 |
| Titel |
Derivation of RCM-driven potential evapotranspiration for hydrological climate change impact analysis in Great Britain: a comparison of methods and associated uncertainty in future projections |
| VerfasserIn |
C. Prudhomme, J. Williamson |
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| ISSN |
1027-5606
|
| Digitales Dokument |
URL |
| Erschienen |
In: Hydrology and Earth System Sciences ; 17, no. 4 ; Nr. 17, no. 4 (2013-04-10), S.1365-1377 |
| Datensatznummer |
250018844
|
| Publikation (Nr.) |
 copernicus.org/hess-17-1365-2013.pdf |
|
|
|
|
|
| Zusammenfassung |
| Potential evapotranspiration (PET) is the water that would be lost by plants
through evaporation and transpiration if water was not limited in the soil,
and it is commonly used in conceptual hydrological modelling in the
calculation of runoff production and hence river discharge. Future changes
of PET are likely to be as important as changes in precipitation patterns in
determining changes in river flows. However PET is not calculated routinely
by climate models so it must be derived independently when the impact of
climate change on river flow is to be assessed. This paper compares PET
estimates from 12 equations of different complexity, driven by the
Hadley Centre's HadRM3-Q0 model outputs representative of 1961–1990, with
MORECS PET, a product used as reference PET in Great Britain. The results
show that the FAO56 version of the Penman–Monteith equations reproduces best
the spatial and seasonal variability of MORECS PET across GB when driven by
HadRM3-Q0 estimates of relative humidity, total cloud, wind speed and
linearly bias-corrected mean surface temperature. This suggests that
potential biases in HadRM3-Q0 climate do not result in significant biases
when the physically based FAO56 equations are used. Percentage changes in
PET between the 1961–1990 and 2041–2070 time slices were also calculated for
each of the 12 PET equations from HadRM3-Q0. Results show a large
variation in the magnitude (and sometimes direction) of changes estimated
from different PET equations, with Turc, Jensen–Haise and calibrated
Blaney–Criddle methods systematically projecting the largest increases
across GB for all months and Priestley–Taylor, Makkink, and Thornthwaite
showing the smallest changes. We recommend the use of the FAO56 equation as,
when driven by HadRM3-Q0 climate data, this best reproduces the reference
MORECS PET across Great Britain for the reference period of 1961–1990.
Further, the future changes of PET estimated by FAO56 are within the range
of uncertainty defined by the ensemble of 12 PET equations. The changes
show a clear northwest–southeast gradient of PET increase with largest
(smallest) changes in the northwest in January (July and October)
respectively. However, the range in magnitude of PET changes due to the
choice of PET method shown in this study for Great Britain suggests that PET
uncertainty is a challenge facing the assessment of climate change impact on
hydrology mostly ignored up to now. |
| |
|
| Teil von |
|
|
|
|
|
|
|
|