Description: Accurate projections of marine particle export production (EP) are crucial for predicting the response of the marine carbon cycle to climate change, yet models show a wide range in both global EP and their responses to climate change. This is, in part, due to EP being the net result of a series of processes, starting with net primary production (NPP) in the sunlit upper ocean, followed by the formation of particulate organic matter and the subsequent sinking and remineralisation of these particles, with each of these processes responding differently to changes in environmental conditions. Here, we compare future projections in EP over the 21st century, generated by four marine ecosystem models under the high emission scenario Representative Concentra- tion Pathways (RCP) 8.5 of the Intergovernmental Panel on Climate Change (IPCC), and determine the processes driving these changes. The models simulate small to modest decreases in global EP between −1 and −12 %. Models differ greatly with regard to the drivers causing these changes. Among them, the formation of particles is the most uncertain process with models not agreeing on either magnitude or the direction of change. The removal of the sinking particles by remineralisation is simulated to increase in the low and intermediate latitudes in three models, driven by either warming-induced increases in remineralisation or slower particle sinking, and show insignificant changes in the remaining model. Changes in ecosystem structure, particularly the relative role of diatoms matters as well, as diatoms produce larger and denser particles that sink faster and are partly protected from remineralisation. Also this controlling factor is afflicted with high uncertainties, particularly since the models differ already substantially with regard to both the initial (present-day) distribution of diatoms (between 11–94 % in the Southern Ocean) and the diatom contribution to particle formation (0.6–3.8 times higher than their contribution to biomass). As a consequence, changes in diatom concentration are a strong driver for EP changes in some models but of low significance in others. Observational and experimental constraints on ecosystem structure and how the fixed carbon is routed through the ecosystem to produce export production are urgently needed in order to improve current generation ecosystem models and their ability to project future changes.

Description: Past model studies have projected a global decrease in marine net primary production (NPP) over the 21st century, but these studies focused on the multi-model mean rather than on the large inter-model differences. Here, we analyze model-simulated changes in NPP for the 21st century under IPCC's high-emission scenario RCP8.5. We use a suite of nine coupled carbon–climate Earth system models with embedded marine ecosystem models and focus on the spread between the different models and the underlying reasons. Globally, NPP decreases in five out of the nine models over the course of the 21st century, while three show no significant trend and one even simulates an increase. The largest model spread occurs in the low latitudes (between 30° S and 30° N), with individual models simulating relative changes between −25 and +40 %. Of the seven models diagnosing a net decrease in NPP in the low latitudes, only three simulate this to be a consequence of the classical interpretation, i.e., a stronger nutrient limitation due to increased stratification leading to reduced phytoplankton growth. In the other four, warming-induced increases in phytoplankton growth outbalance the stronger nutrient limitation. However, temperature-driven increases in grazing and other loss processes cause a net decrease in phytoplankton biomass and reduce NPP despite higher growth rates. One model projects a strong increase in NPP in the low latitudes, caused by an intensification of the microbial loop, while NPP in the remaining model changes by less than 0.5 %. While models consistently project increases NPP in the Southern Ocean, the regional inter-model range is also very substantial. In most models, this increase in NPP is driven by temperature, but it is also modulated by changes in light, macronutrients and iron as well as grazing. Overall, current projections of future changes in global marine NPP are subject to large uncertainties and necessitate a dedicated and sustained effort to improve the models and the concepts and data that guide their development.

Description: In the discussion paper of Weber et al. (2015) a simple model of the anthropogenic carbon cycle is presented. The authors describe a simple linear model, consisting of one ordinary differential equation which describes the changes in CO2 content of the atmosphere over time. The two free parameters of the equation are derived by fitting the results of the model (carbon content of the atmosphere) to the observations or reconstructions covering the last 150 years. The model is then applied to calculate the response of the global carbon cycle to future anthropogenic emissions and some conclusions on the fate of the anthropogenic carbon emissions until the year 2150 are then drawn. The findings show a rather fast reduction in atmospheric CO2 and the conclusions are in contrast to the results of virtually all other global carbon cycle models, (e.g. Meinshausen et al., 2011; Stocker et al., 2013; Friedlingstein et al., 2014), which — in contrast to the simple model presented here — include the current state of understanding of the processes involved in the global carbon cycle. We challenge the overall conclusions of the paper for the following reasons: The simple model (although not perfect) performs well for the anthropogenic period up to today, since the values of the two free parameters in the ordinary differential equation are based on observations (or to be more correct on model-based interpretation of observations). The agreement of the model to the historic atmospheric CO2 record is therefore hardly surprising. Besides the balance of some carbon fluxes in and out of the atmosphere no further theoretical (process-based) understanding is implemented in the simple model. This is a valid approach for simulating the most recent (anthropogenic driven) past, but does not prove that the model contains prognostic value, which justifies its application on future emissions. The model can only be applied to future anthropogenic perturbations on the surmise that the carbon cycle is not fundamentally altered. However, this is clearly not the case for nearly all future emission scenarios, most importantly because the carbon uptake capacity of the ocean depends on the carbonate chemistry (Revelle factor), which is changing at unprecedented speed.