Dr Ingrid Takken - Postdoctoral Research

Postdoctoral Research

My postdoctoral research was part of the ARC-SPIRT project, ‘Forestry practices and water quality management: a new approach for an old problem’. This project had the major objective of delivering "an innovative and cost-effective framework for the protection of water resources in forestry environments".

Unsealed forest roads are commonly regarded as the most significant source of runoff and sediment in forested areas. The probability of runoff and sediments generated on these roads being delivered to the streams depends to a large extent on the degree of connectivity between the road and stream networks. My postdoctoral research focused on the development of a modeling methodology that can be used to evaluate road-to-stream connectivity within catchments and to identify road segments that have a high risk of sediment delivery to streams. The methodology is based on a distinction between three types of delivery pathways (Fig.4), i.e. stream crossings, channelised pathways and diffuse pathways.

Fig.4: Different types of pathways for delivery of road related runoff and sediments to streams.

Various empirical models are used to predict the spatial distribution of the three types of delivery pathways, and the likelihood that runoff will be delivered to the stream through these pathways. The occurrence of gullies at road drains is described in a predictive sense using two parameters: road contributing area and slope. The probability of runoff reaching streams via overland flow pathways can be predicted using the Vbt5 model of Hairsine et al. (2002). The input parameters for the models are relatively few and can be either collected in the field or calculated using a DEM in combination with roads and stream layers.

Field data were collected in four different catchments in New South Wales and Victoria (Australia) and were used to develop and validate the models. Results show that both the degree of road-to-stream connectivity and the spatial distribution of the delivery pathways in a catchment can be highly variable and depends to a large extent on catchment characteristics such as topography, hillslope gradient, road density, the position of the roads, the types of drainage structures and the length of road contributing to the drains (Fig.5).

Fig.5: The Albert River catchment in Victoria (left) and Cuttagee catchment in NSW (right) show great differences in their degree and spatial distribution of road-to-stream connectivity.

The accuracy of the currently available input data, in particular the resolution of DEM's (i.e. 20 or 25 m) limit our capability to precisely define input parameters such as road contributing areas to the existing drain network. However, our model can still be applied to create risk assessment maps that show the general pattern of areas in the catchment that are unsuitable for road construction, or where the spacing between the drains should be very small (Fig.6).

 

Fig.6: Map showing the maximum drain spacing (m) that should be allowed along a road in order to limit gully formation and runoff delivery through diffuse pathways.

Given the huge cost in obtaining more accurate data, the current methodology may represent the most cost-effective approach for state agencies and forest industry to collate available topographic and road infrastructure data and develop catchment-scale risk assessment maps that will limit the delivery of road derived runoff and sediment.

References:

Croke J. and Mockler S., 2001. Gully initiation and road-to-stream linkage in a forested catchment, southeastern Australia. Earth Surface Processes and Landforms, 26, 1-13.

Hairsine P. B., J.C. Croke, H. Mathews, P. Fogarty, and S. P. Mockler (2002). Modelling plumes of overland flow from roads and logging tracks. Hydrological Processes. 16, 2311-2327.

---