Geoscience Reference
In-Depth Information
Fig. 2 Simulated (upper panels) and observed (lower panels) summer climatologies (JJAS) for the period
1961-2000: 2 m temperature [C] (left, REMO and data of Willmott and Matsuura
2009
) and mean sea-
level pressure [hPa] (right, REMO and ERA40). The white circles indicate the heat low region
soil wetness was increased to a critical value in each time step, so that potential ET can
occur. In this way, it is assumed that irrigation is conducted to fulfil optimal conditions for
the vegetation/crops, allowing them to transpire at a potential rate. Note that the main
results will not change if water for irrigation is limited by available water from the rivers
(Saeed et al.
2012
).
In the REMO simulation without irrigation, the overestimation of the heat low (too high
temperature, too low pressure; Fig.
2
) resulted in increased differential heating between
ocean and land, and therefore the overestimation of winds entering into the plains of the
Indian subcontinent from the Arabian Sea. This causes a situation unfavourable for
westward propagating currents from Bay of Bengal to intrude deep into western India and
Pakistan. Therefore, less moisture is advected causing an underestimation of precipitation
over this area as well. When irrigation is accounted for, a more realistic behaviour of the
simulated climate is yielded. Figures
3
and
4
compare the changes in the REMO simu-
lation with irrigation (Fig.
3
b, d) to the reference simulation without irrigation (Fig.
3
a, c).
An improvement of simulated temperature and MSLP can be seen over the whole region,
but statistically significant and most pronounced changes are present over Indus, Ganges
(Fig.
1
b) and southern India (Fig.
4
a, b). For these regions, the standard REMO version
simulated the largest systematic biases (Fig.
2
); hence, the irrigation led to a better
