|
How can we make decisions on land
use that do not destroy our rivers? Part of the answer is to
understand how changes in land use affect water flow. A complete
and sudden change in vegetation, such as from forest to pasture,
throughout a small catchment can lead to reduction in river flows
of around 55%, for flows between the mean annual low flow and the
mean annual flood. However, complete and sudden changes in
vegetation are uncommon. Generally, catchment areas are subject to
gradual, patchy change due to social, climatic or technological
influences or changes in how land is used for economic purposes.
Understanding the impact of gradual change becomes more complex
when the catchments in question have important variations in
rainfall and water holding capacity.
Vegetation map
for Tarawera catchment c.1970, during the forest planting period.
Dark green is forest, light green is scrub and pasture, and blue
is lake. The black line shows the major river channels of the
Tarawera River.
To better understand the impact
of slow and patchy change, NIWA has developed the Topnet catchment
model. Designed to predict the effects on river flow of slow
changes in vegetation, Topnet makes use of GIS (Geographical
Information System) databases and is designed to model a catchment
as a collection of sub-basins linked by a branched river network.
Other database variables such as vegetation, rainfall, and soil
type are used to characterize each sub-basin.
Topnet has been tested by NIWA in
the 900 km2 Tarawera catchment. There has been
significant forest planting in the area since the 1960s and 1970s,
with almost 30% of the catchment changing from scrub to coniferous
forests over a 30-year period. Over the same period a significant
decrease in river flows was also noted.
The Tarawera catchment was
selected by NIWA to test Topnet as at least three major studies
undertaken over the past 20 years have examined the impact of
changing vegetation on river flow in the catchment. By applying
Topnet to the Tarawera catchment, NIWA hydrologists were able to
compare predictions made by the model with the findings of the
three studies.
Annual catchment
runoff from the Tarawera catchment, 1995–1994, from measurements
(blue line) and Topnet model (green line).
“These are the sorts of slow
change related issues faced by resource managers,” says Ross
Woods, of NIWA in Christchurch. “A regional authority or council
may be managing a catchment as a source of water for in-stream
values and out-of-stream uses. Resource managers need to know what
will happen if some of the catchment is converted to pasture, or
housing, or perhaps to forest, and what impact a decision made
today will have in 10, 20 or even 30 years’ time. Topnet is
designed to do precisely this.”
Another critical factor in the
change process is climate. At the same time as the Tarawera
catchment was being converted to harvestable forest, there was a
shift in the local climate resulting in lower rainfall. According
to Ross, Topnet is designed to take events such as these into
consideration. “We apply a complex range of integrating factors to
the model. Rainfall, topography, vegetation and land use are
included and the model mimics reality.”
So did Topnet produce a model of
the Tarawera catchment that mimicked the current status? Ross says
it did and he believes this area of the North Island is an
excellent testing ground as it provides all the complexities of
change required to test the model to the maximum.
“The Tarawera catchment had it
all,” he says. “Major change with scrub being replaced by forest.
Long-term change as initial plantings took 30 years to complete. A
change in rainfall. Plenty of short-term changes as parts of the
forest were harvested then replanted. We also had the usual patchy
change associated with farming and other types of land use.”
The Tarawera catchment also had
that other vital ingredient, time. This is perhaps the greatest
benefit it can offer resource managers, a scientifically based
understanding of how today’s decisions will affect water resources
decades ahead. |
