 |
| Titel |
Hydrological response to deforestation in a large tropical basin in Queensland, Australia |
| VerfasserIn |
Jorge Luis Peña Arancibia, Albert I. J. M. Van Dijk, Juan-Pablo Guerschman, Mark Mulligan, Leendert A. Bruijnzeel |
| Konferenz |
EGU General Assembly 2011
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 13 (2011) |
| Datensatznummer |
250055277
|
|
|
|
|
|
| Zusammenfassung |
| Land use and land cover change (LUCC), particularly deforestation; have been shown to have
influence on hydrology at small scales. Their effects at large scales are more difficult to
discern due to spatial and temporal climatic variability, different patterns of land cover
disturbance, vegetation regrowth and land management. We re-examined the hydrological
response to partial deforestation in the Comet Basin, a large (16,000 km2) sub-basin of the
Fitzroy River in the dry tropics of Queensland, Australia. More than 45% of the basin’s native
vegetation (Acacia harpophylla and eucalypt woodlands) was rapidly cleared for cropping
and cattle ranching in the mid-60s. A previous modelling study concluded that this clearing,
and subsequent land use, may have resulted in a posterior 78% increase in annual
streamflow.
Six approaches were explored to separate climate influences on streamflow from LUCC
effects: (1) Time-series analysis of rainfall and streamflow; (2) a simple conceptual
ecohydrological model based on the long-term coupled water and energy budget at the
inter-annual scale; (3) a top-down modelling approach based on the Budyko framework; (4)
signal analysis of daily streamflow descriptors based on the flow duration curve for
periods before and after LUCC and ; (5) comparison of event runoff for the respective
periods.
Time-series trend analysis showed that interannual changes in streamflow were mostly due to
changes in rainfall. This was reinforced by the results obtained with the conceptual
ecohydrological model. Nevertheless, the period 1967–1970, i.e. immediately after extensive
clearance of native vegetation, showed an upward shift in streamflow reflecting a decrease in
evapotranspiration associated with LUCC.
In order to apply the Budyko model, two periods with similar mean annual rainfall and
rainfall frequency distribution were considered for calibration and prediction, 1919–1949 and
1980–2007 respectively. In contrast to results from small experimental catchments – which in
many cases show an increase in streamflow proportional to the extent of deforestation – the
present modelling results showed a non-statistically significant increment in mean annual
runoff of only 0.5% for the period after LUCC. However, signal analysis of daily
streamflow descriptors revealed a shift after LUCC, with an increased response of the
basin to rainfall, enhanced high flows and decreased flows associated to delayed
runoff pathways. In addition, event runoff response also increased for large rainfall
events.
Based on signal analysis results, it is plausible that the long-term, progressive LUCC in this
semi-arid landscape has changed the water balance and increase runoff response. However, at
this scale the LUCC appear to have a smaller impact than would be expected on the basis of
small-scale experimental catchment results.
Keywords: Deforestation, tropical hydrology, hydrological processes, hydrological models,
streamflow, land use change |
|
|
|
|
|
|