On Validating the Integration of a General Circulation Model into a Limited Area Model for Producing 40-Year Wave Atlas for the Irish Sea
Nandamudi Vijaykumar (1,2),
Robert Devoy (1), Jeremy Gault (1), Declan Dunne (1) Cathal O'Mahony (1)
(1) Coastal and Marine Resources Centre, Environmental Research
Institute,
University College Cork (IE)
(2) Laboratory of Computing and Applied Mathematics, National Institute for
Space Research, SJCampos (BR)
Introduction
According to Intergovernmental Panel on Climate Change (IPCC) Climate Change
Assessments, climate warming for the 21st century is directly associated with
environmental risk. These risks have a huge effect on human populations and
wildlife. The context of this paper is focused towards coastal processes [Devoy,
1994].
It is extremely useful if sensitive coastal areas are determined and assessed
due to climate change influence. By quantifying this influence through some
methodology risk of the locations can be quantified. The methodology should
identify zones with wind-generated waves as these will lead to strong surges,
floods and loss of land. Moreover, analysts must be equipped with adequate
tools to warn the appropriate authorities to take actions to protect the population.
In order to disseminate information in a clear and structured manner, it must
be incorporated into a Geographic Information System (GIS). This aspect needs
an identification of geo-information to assess coastal processes, evaluation
of interoperability issues, organisation of a database to store information
and finally development of tools for displaying and analysing the information.
In order to disseminate the information properly as well as to guarantee its
reliability, rigorous validation procedures must be applied to guarantee the
quality of information that is stored and visualised. GIS can be an efficient
tool for marine and coastal data analysis and therefore will play an important
role in Coastal Zone Management. The paper described here is concentrated
on the validation aspect of the results of an atmospheric model.
Integration of Weather Models
Several studies are in progress to understand and predict the linkage between
climate change and its impact on coastal processes. There is not much information
about the past wave events due to poor observational records that were mostly
based on fewer observations in the past. Added to this is that concepts on
numerical modelling of dynamical processes were limited [WASA Group, 1998].
An ongoing research project is developing a 40-year (1958 to 1997) Wave Atlas
for the Irish Sea. It is also being developed by some other countries to produce
the atlas for other regions. In order to achieve this, weather models have
to be integrated with a wave model by running them backwards in time. Usually,
an atmospheric model is downscaled to coastal sites by integrating a General
Circulation Model with a Limited Area Model to obtain high-quality wind information,
from which wave conditions can be reconstructed. A second phase, then, consists
of using the downscaled results to be integrated with a state-of-the-art wave
model in order to obtain the wave climate and assess their impacts on coastal
processes.
The objective of this paper is focused on the integration of a general circulation
model with a limited area model and more precisely its validation. Within
the context of the production of Wave Atlas for the last 40 years, the limited
area model HIRLAM (High Resolution Limited Area Model) is the model selected
to generate the wind fields for the Irish Sea. However, it is essential that
the outputs generated by HIRLAM runs are reliable and hence the necessity
of applying a validation procedure in order to assess the quality of the generated
results.
The project uses the existing data from NCEP/NCAR global reanalyses [Kalnay
et al., 1996] that were based on running global atmospheric models for reanalysing
observational data back in time. However, both spatial and temporal resolutions
were too coarse for conducting wave studies in coastal areas [Weisse &
Gayer, 2000]. Therefore these reanalyses were used as a forcing of limited
area model, HIRLAM. Additionally, for Irish Sea HIRLAM was nested into REMO
simulations [Soares et a., 2002] to further increase the spatial resolution.
For some other areas, a spectral nudging technique [von Storch et al., 2000]
was applied to enhance lateral boundary conditions by imposing time-variable
large-scale atmospheric states on a regional atmospheric model.
HIRLAM generated wind fields for the 40 years. The next step is to use these
fields as a forcing of the wave model WAM. However, before proceeding to the
next step the wind fields generated by HIRLAM have to be verified for their
quality.
Validation
Validation is essential in verifying the results obtained from general circulation
models. Moreover, in case of this research project several countries are participating
with a common objective of generating a wave atlas for different regions and
it is vital to establish rigorous and common validation procedures to guarantee
a similar quality to the data generated for use in the atlas.
The validation procedure adopted here takes into account available wind reanalyses
data to be compared with the HIRLAM produced wind fields. These fields will
be compared based on correlation coefficient as well as root mean square (rms).
Application of complex correlation [Weisse & Gayer, 2000] is also under
consideration. Observation data as well as ERA-15 and ERA-40 data will also
be explored. Use of GIS in incorporating the Wave Atlas will be discussed.
Results and comments on the procedures as well as future directions in concluding
the research project as a whole will be presented.
References