Abstract:
Within south-eastern Taranaki, New Zealand (Figure 2.1), Crozier and Pillans (1991) identified four separate types of natural dams (Main Valley, Blocked Tributary, Supra, and Crown), defined by their relationship with the landslide mass and the drainage network upon which they impinged. They argued that frequency of major events, including landsliding, ensured that the region was within a condition of dynamic meta-stable equilibrium. The objectives of this thesis has been to establish and quantify where possible, the factors that have determined the formation of those natural dams from landslides. In otherwords, the thrust of this research has been to provide an explanation of why landslides have effectively disrupted the drainage networks of this region.
This thesis put forward suggests that a satisfactory resolution of this problem can be provided from two concepts applicable at different scales of time and space. Firstly, it is hypothesised that the ability of landslides to dam drainage networks for significant periods of time, can be explained with resort to forces existing which tend to resist or remove the landslides mass. This is essentially a contemporary systems approach to the problem. It provides an explanation in limited scales of space and time.
The second hypothesis submits, that the stage of landscape evolution, through its control of such parameters as slope and relief, will control the location of landslide occurrence, which will in turn influence the ability of landslides to effectively block drainage networks.
The methodology used in this thesis, involved indexing the 'Damability' (The ratio of 'Removal' to Resistance Forces') retained through time by landslides that have acted and are continuing to act through as damming mechanisms of these natural dams. It was found that the natural dams studied in this thesis, had formed from landslides where had been able to retain sufficient 'Damability' through time i.e. their 'Resistance Forces' being greater than the 'Removal Forces' of episodically occurring negative feedback mechanisms from the drainage networks that they have dammed. However, the 'Removal Force' parameter by itself provided a more satisfactory explanation of where landslides dams have been able to remain effective.
Investigation of the second hypothesis, through comparing the distribution of natural dam types and types of terrain, indicated that the type of terrain available within the study area was an important factor in the formation of its natural dams. Main valley and Block Tributary type natural dams were found largely confined to parts of the study area where marine terraces were only partially dissected. The remaining natural dam types were located in the 'maturely' dissected hill country. The second hypothesis therefore provides another level of explanation at greater time and spatial scales.
