 |
| Titel |
A simplified GIS-based model for large wood recruitment and connectivity in mountain basins |
| VerfasserIn |
Ana Lucía, Andrea Antonello, Daniela Campana, Marco Cavalli, Stefano Crema, Silvia Franceschi, Enrico Marchese, Martin Niedrist, Stefan Schneiderbauer, Francesco Comiti |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250097533
|
| Publikation (Nr.) |
 EGU/EGU2014-13125.pdf |
|
|
|
|
|
| Zusammenfassung |
| The mobilization of large wood (LW) elements in mountain rivers channels during floods
may increase their hazard potential, especially by clogging narrow sections such as bridges.
However, the prediction of LW transport magnitude during flood events is a challenging
topic. Although some models on LW transport have been recently developed, the objective of
this work was to generate a simplified GIS-based model to identify along the channel network
the most likely LW-related critical sections during high-magnitude flood events in forested
mountain basins.
Potential LW contribution generated by landsliding occurring on hillslopes is
assessed using SHALSTAB stability model coupled to a GIS-based connectivity index,
developed as a modification of the index proposed by Cavalli et al (2013). Connected
slope-derived LW volumes are then summed at each raster cell to LW volumes
generated by bank erosion along the erodibile part of river corridors, where bank
erosion processes are estimated based on user-defined channel widening ratios
stemming from observations following recent extreme events in mountain basins. LW
volume in the channel is then routed through the stream network applying simple
Boolean rules meant to capture the most important limiting transport condition in
these high-energy systems at flood stage, i.e. flow width relative to log length. In
addition, the role of bridges and retention check-dams in blocking floating logs
is accounted for in the model, in particular bridge length and height are used to
characterize their clogging susceptibility for different levels of expected LW volumes and
size.
The model has been tested in the Rienz and Ahr basins (about 630 km2 each), located in
the Eastern Italian Alps. Sixty percent of the basin area is forested, and elevations range
from 811 m a.s.l. to 3488 m a.s.l.. We used a 2.5 m resolution DTM and DSM,
and their difference was used to calculate the canopy height. Data from 35 plots
of the National Forest Inventory were used to estimate forest stand volume by a
semi-empirical model. Ddatabase on shallow landslides along with precipitation depth was
utilized to calibrate the parameters for the SHALSTAB model. Orthophotos (0.5 m
pixel resolution) and existing technical maps were used to delimitate the channel
banks, which were used to calculate automatically channel width for each grid
cell.
The model output provided information about the expected volume and mean size of LW
recruited and transported during a 300 yr flood event in the test basins, as well as the location
of the most probable clogged sections (mostly related to infrastructures) along the channel
network. The model thus shows the capability to assist river managers in identifying the most
critical sections of river networks and to assess the effectiveness and location of different
mitigation options such as wood retention structures or forest management practices. |
|
|
|
|
|
|