 |
| Titel |
Impacts of artificial inundation of ephemeral creek beds on mature riparian eucalypts in semi-arid northwest Australia |
| VerfasserIn |
Rachel Argus, Gerald Page, Pauline Grierson |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250094671
|
| Publikation (Nr.) |
 EGU/EGU2014-10098.pdf |
|
|
|
|
|
| Zusammenfassung |
| The resilience of riparian ecosystems of intermittent rivers to changes in their hydrological
regimes is not well understood. In the Pilbara region of northwest Australia, streams flow
only occasionally, reflecting a highly dynamic and extremely variable cycle of prolonged
droughts punctuated by occasional floods. However, discharge of ground water pumped from
mining activities over recent years has resulted in localised areas with constant surface
water. Here we sought to assess impacts of prolonged saturation on the health and
functioning of two co-occurring eucalypts (Eucalyptus camaldulensis and Eucalyptus
victrix). While riparian vegetation is clearly adapted to partial root-zone hypoxia, we
hypothesised that trees in inundated areas experience reduced root function due to an
energy crisis, which will be reflected by symptoms in the foliage. We expected that
complete saturation of the entire root system for an extended time period reduces
physiological function through lower stomatal conductance and more negative water
potential, results in canopy sparseness and reduces accumulation of foliar nitrogen and
phosphorus.
Trees (n=26) were assessed at two sites with artificially permanent surface water
(discharge sites) and compared to trees (n=21) at a site with a naturally occurring permanent
groundwater fed pool (‘reference site’). Trees were sampled from a range of positions
including the stream bed, the lower bank and the upper bank, in order to determine the extent
of influence of the discharge water. No eucalypts grew in the stream bed at the reference site,
indicating either the stream bed conditions were unsuitable for seedling survival or eucalypts
were outcompeted by the flood tolerant tree Melaleuca argentea (which was absent from the
impact sites). Soil redox potential, an indicator of oxygen availability and other soil
chemical conditions, was measured with platinum redox probes at 25 cm depth. Trees
were assessed for canopy cover, foliage water potential and gas exchange. Leaves
from each tree were also analysed for stable isotope composition (δ13C, δ15N and
δ18O) and nutrients (N and P). Bank trees did not differ between the discharge
and reference sites by any measure. However, trees positioned in the stream bed at
discharge sites were exposed to severely reduced soil redox potential (median =
-189 mV) compared to trees growing on the lower (90 mV) or upper bank (188
mV). Trees in the stream bed were clearly separated from upper or lower bank trees
using principle components analysis (PCA) for all measured attributes. Canopy
cover, δ13C and δ18O contributed most to separating the groups. Canopy cover in
stream bed trees was 41% and 52% sparser compared to bank trees at discharge and
reference sites, respectively. Stream bed tree leaves had more enriched δ13C values but
more depleted δ18O values, indicating leaf gas exchange with the atmosphere was
more restricted than for the trees on the bank. Overall, we conclude that artificially
constant surface water expression significantly changed environmental conditions
in the stream bed and the effect on riparian eucalypt trees was highly localised. |
|
|
|
|
|
|