How to reconstruct past debris flows using tree rings

1. What is a debris flow and what are the associated problems?
Debris flows are a type of mass movement involving rapid transport of water and saturated material (sand, mud, blocks, organic matter, ...) down steep, confined channels (see for example (Iverson, 1997)). Debris flows can occur in many regions where steep drainages and loose material is readily available. They are normally triggered by a sudden input of large amounts of water, such as intense or prolonged rainfall, or the rupture of water pockets in glaciers or moraine-dammed lakes. As their occurrence is rarely predictable, debris flows represent one of the most dangerous of all natural hazards in mountain regions, where they endanger the security of humans and also damage their assets. In order to cope with hazards posed by debris flows, a detailed knowledge of their past occurrence is of crucial importance. However, people tend to remember only the very recent past and archival records on older events are generally rather scarce and incomplete. Therefore, a method for the reconstruction of past debris flows is needed. Dendrochronology, the analysis of tree rings provides such a toool as trees record external disturbances in their tree-ring series and represent therefore a very valuable natural archive. The analysis of past geomorphic events using tree rings is called dendrogeomorphology.

2. How can we obtain data on debris flows from trees?
Tree-ring analyses are based on the fact that trees growing in temperate regions form distinct annual growth rings which are influenced by various tree-specific (e.g. genetic) and environmental (e.g. climate) factors (Schweingruber, 1996). In addition, trees record mechanical disturbances--such as the impact of debris flows--in their tree rings. They will react to these impacts with changes in tree morphology and with internal changes in the tree-ring structure and width.

The most frequent external growth disturbances are given in Figure 1: A. Trees can be injured through boulders or woody material transported in the flowing mass. B. The pressure of the flow can tilt tree stems. C. Debris-flow material can be deposited around the stem base. D. Erosion caused by the flow can denudate tree roots. Trees will respond to these different disturbances with changes in their growth as shown in Figure 2: A. On both sides of the injury, coniferous trees form callus tissue and sometimes tangential rows of traumatic resin ducts to protect themselves against fungi and other impacts. B. After stem tilting, trees will produce so-called reaction wood--normally on the lower side of the stem. C. If nutrient and water supply is reduced as a result of stem burial or root erosion, we will observe a sudden and abrupt decrease in the yearly increment. D. Trees may also react with a growth increase to events, provided that neighboring trees are being eliminated and that the survivors benefit from improved growth conditions. Growth reactions in trees can be analyzed on cross-sections of felled trees or on cores (diameter 6 mm) extracted with an increment borer (Stoffel and Bollschweiler, 2009).

3. What debris-flow features can be reconstructed with tree rings?
The investigation of trees and tree-ring series may provide a plethora of information on earth-surface processes and natural disasters, such as for example the frequency of past events. The question here is how often an event occurred in a specific torrent in the past. By way of example, Figure 3 illustrates the debris-flow activity in a torrent of the Swiss Alps reconstructed through the analysis of 1102 old-growth coniferous trees affected by past events. In total, 124 events could be reconstructed for the period A.D. 1570-2008 which means that on average an event occurred every 3.5 years. This information is of prime importance for the assessment of hazards and the calculation of return periods of events (Bollschweiler et al., 2008; Stoffel et al., 2008).

If data on trees reacting to particular events in the past is illustrated on a map, the spatial spread and reach of debris flows can be determined. Figure 4 provides an example of a debris-flow cone where flow paths of past events were reconstructed. Eleven formerly active channels were identified on the present-day surface of the cone and five flow patterns could be identified for the 40 events reconstructed (two of them are shown in Figure 4). In addition, a clear shift of the trajectory of the main channel was observed. While the main channel passed on the western part of the cone until the 1930s, it started moving its bed to the eastern part of the cone where it stayed to the current days (Bollschweiler et al., 2007). With the same approach, outbreak locations of past events along a channel can be determined.

Tree-ring analyses allow dating of past events with yearly and sometimes even with seasonal or monthly precision. If this data is coupled with meteorological data, triggers of past events can be determined. The rainfall threshold of the reconstructed events can be assessed and this information is subsequently very important for hazard assessments and warning systems.

4. Concluding remarks
Apart from the examples given above, tree rings are also able to provide information on depositional processes on cones, on erosion in the channel and on magnitudes of past events. Their use is not only limited to debris-flow reconstructions but the method can also be used for various other geomorphic processes and natural hazards such as snow avalanches, rockfalls, landslides, floods, volcanic processes, or hurricanes (Stoffel et al., 2010). Therefore, dendrogeomorphology represents a powerful method to obtained data on past geomorphic events.

• Bollschweiler, M., Stoffel, M., Ehmisch, M. and Monbaron, M. 2007: Reconstructing spatio-temporal patterns of debris-flow activity using dendrogeomorphological methods. Geomorphology 87, 337-351.
• Bollschweiler, M., Stoffel, M. and Schneuwly, D.M. 2008: Dynamics in debris-flow activity on a forested cone--A case study using different dendroecological approaches. Catena 72, 67-78.
• Iverson, R.M. 1997: The physics of debris flows. Reviews of Geophysics 35, 245-296.
• Schweingruber, F.H. 1996: Tree rings and environment. Dendroecology. Bern, Stuttgart, Wien: Paul Haupt.
• Stoffel, M. and Bollschweiler, M. 2009: What tree rings can tell about earth-surface processes: teaching the principles of dendrogeomorphology. Geography Compass 3, 1013-1037.
• Stoffel, M., Bollschweiler, M., Butler, D.R. and Luckman, B.H. 2010: Tree rings and natural hazards - a state of the art: Springer.
• Stoffel, M., Conus, D., Grichting, M.A., Lievre, I. and Maitre, G. 2008: Unraveling the patterns of late Holocene debris-flow activity on a cone in the Swiss Alps: Chronology, environment and implications for the future. Global and Planetary Change 60, 222-234.