Open Access Repository

Modelling the effects of forest regeneration on streamflow using forest growth models


Downloads per month over past year

Jaskierniak, D 2011 , 'Modelling the effects of forest regeneration on streamflow using forest growth models', PhD thesis, University of Tasmania.

PDF (Whole thesis)
Jaskierniak_who...pdf | Download (20MB)
Available under University of Tasmania Standard License.

| Preview


Forest regeneration is a dynamic process that affects forest hydrology through
changes in structure and density of natural forests. In Victoria and Tasmania, forest
hydrology models that manage the potential impacts of land cover disturbance on the
water resource are not data-driven with information on vegetation dynamics that
affect forest water use. Current models underutilise forest inventory databases for
managing the forested water resources, even though available evidence suggests an
inverse relationship between forest growth rates and long-term changes in
streamflow. This dissertation is the first published study that uses forest inventory
data to produce spatiotemporal forest growth models to explain vegetation-induced
streamflow trends. The study was undertaken in nine small catchments (7.4 to 52.8
ha) located in Melbourne’s Maroondah water catchment.
The hydrology model runs on an annual time step and partitions streamflow data into
climate-induced noise using a climate filter, and vegetation-induced trend using an
ellipse and gamma (“Kuczera curve”) function. A simulation exercise demonstrates
how well the model structure isolates the vegetation-induced trend from climatic
variability in streamflow using a range of synthesised scenario cases. The model
framework allows for comparison of streamflow trends against a detailed forest
growth model by using the same gamma function to quantify forest growth and
vegetation-induced streamflow trends.
To spatially extrapolate forest growth, field measured stand characteristics were
empirically analysed against LiDAR indices. The indices were produced with
mixture models, which used 11 distribution functions to summarise complex canopy
attributes with bimodal distributions. The LiDAR indices were used to predict
overstorey stand volumes and basal area, and understorey basal area of 18-, 37-, and
70-year old Mountain Ash forest with variable density classes and treatment effects.
Observed versus predicted values of eucalyptus basal area and stand volume were
highly correlated, with bootstrap r2 ranging from 0.61 to 0.89 and 0.67 to 0.88
respectively. Non-eucalyptus basal area r2 ranged from 0.5 to 0.91. To temporally extrapolate stand volumes and basal area, LiDAR indices and
permanent plot data were used in mixed effects models to capture the spatial
heterogeneity in, and temporally polymorphic nature of forest growth. The
spatiotemporal models of forest growth were then lumped to the catchment-scale to
represent changes in growth rates over the stream gauging period. The relationship
between catchment-scale gamma parameters of forest growth and forest water use
were explored, and results demonstrate that forest growth provides useful
information for explaining streamflow trends published in the literature and
quantified in this study.

Item Type: Thesis - PhD
Authors/Creators:Jaskierniak, D
Keywords: forest hydrology, forest growth modelling, forest harvesting, forest productivity, LIDAR indices, mixture models, mixed effects models, plant physiology
Copyright Holders: The Author
Copyright Information:

Copyright 2011 the author

Additional Information:

Chapter 5 is the equivalent of a post-print article published as: Jaskierniak, D., Lane, P.N.J., Robinson, A. & Lucieer, A., 2011. Extracting LiDAR indices to characterise multilayered forest structure using mixture distribution functions. Remote Sensing of Environment, 115 (2), 573-585

- Reference and Appedices are available in the hard copy held by UTAS Library Bib #: 1066888

Related URLs:
Item Statistics: View statistics for this item

Actions (login required)

Item Control Page Item Control Page