Geoscience Reference
In-Depth Information
Table 10.2 Observed and modelled meteorological and hydrological data for the river
Kennet under various climate change scenarios
Observed*
Modelled
Variable
Units
1961-90
1961-90
2020s A2/B2 2050s B2
2050s A2
Annual rainfall
mm yr −1
759
759
778
757
758
Days with:
<0.2 mm
%
56
54.7
55.5
57.7
57.3
<1.0 mm
%
67
66.1
65.9
67.4
67.2
Temperature
Mean
°C
9.2
9.2
10.2
11.0
11.3
Mean daily min
°C
5.1
5.4
6.4
7.2
7.5
Mean daily max
°C
13.0
12.9
14.0
14.8
15.1
Max daily max
°C
33
31.2
33.6
38.1
38.5
Min daily min
°C
−16
−14.8
−12.0
−12.8
−13.5
PET
mm yr −1
536
641
728
750
AET
mm yr −1
459
481
503
512
HER §
mm yr −1
299
298
254
247
Discharge in river Kennet
Annual mean
m 3 s −1
9.60
9.83
9.87
8.38
8.15
Minimum
m 3 s −1
0.93
2.12
1.76
1.29
0.74
Maximum
m 3 s −1
46.7
46.6
61.8
59.3
48.9
5th percentile
m 3 s −1
3.88
3.43
3.38
2.60
2.44
1st percentile
m 3 s −1
2.37
2.65
2.39
1.99
1.62
*Meteorological data are catchment means from the UK Meteorological Office at http://www.
metoffice.gov.uk/climate/uk/averages/ukmapavge.html#. Hydrological data are from the UK
National River Flow Archive at http://www.nwl.ac.uk/ih/nrfa/webdata/039016/g.html. INCA-
predicted discharge has been adjusted to take into account drinking water abstraction from the
catchment.
PET, potential evapotranspiration (calculated without restrictions due to water availability).
AET, actual evapotranspiration.
§ HER, hydrologically effective rainfall (rainfall potentially available to recharge rivers).
questions requires a long chain of models with associated uncertainties, suggesting
that there might be some benefit in developing methods to reduce the number of
models needed. Tisseuil et al . (2009) had reasonable success in predicting river
flows in a variety of river types in the Garonne Basin by direct downscaling from
GCMs using a variety of statistical models.
The results of the Kennet study (Table 10.2) show that when comparing
observed and modelled data for the validation period of 1961-90, EARWIG is
successful in reproducing the observed mean and distribution of temperature
and rainfall. Likewise, the INCA-N model successfully uses these data to
reproduce the observed mean and distribution of discharge in the river, except
that the observed absolute minimum values, due to the exceptionally dry period
of 1975-6, are not simulated. The success at reproducing the observed data with
minimal calibration gives some confidence in the future predictions (Table 10.2).
Search WWH ::




Custom Search