 |
| Titel |
A model for water discharge based on energy consumption data (WATEN). |
| VerfasserIn |
María Carmen Moyano, Lucía Tornos, Luis Juana |
| Konferenz |
EGU General Assembly 2014
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 16 (2014) |
| Datensatznummer |
250086848
|
| Publikation (Nr.) |
 EGU/EGU2014-783.pdf |
|
|
|
|
|
| Zusammenfassung |
| As the need for water conservation is becoming a major water concern, a lumped model
entitled WATEN has been proposed to analyse the water balance in the B-XII Irrigation
Sector of the Lower Guadalquivir Irrigated Area, one of the largest irrigated areas in Spain.
The aim of this work is to approach the hydrological study of an irrigation district lacking of
robust data in such a manner that the water balance is performed from less to more process
complexity.
WATEN parameters are the total and readily available moisture in the soil, a fix
percentage for effective precipitation, and the irrigation efficiency. The Sector presents six
different drainage pumping stations, with particular pumping groups and with no water flow
measurement devices. Energy consumption depends on the working pumping stations and
groups, and on the variable water level to discharge. Energy consumed in the drainage
pumping stations has been used for calibration
The study has relied on two monthly series of data: the volume of drainage
obtained from the model and the energy consumed in the pumping stations. A double
mass analysis has permitted the detection of data tendencies. The two resulting
series of data have been compared to assess model performance, particularly the
Pearson’s product moment correlation coefficient and the Nash-Sutcliffe coefficient of
efficiency, e2, determined for monthly data and for annual and monthly average
data.
For model calibration, we have followed a classical approach based on objective functions
optimization, and a robust approach based on Markov chain Monte Carlo simulation process,
driven in a similar manner to genetic algorithms, entitled Parameters Estimation on Driven
Trials (PEDT), and aiming to reduce computational requirements.
WATEN has been parameterised maintaining its physical and conceptual rationality. The
study approach is outlined as a progressive introduction of data. In this manner, we can
observe its effect on the studied objective functions, and visualize if new data adds significant
improvements to model results. The model attained an average Nash-Sutcliffe coefficient
e2~= 0.90 between based on energy drainage observations and estimated drainage
discharge.
The study has shown that the Sector crop evapotranspiration, is lower than the expected
value in pristine conditions. This reduction would be more noticeable at the end of the
summer months, attaining as far as a 40% reduction. Average drainage in the studied period,
is about 3700 m3/ha/year.
This methodology is thought to be the basis for similar worldwide studies comprising
scarce-data irrigation districts with drainage discharge to receiving water bodies, and serve as
a guide for future alike applications. |
|
|
|
|
|
|