 |
| Titel |
Mathematical modeling riparian vegetation zonation in semiarid conditions based on a transpiration index. |
| VerfasserIn |
Joaquin Real, Marco Morales, Alicia Garcia, Virginia Garofano, Francisco Martinez-Capel, Félix Francés |
| Konferenz |
EGU General Assembly 2010
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 12 (2010) |
| Datensatznummer |
250036392
|
|
|
|
|
|
| Zusammenfassung |
| Initially riparian vegetation modeling was focused on the study of ecological patches without
taking into account the interactive effects of structures and processes in between them
(Tabacchi et al., 1998). One of the greatest challenges, when carrying out a riparian
ecosystem restoration, is to understand the physical and ecological processes of a system and
the interaction and feedback within these processes. Jorde (2002) pointed out the importance
of addressing complex linkages between processes and biotic interactions in research and in
the development of restoration projects over larger spatial and temporal scales in the
future.
According to Tabacchi et al. (2000), the water cycle in riparian zones depends on three
important relations: the water absorption by the plants, water storage and atmospherical
return by evaporation. During recent years a variety of ecological models have taken
into account the changes in the plant species as consequence of changes in the
environmental variables and hydrological alterations (Baptist, 2005; Braatne et al., 2002;
Glenz, 2005; Hooke et al., 2005; Murphy et al., 2006). Most of these models are
based on functional relationships between river hydrology and vegetation species or
communities. In semiarid regions we make the hypothesis transpiration will be
one of the key factors determining the riparian vegetation presence and therefore,
we will not consider in our model other factors as recruitment, flood damages,
etc.
The objectives of this work are: firstly to develop a model capable of simulating several
riparian vegetation types which can be applied in a wide range of conditions across
Mediterranean environments; and secondly to calibrate and to validate the model in several
Mediterranean river stretches of the Iberian Peninsula, both in undisturbed and disturbed flow
regimes.
To achieve these objectives the following methodology has been applied. The model has
been conceptualized as a static tank flow model based on the actual evapotranspiration of the
riparian plants. This tank represents a portion of soil of the superficial root layer. The lower
capacity limit of this tank is the permanent wilting moisture of the soil sample. On the other
hand the upper capacity limit is the field capacity moisture. The tank’s input flows are the
precipitation, the root water rise and the capillary water rise. In contrast output flows are
the actual evapotranspiration and the excess water of the tank. The most relevant
model parameters are the soil retention curves, vegetation functional type parameters
(specially related to root depths and the transpiration efficiency factors) and the daily
hidro-meteorological data, which are water table elevation, precipitation and potential
evapotranspiration.
The model runs for a limited amount of vegetation functional types. In our simulations the
following four functional types were used: Riparian Herbs; Riparian Juveniles and Small
Scrubs, Riparian Trees and Big Shrubs; and Terrestrial Vegetation. The general model output
variable is an evapotranspiration index based in the quotient between the current and the
potential evapotranspiration. This index is used to determine the suitability of the
simulated vegetation functional types to certain environmental conditions. Secondly, a
sensitivity analysis was made for determining the most relevant model parameters.
Finally the model has been calibrated and validated using as objective function a
confusion matrix which compares the observed and the simulated riparian vegetation
zonation.
The calibration/validation processes have been carried out in seven study sites of the Jucar
River Basin District. Four of those sites have a natural flow regime and three of them a
regulated flow regime due to the presence of dams. Results have shown that the model is
capable of providing effective simulations in compared to the observed riparian vegetation. |
|
|
|
|
|
|