 |
| Titel |
Effects of site characteristics on cumulative frequency distribution of water table depth in peatlands |
| VerfasserIn |
Michel Bechtold, Bärbel Tiemeyer, Enrico Frahm, Niko Roßkopf |
| Konferenz |
EGU General Assembly 2013
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 15 (2013) |
| Datensatznummer |
250082097
|
|
|
|
|
|
| Zusammenfassung |
| Previous studies demonstrated strong dependency of vegetation development and GHG
emissions from peatlands on annual mean water table depth. It is also proposed that the
duration of ponding and low water level periods are important indicators for CH4 emissions
and the presence of specific plant species. Better understanding of the annual water table
dynamics and the influence of site characteristics helps to explain variability of vegetation
and emissions at the plot scale. It also provides essential information for a nation-wide
upscaling of local gas flux measurements and for estimating the impact of regional adaption
strategies.
In this study, we analyze the influence of site characteristics on the cumulative frequency
distribution of water table depth in a peatland. On the basis of data from about 100 sites we
evaluate how distribution functions, e.g. the beta distribution function, are a tool for
the systematic analysis of the site-specific frequency distribution of water table
depth. Our analysis shows that it is possible to differentiate different shape types
of frequency distributions, in particular left-skewed (bias towards the water table
minimum), right-skewed (bias towards the water table maximum), and ’S’-shaped
distributions (bias towards the mid of min and max). The shape is primarily dependent
on the annual mean water table depth, but also shows dependencies on land use,
peatland type, catchment size and soil properties. Forest soils are for example all
characterized by a ’S’-shaped distribution. Preliminary results indicate that data
sets that do not show a beta distribution are mostly from observation wells that are
located close to drainage courses and/or are from sites characterized by strong water
management (e.g. abruptly changing weir levels). The beta distribution might thus be a
tool to identify sites with a ’non-natural’ frequency distribution or erroneous data
sets. Because the parameters of the beta distribution show a dependency on site
characteristics, they can be used for the regionalization of threshold exceedance
probabilities. |
|
|
|
|
|
|