18
>studies. Both Gill and the researcher's studying the oxygen
>isotopes place the drought in the 9th and 10th centuries, yet the
>population decline in the southern lowlands occurred in the 8th
>"And, in regard to the isotope data, the isotope information is
>all from the northern lowlands, and pinpoints to a "drought" in
>the 9th and 10th centuries in the north. Yet, the north was
>undergoing a resurgence at this time. The Puuc was seeing a
>population explosion, Chichen was seeing large scale building
>going on, Dzibilchaltun was undergoing a population explosion.
>"The data on population changes in the 8th, 9th and 10th centuries
>simply does not fit the drought model."
>I disagree.This is the first chance I've had to respond to the many comments
>Hodell et al (Science 292 p1367) have the major drought starting
>around AD 750, ie the 8th century, in the central Yucatan. One
>might expect that the main population effects of a severe drought
>would show up at the beginning of the period as the system was
>caught off guard with several failed harvests in succession, and
>that after no more than a few decades, the survivors would have
>learned to live under the new regime, for instance by learning to
>use the groundwater aquifers instead of rainfall to irrigate
>It also seems to me that the climate data from lake cores is
>likely more accurate than current population estimates. It may be
>for instance that as the climate dried out, scattered farmers
>moved to locations with cenotes, giving rise a population
>increase at these centres, while the corresponding depopulation
>of the hinterland would go unnoticed using current archaological
on this topic.
Subsequent work by Curtis, Hodell and company has placed a peak
in the drought at A.D. 862, with a peak in humidity (i.e. high
rainfall) 50 - 75 years before the drought, or at the end of the
8th century (Curtis et al. 1996).
This underscores another problem with the isotopic studies, the
inconsistency in the climatic data coming from different lakes.
The extreme example of this can be found in a comparison of Lakes
Coba and Punta Laguna, which are separated by a mere 20 km. Over
the last 1500 years, the isotopic data from Lake Coba appears to
show a drying trend (Whitmore et al. 1996), while the last 1000
years of oxygen isotope data from Punta Laguna seems to indicate
a period of increasing humidity (Curtis 1996).
An examination of modern rainfall and the O-18 ratios of modern
lakewater does not provide support for a relationship between
O-18 ratios and precipitation. The following presents several
Yucatecan lakes, modern rainfall levels and the O-18 ratio of the
San Jose Chulchaca: 500 - 900 mm/yr; 0.5 ppt.
Lake Sayaucil: 1000 mm/yr; 5.3 ppt.
Lake Chichancanab: 1300 mm/yr; 3.5 -5.4 ppt.
Punta Laguna: 1519 mm/yr; 0.93 ppt.
Lake Coba: >1500 mm/yr; 1.18 ppt.
Based upon the oxygen isotope ratios, one would assume that Lake
Sayaucil would have the highest rainfall of any of the lakes, yet
it has the second lowest rainfall total. Punta Laguna, with one
of the highest rainfall totals, has the second lowest isotopic
ratio, not one of the highest. It is clear that there is no
correlation between modern rainfall and the O-18 levels of modern
lakewater (Whitmore et al. 1996: 285), yet researchers assume
that there is a correlation between rainfall and O-18 levels in
I think much of the problem stems from the number of variables
that can influence the O-18 of ostracod and gastropod shells. A
total of 13 variables can influence the O-18 in the shells, yet
researchers assume that the only significant variable is the
Evaporation/Precipation ratio (E/P) (Baker 2002). Even if the
other 12 variables are relatively minor by themselves, combined
they could have a significant effect on the fossil O-18 signal.
Turning specifically to surveys in the northern Yucatan:
Current archaeological methods do study rural households. There
have been numerous surveys over the past thirty years on
hinterland areas in both the northern and southern lowlands (see
Culbert and Rice 1990). In the Puuc area, Dunning's (1992) work
did involve an examination of rural households. Similarly, Vlcek
et al.'s (1978) work at Dzibilchaltun extended outside of the
urban area into the rural areas surrounding Dzibilchaltun. The
population patterns I mentioned above are not based solely upon
site center studies, a point supported by Jerry Ek, David Hixson
and Clifford Brown. The evidence we have for the northern Yucatan
clearly indicates that the countrysides were NOT abandoned in the
9th and 10th centuries.
Regarding water tables in the Puuc:
Martin Peach wrote:
"In the context of a collapse of larger scale social organization,
lowlands would not be safe places to live. Also, the water table,
despite its name, is not flat, but follows the topography."
There are places in the Puuc where the water table is close to
the surface, but there are also many places where the water table
is much deeper. However, this is similar to the situation in the
Peten. Though everyone is aware of the failure of the Tikal
project to dig a well deep enough to tap into the water table,
most people are unaware of many handdug wells in the Peten
physiographic region that tap into shallow water tables. There is
one that is not far from the site of La Milpa at an elevation of
app. 200 meters above sea level (while La Milpa is not the
Department of the Peten, it is within the Peten physiographic
region). In a presented at the SAA' several years ago, Hugh
Robicheaux mentioned the presence of a series of wells in the
Peten that were hand dug and tapped into shallow water tables.
The water table in the Puuc does not seem to be significantly
different from that in the Peten. And, despite being termed the
lowlands, the Peten is far from a flat terrain. The Peten
physiographic region is characterized by steep hills with
interspersed flat land, similar to the Puuc. But, in the Peten,
the sites are located on the hills, while the flat land is
unoccupied. From my understanding of Dunning's work in the Puuc,
the exact opposite situation is present there.
A couple final points relate to Clifford Brown's e-mail:
"There is no doubt that Jerry and David are right. The timing of
the drought documented in the lake cores is now admitted by all
to not correspond terribly well with the Collapse (if that's not
a misnomer, and many think it is). It is also true that the Puuc
region has the least surface water and the fewest water sources
of any part of the Maya lowlands. Nevertheless, it was very
densely populated during the Late and Terminal Classic periods.
We need to be looking at other climate proxies, like stalactites,
tree rings, and otoliths, to understand the climate patterns
In research I've conducted at the site of Sierra de Agua in
northwestern Belize, we cored a small, shallow aguada. This
aguada was cleaned out during the Late Classic Period, but has a
continuous pollen record for the last 1200+ years. During the
drought years of 1995 and 1997, there was no surface water in the
aguada during the dry season, but the sediments were moist.
A prolonged, severe drought, such as that postulated for the 9th
and 10th centuries would, I think, have dried these sediments out
during the dry season.
During the rainy season, the aguada would have filled with water.
This alternating saturation desiccation would have destroyed
pollen and organic matter within the aguada. There is no evidence
for such an event occurring at Sierra de Aguada. I should note at
this point, that the water table in the vicinity of the aguada is
Dr. Brown also indicated that the timing of the drought is "now
admitted by all to not correspond terribly well with the Collapse."
I wish this was true. At the recent AAA meetings (in mid-November),
during one symposium, one researcher have another was mentioning
the 'overwhelming' and/or 'conclusive' evidence for a drought at
the end of the Late Classic.
Baker, Jeffrey L., 2002, Maya Wetlands: Ecology and Pre-hispanic
Utilization of Wetlands in Northwestern Belize. Ph.D Dissertation,
University of Arizona.
Culbert, T. Patrick and Don S. Rice (eds), 1990, Precolumbian
Population History in the Maya Lowlands. University of New Mexico
Curtis, Jason H., David A. Hodell and Mark Brenner, 1996, Climate
Variability on the Yucatan Peninsula (Mexico) during the Past 3500
Years, and Implications for Maya Cultural Evolution. Quaternary
Research 46: 37-47.
Dunning, Nicholas, 1992 Lords of the Hills: Ancient Maya
Settlement in the Puuc Region, Yucatan, Mexico. Prehistory Press,
Vlcek, David T., Sylvia Garza de Gonzalez and Edward B. Kurjack,
1978, Contemporary Farming and Ancient Maya Settlements: Some
Disconcerting Evidence. In Pre-Hispanic Maya Agriculture, edited
by Peter D. Harrison and B. L. Turner II, pp. 211 ? 223.
University of New Mexico Press, Albuquerque.
Whitmore, Thomas J., Mark Brenner, Jason H. Curtis, Bruce H.
Dahlin and Barbara W. Leyden, 1996, Holocene Climatic and Human
Influences on Lakes of the Yucatan Peninsula, Mexico: An
Interdisciplinary Paleolimnological Approach. The Holocene 6:
273 ? 287.
Copyright � AZTLAN <AZTLAN@...> 2003
All rights reserved.
- << Previous post in topic