In a recent paper, JA Dearing and colleagues (J. Paleolimnology 40: 3-31) use paleolimnological techniques to explore the long-term history of the region around Erhai Lake in Yunnan Province. Lake sediment cores (which can explain catchment vegetation, flooding, soil erosion, sediment sources and metal workings) are complemented by independent regional climate time-series from speleothems, archaeological records of human habitation, and a detailed documented environmental history. The authors integrate these data to “provide a Holocene scale record of environmental change and human–environment interactions.”

They use these data to ask:

• “How sensitive are the studied environmental system processes to climate and human drivers of change?”
• “Can we observe long-term trajectories of socio-environmental interactions, or periods of social collapse and recovery?”

The authors identify a number of points at which there were major changes in the human interaction with the landscape, including ~9000 cal year BP, when sediment records show a ‘human-affected environment’, ~4800 cal year BP, when major deforestation for grazing led to the extirpation of forest species and some functional units, and ~2000 cal year BP at the introduction of paddy field irrigated farming, and ~1600 cal year BP at which point surface erosion and gullying were caused by increased exploitation of mountain slopes. They go on to suggest that these records indicate several major ‘periods’ in human-environment interactions in this area:

The earliest of these cases probably represents the dispersion of the population away from the established sedentary agricultural units on alluvial fans to the more inhospitable margins of the lake and the valleys. This perhaps signifies the end of the ‘nature dominated’ phase (Messerli et al.) where society could cause significant modification of the landscape but was still vulnerable to the main risks of drought and flood (though the evidence for climate determinism is weak). In contrast, the introduction of irrigation is associated with a trend of weakening monsoon intensity, increasing numbers of centennial scale dry phases, and population growth. It represents an agrarian society in transition, using technological innovation to raise carrying capacities without increasing greatly the vulnerability to drought or flood. The third period is linked to natural population growth, inward migration and metal extraction brought about by the rise of Nanzhao/Dali as a major center”

The authors then ask at what stage of the adaptive cycle the modern Erhai socio-ecological system exists:

At Erhai, the slow processes of weathering and soil accumulation, in association with vegetation cover held fairly constant by a benign early-mid Holocene climate, were interrupted by fast processes of anthropogenic modification of vegetation. For many centuries, this concatenation of ‘slow–long’ and ‘fast–short’ processes led to a resilient land use-soil system (cf. Gunderson and Holling). But increasing perturbations led to system failure, and we can observe that the late Ming environmental crisis represents the end of the last release phase. Thus, the modern landscape may be approaching a conservation phase (K) characterised by minimum resilience.

Dearing and colleagues explore the meanings of this research for current sustainability and conclude that the main threat to the region is high magnitude-low frequency flooding of the agricultural plain and low terraces, which is exacerbated by:

1. continued use of high altitude and steep slopes for grazing and cultivation that generate high runoff from unprotected slopes and maintain active gully systems, particularly in the northern basins;
2. reduction or poor maintenance of paddy field systems, engineered flood defences, river channels and terraces; [and]
3. increased intensities of the summer monsoon.

This fascinating paper is an excellent example of how historical data sources can be integrated to provide a new perspective on social and ecological change over long periods of time.