EGU Journal Logos (RGB) Open Access Open Access Open Access Advances in Annales Nonlinear Processes Geosciences Geophysicae in Geophysics Open Access Open Access Natural Hazards Natural Hazards and Earth System and Earth System Sciences Sciences Discussions Open Access Open Access Atmospheric Atmospheric Chemistry Chemistry and Physics and Physics Discussions Open Access Open Access Atmospheric Atmospheric Measurement Measurement Techniques Techniques Discussions Open Access Open Access Biogeosciences Biogeosciences Discussions Open Access Open Access Climate Climate of the Past of the Past Discussions Open Access Open Access Earth System Earth System Dynamics Dynamics Discussions Open Access Geoscientific Geoscientific Open Access Instrumentation Instrumentation Methods and Methods and Data Systems Data Systems Discussions Open Access Open Access Geosci. Model Dev., 6, 301–325, 2013 Geoscientific www.geosci-model-dev.net/6/301/2013/ Geoscientific doi:10.5194/gmd-6-301-2013 Model Development Model Development © Author(s) 2013. CC Attribution 3.0 License. Discussions Open Access Open Access Hydrology and Hydrology and Earth System Earth System Sciences Sciences Evaluation of the carbon cycle components in the Norwegian Earth Discussions Open Access Open Access System Model (NorESM) Ocean Science Ocean Science Discussions J. F. Tjiputra1,2,3, C. Roelandt1,3, M. Bentsen1,3, D. M. Lawrence4, T. Lorentzen1,3, J. Schwinger2,3, Ø. Seland5, and C. Heinze1,2,3 1 Open Access Uni Climate, Uni Research, Bergen, Norway Open Access 2University of Bergen, Geophysical Institute, Bergen, Norway 3 Solid Earth Bjerknes Centre for Climate Research, Bergen, Norway Solid Earth 4National Center for Atmospheric Research, Boulder, Colorado, USA Discussions 5Norwegian Meteorological Institute, Oslo, Norway Correspondence to: J. F. Tjiputra ([email protected]) Open Access Open Access Received: 23 July 2012 – Published in Geosci. Model Dev. Discuss.: 4 October 2012 The Cryosphere The Cryosphere Revised: 6 February 2013 – Accepted: 11 February 2013 – Published: 4 March 2013 Discussions Abstract. The recently developed Norwegian Earth System vegetation gross primary productivity (GPP) when com- Model (NorESM) is employed for simulations contribut- pared to the observationally based values derived from the ing to the CMIP5 (Coupled Model Intercomparison Project FLUXNET network of eddy covariance towers. While the phase 5) experiments and the fifth assessment report of the model simulates well the vegetation carbon pool, the soil car- Intergovernmental Panel on Climate Change (IPCC-AR5). bon pool is smaller by a factor of three relative to the observa- In this manuscript, we focus on evaluating the ocean and tional based estimates. The simulated annual mean terrestrial land carbon cycle components of the NorESM, based on GPP and total respiration are slightly larger than observed, the preindustrial control and historical simulations. Many of but the difference between the global GPP and respiration is the observed large scale ocean biogeochemical features are comparable. Model-data bias in GPP is mainly simulated in reproduced satisfactorily by the NorESM. When compared the tropics (overestimation) and in high latitudes (underesti- to the climatological estimates from the World Ocean At- mation). Within the NorESM framework, both the ocean and las (WOA), the model simulated temperature, salinity, oxy- terrestrial carbon cycle models simulate a steady increase in gen, and phosphate distributions agree reasonably well in carbon uptake from the preindustrial period to the present- both the surface layer and deep water structure. However, day. The land carbon uptake is noticeably smaller than the the model simulates a relatively strong overturning circula- observations, which is attributed to the strong nitrogen limi- tion strength that leads to noticeable model-data bias, es- tation formulated by the land model. pecially within the North Atlantic Deep Water (NADW). This strong overturning circulation slightly distorts the struc- ture of the biogeochemical tracers at depth. Advancements in simulating the oceanic mixed layer depth with respect 1 Introduction to the previous generation model particularly improve the surface tracer distribution as well as the upper ocean bio- In addition to the atmospheric radiative properties, global cli- geochemical processes, particularly in the Southern Ocean. mate dynamics also depend on the complex simultaneous in- Consequently, near-surface ocean processes such as biologi- teractions between the atmosphere, ocean, and land. These cal production and air–sea gas exchange, are in good agree- interactions are not only non-linear, but also introduce feed- ment with climatological observations. The NorESM adopts backs of different magnitude and signs to the climate system. the same terrestrial model as the Community Earth System In order to understand the sophisticated interplay between Model (CESM1). It reproduces the general pattern of land- the different components, Earth system models have been de- veloped by the geoscientific community in recent years. The Published by Copernicus Publications on behalf of the European Geosciences Union. 302 J. F. Tjiputra et al.: NorESM carbon cycle last Intergovernmental Panel for Climate Change Assessment ulated as part of the NorESM or CESM1 frameworks (e.g., Report (IPCC-AR4) stated that in order to produce a reliable Arora et al., 2013; Jones et al., 2013). For more in-depth re- future climate projection, such models are required (Denman views and evaluation of the CLM4, the readers are referred to et al., 2007). recent publications by CLM4 developers (e.g., Oleson et al., An Earth system model typically consists of a global phys- 2010; Lawrence et al., 2011; Gent et al., 2011). ical climate model coupled with land and ocean biogeochem- In this manuscript, we focus on reviewing the basic per- ical models (Bretherton, 1985), but can be extended to in- formance of the ocean and land carbon cycle components of clude further processes and reservoirs (e.g. anthropogenic the NorESM. In order to assess the quality of model projec- interactions). As an integrated global model system, such tions, it is necessary to evaluate the respective model simu- model does not only simulate the change in climate physi- lations against the available present-day climate and biogeo- cal variability due to anthropogenic drivers, but also includes chemical states. The biogeochemical states simulated by an climate feedbacks associated with the global carbon cycle. Earth system model strongly depend on the quality of the These feedback processes include changes in terrestrial and physical fields in the model. Therefore, we will first analyze oceanic carbon uptake due to anthropogenic CO2 emissions, statistically how well the model simulates the climatologi- perturbed surface temperature, precipitation, ocean circula- cal states of sea surface temperature and salinity. Next, the tion, sea ice extent, biological productivity, etc. A new Nor- model simulated mean state of ocean biogeochemical tracers, wegian Earth System Model (NorESM) was recently devel- such as oxygen, phosphate, and air–sea CO2 gas exchange oped (Bentsen et al., 2012). The NorESM is among the many are compared with the observations from the World Ocean models used worldwide to project future climate change and Atlas (WOA) and other observationally based estimates. Fi- is used for the upcoming IPCC-AR5. The ocean carbon cycle nally, we compare the land carbon pools, vegetation produc- model in NorESM is unique compared to most other models tivity, and respiration simulated by NorESM with the obser- due to its coupling with an isopycnic ocean general circu- vationally based values (e.g. from the FLUXNET network of lation model. One of the advantages of such a coupling is eddy covariance towers, among others). Note that the physi- more accurate representation of the transport and mixing of cal parameters affecting the terrestrial biogeochemistry, such biogeochemical tracers along the isopycnals in the interior as mean surface sensible and latent heat flux from land, sur- ocean. The isopycnic model also avoids physically inappro- face air temperature over land, spatial distribution of cloud priate splitting of transport and diffusion processes in hori- fraction, and mean precipitation are separately discussed in zontal and vertical components as done in a more common an accompanying manuscript by Bentsen et al. (2012). z-coordinate model (Bleck, 1998; Haidvogel and Beckmann, The model description is presented in Sect. 2. Section 3 1999). Through the vertically adaptive grid, areas of high describes the model experiment setup. The results of the ex- horizontal and vertical density gradients can be simulated periment are discussed in Sect. 4. Finally, conclusions are well by the model. An earlier study by Assmann et al. (2010) summarized in Sect. 5. also shows that higher spatial gradients in tracer distributions can be achieved. On the other hand, depending on the num- ber of density surfaces, an isopycnic coordinate model may 2 Model description or may not represent well the buoyancy driven circulation. The Norwegian Earth System Model (NorESM) is partly In areas of low density gradients, the model cannot simulate based on the recently released Community Climate System velocity shear and surface processes are also more difficult Model (CCSM4, Gent et al., 2011), which is maintained by to simulate in outcropping layers, which is avoided