 |
| Titel |
Setting up an atmospheric-hydrologic model for seasonal forecasts of water flow into dams in a mountainous semi-arid environment (Cyprus) |
| VerfasserIn |
Corrado Camera, Adriana Bruggeman, Georgios Zittis, Panos Hadjinicolaou |
| Konferenz |
EGU General Assembly 2017
|
| Medientyp |
Artikel
|
| Sprache |
en
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 19 (2017) |
| Datensatznummer |
250138360
|
| Publikation (Nr.) |
 EGU/EGU2017-1344.pdf |
|
|
|
|
|
| Zusammenfassung |
| Due to limited rainfall concentrated in the winter months and long dry summers, storage and
management of water resources is of paramount importance in Cyprus. For water storage
purposes, the Cyprus Water Development Department is responsible for the operation of 56
large dams total volume of 310 Mm3) and 51 smaller reservoirs (total volume of 17 Mm3)
over the island. Climate change is also expected to heavily affect Cyprus water
resources with a 1.5%-12% decrease in mean annual rainfall (Camera et al., 2016)
projected for the period 2020-2050, relative to 1980-2010. This will make reliable
seasonal water inflow forecasts even more important for water managers. The overall
aim of this study is to set-up the widely used Weather Research and Forecasting
(WRF) model with its hydrologic extension (WRF-hydro), for seasonal forecasts of
water inflow in dams located in the Troodos Mountains of Cyprus. The specific
objectives of this study are: i) the calibration and evaluation of WRF-Hydro for the
simulation of stream flows, in the Troodos Mountains, for past rainfall seasons; ii) a
sensitivity analysis of the model parameters; iii) a comparison of the application of the
atmospheric-hydrologic modelling chain versus the use of climate observations as forcing.
The hydrologic model is run in its off-line version with daily forcing over a 1-km grid, while
the overland and channel routing is performed on a 100-m grid with a time-step
of 6 seconds. Model outputs are exported on a daily base. First, WRF-Hydro is
calibrated and validated over two 1-year periods (October-September), using a 1-km
gridded observational precipitation dataset (Camera et al., 2014) as input. For the
calibration and validation periods, years with annual rainfall close to the long-term
average and with the presence of extreme rainfall and flow events were selected.
A sensitivity analysis is performed, for the following parameters: partitioning of
rainfall into runoff and infiltration (REFKDT), the partitioning of deep percolation
between losses and baseflow contribution (LOSS_BASE), water retention depth
(RETDEPRTFAC), overland roughness (OVROUGHRTFAC), and channel manning
coefficients (MANN). The calibrated WRF-Hydro shows a good ability to reproduce
annual total streamflow (-19% error) and total peak discharge volumes (+3% error),
although very high values of MANN were used to match the timing of the peak and
get positive values of Nash-Sutcliffe efficiency coefficient (0.13). The two most
sensitive parameters for the modeled seasonal flow were REFKDT and LOSS_BASE.
Simulations of the calibrated WRF-Hydro with WRF modelled atmospheric forcing
showed high errors in comparison with those forced with observations, which can be
corrected only by modifying the most sensitive parameters by at least one order of
magnitude.
This study has received funding from the EU H2020 BINGO Project (GA 641739).
Camera C., Bruggeman A., Hadjinicolaou P., Pashiardis S., Lange M.A., 2016. Evaluation of
interpolation techniques for the creation of gridded daily precipitation (1 × 1 km2); Cyprus,
1980–2010. J Geophys Res Atmos 119, 693–712, DOI:10.1002/2013JD020611
Camera C., Bruggeman A., Hadjinicolaou P., Michaelides S., Lange M.A., 2016.
Evaluation of a spatial rainfall generator for generating high resolution precipitation
projections over orographically complex terrain. Stoch Environ Res Risk Assess, DOI
10.1007/s00477-016-1239-1 |
|
|
|
|
|
|