Potential influence of the Pacific Ocean on the Indian summer monsoon and Harappan decline

https://doi.org/10.1016/j.quaint.2009.11.012Get rights and content

Abstract

Harappan agriculture and urban water usage were likely sensitive to variations in Indian Summer Monsoon (ISM) rainfall. The strength of the ISM displays negative correlations with sea surface temperatures (SST's) in the eastern and central equatorial Pacific Ocean and with El Niño events and is positively correlated with temperatures in the western equatorial Pacific. The development, decline and disappearance of Harappan civilization between ∼9000 and 3000 cal BP occurred against an environmental backdrop of decreasing northern hemisphere insolation, decreasing western equatorial Pacific SST's and increasing frequency and amplitude of ENSO events. Macro-regional paleohydrological records sensitive to the strength of the Indian Monsoon generally show increasing aridity and variability over this period consistent with the changes in insolation and the Pacific Ocean. Evidence also exists for a pronounced increase in aridity in the late Holocene and a particularly steep drought at ∼4200 cal BP. However, the initiation of transition to the deurbanized Late Harappan Phase may have commenced some 300 years prior to that event and the persistence of the Late Harappan Phase until 3000 cal BP suggests a more complex story than societal collapse and disappearance induced by one discrete environmental event. The continued long-term trend towards greater aridity and interannual variability during the Late Harappan Phase would have continued to present water resource challenges beyond the 4200 cal BP event and could have been a contributing factor to Harappan decline and ultimate disappearance.

Introduction

Over the eight decades following the announcement its discovery by Sir John Marshall, the Bronze Age Harappan Civilization of the greater Indus Valley has remained one of the outstanding enigmas in archaeology. What is known from the numerous excavations since the time of Marshall is that during the Mature Harappan Phase the civilization extended along the axis of the Indus Valley from foothills at the base of the Himalaya to the Arabian Sea (Fig. 1), was agrarian – using wheat, barley, cattle and other domesticates, included a number of large cities and numerous smaller villages, displayed social stratification, had craft-based industry and arguably possessed a written language based upon a logo-syllabic script (Bryant, 2000, Mahadevan, 2002, Possehl, 2002, Farmer et al., 2004, Ratnagar, 2006, McIntosh, 2007). The Harappan writing unfortunately remains undeciphered and the language of the Bronze Age Indus Valley remains one of the great mysteries. From the presence of Harappan artifacts in Mesopotamia and Oman, it is clear that Harappan trade linkages not only extended up and down the Indus Valley, but reached much farther, touching the Bronze Age Akkadian empire of Mesopotamia (Possehl, 2002, Ray, 2003, Ratnagar, 2006, McIntosh, 2007). To the Akkadians, the Harrapans were likely known as the Maluhha, and if this is correct then the trade-goods regularly arriving in Mesopotamia from the Indus Valley included carnelian, pearls, lapis-lazuli, wood, plants and other items (Ray, 2003). One of the most impressive remains of the Harappan sea-faring and trading infrastructure is the ruins of a port at Lothal which lies in Gujarat near the southeastern edge of the Harappan civilization (Fig. 1, Fig. 2) (Rao, 2000, Khadkikar et al., 2004). The precise angular layout of much of Lothal, its sophisticated water systems and the scale of what is interpreted as possibly its docking basin remains impressive today – almost 4000 years after the decline and eventual abandonment of the port.

The causes of the abandonment of Lothal, along with the decline and abandonment of all the great Harappan cities such as those at Harappa and Mohenjo-Daro in Pakistan and Dholovira in India in the late 3rd and early 2nd millennium BC remains perhaps the greatest mystery pertaining to the Harappa. Not only did the urban civilization disappear, but so did the writing and most of the unique manifestations of the material culture (Lahiri, 2000, Possehl, 2002, Ratnagar, 2000, Ratnagar, 2006, Madella and Fuller, 2006, McIntosh, 2007). Indeed, so thorough was the disappearance of the Harappa that the presence of an urbanized Bronze Age society in the Indus Valley was unsuspected until the time of Marshall. There are differing subdivisions applied to the Indus archaeological record, but following the chronology of Possehl (2002) early phases in the development of agriculture and village life in the region are recognized between ∼9000 and 5200 cal BP (calendar years before AD 1950); followed by the Early Harappan Phase between ∼5200 and 4500 cal BP; a Mature Harappan Phase between 4500 and 3900 cal BP represented by the most abundant evidence of large urban complexes (Fig. 2), standardized seals, standardized trade weights, some standardized aspects of city planning, and trade linkages with Mesopotamia and Arabia; and a Late Harappan Phase between ∼3900 and 3000 cal BP, marked by de-urbanization and eventual disappearance of distinctive Harappan artifacts. This final stage was transitional and appears to have included increased movement to village life in some regions and occasional small-scale reoccupation of some former Harappan city sites. There is also a geographic pattern of abandonment of the Indus Valley with declining occupation in the west in favor of the northern hill region, northwestern India including the westernmost Yamuna–Ganges rivers region, and Gujarat in the southwest (Fig. 1). Lothal (Fig. 2) lies in this final southeastern redoubt.

Over the past 80 years, many causes have been proposed for the decline and disappearance of the Harappan civilization. These range from Aryan invasion, to hydrological calamities related to floods, changing river courses and sea-levels, social instabilities and trade decline (Lahiri, 2000, Ratnagar, 2000, Ratnagar, 2006, Possehl, 2002, McIntosh, 2007). From the perspectives of paleolimnology and paleoclimatology, perhaps the most intriguing debates revolve around the hypothesis that the disappearance of urbanized Harappan civilization was the result of prolonged and severe drought. An exposition of the drought theory based upon paleoecological data from lake sediments arose from the work of Singh et al., 1974, Singh et al., 1990 at Lake Didwana in the Thar Desert of western India (Fig. 1). In their pioneering work, Singh and his colleagues posited on the basis of palynological evidence that the florescence of the Mature Harappan civilization occurred under the favorable influence of increased precipitation and water availability during the 3rd millennium BC and the decline was brought about by subsequent increases in aridity. In a recent influential paper, Weiss and Bradley (2001) speculated that the Harappan decline may have been linked to a larger-scale climatic event at 4200 cal BP that may have produced cooling, drought and societal collapse throughout the Bronze Age world including the Akkadian empire, Old Kingdom of Egypt, the Early Bronze Age civilizations of Greece and Crete and the Harappans. The collapse of the Yangtze Civilization in China at about this time has also been attributed to the 4200 cal BP climatic event (Yasuda et al., 2004, Yasuda, 2008). However, analysis of the lacustrine sedimentological records and paleolimnoligical history of Lake Lunkaransar by Enzel et al. (1999 – Fig. 1) concluded that drying there commenced some 1000 years prior to the Harappan decline, and furthermore the peak of the Mature Harappan stage actually corresponded to an arid period typified by phenomena such as sand dune destabilization. To quote Enzel et al., “The major Harrapan-Indus civilization began and flourished in this region 1000 years after desiccation of the lake during arid climate and was not synchronous with the lacustral phase.” (Enzel et al., 1999 p 125). There have been numerous studies using fossil pollen, charcoal, wood, paleolimnological, pedological data and geomorphology to examine Harappan-environment relations. As recently reviewed by Schuldenrein, 2002, Madella and Fuller, 2006, and Wright et al. (2008), the terrestrial data often provide unclear or conflicting evidence when the timing of climatic changes are compared to the history of the Harappan civilization. Alternative approaches for providing evidence of linkages between climate and Harappan history include the use of marine sediment records from the Arabian Sea and climate model simulations. Using stable isotope records from foraminifera taken from a core near the Indus Delta Staubwasser et al. (2003) produced a paleodischarge record for the Indus River and suggested that the Harappan decline was driven by a sharp drought at 4200 cal BP followed by the establishment of centennial-scale (200–800 year) drought cycles. Wright et al. (2008) used the ‘Macrophysical Climate Model’ to reconstruct the Holocene flow of the Beas River, which is a tributary of the Indus and has a concentration of Harappan sites. They concluded that flow in the river increased around 5800 cal BP and fell abruptly at 4100 BP, thus “correlates nicely with the brief flourishing of Harappa” (Wright et al., 2008, p. 37). However, with the exception of drying at around 4200–4100 cal BP, the general pattern of the Beas paleohydrology as reconstructed by Wright et al. (2008) does not match well the overall flow of the Indus as reconstructed by Staubwasser et al. (2003). Although this may be due to differences in source areas and climatologies for the Beas and other Indus tributaries, it must be remembered that Harappan decline took place across the entire Indus Valley region and likely reflects causes that had a wide, rather than local geographic scope. In their fulsome and thoughtful review of the evidence, Madella and Fuller (2006 p 1283) conclude that the current body of evidence supports the view that “No climatic event can be blamed for a precipitous end of this civilisation, although strategic local shifts in agriculture that may have begun in response to prolonged droughts at ca 2200BC may have contributed to the de-urbanisation process and the restructuring of human communities over the following 200–300 yr.”.

It may be argued that one element missing in most considerations of the climatic history of the greater Indus region and the Harappan civilization is the potential role that the Pacific Ocean played in climatic change and climatic variability. The recent discussions of climate and the Harappan civilization are generally silent about Pacific and how changes there may have influenced the strength or variability of the Indian Summer Monsoon (ISM) and resulting hydroclimatology of western India and Pakistan. In recent years there has been a growth in knowledge regarding the relationship of the Pacific Ocean to the strength of the ISM and the impacts this linkage on modern agriculture in India. There is also increased knowledge of the Holocene history of sea surface temperatures (SST's) and El Niño Southern Oscillation (ENSO) variability in the Pacific. This preliminary consideration briefly outlines the relationship of the ISM to precipitation and summer and winter crops in the Harappan region, and also comments the role that Pacific SST's appear to play in the strength of the ISM today as deduced from the instrumental climate record. It will then consider Holocene records of hydrology in the Harappan region and beyond, records of Pacific Ocean conditions and how these might link to Holocene changes in aridity and Harappan prehistory.

Section snippets

The Harappan settlement region and the importance of the Indian summer monsoon

Today the greater Indus Valley region and core of the former Harrapan settlement area (Fig. 1), lies in a zone of marked gradients in average annual precipitation and the importance of ISM precipitation (taken here as precipitation during June, July, August and September). The strike of the axes of these gradients runs roughly parallel to the Indus Valley itself. The average daily rates of precipitation (1979–2005) for the entire year and those for the ISM period highlight the geographic and

The relationship of the Indian summer monsoon to the Pacific Ocean today

Conditions consisting of warmer than normal SST's in the eastern equatorial Pacific, relative cooling of the western equatorial Pacific and the resulting positive states (El Niño) of the ENSO index, have been shown to be one of the most important external forces acting upon ISM rainfall variability (Ihara et al., 2007). El Niño events typically result in increased subsidence of air over India and decreased summer monsoon precipitation (Kumar et al., 2006). Specifically, the thermocline depth in

Paleohydrology during Harappan times

The work of Singh and colleagues on climate and Harappan collapse (Singh et al., 1974, Singh et al., 1990) was based upon analysis of lake deposits from the Thar Desert of Rajasthan and some of the most striking paleorecords of changes in Indian hydrology over the Holocene comes from the Thar and adjacent areas (eg. Singh et al., 1974, Singh et al., 1990, Bryson and Swain, 1981, Swain et al., 1983, Wasson et al., 1984, Prasad et al., 1997, Enzel et al., 1999). However, the archaeological record

Pacific Ocean variability and regional records of paleohydrology

Paleoclimatic and paleohydrological records from several different sources can be used to provide a broader view of aridity changes in western India and Pakistan and provide a basis for comparison with paleoceanographic records from the Pacific Ocean. Seasonal changes in insolation due to variations in the Earth's orbit likely affected the strength of the ISM directly and also influenced Pacific Ocean SSTs. Taking the broad approach of considering orbital forcing, Pacific Ocean conditions and

Pacific Ocean variability and Harappan history

The evidence reviewed above suggests that changes in Pacific Ocean SST's could have played two roles in influencing the ISM related hydrology of western India and Pakistan during the Harappan period. First, throughout most of the Holocene there has been a gradual decline in summer insolation in the northern hemisphere and a general diminishment in western tropical SST's. The declining summer insolation and decreasing SST's in the western equatorial Pacific would have contributed to a

Acknowledgements

The research that resulted in this paper was supported by a 2008 John Simon Guggenheim Fellowship and a 2009 Christensen Visiting Fellowship at Saint Catherine's College, Oxford. I owe a particular debt of gratitude to Dr. Bahadur Kotlia who organized the 3rd LIMPACS Conference and Chandigarh, India that led to production of this manuscript. I also thank the many Indian archaeologists, geographers and geologists who shared their thoughts on Indian climate and the Harappan civilization with me.

References (47)

  • Y. Yasuda et al.

    Environmental archaeology at the Chengtoushan site, Hunan province, China and implications for environmental change and the rise and fall of the Yangtze river civilization

    Quaternary International

    (2004)
  • N.J. Abram et al.

    Seasonal characteristics of the Indian Ocean dipole during the Holocene epoch

    Nature

    (2007)
  • K. Ashok et al.

    Impact on the Indian Ocean dipole on the relationship between the Indian monsoon rainfall and ENSO

    Geophysical Research Letters

    (2001)
  • E. Bryant

    The Quest for the Origins of Vedic Culture: the Indo-Aryan Migration Debate

    (2000)
  • Y. Enzel et al.

    High resolution Holocene environmental changes in the Thar Desert, northwestern India

    Science

    (1999)
  • S. Farmer et al.

    The Collapse of the Indus-script thesis: the myth of a literate Harappan Civilization

    Electronic Journal of Vedic Studies

    (2004)
  • A.S. Fischer et al.

    Two independent triggers for the Indian Ocean Dipole/Zonal Mode in a coupled GCM

    Journal of Climate

    (2005)
  • D. Fleitmann et al.

    Holocene forcing of the Indian monsoon recorded in a stalagmite from Southern Oman

    Science

    (2003)
  • A.K. Gupta et al.

    Abrupt changes in the Asian southwest monsoon during the Holocene and their links to the North Atlantic Ocean

    Nature

    (2003)
  • C. Ihara et al.

    Indian summer monsoon rainfall and its link with ENSO and the Indian Ocean climate indices

    International Journal of Climatology

    (2007)
  • C. Ihara et al.

    July droughts over Homogeneous Indian monsoon region and Indian Ocean dipole during El Niño events

    International Journal of Climatology

    (2008)
  • A.S. Khadkikar et al.

    Palaeogeography around the Harappa port of Lothal, Gujarat, western India

    Antiquity

    (2004)
  • A. Koutavas et al.

    El Niño-like pattern in ice age tropical Pacific sea surface temperature

    Science

    (2002)
  • Cited by (53)

    • Tropical Indian Ocean basin hydroclimate at the Mid- to Late-Holocene transition and the double drying hypothesis

      2023, Quaternary Science Reviews
      Citation Excerpt :

      As a result, they are likely biased towards local summer monsoonal rainfall. A shift in tropical hydroclimate beginning around 4 kyr BP is widely recognised in other Indian Ocean tropical hydroclimate records (de Boer et al., 2014; Denniston et al., 2013; Gagan et al., 2004; Giosan et al., 2018; Li et al., 2018; MacDonald, 2011; Toth et al., 2012). The inferred mechanism is an abrupt change in El Niño-Southern Oscillation (ENSO) behavior around 4.0 kyr BP with an increase in the frequency or magnitude of El Niño events (de Boer et al., 2014; Denniston et al., 2013; Gagan et al., 2004; MacDonald, 2011; Toth et al., 2012), and/or a narrowing of the tropical rainbelt (Li et al., 2018).

    View all citing articles on Scopus
    View full text