Rapid Communication
Early rice farming and anomalous methane trends

https://doi.org/10.1016/j.quascirev.2008.03.007Get rights and content

Abstract

The anthropogenic explanation for the increase in atmospheric methane concentration during the last 5000 years requires large CH4 emissions from human activities beginning early in the Bronze Age. This paper presents a compilation of 311 archeological sites in rice-growing regions of China. The number of new sites between 6000 and 4000 years ago increased almost ten-fold compared with those during previous millennia. This early spread of rice production across most of the area in China where irrigated rice is grown today supports the hypothesis that early farming caused the anomalous methane reversal.

Introduction

Atmospheric methane concentrations decreased during the first half of the Holocene, but then reversed direction and began to increase 5000 years ago (Fig. 1). This reversal and subsequent increase have been variously attributed to natural and anthropogenic factors.

Schmidt et al. (2004) proposed that the trend resulted mainly from an increase in natural CH4 emissions from expanding circum-Arctic wetlands. Previously, however, Chappellaz et al. (1997) and Brook et al. (2000) had ruled out a boreal source based on a late-Holocene decrease in the interhemispheric CH4 gradient (recently confirmed by Brook and Mitchell, 2007). This gradient is an index of the difference between CH4 concentrations recorded in Greenland and Antarctic ice. Greater releases from boreal sources boost CH4 concentrations more in nearby Greenland ice than in distant Antarctic ice, thereby increasing the gradient, and conversely. Between ∼3750 and ∼750 years ago, the gradient decreased, and box modeling of this change indicates reduced emissions from circum-Arctic sources (Chappellaz et al., 1997). Despite the fact that wetlands were still slowly expanding in north-polar regions during the late Holocene (Smith et al., 2004), the decrease in boreal emissions appears to have been caused by the onset of cooler summers (COHMAP Members, 1988; Koerner and Fisher, 1990) and by a change toward drier bog types (MacDonald et al., 2006).

The reduced interhemispheric CH4 gradient indicates that the increased methane emissions must have come from the subtropics and tropics (Chappellaz, et al., 1997). Although low-latitude wetlands are the world's largest natural methane source, however, they were not the cause of the late-Holocene CH4 increase. During the last 9000 years, a progressive weakening of the orbitally driven summer monsoon has caused methane-emitting wetlands and lakes in the tropics and subtropics to shrink or disappear (Kutzbach, 1981; COHMAP Members, 1988). Large-amplitude δ18O trends in speleothem calcite show a progressive weakening of summer monsoons during the middle and late Holocene in both Asia (Yuan et al., 2004) and in the Indian monsoon sector (Fleitmann et al., 2003).

The only natural source remaining in the tropics is wetlands in river deltas. Schmidt et al. (2004) hypothesized that the growth of tropical river deltas played a role in the late-Holocene methane increase. Much of this delta growth, however, was not natural in origin. By the time of the Bronze Age, extensive forest clearing for agriculture was causing widespread erosion that led to increased sediment loads in rivers and to enlargement of deltas (Roberts, 1998; Stefani and Vincenz, 2005; Vella et al., 2005). Anthropogenic influences thus account for a substantial and possibly dominant fraction of any late-Holocene increase in CH4 fluxes from deltaic areas.

The rising methane trend of the last 5000 years is anomalous compared to the downward CH4 trends during previous early interglacial intervals (Ruddiman and Thomson, 2001; Ruddiman, 2003, Ruddiman, 2007). Because these earlier downward trends were unquestionably of natural origin, the upward trend in the late Holocene is anomalous by comparison.

With CH4 emissions from natural sources decreasing rather than rising during the last 5000 years, the only plausible low-latitude source of methane to explain the rising concentrations in the atmosphere is early agricultural activity (Ruddiman and Thomson, 2001; Ruddiman, 2003). Early farming activities that caused increased methane emissions include: irrigating rice paddies (which in effect creates anthropogenic wetlands), tending methane-emitting livestock, burning seasonal (grass) biomass, and generating human waste. An obvious question is whether or not human activities could have become sufficiently extensive by 5000 years ago to reverse the downward trend in atmospheric methane concentrations (Fig. 1). This paper addresses that issue by examining the extent of early rice irrigation in China based on archeological evidence.

Section snippets

The spread of rice irrigation in China

The broad outlines of the history of early rice farming and the associated ‘rice cultures’ of China are known (Yan, 1982; You, 1995; Wu, 1998; An, 1999). Evidence of rice use prior to 9000 years ago has been found in warm, summer-wet regions of south-central China (Fig. 2) (Peng, 1998; Xiang, 2005; Sheng et al., 2006). The time of first planting of rice is somewhat uncertain, but was unambiguously underway by 6500–6000 years ago (Fuller et al., 2007). Irrigation of wet-adapted rice strains in

Link to Holocene changes in atmospheric methane

Today, the ∼0.3 million km2 of irrigated land in China account for ∼20% of the global total (IRRI, 2007). Because irrigation started early in China, it could have accounted for a larger percentage of global irrigation and methane emissions during these earlier millennia. The rice-irrigation trends in Fig. 2, Fig. 3 must have figured prominently in the anomalous reversal and subsequent rise of atmospheric CH4 concentrations shown in Fig. 1.

Unfortunately, this compilation of archeological sites

Acknowledgments

This work was supported by the Paleoclimate Program of the Atmospheric Sciences Division of the US National Science Foundation and by the National Natural Science Foundation of China and the Chinese Academy of Sciences. We thank Meiyan Liang and Richard Nevle for helpful discussions and two anonymous reviewers.

References (42)

  • D. Fleitmann

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

    Science

    (2003)
  • D.Q. Fuller

    Agricultural origins and frontiers in South Asia: a working synthesis

    Journal of World Prehistory

    (2006)
  • D.Q. Fuller et al.

    Presumed domestication? Evidence for wild rice cultivation and domestication in the fifth millennium BC of the Lower Yangtze region

    Antiquity

    (2007)
  • I.C. Glover et al.

    New evidence for early rice cultivation in South, Southeast and East Asia

  • Z.T. Gong

    The temporal and spatial distribution and the significance of ancient rice in China

    Chinese Science Bulletin

    (2007)
  • Hunan Provincial Institute of Cultural Relics and Archaeology

    The 1997 and 1998 seasons of excavation of a city site at Chengtoushan, Lixian, Hunan

    Cultural Relics

    (1999)
  • IRRI, 2007. International Rice Research Institute. Available at...
  • H. Jia

    Research on terrace in China

    Agricultural Archeology

    (2003)
  • R.M. Koerner et al.

    A record of Holocene summer climate from a Canadian high-Arctic ice core

    Nature

    (1990)
  • J.E. Kutzbach

    Monsoon climate of the Early Holocene: climate experiment with Earth's orbital parameters for 9000 years ago

    Science

    (1981)
  • Li, X., Dodson, J., Zhou, J., Zhou, X.Y. Increases of population and expansion of rice agriculture in Asia, and methane...
  • Cited by (0)

    View full text