 |
| Titel |
Turbulence and intermittency in lakes, a multifractal correlation analysis |
| VerfasserIn |
Y. Mezemate, V. Rambach, I. Tchiguirinskaia, B. Lemaire, B. Vincon Leite, C. Bonhomme, D. Schertzer, S. Lovejoy |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250070808
|
|
|
|
|
|
| Zusammenfassung |
| Cyanobacteria proliferation in lakes, such as those occuring frequently in Lake Bourget, is an
environmental issue, as well as a matter of public health. In the frame of the PROLIPHYC
project, an alarm system to warn about sudden proliferations of cyanobacteria was developed.
The system is based on high-frequency measurements of water temperature and of the
fluorescence of different phytoplankton classes, a proxy of their concentrations. In an
attempt to simplify the warning system, we investigate the predictability of the
proliferation of cyanobacteria from measurements of precursor variables, which are
easy to measure (e.g. water temperature). The classical approach would be to find
the correlations at a given scale of cyanobacteria fluorescence with the precursor
fields. However, this is rather in contradiction with the multiscale variability of lake
dynamics, which results from nonlinear interactions between different scales and
processes.
To find an alternative, we first performed a spectral analysis of the velocity, temperature
and cyanobacteria fluorescence measurements to have a first estimate of their scaling ranges.
For instance, the temperature spectrum over small scales is rather similar to that of a passive
scalar. To go beyond the limitations of a second order statistical analysis, we performed
multifractal analyses (Trace Moment, Double Trace Moment) to estimate the scaling
behaviour of statistical moments of different orders q′s . The nonlinear behaviour of their
scaling moment function K(q) shows that the fields are indeed multifractal and therefore
intermittent. As a consequence we proceed to a multifractal correlation analysis between
these fields, i.e. a correlation analysis across scales, in particular between temperature and
cyanobacteria fluorescence.
In conclusion, we discuss the predictability of the proliferation of cyanobacteria
from temperature measurements, which results from their multifractal correlation. |
|
|
|
|
|
|