Geoscience Reference
In-Depth Information
rainfall on soil moisture on the seasonal timescale. Douville et al. (
2001
) showed that water
recycling (where increased evaporation from wetter soil enhances precipitation) has a
stronger influence on the West African monsoon than the South Asian monsoon because
the moisture convergence in the latter region is much larger and itself responds to increased
soil moisture in such a way as to counteract the local enhancement of precipitation. This
illustrates that the soil-precipitation feedback on the monsoons is regionally dependent.
Studies such as that of Turner and Slingo (
2011
) have indicated that the cooling
associated with increased Eurasian snow cover in boreal spring can be associated with
weakening of the Tibetan anticyclone and weaker upper level easterly winds over the
monsoon region in summer, as a result of the reduction in the tropospheric meridional
temperature gradient. However, Turner and Slingo (
2011
) noted that this response could be
reversed if the opposing changes over the Himalayas and Tibetan Plateau were dominant.
Martin and Levine (
2012
) demonstrated that the inclusion of dynamic vegetation in the
HadGEM2 model family generated a similar response through changes in the distribution
of needleleaf trees, and the resulting impact on snow cover, over northeast Eurasia.
Simulating monsoons, therefore, requires adequate representation of hydrological pro-
cesses in nearly all components of the climate system. Despite many decades of research
focused on these phenomena, they remain a challenge for model development.
3 Summary
The challenge to quantify and reduce uncertainty in the large-scale response of the global
water cycle is immense, not least because the hydrological cycle involves almost every
component of the climate system. In this review, we have attempted to demonstrate the
vast amount of research which is currently underway to improve our understanding of
hydrological cycle processes and their representation in models. This includes observa-
tional analyses, systematic increases in model resolution, parametrisation development,
inclusion of Earth system processes and idealised modeling studies.
The role of resolution in the representation of the hydrological cycle is complex: there is
evidence that global models require a horizontal resolution of at least 50 km in order to
represent the global hydrological cycle, while km-scale resolution facilitates realistic
representation of rainfall distribution and variability on a local scale, and may be required
for other areas of the hydrological cycle such as the storage and movement of water
through the land surface. Increased vertical resolution may also be required in order that
layer cloud processes can be represented.
The realism of rainfall in a model is a key indicator of its skill in representing the
underlying physical processes, and hence for projecting future changes in rainfall. In par-
ticular, the spatial and temporal structure of rainfall is arguably more important than the
absolute rainfall amount, which is typically used to assess model skill. We have highlighted
studies which indicate that explicit representation of convective processes can improve the
spatial and temporal structure of rainfall, but such models are still very expensive to run.
Therefore, there is a need for such experiments to inform the development of improved
parametrisations. Of particular interest in future will be whether models with explicit treat-
ment of convection show a change in the spatial and temporal characteristics of heavy rainfall
in a warmer climate, since this is unlikely to be captured by coarser resolution models.
Accurately predicting such changes is essential for estimating changes in flood risk.
Development and evaluation of new physical parametrisations across timescales from
daily/monthly/seasonal/decadal/centennial has been shown to be beneficial in several
