She et al. 
Operational Oceanography and Earth System Science 
Frontiers In Earth Science | www.frontlersln.org 
10 
February 2020 | Volume 8 | Article 7 
monitoring program, fishery monitoring and some research 
projects, which has a much better coverage then the BOOS 
network in the open Baltic Sea. Most of the research observations 
are not in real or near real time. The research observing program 
can be made to fit for the operational application through 
open data policy and NRT data delivery, which will fill the 
observational data gaps. For example, research data from classical 
observations like CTD can be made available in NRT before the 
final processing for (climate) research. Vice versa, good quality 
operational observations can be used for (climate) research 
after appropriate quality control and exposing it to appropriate 
processing in delayed mode. 
Operational Modeling 
The research on the six GCs by the Baltic Earth community is 
highly relevant to improve the operational modeling capacity. By 
revealing important factors controlling the salinity and sea level 
dynamics, results from the GCs 1,4, and 5 can be used to improve 
the long-term performance of operational ocean models. The 
GC2 research can be used to improve the downscaled models 
for resolving coastal-estuary continuum. The GC3 may find new 
features and knowledge regarding to the extreme events, which 
are always challenge cases in the operational modeling. The GC6 
and GC2 research may benefit emerging areas of operational 
modeling, e.g., operational ecological modeling, coupled ocean- 
hydrological modeling and sediment transport modeling etc. 
In order to benefit the operational modeling, there should be 
a platform to transform the Baltic Earth research results into the 
operational models. This is similar to the “Service Evolution” 
element in the CMEMS where dedicated, short-term mini-R&D 
projects are funded to transfer the best practice in modeling and 
observing research into CMEMS system. 
DISCUSSIONS AND RECOMMENDATIONS 
The Baltic operational oceanography community and the 
BALTEX/Baltic Earth community have co-existed in the 
past two/three decades. However, only very preliminary 
interactions have been carried out because differing objectives 
of both communities hampered an intensive collaboration as 
outlined below. 
Through analysis of the state-of-the-art of operational 
oceanography and regional Earth system research, it was found 
that the operational observations, modeling platforms, and 
products can significantly benefit the Baltic Earth research, 
e.g., in the areas of the six grand challenges, while the Baltic 
Earth research can also benefit operational oceanography. Most 
of the research databases such as the Baltic Environmental 
Database (BED, http://nest.su.se/bed) at Stockholm University, 
the Swedish Ocean Archive (SHARK, http://sharkweb.smhi. 
se) and all other environmental databases operated by the 
Swedish Meteorological and Hydrological Institute (SMHI) 
and the German Baltic Sea monitoring data archive (IOWDB, 
http://iowmeta.io-warnemuende.de and https://odin2.io- 
warnemuende.de/) operated by the Leibniz Institute for Baltic 
Sea Research Warnemunde (IOW) are already open access. 
In the future, a common database including both operational 
and research data may be established. In Europe, EMODnet 
has integrated marine observations, both online and offline 
data, in the entire parameter domain ranging from physical, 
biogeochemical, biological to human activities. In future, it 
may play a more important and active role for linking research 
observations and operational oceanography. However, for the 
research of, for instance, detecting ocean changes high accuracy 
of measurements is needed which is today not assured by all 
operational data products. The calibration of measurement 
devices is time consuming and expensive. Quality control is 
also the reason why not all research data of IOWDB and other 
databases can be provided in real or near real time. 
Research observations can be an important complementary 
in emerging observations for developing operational ecology, 
predicting the fate of visible, and invisible marine plastics in 
the Baltic Sea, modeling and forecasting sediment transport, 
underwater noise etc. For these areas, operational observing 
capacity has not been established yet. They will need to heavily 
rely on data from research projects. 
As mentioned above, reanalysis data sets are useful for climate 
analysis and for the evaluation of climate models used within 
Baltic Earth. However, current reanalysis data sets are usually not 
based on first principles, i.e., the conservation of mass, energy 
and momentum. Hence, many data assimilation schemes may 
cause problems, for instance, for budget and flux calculations of 
nutrients (Liu et al., 2017). This may lead to problems when using 
reanalysis for trend and long-term variability analyses. Hence, 
reanalysis products should be improved for the climate research 
purpose by using more mass and energy conserved assimilation 
method, e.g., 4D variational assimilation. 
Research observing infrastructure should be made more 
usable for various applications. Examples of observing systems 
that fulfill such a criterion are the Australian IMOS (Integrated 
Marine Observing System, Hill et al., 2009) system, the German 
COSYNA (Coastal Observing System for the Northern and 
Arctic Seas, Baschek et al., 2017) and the IOW long-term 
monitoring program outside the German territorial waters. 
Although funded as a research infrastructure, the programs 
provide openly accessible observations, which can be used for 
many other purposes. These systems have been operated for more 
than 10 years. 
The new knowledge made from the Baltic Earth research, 
both on the processes and model system development, can 
be transferred to the operational models, which will fill the 
knowledge and technological gaps in the operational modeling. 
The two communities should have regular joint meetings to 
identify topics with common interests and make the technology 
transfer from regional Earth system research to operational 
oceanography. For such a working group Baltic Earth would 
provide an ideal discussion platform. Unfortunately, state-of-the- 
art climate and operational ocean model versions usually differ 
although some institutes aim to have only one version for both 
applications. However, in reality two model versions are still 
needed because not all processes are well-enough understood 
such that their parameterizations fulfill the requirements of 
both applications on short and long time scales. An example 
is diapycnical mixing. Climate models require that long-term